ViewVC logotype

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 469 - (hide annotations) (download)
Mon Oct 19 14:38:48 2009 UTC (5 years, 5 months ago) by ph10
File size: 174788 byte(s)
Final doc and source tidies for 8.00

1 nigel 3 ChangeLog for PCRE
2     ------------------
4 ph10 469 Version 8.00 19-Oct-09
5 ph10 418 ----------------------
7     1. The table for translating pcre_compile() error codes into POSIX error codes
8 ph10 461 was out-of-date, and there was no check on the pcre_compile() error code
9     being within the table. This could lead to an OK return being given in
10 ph10 418 error.
11 ph10 461
12     2. Changed the call to open a subject file in pcregrep from fopen(pathname,
13     "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
14     in a Windows environment.
16 ph10 420 3. The pcregrep --count option prints the count for each file even when it is
17     zero, as does GNU grep. However, pcregrep was also printing all files when
18     --files-with-matches was added. Now, when both options are given, it prints
19     counts only for those files that have at least one match. (GNU grep just
20 ph10 461 prints the file name in this circumstance, but including the count seems
21     more useful - otherwise, why use --count?) Also ensured that the
22 ph10 420 combination -clh just lists non-zero counts, with no names.
23 ph10 461
24     4. The long form of the pcregrep -F option was incorrectly implemented as
25     --fixed_strings instead of --fixed-strings. This is an incompatible change,
26     but it seems right to fix it, and I didn't think it was worth preserving
27     the old behaviour.
29     5. The command line items --regex=pattern and --regexp=pattern were not
30 ph10 422 recognized by pcregrep, which required --regex pattern or --regexp pattern
31 ph10 461 (with a space rather than an '='). The man page documented the '=' forms,
32 ph10 422 which are compatible with GNU grep; these now work.
33 ph10 461
34     6. No libpcreposix.pc file was created for pkg-config; there was just
35 ph10 423 libpcre.pc and libpcrecpp.pc. The omission has been rectified.
36 ph10 461
37 ph10 425 7. Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
38 ph10 461 when UCP support is not needed, by modifying the Python script that
39 ph10 425 generates it from Unicode data files. This should not matter if the module
40     is correctly used as a library, but I received one complaint about 50K of
41     unwanted data. My guess is that the person linked everything into his
42     program rather than using a library. Anyway, it does no harm.
43 ph10 461
44 ph10 426 8. A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
45 ph10 461 was a minimum greater than 1 for a wide character in a possessive
46 ph10 431 repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
47     which had an unlimited repeat of a nested, fixed maximum repeat of a wide
48     character. Chaos in the form of incorrect output or a compiling loop could
49     result.
50 ph10 461
51 ph10 426 9. The restrictions on what a pattern can contain when partial matching is
52 ph10 461 requested for pcre_exec() have been removed. All patterns can now be
53 ph10 426 partially matched by this function. In addition, if there are at least two
54 ph10 435 slots in the offset vector, the offset of the earliest inspected character
55     for the match and the offset of the end of the subject are set in them when
56 ph10 461 PCRE_ERROR_PARTIAL is returned.
58 ph10 428 10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
59     synonymous with PCRE_PARTIAL, for backwards compatibility, and
60     PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
61 ph10 462 and may be more useful for multi-segment matching.
62 ph10 461
63     11. Partial matching with pcre_exec() is now more intuitive. A partial match
64     used to be given if ever the end of the subject was reached; now it is
65     given only if matching could not proceed because another character was
66     needed. This makes a difference in some odd cases such as Z(*FAIL) with the
67     string "Z", which now yields "no match" instead of "partial match". In the
68     case of pcre_dfa_exec(), "no match" is given if every matching path for the
69     final character ended with (*FAIL).
71 ph10 428 12. Restarting a match using pcre_dfa_exec() after a partial match did not work
72 ph10 461 if the pattern had a "must contain" character that was already found in the
73 ph10 428 earlier partial match, unless partial matching was again requested. For
74     example, with the pattern /dog.(body)?/, the "must contain" character is
75     "g". If the first part-match was for the string "dog", restarting with
76 ph10 435 "sbody" failed. This bug has been fixed.
77 ph10 461
78     13. The string returned by pcre_dfa_exec() after a partial match has been
79     changed so that it starts at the first inspected character rather than the
80     first character of the match. This makes a difference only if the pattern
81     starts with a lookbehind assertion or \b or \B (\K is not supported by
82     pcre_dfa_exec()). It's an incompatible change, but it makes the two
83 ph10 435 matching functions compatible, and I think it's the right thing to do.
84 ph10 461
85 ph10 435 14. Added a pcredemo man page, created automatically from the pcredemo.c file,
86 ph10 461 so that the demonstration program is easily available in environments where
87     PCRE has not been installed from source.
89 ph10 435 15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
90 ph10 430 libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
91     library.
92 ph10 461
93 ph10 435 16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
94 ph10 432 It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
95 ph10 461 is not the first non-POSIX option to be added. Clearly some people find
96 ph10 433 these options useful.
97 ph10 461
98     17. If a caller to the POSIX matching function regexec() passes a non-zero
99 ph10 438 value for nmatch with a NULL value for pmatch, the value of
100 ph10 461 nmatch is forced to zero.
102 ph10 437 18. RunGrepTest did not have a test for the availability of the -u option of
103 ph10 461 the diff command, as RunTest does. It now checks in the same way as
104 ph10 437 RunTest, and also checks for the -b option.
105 ph10 461
106 ph10 438 19. If an odd number of negated classes containing just a single character
107     interposed, within parentheses, between a forward reference to a named
108 ph10 461 subpattern and the definition of the subpattern, compilation crashed with
109     an internal error, complaining that it could not find the referenced
110 ph10 438 subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
111 ph10 461 [The bug was that it was starting one character too far in when skipping
112     over the character class, thus treating the ] as data rather than
113     terminating the class. This meant it could skip too much.]
115 ph10 442 20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
116 ph10 461 /g option in pcretest when the pattern contains \K, which makes it possible
117 ph10 442 to have an empty string match not at the start, even when the pattern is
118 ph10 461 anchored. Updated pcretest and pcredemo to use this option.
120 ph10 446 21. If the maximum number of capturing subpatterns in a recursion was greater
121 ph10 461 than the maximum at the outer level, the higher number was returned, but
122     with unset values at the outer level. The correct (outer level) value is
123 ph10 446 now given.
124 ph10 461
125 ph10 447 22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
126     PCRE did not set those parentheses (unlike Perl). I have now found a way to
127     make it do so. The string so far is captured, making this feature
128     compatible with Perl.
129 ph10 461
130     23. The tests have been re-organized, adding tests 11 and 12, to make it
131 ph10 448 possible to check the Perl 5.10 features against Perl 5.10.
132 ph10 461
133 ph10 454 24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
134 ph10 461 pattern matches a fixed length string. PCRE did not allow this; now it
135     does. Neither allows recursion.
137     25. I finally figured out how to implement a request to provide the minimum
138     length of subject string that was needed in order to match a given pattern.
139     (It was back references and recursion that I had previously got hung up
140     on.) This code has now been added to pcre_study(); it finds a lower bound
141 ph10 455 to the length of subject needed. It is not necessarily the greatest lower
142     bound, but using it to avoid searching strings that are too short does give
143     some useful speed-ups. The value is available to calling programs via
144     pcre_fullinfo().
145 ph10 461
146 ph10 455 26. While implementing 25, I discovered to my embarrassment that pcretest had
147     not been passing the result of pcre_study() to pcre_dfa_exec(), so the
148     study optimizations had never been tested with that matching function.
149     Oops. What is worse, even when it was passed study data, there was a bug in
150     pcre_dfa_exec() that meant it never actually used it. Double oops. There
151     were also very few tests of studied patterns with pcre_dfa_exec().
152 ph10 461
153 ph10 457 27. If (?| is used to create subpatterns with duplicate numbers, they are now
154     allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
155     on the other side of the coin, they are no longer allowed to have different
156     names, because these cannot be distinguished in PCRE, and this has caused
157     confusion. (This is a difference from Perl.)
158 ph10 461
159     28. When duplicate subpattern names are present (necessarily with different
160     numbers, as required by 27 above), and a test is made by name in a
161     conditional pattern, either for a subpattern having been matched, or for
162     recursion in such a pattern, all the associated numbered subpatterns are
163 ph10 459 tested, and the overall condition is true if the condition is true for any
164     one of them. This is the way Perl works, and is also more like the way
165     testing by number works.
166 ph10 418
167 ph10 461
168 ph10 412 Version 7.9 11-Apr-09
169 ph10 376 ---------------------
171 ph10 392 1. When building with support for bzlib/zlib (pcregrep) and/or readline
172 ph10 376 (pcretest), all targets were linked against these libraries. This included
173     libpcre, libpcreposix, and libpcrecpp, even though they do not use these
174     libraries. This caused unwanted dependencies to be created. This problem
175 ph10 392 has been fixed, and now only pcregrep is linked with bzlib/zlib and only
176 ph10 376 pcretest is linked with readline.
177 ph10 392
178 ph10 376 2. The "typedef int BOOL" in pcre_internal.h that was included inside the
179     "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
180     moved outside it again, because FALSE and TRUE are already defined in AIX,
181     but BOOL is not.
182 ph10 392
183     3. The pcre_config() function was treating the PCRE_MATCH_LIMIT and
184 ph10 409 PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
185 ph10 392
186 ph10 376 4. The pcregrep documentation said spaces were inserted as well as colons (or
187     hyphens) following file names and line numbers when outputting matching
188 ph10 392 lines. This is not true; no spaces are inserted. I have also clarified the
189 ph10 376 wording for the --colour (or --color) option.
190 ph10 392
191 ph10 378 5. In pcregrep, when --colour was used with -o, the list of matching strings
192     was not coloured; this is different to GNU grep, so I have changed it to be
193     the same.
194 ph10 392
195 ph10 378 6. When --colo(u)r was used in pcregrep, only the first matching substring in
196 ph10 392 each matching line was coloured. Now it goes on to look for further matches
197     of any of the test patterns, which is the same behaviour as GNU grep.
199     7. A pattern that could match an empty string could cause pcregrep to loop; it
200     doesn't make sense to accept an empty string match in pcregrep, so I have
201 ph10 379 locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
202     seems to be how GNU grep behaves.
203 ph10 392
204 ph10 380 8. The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
205 ph10 392 start or after a newline", because the conditional assertion was not being
206     correctly handled. The rule now is that both the assertion and what follows
207     in the first alternative must satisfy the test.
209 ph10 405 9. If auto-callout was enabled in a pattern with a conditional group whose
210 ph10 399 condition was an assertion, PCRE could crash during matching, both with
211     pcre_exec() and pcre_dfa_exec().
212 ph10 392
213     10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
214     used for matching.
216     11. Unicode property support in character classes was not working for
217 ph10 385 characters (bytes) greater than 127 when not in UTF-8 mode.
218 ph10 388
219 ph10 392 12. Added the -M command line option to pcretest.
221 ph10 388 14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
222 ph10 389
223     15. Added the PCRE_NO_START_OPTIMIZE match-time option.
224 ph10 390
225 ph10 392 16. Added comments and documentation about mis-use of no_arg in the C++
226 ph10 390 wrapper.
227 ph10 392
228 ph10 391 17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
229 ph10 392 from Martin Jerabek that uses macro names for all relevant character and
230 ph10 391 string constants.
231 ph10 405
232 ph10 393 18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
233     SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
234 ph10 405 SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
235     these, but not everybody uses configure.
237     19. A conditional group that had only one branch was not being correctly
238     recognized as an item that could match an empty string. This meant that an
239     enclosing group might also not be so recognized, causing infinite looping
240     (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
241     with the subject "ab", where knowledge that the repeated group can match
242 ph10 396 nothing is needed in order to break the loop.
243 ph10 405
244 ph10 397 20. If a pattern that was compiled with callouts was matched using pcre_dfa_
245 ph10 405 exec(), but without supplying a callout function, matching went wrong.
247     21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
248     leak if the size of the offset vector was greater than 30. When the vector
249     is smaller, the saved offsets during recursion go onto a local stack
250     vector, but for larger vectors malloc() is used. It was failing to free
251     when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
252 ph10 400 error, in fact).
253 ph10 405
254 ph10 402 22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
255 ph10 405 heapframe that is used only when UTF-8 support is enabled. This caused no
256     problem, but was untidy.
257 ph10 376
258 ph10 405 23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
259     CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
260     included within another project.
261 ph10 411
262 ph10 407 24. Steven Van Ingelgem's patches to add more options to the CMake support,
263     slightly modified by me:
264 ph10 411
265     (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
266     not building pcregrep.
268 ph10 407 (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
269 ph10 411 if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
271 ph10 408 25. Forward references, both numeric and by name, in patterns that made use of
272     duplicate group numbers, could behave incorrectly or give incorrect errors,
273     because when scanning forward to find the reference group, PCRE was not
274     taking into account the duplicate group numbers. A pattern such as
275     ^X(?3)(a)(?|(b)|(q))(Y) is an example.
276 ph10 411
277     26. Changed a few more instances of "const unsigned char *" to USPTR, making
278     the feature of a custom pointer more persuasive (as requested by a user).
279 ph10 416
280 ph10 411 27. Wrapped the definitions of fileno and isatty for Windows, which appear in
281 ph10 416 pcretest.c, inside #ifndefs, because it seems they are sometimes already
282     pre-defined.
283 ph10 392
284 ph10 416 28. Added support for (*UTF8) at the start of a pattern.
286 ph10 413 29. Arrange for flags added by the "release type" setting in CMake to be shown
287     in the configuration summary.
288 ph10 405
289 ph10 411
290 ph10 374 Version 7.8 05-Sep-08
291 ph10 349 ---------------------
293     1. Replaced UCP searching code with optimized version as implemented for Ad
294     Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
295 ph10 351 stage table and inline lookup instead of a function, giving speed ups of 2
296     to 5 times on some simple patterns that I tested. Permission was given to
297     distribute the MultiStage2.py script that generates the tables (it's not in
298     the tarball, but is in the Subversion repository).
299 ph10 349
300 ph10 351 2. Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
301     scripts.
302 ph10 358
303 ph10 353 3. Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
304     a group with a zero qualifier. The result of the study could be incorrect,
305 ph10 358 or the function might crash, depending on the pattern.
307     4. Caseless matching was not working for non-ASCII characters in back
308 ph10 354 references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
309 ph10 358 It now works when Unicode Property Support is available.
311 ph10 355 5. In pcretest, an escape such as \x{de} in the data was always generating
312     a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
313     non-UTF-8 mode. If the value is greater than 255, it gives a warning about
314 ph10 358 truncation.
315 ph10 349
316 ph10 358 6. Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
318 ph10 357 7. Added two (int) casts to pcregrep when printing the difference of two
319     pointers, in case they are 64-bit values.
320 ph10 371
321     8. Added comments about Mac OS X stack usage to the pcrestack man page and to
322     test 2 if it fails.
324 ph10 359 9. Added PCRE_CALL_CONVENTION just before the names of all exported functions,
325 ph10 371 and a #define of that name to empty if it is not externally set. This is to
326     allow users of MSVC to set it if necessary.
328     10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
329 ph10 359 the convenience functions in the pcre_get.c source file.
330 ph10 371
331 ph10 360 11. An option change at the start of a pattern that had top-level alternatives
332     could cause overwriting and/or a crash. This command provoked a crash in
333 ph10 371 some environments:
335     printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
337 ph10 360 This potential security problem was recorded as CVE-2008-2371.
338 ph10 371
339 ph10 361 12. For a pattern where the match had to start at the beginning or immediately
340     after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
341 ph10 371 pcre_dfa_exec() could read past the end of the passed subject if there was
342 ph10 361 no match. To help with detecting such bugs (e.g. with valgrind), I modified
343     pcretest so that it places the subject at the end of its malloc-ed buffer.
344 ph10 371
345 ph10 364 13. The change to pcretest in 12 above threw up a couple more cases when pcre_
346 ph10 371 exec() might read past the end of the data buffer in UTF-8 mode.
348 ph10 364 14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
349 ph10 371 the data contained the byte 0x85 as part of a UTF-8 character within its
350     first line. This applied both to normal and DFA matching.
352 ph10 366 15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
353 ph10 371 /^[^d]*?$/8 failed to match "abc".
354 ph10 351
355 ph10 371 16. Added a missing copyright notice to pcrecpp_internal.h.
357     17. Make it more clear in the documentation that values returned from
358 ph10 368 pcre_exec() in ovector are byte offsets, not character counts.
359 ph10 357
360 ph10 371 18. Tidied a few places to stop certain compilers from issuing warnings.
361 ph10 368
362 ph10 373 19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
363     supplied by Stefan Weber. I made a further small update for 7.8 because
364 ph10 374 there is a change of source arrangements: the pcre_searchfuncs.c module is
365     replaced by pcre_ucd.c.
366 ph10 371
367 ph10 373
368 ph10 347 Version 7.7 07-May-08
369 ph10 321 ---------------------
371     1. Applied Craig's patch to sort out a long long problem: "If we can't convert
372 ph10 345 a string to a long long, pretend we don't even have a long long." This is
373 ph10 321 done by checking for the strtoq, strtoll, and _strtoi64 functions.
374 ph10 345
375 ph10 322 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
376     pre-7.6 versions, which defined a global no_arg variable instead of putting
377 ph10 332 it in the RE class. (See also #8 below.)
378 ph10 345
379     3. Remove a line of dead code, identified by coverity and reported by Nuno
380     Lopes.
382 ph10 324 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
383 ph10 321
384 ph10 324 (1) The include/exclude patterns were being applied to the whole pathnames
385 ph10 345 of files, instead of just to the final components.
387 ph10 324 (2) If there was more than one level of directory, the subdirectories were
388     skipped unless they satisfied the include/exclude conditions. This is
389     inconsistent with GNU grep (and could even be seen as contrary to the
390     pcregrep specification - which I improved to make it absolutely clear).
391     The action now is always to scan all levels of directory, and just
392     apply the include/exclude patterns to regular files.
393 ph10 345
394 ph10 325 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
395 ph10 345 --exclude_dir in the tests to avoid scanning .svn directories.
397 ph10 326 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
398 ph10 345 NUL character as backslash + 0 rather than backslash + NUL, because PCRE
399 ph10 326 doesn't support NULs in patterns.
400 ph10 345
401     7. Added some missing "const"s to declarations of static tables in
402     pcre_compile.c and pcre_dfa_exec.c.
404 ph10 328 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
405 ph10 329 caused by fix #2 above. (Subsequently also a second patch to fix the
406 ph10 345 first patch. And a third patch - this was a messy problem.)
407 ph10 321
408 ph10 345 9. Applied Craig's patch to remove the use of push_back().
410 ph10 332 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
411     matching function regexec().
412 ph10 345
413     11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
414     which, however, unlike Perl's \g{...}, are subroutine calls, not back
415 ph10 333 references. PCRE supports relative numbers with this syntax (I don't think
416     Oniguruma does).
417 ph10 345
418     12. Previously, a group with a zero repeat such as (...){0} was completely
419 ph10 335 omitted from the compiled regex. However, this means that if the group
420     was called as a subroutine from elsewhere in the pattern, things went wrong
421 ph10 345 (an internal error was given). Such groups are now left in the compiled
422     pattern, with a new opcode that causes them to be skipped at execution
423 ph10 335 time.
424 ph10 345
425 ph10 341 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
426     to the way PCRE behaves:
427 ph10 345
428 ph10 336 (a) A lone ] character is dis-allowed (Perl treats it as data).
429 ph10 345
430     (b) A back reference to an unmatched subpattern matches an empty string
431 ph10 336 (Perl fails the current match path).
432 ph10 345
433 ph10 341 (c) A data ] in a character class must be notated as \] because if the
434 ph10 345 first data character in a class is ], it defines an empty class. (In
435 ph10 341 Perl it is not possible to have an empty class.) The empty class []
436     never matches; it forces failure and is equivalent to (*FAIL) or (?!).
437     The negative empty class [^] matches any one character, independently
438     of the DOTALL setting.
439 ph10 345
440     14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
441 ph10 340 non-existent subpattern following a character class starting with ']' and
442     containing () gave an internal compiling error instead of "reference to
443     non-existent subpattern". Fortunately, when the pattern did exist, the
444 ph10 345 compiled code was correct. (When scanning forwards to check for the
445 ph10 340 existencd of the subpattern, it was treating the data ']' as terminating
446 ph10 345 the class, so got the count wrong. When actually compiling, the reference
447 ph10 340 was subsequently set up correctly.)
448 ph10 345
449 ph10 341 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
450 ph10 345 it was being rejected as not supported by pcre_dfa_exec(), even though
451     other assertions are supported. I have made pcre_dfa_exec() support
452     (*FAIL).
454 ph10 342 16. The implementation of 13c above involved the invention of a new opcode,
455 ph10 345 OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
456     cannot be changed at match time, I realized I could make a small
457     improvement to matching performance by compiling OP_ALLANY instead of
458     OP_ANY for "." when DOTALL was set, and then removing the runtime tests
459     on the OP_ANY path.
461     17. Compiling pcretest on Windows with readline support failed without the
462 ph10 343 following two fixes: (1) Make the unistd.h include conditional on
463 ph10 345 HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
465 ph10 344 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
466 ph10 345 ncurses library to be included for pcretest when ReadLine support is
467     requested, but also to allow for it to be overridden. This patch came from
468     Daniel Bergström.
469 ph10 347
470 ph10 346 19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
471 ph10 347 as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
472     any errors with the current Unicode tables. Thanks to Peter Kankowski for
473     spotting this.
474 ph10 324
475 ph10 332
476 ph10 319 Version 7.6 28-Jan-08
477 ph10 300 ---------------------
479 ph10 302 1. A character class containing a very large number of characters with
480     codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
481     overflow.
482 ph10 309
483     2. Patch to cut out the "long long" test in pcrecpp_unittest when
484     HAVE_LONG_LONG is not defined.
486 ph10 303 3. Applied Christian Ehrlicher's patch to update the CMake build files to
487 ph10 304 bring them up to date and include new features. This patch includes:
488 ph10 309
489 ph10 304 - Fixed PH's badly added libz and libbz2 support.
490     - Fixed a problem with static linking.
491 ph10 312 - Added pcredemo. [But later removed - see 7 below.]
492 ph10 304 - Fixed dftables problem and added an option.
493     - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
495     - Added readline support for pcretest.
496 ph10 309 - Added an listing of the option settings after cmake has run.
498 ph10 314 4. A user submitted a patch to Makefile that makes it easy to create
499     "pcre.dll" under mingw when using Configure/Make. I added stuff to
500     Makefile.am that cause it to include this special target, without
501     affecting anything else. Note that the same mingw target plus all
502     the other distribution libraries and programs are now supported
503     when configuring with CMake (see 6 below) instead of with
504     Configure/Make.
505 ph10 309
506 ph10 308 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
507     This is an attempt to solve the reported problem "pcrecpp::no_arg is not
508 ph10 309 exported in the Windows port". It has not yet been confirmed that the patch
509     solves the problem, but it does no harm.
510 ph10 313
511 ph10 311 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
512 ph10 319 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
513 ph10 314 with CMake, and also correct the comment about stack recursion.
514 ph10 313
515 ph10 312 7. Remove the automatic building of pcredemo from the ./configure system and
516 ph10 313 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
517     of a program that users should build themselves after PCRE is installed, so
518     building it automatically is not really right. What is more, it gave
519 ph10 312 trouble in some build environments.
520 ph10 300
521 ph10 319 8. Further tidies to CMakeLists.txt from Sheri and Christian.
522 ph10 308
523 ph10 319
524 ph10 298 Version 7.5 10-Jan-08
525 ph10 263 ---------------------
527     1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
528     values in parens when parsing an RE using the C++ wrapper."
529 ph10 286
530 ph10 264 2. Negative specials like \S did not work in character classes in UTF-8 mode.
531     Characters greater than 255 were excluded from the class instead of being
532     included.
533 ph10 286
534     3. The same bug as (2) above applied to negated POSIX classes such as
535 ph10 265 [:^space:].
536 ph10 286
537 ph10 267 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
538     defined or documented. It seems to have been a typo for PCRE_STATIC, so
539 ph10 286 I have changed it.
541 ph10 268 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
542 ph10 286 first named subpattern) and a construct such as (?&a) would reference the
543     first named subpattern whose name started with "a" (in other words, the
544 ph10 272 length check was missing). Both these problems are fixed. "Subpattern name
545     expected" is now given for (?&) (a zero-length name), and this patch also
546     makes it give the same error for \k'' (previously it complained that that
547     was a reference to a non-existent subpattern).
548 ph10 286
549 ph10 269 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
550 ph10 286 this is right because (?- can be followed by option settings as well as by
551 ph10 269 digits. I have, however, made the messages clearer.
552 ph10 286
553 ph10 270 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
554     than the number used in the conditional) now cause a compile-time error.
555     This is actually not compatible with Perl, which accepts such patterns, but
556     treats the conditional as always being FALSE (as PCRE used to), but it
557     seems to me that giving a diagnostic is better.
558 ph10 286
559 ph10 274 8. Change "alphameric" to the more common word "alphanumeric" in comments
560 ph10 275 and messages.
561 ph10 286
562     9. Fix two occurrences of "backslash" in comments that should have been
563     "backspace".
565     10. Remove two redundant lines of code that can never be obeyed (their function
566     was moved elsewhere).
568     11. The program that makes PCRE's Unicode character property table had a bug
569     which caused it to generate incorrect table entries for sequences of
570 ph10 277 characters that have the same character type, but are in different scripts.
571 ph10 286 It amalgamated them into a single range, with the script of the first of
572 ph10 277 them. In other words, some characters were in the wrong script. There were
573     thirteen such cases, affecting characters in the following ranges:
574 ph10 286
575 ph10 277 U+002b0 - U+002c1
576     U+0060c - U+0060d
577 ph10 286 U+0061e - U+00612
578 ph10 277 U+0064b - U+0065e
579     U+0074d - U+0076d
580     U+01800 - U+01805
581     U+01d00 - U+01d77
582     U+01d9b - U+01dbf
583     U+0200b - U+0200f
584     U+030fc - U+030fe
585     U+03260 - U+0327f
586     U+0fb46 - U+0fbb1
587     U+10450 - U+1049d
588 ph10 286
589 ph10 279 12. The -o option (show only the matching part of a line) for pcregrep was not
590     compatible with GNU grep in that, if there was more than one match in a
591     line, it showed only the first of them. It now behaves in the same way as
592     GNU grep.
593 ph10 286
594     13. If the -o and -v options were combined for pcregrep, it printed a blank
595     line for every non-matching line. GNU grep prints nothing, and pcregrep now
596     does the same. The return code can be used to tell if there were any
597     non-matching lines.
599 ph10 289 14. Added --file-offsets and --line-offsets to pcregrep.
601     15. The pattern (?=something)(?R) was not being diagnosed as a potentially
602 ph10 282 infinitely looping recursion. The bug was that positive lookaheads were not
603 ph10 286 being skipped when checking for a possible empty match (negative lookaheads
604     and both kinds of lookbehind were skipped).
606 ph10 289 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
607 ph10 284 inclusion of <windows.h> to before rather than after the definition of
608     INVALID_FILE_ATTRIBUTES (patch from David Byron).
609 ph10 263
610 ph10 289 17. Specifying a possessive quantifier with a specific limit for a Unicode
611 ph10 286 character property caused pcre_compile() to compile bad code, which led at
612     runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
613     are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
614     caused the error; without that there was no problem.
615 ph10 263
616 ph10 289 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
617 ph10 286
618 ph10 289 19. Added --enable-pcretest-libreadline.
619 ph10 286
620 ph10 289 20. In pcrecpp.cc, the variable 'count' was incremented twice in
621 ph10 288 RE::GlobalReplace(). As a result, the number of replacements returned was
622 ph10 298 double what it should be. I removed one of the increments, but Craig sent a
623     later patch that removed the other one (the right fix) and added unit tests
624     that check the return values (which was not done before).
625 ph10 292
626 ph10 291 21. Several CMake things:
627 ph10 287
628 ph10 291 (1) Arranged that, when cmake is used on Unix, the libraries end up with
629 ph10 292 the names libpcre and libpcreposix, not just pcre and pcreposix.
630 ph10 288
631 ph10 292 (2) The above change means that pcretest and pcregrep are now correctly
632     linked with the newly-built libraries, not previously installed ones.
633 ph10 291
635 ph10 298
636     22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
637     crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
638     UTF-8 newline character). The key issue is that the pattern starts .*;
639     this means that the match must be either at the beginning, or after a
640     newline. The bug was in the code for advancing after a failed match and
641     checking that the new position followed a newline. It was not taking
642 ph10 294 account of UTF-8 characters correctly.
643 ph10 298
644     23. PCRE was behaving differently from Perl in the way it recognized POSIX
645     character classes. PCRE was not treating the sequence [:...:] as a
646     character class unless the ... were all letters. Perl, however, seems to
647     allow any characters between [: and :], though of course it rejects as
648     unknown any "names" that contain non-letters, because all the known class
649     names consist only of letters. Thus, Perl gives an error for [[:1234:]],
650     for example, whereas PCRE did not - it did not recognize a POSIX character
651     class. This seemed a bit dangerous, so the code has been changed to be
652     closer to Perl. The behaviour is not identical to Perl, because PCRE will
653     diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
654     treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
655     Perl does, and where it didn't before.
657 ph10 296 24. Rewrite so as to remove the single use of %n from pcregrep because in some
658     Windows environments %n is disabled by default.
659 ph10 292
661 ph10 260 Version 7.4 21-Sep-07
662 ph10 230 ---------------------
664     1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
665 ph10 231 means that a class such as [\s] counted as "explicit reference to CR or
666     LF". That isn't really right - the whole point of the change was to try to
667     help when there was an actual mention of one of the two characters. So now
668     the change happens only if \r or \n (or a literal CR or LF) character is
669 ph10 230 encountered.
670 ph10 231
671     2. The 32-bit options word was also used for 6 internal flags, but the numbers
672     of both had grown to the point where there were only 3 bits left.
673     Fortunately, there was spare space in the data structure, and so I have
674     moved the internal flags into a new 16-bit field to free up more option
675 ph10 230 bits.
676 ph10 231
677     3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
678     but did not set the internal JCHANGED flag - either of these is enough to
679     control the way the "get" function works - but the PCRE_INFO_JCHANGED
680     facility is supposed to tell if (?J) was ever used, so now (?J) at the
681 ph10 230 start sets both bits.
683 ph10 231 4. Added options (at build time, compile time, exec time) to change \R from
684     matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
685 ph10 230
686 ph10 243 5. doc/pcresyntax.html was missing from the distribution.
688     6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
689 ph10 233 compatibility, even though it is no longer used.
690 ph10 243
691     7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
692 ph10 254 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
693 ph10 259 windows.h file is present (where different names are used). [This was
694 ph10 257 reversed later after testing - see 16 below.]
695 ph10 243
696     8. Changed all #include <config.h> to #include "config.h". There were also
697 ph10 236 some further <pcre.h> cases that I changed to "pcre.h".
698 ph10 243
699 ph10 239 9. When pcregrep was used with the --colour option, it missed the line ending
700 ph10 243 sequence off the lines that it output.
702     10. It was pointed out to me that arrays of string pointers cause lots of
703     relocations when a shared library is dynamically loaded. A technique of
704     using a single long string with a table of offsets can drastically reduce
705     these. I have refactored PCRE in four places to do this. The result is
706 ph10 240 dramatic:
707 ph10 243
708 ph10 240 Originally: 290
709     After changing UCP table: 187
710 ph10 243 After changing error message table: 43
711 ph10 240 After changing table of "verbs" 36
712     After changing table of Posix names 22
713 ph10 243
714 ph10 240 Thanks to the folks working on Gregex for glib for this insight.
715 ph10 247
716 ph10 244 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
717 ph10 247 unicode-properties was also set.
719 ph10 248 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
720 ph10 231
721 ph10 253 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
722     checked only for CRLF.
723 ph10 233
724 ph10 259 14. Added casts to pcretest.c to avoid compiler warnings.
726 ph10 256 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
727 ph10 253
728 ph10 257 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
729 ph10 259 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
730     entirely. This removes changes made in 7 above.
731 ph10 256
732 ph10 261 17. The CMake files have been updated, and there is now more information about
733     building with CMake in the NON-UNIX-USE document.
734 ph10 257
735 ph10 261
736 ph10 228 Version 7.3 28-Aug-07
737 ph10 157 ---------------------
739 ph10 189 1. In the rejigging of the build system that eventually resulted in 7.1, the
740     line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
741     brackets there is not right, since it causes compilers to look for an
742     installed pcre.h, not the version that is in the source that is being
743     compiled (which of course may be different). I have changed it back to:
744 ph10 197
745 ph10 189 #include "pcre.h"
746 ph10 197
747     I have a vague recollection that the change was concerned with compiling in
748     different directories, but in the new build system, that is taken care of
749     by the VPATH setting the Makefile.
751 ph10 190 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
752     when the subject happened to end in the byte 0x85 (e.g. if the last
753     character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
754     characters but of course it shouldn't be taken as a newline when it is part
755     of another character. The bug was that, for an unlimited repeat of . in
756     not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
757     characters when looking for a newline.
758 ph10 197
759     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
761     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
762     in debug output.
764 ph10 201 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
765     long printing in the pcrecpp unittest when running under MinGW.
766 ph10 189
767 ph10 197 6. ESC_K was left out of the EBCDIC table.
768 ph10 189
769 ph10 197 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
770     parentheses; I made it 1000, which seemed large enough. Unfortunately, the
771     limit also applies to "virtual nesting" when a pattern is recursive, and in
772     this case 1000 isn't so big. I have been able to remove this limit at the
773     expense of backing off one optimization in certain circumstances. Normally,
774     when pcre_exec() would call its internal match() function recursively and
775     immediately return the result unconditionally, it uses a "tail recursion"
776     feature to save stack. However, when a subpattern that can match an empty
777     string has an unlimited repetition quantifier, it no longer makes this
778     optimization. That gives it a stack frame in which to save the data for
779     checking that an empty string has been matched. Previously this was taken
780     from the 1000-entry workspace that had been reserved. So now there is no
781     explicit limit, but more stack is used.
782 ph10 189
783 ph10 200 8. Applied Daniel's patches to solve problems with the import/export magic
784     syntax that is required for Windows, and which was going wrong for the
785     pcreposix and pcrecpp parts of the library. These were overlooked when this
786     problem was solved for the main library.
787 ph10 197
788 ph10 202 9. There were some crude static tests to avoid integer overflow when computing
789     the size of patterns that contain repeated groups with explicit upper
790     limits. As the maximum quantifier is 65535, the maximum group length was
791     set at 30,000 so that the product of these two numbers did not overflow a
792     32-bit integer. However, it turns out that people want to use groups that
793     are longer than 30,000 bytes (though not repeat them that many times).
794     Change 7.0/17 (the refactoring of the way the pattern size is computed) has
795     made it possible to implement the integer overflow checks in a much more
796     dynamic way, which I have now done. The artificial limitation on group
797     length has been removed - we now have only the limit on the total length of
798     the compiled pattern, which depends on the LINK_SIZE setting.
799 ph10 208
800     10. Fixed a bug in the documentation for get/copy named substring when
801     duplicate names are permitted. If none of the named substrings are set, the
802     functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
803     empty string.
805     11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
806     instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
807     because the ] is interpreted as the first data character and the
808     terminating ] is not found. PCRE has been made compatible with Perl in this
809     regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
810     cause memory overwriting.
812 ph10 206 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
813     string has been matched (to stop an infinite loop). It was not recognizing
814 ph10 208 a conditional subpattern that could match an empty string if that
815 ph10 206 subpattern was within another subpattern. For example, it looped when
816 ph10 208 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
817 ph10 206 condition was not nested. This bug has been fixed.
818 ph10 208
819 ph10 207 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
820     past the start of the subject in the presence of bytes with the top bit
821     set, for example "\x8aBCD".
822 ph10 211
823 ph10 210 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
824     (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
825 ph10 200
826 ph10 211 14. Optimized (?!) to (*FAIL).
827 ph10 202
828 ph10 212 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
829     This restricts code points to be within the range 0 to 0x10FFFF, excluding
830     the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
831     full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
832     does: it's just the validity check that is more restrictive.
833 ph10 220
834     16. Inserted checks for integer overflows during escape sequence (backslash)
835     processing, and also fixed erroneous offset values for syntax errors during
836     backslash processing.
838 ph10 214 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
839 ph10 220 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
841 ph10 215 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
842     caused an overrun.
843 ph10 220
844     19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
845     something other than just ASCII characters) inside a group that had an
846     unlimited repeat caused a loop at compile time (while checking to see
847     whether the group could match an empty string).
849 ph10 217 20. Debugging a pattern containing \p or \P could cause a crash. For example,
850     [\P{Any}] did so. (Error in the code for printing property names.)
851 ph10 210
852 ph10 220 21. An orphan \E inside a character class could cause a crash.
854     22. A repeated capturing bracket such as (A)? could cause a wild memory
855 ph10 218 reference during compilation.
856 ph10 220
857     23. There are several functions in pcre_compile() that scan along a compiled
858     expression for various reasons (e.g. to see if it's fixed length for look
859 ph10 218 behind). There were bugs in these functions when a repeated \p or \P was
860 ph10 220 present in the pattern. These operators have additional parameters compared
861     with \d, etc, and these were not being taken into account when moving along
862 ph10 218 the compiled data. Specifically:
863 ph10 220
864     (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
865     length.
867     (b) An item such as \pL+ within a repeated group could cause crashes or
868 ph10 218 loops.
869 ph10 220
870 ph10 218 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
871 ph10 220 "reference to non-existent subpattern" error.
873 ph10 221 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
875 ph10 219 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
876 ph10 220 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
877 ph10 211
878 ph10 222 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
879 ph10 218
880 ph10 227 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
881 ph10 223 character were causing crashes (broken optimization).
882 ph10 227
883 ph10 224 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
884 ph10 227 \p or \P) caused a compile-time loop.
886 ph10 226 28. More problems have arisen in unanchored patterns when CRLF is a valid line
887     break. For example, the unstudied pattern [\r\n]A does not match the string
888     "\r\nA" because change 7.0/46 below moves the current point on by two
889     characters after failing to match at the start. However, the pattern \nA
890 ph10 227 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
891     the same is true. There doesn't seem any very clean way out of this, but
892     what I have chosen to do makes the common cases work: PCRE now takes note
893     of whether there can be an explicit match for \r or \n anywhere in the
894     pattern, and if so, 7.0/46 no longer applies. As part of this change,
895     there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
896     pattern has explicit CR or LF references.
897 ph10 222
898 ph10 227 29. Added (*CR) etc for changing newline setting at start of pattern.
899 ph10 223
900 ph10 227
901 ph10 189 Version 7.2 19-Jun-07
902     ---------------------
904 ph10 157 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
905     which is apparently normally available under Windows.
906 ph10 159
907     2. Re-jig the pcregrep tests with different newline settings in an attempt
908     to make them independent of the local environment's newline setting.
909 ph10 157
910 ph10 160 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
911 ph10 165
912 ph10 161 4. Some of the "internals" tests were previously cut out when the link size
913     was not 2, because the output contained actual offsets. The recent new
914 ph10 165 "Z" feature of pcretest means that these can be cut out, making the tests
915     usable with all link sizes.
917 ph10 164 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
918     stack recursion. This gives a massive performance boost under BSD, but just
919 ph10 165 a small improvement under Linux. However, it saves one field in the frame
920 ph10 164 in all cases.
921 ph10 172
922 ph10 166 6. Added more features from the forthcoming Perl 5.10:
923 ph10 172
924 ph10 166 (a) (?-n) (where n is a string of digits) is a relative subroutine or
925     recursion call. It refers to the nth most recently opened parentheses.
926 ph10 172
927 ph10 166 (b) (?+n) is also a relative subroutine call; it refers to the nth next
928 ph10 172 to be opened parentheses.
930     (c) Conditions that refer to capturing parentheses can be specified
931 ph10 167 relatively, for example, (?(-2)... or (?(+3)...
932 ph10 172
933 ph10 168 (d) \K resets the start of the current match so that everything before
934 ph10 172 is not part of it.
936 ph10 171 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
937 ph10 172
938 ph10 171 (f) \g{name} is another synonym - part of Perl 5.10's unification of
939 ph10 172 reference syntax.
940 ph10 182
941 ph10 175 (g) (?| introduces a group in which the numbering of parentheses in each
942 ph10 182 alternative starts with the same number.
943 ph10 172
944 ph10 182 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
946 ph10 172 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
949     8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
950     terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
951 ph10 170 for detecting groups that can match an empty string.
952 ph10 159
953 ph10 172 9. A pattern with a very large number of alternatives (more than several
954     hundred) was running out of internal workspace during the pre-compile
955     phase, where pcre_compile() figures out how much memory will be needed. A
956     bit of new cunning has reduced the workspace needed for groups with
957     alternatives. The 1000-alternative test pattern now uses 12 bytes of
958     workspace instead of running out of the 4096 that are available.
959 ph10 182
960 ph10 176 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
961 ph10 172
962 ph10 179 11. Applied patch from Google to remove an optimization that didn't quite work.
963     The report of the bug said:
964 ph10 182
965 ph10 179 pcrecpp::RE("a*").FullMatch("aaa") matches, while
966     pcrecpp::RE("a*?").FullMatch("aaa") does not, and
967 ph10 182 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
968 ph10 185
969 ph10 184 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
970 ph10 185 it matched the wrong number of bytes.
971 ph10 172
972 ph10 179
973 ph10 155 Version 7.1 24-Apr-07
974 ph10 98 ---------------------
976 ph10 111 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
977 ph10 122 that is more "standard", making use of automake and other Autotools. There
978 ph10 99 is some re-arrangement of the files and adjustment of comments consequent
979     on this.
980 ph10 111
981     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
982     for recursive directory scanning broke on some systems because the files
983     are not scanned in any specific order and on different systems the order
984     was different. A call to "sort" has been inserted into RunGrepTest for the
985     approprate test as a short-term fix. In the longer term there may be an
986 ph10 100 alternative.
987 ph10 111
988 ph10 100 3. I had an email from Eric Raymond about problems translating some of PCRE's
989 ph10 111 man pages to HTML (despite the fact that I distribute HTML pages, some
990     people do their own conversions for various reasons). The problems
991     concerned the use of low-level troff macros .br and .in. I have therefore
992     removed all such uses from the man pages (some were redundant, some could
993 ph10 113 be replaced by .nf/.fi pairs). The 132html script that I use to generate
994     HTML has been updated to handle .nf/.fi and to complain if it encounters
995     .br or .in.
996 ph10 111
997 ph10 100 4. Updated comments in configure.ac that get placed in config.h.in and also
998 ph10 123 arranged for config.h to be included in the distribution, with the name
999 ph10 111 config.h.generic, for the benefit of those who have to compile without
1000     Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1002     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1003     Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1004 ph10 127 makevp.bat for the current PCRE, using the additional files
1005 ph10 135 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1006 ph10 111
1007     6. A Windows user reported a minor discrepancy with test 2, which turned out
1008     to be caused by a trailing space on an input line that had got lost in his
1009 ph10 102 copy. The trailing space was an accident, so I've just removed it.
1010 ph10 111
1011 ph10 104 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1012 ph10 111 that is needed.
1014 ph10 105 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1015     as "const" (a) because they are and (b) because it helps the PHP
1016     maintainers who have recently made a script to detect big data structures
1017 ph10 111 in the php code that should be moved to the .rodata section. I remembered
1018     to update Builducptable as well, so it won't revert if ucptable.h is ever
1019 ph10 105 re-created.
1020 ph10 111
1021     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1022 ph10 107 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1023 ph10 111 order to be able to cut out the UTF-8 tables in the latter when UTF-8
1024     support is not required. This saves 1.5-2K of code, which is important in
1025 ph10 107 some applications.
1026 ph10 111
1027 ph10 110 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1028 ph10 111 so as not to refer to the tables, even though these functions will never be
1029     called when UTF-8 support is disabled. Otherwise there are problems with a
1030     shared library.
1032 ph10 118 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1034     (a) It was defining its arguments as char * instead of void *.
1036 ph10 123 (b) It was assuming that all moves were upwards in memory; this was true
1037     a long time ago when I wrote it, but is no longer the case.
1039 ph10 118 The emulated memove() is provided for those environments that have neither
1040 ph10 123 memmove() nor bcopy(). I didn't think anyone used it these days, but that
1041 ph10 118 is clearly not the case, as these two bugs were recently reported.
1042 ph10 123
1043 ph10 111 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1044 ph10 123 and Detrail to create the HTML documentation, the .txt form of the man
1045     pages, and it removes trailing spaces from listed files. It also creates
1046     pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1047     case, it wraps all the #defines with #ifndefs. This script should be run
1048 ph10 111 before "make dist".
1049 ph10 123
1050 ph10 115 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1051     with Unicode property support.
1052 ph10 123
1053     (a) For a maximizing quantifier, if the two different cases of the
1054     character were of different lengths in their UTF-8 codings (there are
1055     some cases like this - I found 11), and the matching function had to
1056 ph10 115 back up over a mixture of the two cases, it incorrectly assumed they
1057     were both the same length.
1058 ph10 123
1059     (b) When PCRE was configured to use the heap rather than the stack for
1060     recursion during matching, it was not correctly preserving the data for
1061     the other case of a UTF-8 character when checking ahead for a match
1062     while processing a minimizing repeat. If the check also involved
1063 ph10 115 matching a wide character, but failed, corruption could cause an
1064     erroneous result when trying to check for a repeat of the original
1065     character.
1066 ph10 123
1067 ph10 116 13. Some tidying changes to the testing mechanism:
1068 ph10 98
1069 ph10 116 (a) The RunTest script now detects the internal link size and whether there
1070     is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1071 ph10 123 values substituted by "configure". (The RunGrepTest script already did
1072     this for UTF-8.) The configure.ac script no longer substitutes the
1073     relevant variables.
1075 ph10 116 (b) The debugging options /B and /D in pcretest show the compiled bytecode
1076     with length and offset values. This means that the output is different
1077     for different internal link sizes. Test 2 is skipped for link sizes
1078     other than 2 because of this, bypassing the problem. Unfortunately,
1079     there was also a test in test 3 (the locale tests) that used /B and
1080 ph10 123 failed for link sizes other than 2. Rather than cut the whole test out,
1081     I have added a new /Z option to pcretest that replaces the length and
1082     offset values with spaces. This is now used to make test 3 independent
1083 ph10 122 of link size. (Test 2 will be tidied up later.)
1084 ph10 123
1085     14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1086 ph10 122 segmentation fault instead of returning the appropriate error message.
1087 ph10 134
1088 ph10 131 15. In multiline mode when the newline sequence was set to "any", the pattern
1089 ph10 134 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1090     This doesn't seem right; it now treats the CRLF combination as the line
1091     ending, and so does not match in that case. It's only a pattern such as ^$
1092     that would hit this one: something like ^ABC$ would have failed after \r
1093     and then tried again after \r\n.
1095 ph10 131 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1096 ph10 134 in an attempt to make files that differ only in their line terminators
1097     compare equal. This works on Linux.
1098 ph10 142
1099 ph10 141 17. Under certain error circumstances pcregrep might try to free random memory
1100     as it exited. This is now fixed, thanks to valgrind.
1101 ph10 142
1102 ph10 141 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1103 ph10 142 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1104     was because its rules for how to advance for /g after matching an empty
1105 ph10 143 string at the end of a line did not allow for this case. They now check for
1106     it specially.
1107 ph10 150
1108     20. pcretest is supposed to handle patterns and data of any length, by
1109     extending its buffers when necessary. It was getting this wrong when the
1110 ph10 147 buffer for a data line had to be extended.
1111 ph10 150
1112 ph10 149 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1113 ph10 150 CRLF as a newline sequence.
1114 ph10 152
1115 ph10 151 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1116 ph10 152 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1117     I have nevertheless tidied it up.
1118 ph10 111
1119 ph10 152 23. Added some casts to kill warnings from HP-UX ia64 compiler.
1120 ph10 123
1121 ph10 153 24. Added a man page for pcre-config.
1122 ph10 152
1123 ph10 153
1124 nigel 93 Version 7.0 19-Dec-06
1125     ---------------------
1127     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1128     moving to gcc 4.1.1.
1130     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1131     sys/time.h, which is documented as needed for this function. It doesn't
1132     seem to matter on Linux, but it showed up on some releases of OS X.
1134     3. It seems that there are systems where bytes whose values are greater than
1135     127 match isprint() in the "C" locale. The "C" locale should be the
1136     default when a C program starts up. In most systems, only ASCII printing
1137     characters match isprint(). This difference caused the output from pcretest
1138     to vary, making some of the tests fail. I have changed pcretest so that:
1140     (a) When it is outputting text in the compiled version of a pattern, bytes
1141     other than 32-126 are always shown as hex escapes.
1143     (b) When it is outputting text that is a matched part of a subject string,
1144     it does the same, unless a different locale has been set for the match
1145     (using the /L modifier). In this case, it uses isprint() to decide.
1147     4. Fixed a major bug that caused incorrect computation of the amount of memory
1148     required for a compiled pattern when options that changed within the
1149     pattern affected the logic of the preliminary scan that determines the
1150     length. The relevant options are -x, and -i in UTF-8 mode. The result was
1151     that the computed length was too small. The symptoms of this bug were
1152     either the PCRE error "internal error: code overflow" from pcre_compile(),
1153     or a glibc crash with a message such as "pcretest: free(): invalid next
1154     size (fast)". Examples of patterns that provoked this bug (shown in
1155     pcretest format) are:
1157     /(?-x: )/x
1158     /(?x)(?-x: \s*#\s*)/
1159     /((?i)[\x{c0}])/8
1160     /(?i:[\x{c0}])/8
1162     HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1163     is now done differently.
1165     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1166     wrapper classes; (b) implement a new function in the C++ scanner that is
1167     more efficient than the old way of doing things because it avoids levels of
1168     recursion in the regex matching; (c) add a paragraph to the documentation
1169     for the FullMatch() function.
1171     6. The escape sequence \n was being treated as whatever was defined as
1172     "newline". Not only was this contrary to the documentation, which states
1173     that \n is character 10 (hex 0A), but it also went horribly wrong when
1174     "newline" was defined as CRLF. This has been fixed.
1176     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1177     was being set to -1 for the "end of line" case (supposedly a value that no
1178     character can have). Though this value is never used (the check for end of
1179     line is "zero bytes in current character"), it caused compiler complaints.
1180     I've changed it to 0xffffffff.
1182     8. In pcre_version.c, the version string was being built by a sequence of
1183     C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1184     string (as it is for production releases) called a macro with an empty
1185     argument. The C standard says the result of this is undefined. The gcc
1186     compiler treats it as an empty string (which was what was wanted) but it is
1187     reported that Visual C gives an error. The source has been hacked around to
1188     avoid this problem.
1190     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1191     builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1192     instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1193     of them did).
1195     10. Originally, pcretest opened its input and output without "b"; then I was
1196     told that "b" was needed in some environments, so it was added for release
1197     5.0 to both the input and output. (It makes no difference on Unix-like
1198     systems.) Later I was told that it is wrong for the input on Windows. I've
1199     now abstracted the modes into two macros, to make it easier to fiddle with
1200     them, and removed "b" from the input mode under Windows.
1202     11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1204     12. Added -help and --help to pcretest as an official way of being reminded
1205     of the options.
1207     13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1208     and pcrecpp.cc because they annoy compilers at high warning levels.
1210     14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1212     15. Fixed an occurrence of == in configure.ac that should have been = (shell
1213     scripts are not C programs :-) and which was not noticed because it works
1214     on Linux.
1216     16. pcretest is supposed to handle any length of pattern and data line (as one
1217     line or as a continued sequence of lines) by extending its input buffer if
1218     necessary. This feature was broken for very long pattern lines, leading to
1219     a string of junk being passed to pcre_compile() if the pattern was longer
1220     than about 50K.
1222     17. I have done a major re-factoring of the way pcre_compile() computes the
1223     amount of memory needed for a compiled pattern. Previously, there was code
1224     that made a preliminary scan of the pattern in order to do this. That was
1225     OK when PCRE was new, but as the facilities have expanded, it has become
1226     harder and harder to keep it in step with the real compile phase, and there
1227     have been a number of bugs (see for example, 4 above). I have now found a
1228     cunning way of running the real compile function in a "fake" mode that
1229     enables it to compute how much memory it would need, while actually only
1230     ever using a few hundred bytes of working memory and without too many
1231     tests of the mode. This should make future maintenance and development
1232     easier. A side effect of this work is that the limit of 200 on the nesting
1233     depth of parentheses has been removed (though this was never a serious
1234     limitation, I suspect). However, there is a downside: pcre_compile() now
1235     runs more slowly than before (30% or more, depending on the pattern). I
1236     hope this isn't a big issue. There is no effect on runtime performance.
1238     18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1239     newline (only possible for the last line of a file) and it was a
1240     pattern that set a locale (followed by /Lsomething), pcretest crashed.
1242     19. Added additional timing features to pcretest. (1) The -tm option now times
1243     matching only, not compiling. (2) Both -t and -tm can be followed, as a
1244     separate command line item, by a number that specifies the number of
1245     repeats to use when timing. The default is 50000; this gives better
1246     precision, but takes uncomfortably long for very large patterns.
1248     20. Extended pcre_study() to be more clever in cases where a branch of a
1249     subpattern has no definite first character. For example, (a*|b*)[cd] would
1250     previously give no result from pcre_study(). Now it recognizes that the
1251     first character must be a, b, c, or d.
1253     21. There was an incorrect error "recursive call could loop indefinitely" if
1254     a subpattern (or the entire pattern) that was being tested for matching an
1255     empty string contained only one non-empty item after a nested subpattern.
1256     For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1257     incorrectly, because the \d was being skipped in the check.
1259     22. The pcretest program now has a new pattern option /B and a command line
1260     option -b, which is equivalent to adding /B to every pattern. This causes
1261     it to show the compiled bytecode, without the additional information that
1262     -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1263     is the same as /B/I).
1265     23. A new optimization is now able automatically to treat some sequences such
1266     as a*b as a*+b. More specifically, if something simple (such as a character
1267     or a simple class like \d) has an unlimited quantifier, and is followed by
1268     something that cannot possibly match the quantified thing, the quantifier
1269     is automatically "possessified".
1271     24. A recursive reference to a subpattern whose number was greater than 39
1272     went wrong under certain circumstances in UTF-8 mode. This bug could also
1273     have affected the operation of pcre_study().
1275     25. Realized that a little bit of performance could be had by replacing
1276     (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1278     26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1280     27. Possessive quantifiers such as a++ were previously implemented by turning
1281     them into atomic groups such as ($>a+). Now they have their own opcodes,
1282     which improves performance. This includes the automatically created ones
1283     from 23 above.
1285     28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1286     lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1287     the first matched character to be a colon. This applied both to named and
1288     numbered groups.
1290     29. The ucpinternal.h header file was missing its idempotency #ifdef.
1292     30. I was sent a "project" file called libpcre.a.dev which I understand makes
1293     building PCRE on Windows easier, so I have included it in the distribution.
1295     31. There is now a check in pcretest against a ridiculously large number being
1296     returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1297     loop, the loop is abandoned.
1299     32. Forward references to subpatterns in conditions such as (?(2)...) where
1300     subpattern 2 is defined later cause pcre_compile() to search forwards in
1301     the pattern for the relevant set of parentheses. This search went wrong
1302     when there were unescaped parentheses in a character class, parentheses
1303     escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1305     33. "Subroutine" calls and backreferences were previously restricted to
1306     referencing subpatterns earlier in the regex. This restriction has now
1307     been removed.
1309     34. Added a number of extra features that are going to be in Perl 5.10. On the
1310     whole, these are just syntactic alternatives for features that PCRE had
1311     previously implemented using the Python syntax or my own invention. The
1312     other formats are all retained for compatibility.
1314     (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1315     as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1316     also .NET compatible.
1318     (b) A recursion or subroutine call to a named group can now be defined as
1319     (?&name) as well as (?P>name).
1321     (c) A backreference to a named group can now be defined as \k<name> or
1322     \k'name' as well as (?P=name). The new forms, as well as being in Perl
1323     5.10, are also .NET compatible.
1325     (d) A conditional reference to a named group can now use the syntax
1326     (?(<name>) or (?('name') as well as (?(name).
1328     (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1329     groups (named and numbered) that are never evaluated inline, but can be
1330     called as "subroutines" from elsewhere. In effect, the DEFINE condition
1331     is always false. There may be only one alternative in such a group.
1333     (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1334     as the simple (?(R). The condition is true only if the most recent
1335     recursion is that of the given number or name. It does not search out
1336     through the entire recursion stack.
1338     (g) The escape \gN or \g{N} has been added, where N is a positive or
1339     negative number, specifying an absolute or relative reference.
1341     35. Tidied to get rid of some further signed/unsigned compiler warnings and
1342     some "unreachable code" warnings.
1344     36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1345     things, this adds five new scripts.
1347     37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1348     There were also incompatibilities regarding the handling of \Q..\E inside
1349     character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1350     hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1352     38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1353     matches an empty string, and forcibly breaks the loop. There were bugs in
1354     this code in non-simple cases. For a pattern such as ^(a()*)* matched
1355     against aaaa the result was just "a" rather than "aaaa", for example. Two
1356     separate and independent bugs (that affected different cases) have been
1357     fixed.
1359     39. Refactored the code to abolish the use of different opcodes for small
1360     capturing bracket numbers. This is a tidy that I avoided doing when I
1361     removed the limit on the number of capturing brackets for 3.5 back in 2001.
1362     The new approach is not only tidier, it makes it possible to reduce the
1363     memory needed to fix the previous bug (38).
1365     40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1366     sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1367     processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1368     mode.
1370     41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1371     report.
1373     42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1374     copy construction and assignment in the C++ wrapper.
1376     43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1377     couple of bugs that could have given wrong results in the "--newline=crlf"
1378     case.
1380     44. Added a number of casts and did some reorganization of signed/unsigned int
1381     variables following suggestions from Dair Grant. Also renamed the variable
1382     "this" as "item" because it is a C++ keyword.
1384     45. Arranged for dftables to add
1386     #include "pcre_internal.h"
1388     to pcre_chartables.c because without it, gcc 4.x may remove the array
1389     definition from the final binary if PCRE is built into a static library and
1390     dead code stripping is activated.
1392     46. For an unanchored pattern, if a match attempt fails at the start of a
1393     newline sequence, and the newline setting is CRLF or ANY, and the next two
1394     characters are CRLF, advance by two characters instead of one.
1397 nigel 91 Version 6.7 04-Jul-06
1398     ---------------------
1400     1. In order to handle tests when input lines are enormously long, pcretest has
1401     been re-factored so that it automatically extends its buffers when
1402     necessary. The code is crude, but this _is_ just a test program. The
1403     default size has been increased from 32K to 50K.
1405     2. The code in pcre_study() was using the value of the re argument before
1406     testing it for NULL. (Of course, in any sensible call of the function, it
1407     won't be NULL.)
1409     3. The memmove() emulation function in pcre_internal.h, which is used on
1410     systems that lack both memmove() and bcopy() - that is, hardly ever -
1411     was missing a "static" storage class specifier.
1413     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1414     containing an extended class (one that cannot be represented by a bitmap
1415     because it contains high-valued characters or Unicode property items, e.g.
1416     [\pZ]). Almost always one would set UTF-8 mode when processing such a
1417     pattern, but PCRE should not loop if you do not (it no longer does).
1418     [Detail: two cases were found: (a) a repeated subpattern containing an
1419     extended class; (b) a recursive reference to a subpattern that followed a
1420     previous extended class. It wasn't skipping over the extended class
1421     correctly when UTF-8 mode was not set.]
1423     5. A negated single-character class was not being recognized as fixed-length
1424     in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1425     compile error "lookbehind assertion is not fixed length".
1427     6. The RunPerlTest auxiliary script was showing an unexpected difference
1428     between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1429     write a Perl script that can interpret lines of an input file either as
1430     byte characters or as UTF-8, which is what "perltest" was being required to
1431     do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1432     can't do is switch easily at run time between having the "use utf8;" pragma
1433     or not. In the end, I fudged it by using the RunPerlTest script to insert
1434     "use utf8;" explicitly for the UTF-8 tests.
1436     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1437     the end of the subject string, contrary to the documentation and to what
1438     Perl does. This was true of both matching functions. Now it matches only at
1439     the start of the subject and immediately after *internal* newlines.
1441     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1442     a pointer to an int instead of a pointer to an unsigned long int. This
1443     caused problems on 64-bit systems.
1445     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1446     instance of the 'standard' template library not being so standard".
1448     10. There was no check on the number of named subpatterns nor the maximum
1449     length of a subpattern name. The product of these values is used to compute
1450     the size of the memory block for a compiled pattern. By supplying a very
1451     long subpattern name and a large number of named subpatterns, the size
1452     computation could be caused to overflow. This is now prevented by limiting
1453     the length of names to 32 characters, and the number of named subpatterns
1454     to 10,000.
1456     11. Subpatterns that are repeated with specific counts have to be replicated in
1457     the compiled pattern. The size of memory for this was computed from the
1458     length of the subpattern and the repeat count. The latter is limited to
1459     65535, but there was no limit on the former, meaning that integer overflow
1460     could in principle occur. The compiled length of a repeated subpattern is
1461     now limited to 30,000 bytes in order to prevent this.
1463     12. Added the optional facility to have named substrings with the same name.
1465     13. Added the ability to use a named substring as a condition, using the
1466     Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1467     are numbers (not recommended). Forward references are permitted.
1469     14. Added forward references in named backreferences (if you see what I mean).
1471     15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1472     pattern could run off the end of the subject. For example, the pattern
1473     "(?s)(.{1,5})"8 did this with the subject "ab".
1475     16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1476     PCRE_CASELESS was set when matching characters that were quantified with ?
1477     or *.
1479     17. A character class other than a single negated character that had a minimum
1480     but no maximum quantifier - for example [ab]{6,} - was not handled
1481     correctly by pce_dfa_exec(). It would match only one character.
1483     18. A valid (though odd) pattern that looked like a POSIX character
1484     class but used an invalid character after [ (for example [[,abc,]]) caused
1485     pcre_compile() to give the error "Failed: internal error: code overflow" or
1486     in some cases to crash with a glibc free() error. This could even happen if
1487     the pattern terminated after [[ but there just happened to be a sequence of
1488     letters, a binary zero, and a closing ] in the memory that followed.
1490     19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1491     over the years. Originally (before any Unicode support), just the bottom 8
1492     bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1493     output from "man perlunicode" includes this:
1495     The regular expression compiler produces polymorphic opcodes. That
1496     is, the pattern adapts to the data and automatically switches to
1497     the Unicode character scheme when presented with Unicode data--or
1498     instead uses a traditional byte scheme when presented with byte
1499     data.
1501     Sadly, a wide octal escape does not cause a switch, and in a string with
1502     no other multibyte characters, these octal escapes are treated as before.
1503     Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
1504     /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1505     Unicode string.
1507     I have not perpetrated such confusion in PCRE. Up till now, it took just
1508     the bottom 8 bits, as in old Perl. I have now made octal escapes with
1509     values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1510     translate to the appropriate multibyte character.
1512     29. Applied some refactoring to reduce the number of warnings from Microsoft
1513     and Borland compilers. This has included removing the fudge introduced
1514     seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1515     a warning about an unused variable.
1517     21. PCRE has not included VT (character 0x0b) in the set of whitespace
1518     characters since release 4.0, because Perl (from release 5.004) does not.
1519     [Or at least, is documented not to: some releases seem to be in conflict
1520     with the documentation.] However, when a pattern was studied with
1521     pcre_study() and all its branches started with \s, PCRE still included VT
1522     as a possible starting character. Of course, this did no harm; it just
1523     caused an unnecessary match attempt.
1525     22. Removed a now-redundant internal flag bit that recorded the fact that case
1526     dependency changed within the pattern. This was once needed for "required
1527     byte" processing, but is no longer used. This recovers a now-scarce options
1528     bit. Also moved the least significant internal flag bit to the most-
1529     significant bit of the word, which was not previously used (hangover from
1530     the days when it was an int rather than a uint) to free up another bit for
1531     the future.
1533     23. Added support for CRLF line endings as well as CR and LF. As well as the
1534     default being selectable at build time, it can now be changed at runtime
1535     via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1536     specify that it is scanning data with non-default line endings.
1538     24. Changed the definition of CXXLINK to make it agree with the definition of
1539     LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1541     25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1542     recursions. This makes a big different to stack usage for some patterns.
1544     26. If a subpattern containing a named recursion or subroutine reference such
1545     as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1546     the space required for the compiled pattern went wrong and gave too small a
1547     value. Depending on the environment, this could lead to "Failed: internal
1548     error: code overflow at offset 49" or "glibc detected double free or
1549     corruption" errors.
1551     27. Applied patches from Google (a) to support the new newline modes and (b) to
1552     advance over multibyte UTF-8 characters in GlobalReplace.
1554     28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1555     difference for some implementation of PCRE in some Windows version.
1557     29. Added some extra testing facilities to pcretest:
1559     \q<number> in a data line sets the "match limit" value
1560     \Q<number> in a data line sets the "match recursion limt" value
1561     -S <number> sets the stack size, where <number> is in megabytes
1563     The -S option isn't available for Windows.
1566 nigel 89 Version 6.6 06-Feb-06
1567     ---------------------
1569     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1570     in pcreposix.h. I have copied the definition from pcre.h.
1572     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1573     because pcre.h is no longer a built file.
1575     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1576     not normally included in the compiled code.
1579 nigel 87 Version 6.5 01-Feb-06
1580     ---------------------
1582     1. When using the partial match feature with pcre_dfa_exec(), it was not
1583     anchoring the second and subsequent partial matches at the new starting
1584     point. This could lead to incorrect results. For example, with the pattern
1585     /1234/, partially matching against "123" and then "a4" gave a match.
1587     2. Changes to pcregrep:
1589     (a) All non-match returns from pcre_exec() were being treated as failures
1590     to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1591     error message is output. Some extra information is given for the
1593     probably the only errors that are likely to be caused by users (by
1594     specifying a regex that has nested indefinite repeats, for instance).
1595     If there are more than 20 of these errors, pcregrep is abandoned.
1597     (b) A binary zero was treated as data while matching, but terminated the
1598     output line if it was written out. This has been fixed: binary zeroes
1599     are now no different to any other data bytes.
1601     (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1602     used to set a locale for matching. The --locale=xxxx long option has
1603     been added (no short equivalent) to specify a locale explicitly on the
1604     pcregrep command, overriding the environment variables.
1606     (d) When -B was used with -n, some line numbers in the output were one less
1607     than they should have been.
1609     (e) Added the -o (--only-matching) option.
1611     (f) If -A or -C was used with -c (count only), some lines of context were
1612     accidentally printed for the final match.
1614     (g) Added the -H (--with-filename) option.
1616     (h) The combination of options -rh failed to suppress file names for files
1617     that were found from directory arguments.
1619     (i) Added the -D (--devices) and -d (--directories) options.
1621     (j) Added the -F (--fixed-strings) option.
1623     (k) Allow "-" to be used as a file name for -f as well as for a data file.
1625     (l) Added the --colo(u)r option.
1627     (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1628     is not present by default.
1630     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1631     items such as (?R) or (?1), when the recursion could match a number of
1632     alternatives. If it matched one of the alternatives, but subsequently,
1633     outside the recursion, there was a failure, the code tried to back up into
1634     the recursion. However, because of the way PCRE is implemented, this is not
1635     possible, and the result was an incorrect result from the match.
1637     In order to prevent this happening, the specification of recursion has
1638     been changed so that all such subpatterns are automatically treated as
1639     atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1641     4. I had overlooked the fact that, in some locales, there are characters for
1642     which isalpha() is true but neither isupper() nor islower() are true. In
1643     the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1644     and ordfeminine) are like this. This affected the treatment of \w and \W
1645     when they appeared in character classes, but not when they appeared outside
1646     a character class. The bit map for "word" characters is now created
1647     separately from the results of isalnum() instead of just taking it from the
1648     upper, lower, and digit maps. (Plus the underscore character, of course.)
1650     5. The above bug also affected the handling of POSIX character classes such as
1651     [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1652     permanent tables. Instead, the bit maps for such a class were previously
1653     created as the appropriate unions of the upper, lower, and digit bitmaps.
1654     Now they are created by subtraction from the [[:word:]] class, which has
1655     its own bitmap.
1657     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1658     It is created by subtracting the vertical space characters (\x09, \x0a,
1659     \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1660     subtraction was done in the overall bitmap for a character class, meaning
1661     that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1662     be recognized. This bug has been fixed.
1664     7. Patches from the folks at Google:
1666     (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1667     real life, but is still worth protecting against".
1669     (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1670     regular expressions".
1672     (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1673     have it.
1675     (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1676     "configure" and the latter not, in order to fix a problem somebody had
1677     with compiling the Arg class on HP-UX.
1679     (e) Improve the error-handling of the C++ wrapper a little bit.
1681     (f) New tests for checking recursion limiting.
1683     8. The pcre_memmove() function, which is used only if the environment does not
1684     have a standard memmove() function (and is therefore rarely compiled),
1685     contained two bugs: (a) use of int instead of size_t, and (b) it was not
1686     returning a result (though PCRE never actually uses the result).
1688     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1689     large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1690     returned instead of calling malloc() with an overflowing number that would
1691     most likely cause subsequent chaos.
1693     10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1695     11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1696     with this option is matched, the nmatch and pmatch options of regexec() are
1697     ignored.
1699     12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1700     provided in case anyone wants to the the POSIX interface with UTF-8
1701     strings.
1703     13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1704     C++ linking (needed for some HP-UX environments).
1706     14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1707     (unused parameter) and in the pcre_printint() function (omitted "default"
1708     switch label when the default is to do nothing).
1710     15. Added some code to make it possible, when PCRE is compiled as a C++
1711     library, to replace subject pointers for pcre_exec() with a smart pointer
1712     class, thus making it possible to process discontinuous strings.
1714     16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1715     much the same function. They were added by different people who were trying
1716     to make PCRE easy to compile on non-Unix systems. It has been suggested
1717     that PCRE_EXPORT be abolished now that there is more automatic apparatus
1718     for compiling on Windows systems. I have therefore replaced it with
1719     PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1720     defaults to "extern" for C or "extern C" for C++, which works fine on
1721     Unix-like systems. It is now possible to override the value of PCRE_DATA_
1722     SCOPE with something explicit in config.h. In addition:
1724     (a) pcreposix.h still had just "extern" instead of either of these macros;
1725     I have replaced it with PCRE_DATA_SCOPE.
1727     (b) Functions such as _pcre_xclass(), which are internal to the library,
1728     but external in the C sense, all had PCRE_EXPORT in their definitions.
1729     This is apparently wrong for the Windows case, so I have removed it.
1730     (It makes no difference on Unix-like systems.)
1732     17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1733     of recursive calls to match(). This is different to MATCH_LIMIT because
1734     that limits the total number of calls to match(), not all of which increase
1735     the depth of recursion. Limiting the recursion depth limits the amount of
1736     stack (or heap if NO_RECURSE is set) that is used. The default can be set
1737     when PCRE is compiled, and changed at run time. A patch from Google adds
1738     this functionality to the C++ interface.
1740     18. Changes to the handling of Unicode character properties:
1742     (a) Updated the table to Unicode 4.1.0.
1744     (b) Recognize characters that are not in the table as "Cn" (undefined).
1746     (c) I revised the way the table is implemented to a much improved format
1747     which includes recognition of ranges. It now supports the ranges that
1748     are defined in UnicodeData.txt, and it also amalgamates other
1749     characters into ranges. This has reduced the number of entries in the
1750     table from around 16,000 to around 3,000, thus reducing its size
1751     considerably. I realized I did not need to use a tree structure after
1752     all - a binary chop search is just as efficient. Having reduced the
1753     number of entries, I extended their size from 6 bytes to 8 bytes to
1754     allow for more data.
1756     (d) Added support for Unicode script names via properties such as \p{Han}.
1758     19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1759     matching that character.
1761     20. When matching a repeated Unicode property with a minimum greater than zero,
1762     (for example \pL{2,}), PCRE could look past the end of the subject if it
1763     reached it while seeking the minimum number of characters. This could
1764     happen only if some of the characters were more than one byte long, because
1765     there is a check for at least the minimum number of bytes.
1767     21. Refactored the implementation of \p and \P so as to be more general, to
1768     allow for more different types of property in future. This has changed the
1769     compiled form incompatibly. Anybody with saved compiled patterns that use
1770     \p or \P will have to recompile them.
1772     22. Added "Any" and "L&" to the supported property types.
1774     23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1775     but give a compile time error if the value is greater than 0xff.
1777     24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1778     accidentally not being installed or uninstalled.
1780     25. The pcre.h file was built from pcre.h.in, but the only changes that were
1781     made were to insert the current release number. This seemed silly, because
1782     it made things harder for people building PCRE on systems that don't run
1783     "configure". I have turned pcre.h into a distributed file, no longer built
1784     by "configure", with the version identification directly included. There is
1785     no longer a pcre.h.in file.
1787     However, this change necessitated a change to the pcre-config script as
1788     well. It is built from pcre-config.in, and one of the substitutions was the
1789     release number. I have updated configure.ac so that ./configure now finds
1790     the release number by grepping pcre.h.
1792     26. Added the ability to run the tests under valgrind.
1795 nigel 85 Version 6.4 05-Sep-05
1796     ---------------------
1798     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1799     "--" to be printed when multiple files were scanned, even when none of the
1800     -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1801     consider it to be a bug, and have restored the previous behaviour.
1803     2. A couple of code tidies to get rid of compiler warnings.
1805     3. The pcretest program used to cheat by referring to symbols in the library
1806     whose names begin with _pcre_. These are internal symbols that are not
1807     really supposed to be visible externally, and in some environments it is
1808     possible to suppress them. The cheating is now confined to including
1809     certain files from the library's source, which is a bit cleaner.
1811     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1812     file's purpose clearer.
1814     5. Reorganized pcre_ucp_findchar().
1817 nigel 83 Version 6.3 15-Aug-05
1818     ---------------------
1820     1. The file libpcre.pc.in did not have general read permission in the tarball.
1822     2. There were some problems when building without C++ support:
1824     (a) If C++ support was not built, "make install" and "make test" still
1825     tried to test it.
1827     (b) There were problems when the value of CXX was explicitly set. Some
1828     changes have been made to try to fix these, and ...
1830     (c) --disable-cpp can now be used to explicitly disable C++ support.
1832     (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1833     backslash in a target when C++ was disabled. This confuses some
1834     versions of "make", apparently. Using an intermediate variable solves
1835     this. (Same for CPP_LOBJ.)
1837     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1838     (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1839     necessary on certain architectures.
1841     4. Added a setting of -export-symbols-regex to the link command to remove
1842     those symbols that are exported in the C sense, but actually are local
1843     within the library, and not documented. Their names all begin with
1844     "_pcre_". This is not a perfect job, because (a) we have to except some
1845     symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1846     available (and never for static libraries). I have made a note to try to
1847     find a way round (a) in the future.
1850 nigel 81 Version 6.2 01-Aug-05
1851     ---------------------
1853     1. There was no test for integer overflow of quantifier values. A construction
1854     such as {1111111111111111} would give undefined results. What is worse, if
1855     a minimum quantifier for a parenthesized subpattern overflowed and became
1856     negative, the calculation of the memory size went wrong. This could have
1857     led to memory overwriting.
1859     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1861     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1862     operating environments where this matters.
1864     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1865     PCRE options from within the C++ wrapper.
1867     5. Named capturing subpatterns were not being correctly counted when a pattern
1868     was compiled. This caused two problems: (a) If there were more than 100
1869     such subpatterns, the calculation of the memory needed for the whole
1870     compiled pattern went wrong, leading to an overflow error. (b) Numerical
1871     back references of the form \12, where the number was greater than 9, were
1872     not recognized as back references, even though there were sufficient
1873     previous subpatterns.
1875     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1876     versions of gcc, e.g. 2.95.4.
1879 nigel 79 Version 6.1 21-Jun-05
1880     ---------------------
1882     1. There was one reference to the variable "posix" in pcretest.c that was not
1883     surrounded by "#if !defined NOPOSIX".
1885     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1886     the cross-check on the old pcre_info() function, for the benefit of the
1887     cut-down version of PCRE that is currently imported into Exim.
1889     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1890     allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1891     patterns that start (?i) but I don't think that matters. The use of (?i) is
1892     just an example; this all applies to the other options as well.
1894     4. Since libtool seems to echo the compile commands it is issuing, the output
1895     from "make" can be reduced a bit by putting "@" in front of each libtool
1896     compile command.
1898     5. Patch from the folks at Google for configure.in to be a bit more thorough
1899     in checking for a suitable C++ installation before trying to compile the
1900     C++ stuff. This should fix a reported problem when a compiler was present,
1901     but no suitable headers.
1903     6. The man pages all had just "PCRE" as their title. I have changed them to
1904     be the relevant file name. I have also arranged that these names are
1905     retained in the file doc/pcre.txt, which is a concatenation in text format
1906     of all the man pages except the little individual ones for each function.
1908     7. The NON-UNIX-USE file had not been updated for the different set of source
1909     files that come with release 6. I also added a few comments about the C++
1910     wrapper.
1913 nigel 77 Version 6.0 07-Jun-05
1914     ---------------------
1916     1. Some minor internal re-organization to help with my DFA experiments.
1918     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1919     didn't matter for the library itself when fully configured, but did matter
1920     when compiling without UCP support, or within Exim, where the ucp files are
1921     not imported.
1923     3. Refactoring of the library code to split up the various functions into
1924     different source modules. The addition of the new DFA matching code (see
1925     below) to a single monolithic source would have made it really too
1926     unwieldy, quite apart from causing all the code to be include in a
1927     statically linked application, when only some functions are used. This is
1928     relevant even without the DFA addition now that patterns can be compiled in
1929     one application and matched in another.
1931     The downside of splitting up is that there have to be some external
1932     functions and data tables that are used internally in different modules of
1933     the library but which are not part of the API. These have all had their
1934     names changed to start with "_pcre_" so that they are unlikely to clash
1935     with other external names.
1937     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1938     a different (DFA) algorithm. Although it is slower than the original
1939     function, it does have some advantages for certain types of matching
1940     problem.
1942     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1943     including restarting after a partial match.
1945     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1946     defined when compiling for Windows was sent to me. I have put it into the
1947     code, though I have no means of testing or verifying it.
1949     7. Added the pcre_refcount() auxiliary function.
1951     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1952     match before or at the first newline in the subject string. In pcretest,
1953     the /f option on a pattern can be used to set this.
1955     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1956     would behave wrongly. This has been present in PCRE since release 4.0.
1958     10. A number of changes to the pcregrep command:
1960     (a) Refactored how -x works; insert ^(...)$ instead of setting
1961     PCRE_ANCHORED and checking the length, in preparation for adding
1962     something similar for -w.
1964     (b) Added the -w (match as a word) option.
1966     (c) Refactored the way lines are read and buffered so as to have more
1967     than one at a time available.
1969     (d) Implemented a pcregrep test script.
1971     (e) Added the -M (multiline match) option. This allows patterns to match
1972     over several lines of the subject. The buffering ensures that at least
1973     8K, or the rest of the document (whichever is the shorter) is available
1974     for matching (and similarly the previous 8K for lookbehind assertions).
1976     (f) Changed the --help output so that it now says
1978     -w, --word-regex(p)
1980     instead of two lines, one with "regex" and the other with "regexp"
1981     because that confused at least one person since the short forms are the
1982     same. (This required a bit of code, as the output is generated
1983     automatically from a table. It wasn't just a text change.)
1985     (g) -- can be used to terminate pcregrep options if the next thing isn't an
1986     option but starts with a hyphen. Could be a pattern or a path name
1987     starting with a hyphen, for instance.
1989     (h) "-" can be given as a file name to represent stdin.
1991     (i) When file names are being printed, "(standard input)" is used for
1992     the standard input, for compatibility with GNU grep. Previously
1993     "<stdin>" was used.
1995     (j) The option --label=xxx can be used to supply a name to be used for
1996     stdin when file names are being printed. There is no short form.
1998     (k) Re-factored the options decoding logic because we are going to add
1999     two more options that take data. Such options can now be given in four
2000     different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2002     (l) Added the -A, -B, and -C options for requesting that lines of context
2003     around matches be printed.
2005     (m) Added the -L option to print the names of files that do not contain
2006     any matching lines, that is, the complement of -l.
2008     (n) The return code is 2 if any file cannot be opened, but pcregrep does
2009     continue to scan other files.
2011     (o) The -s option was incorrectly implemented. For compatibility with other
2012     greps, it now suppresses the error message for a non-existent or non-
2013     accessible file (but not the return code). There is a new option called
2014     -q that suppresses the output of matching lines, which was what -s was
2015     previously doing.
2017     (p) Added --include and --exclude options to specify files for inclusion
2018     and exclusion when recursing.
2020     11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2021     Hopefully, it now does.
2023     12. Missing cast in pcre_study().
2025     13. Added an "uninstall" target to the makefile.
2027     14. Replaced "extern" in the function prototypes in Makefile.in with
2028     "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2029     world, but is set differently for Windows.
2031     15. Added a second compiling function called pcre_compile2(). The only
2032     difference is that it has an extra argument, which is a pointer to an
2033     integer error code. When there is a compile-time failure, this is set
2034     non-zero, in addition to the error test pointer being set to point to an
2035     error message. The new argument may be NULL if no error number is required
2036     (but then you may as well call pcre_compile(), which is now just a
2037     wrapper). This facility is provided because some applications need a
2038     numeric error indication, but it has also enabled me to tidy up the way
2039     compile-time errors are handled in the POSIX wrapper.
2041     16. Added VPATH=.libs to the makefile; this should help when building with one
2042     prefix path and installing with another. (Or so I'm told by someone who
2043     knows more about this stuff than I do.)
2045     17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2046     passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2047     match everything, including newlines. This is not POSIX-compatible, but
2048     somebody wanted the feature. From pcretest it can be activated by using
2049     both the P and the s flags.
2051     18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2053     19. libpcre.pc was being incorrectly installed as executable.
2055     20. A couple of places in pcretest check for end-of-line by looking for '\n';
2056     it now also looks for '\r' so that it will work unmodified on Windows.
2058     21. Added Google's contributed C++ wrapper to the distribution.
2060     22. Added some untidy missing memory free() calls in pcretest, to keep
2061     Electric Fence happy when testing.
2065 nigel 75 Version 5.0 13-Sep-04
2066     ---------------------
2068     1. Internal change: literal characters are no longer packed up into items
2069     containing multiple characters in a single byte-string. Each character
2070     is now matched using a separate opcode. However, there may be more than one
2071     byte in the character in UTF-8 mode.
2073     2. The pcre_callout_block structure has two new fields: pattern_position and
2074     next_item_length. These contain the offset in the pattern to the next match
2075     item, and its length, respectively.
2077     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2078     insertion of callouts before each pattern item. Added the /C option to
2079     pcretest to make use of this.
2081     4. On the advice of a Windows user, the lines
2083     #if defined(_WIN32) || defined(WIN32)
2084     _setmode( _fileno( stdout ), 0x8000 );
2085     #endif /* defined(_WIN32) || defined(WIN32) */
2087     have been added to the source of pcretest. This apparently does useful
2088     magic in relation to line terminators.
2090     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2091     for the benefit of those environments where the "b" makes a difference.
2093     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2094     to know about it. I have put a hack into configure.in that adds in code
2095     to set GCC=yes if CC=icc. This seems to end up at a point in the
2096     generated configure script that is early enough to affect the setting of
2097     compiler options, which is what is needed, but I have no means of testing
2098     whether it really works. (The user who reported this had patched the
2099     generated configure script, which of course I cannot do.)
2101     LATER: After change 22 below (new libtool files), the configure script
2102     seems to know about icc (and also ecc). Therefore, I have commented out
2103     this hack in configure.in.
2105     7. Added support for pkg-config (2 patches were sent in).
2107     8. Negated POSIX character classes that used a combination of internal tables
2108     were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2109     [[:^ascii]]. Typically, they would match almost any characters. The other
2110     POSIX classes were not broken in this way.
2112     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2113     to find the match, as PCRE was deluded into thinking that the match had to
2114     start at the start point or following a newline. The same bug applied to
2115     patterns with negative forward assertions or any backward assertions
2116     preceding ".*" at the start, unless the pattern required a fixed first
2117     character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2119     10. In UTF-8 mode, when moving forwards in the subject after a failed match
2120     starting at the last subject character, bytes beyond the end of the subject
2121     string were read.
2123     11. Renamed the variable "class" as "classbits" to make life easier for C++
2124     users. (Previously there was a macro definition, but it apparently wasn't
2125     enough.)
2127     12. Added the new field "tables" to the extra data so that tables can be passed
2128     in at exec time, or the internal tables can be re-selected. This allows
2129     a compiled regex to be saved and re-used at a later time by a different
2130     program that might have everything at different addresses.
2132     13. Modified the pcre-config script so that, when run on Solaris, it shows a
2133     -R library as well as a -L library.
2135     14. The debugging options of pcretest (-d on the command line or D on a
2136     pattern) showed incorrect output for anything following an extended class
2137     that contained multibyte characters and which was followed by a quantifier.
2139     15. Added optional support for general category Unicode character properties
2140     via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2141     support. It adds about 90K to the size of the library. The meanings of the
2142     inbuilt class escapes such as \d and \s have NOT been changed.
2144     16. Updated pcredemo.c to include calls to free() to release the memory for the
2145     compiled pattern.
2147     17. The generated file chartables.c was being created in the source directory
2148     instead of in the building directory. This caused the build to fail if the
2149     source directory was different from the building directory, and was
2150     read-only.
2152     18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2153     file. No doubt somebody will tell me if they don't make sense... Also added
2154     Dan Mooney's comments about building on OpenVMS.
2156     19. Added support for partial matching via the PCRE_PARTIAL option for
2157     pcre_exec() and the \P data escape in pcretest.
2159     20. Extended pcretest with 3 new pattern features:
2161     (i) A pattern option of the form ">rest-of-line" causes pcretest to
2162     write the compiled pattern to the file whose name is "rest-of-line".
2163     This is a straight binary dump of the data, with the saved pointer to
2164     the character tables forced to be NULL. The study data, if any, is
2165     written too. After writing, pcretest reads a new pattern.
2167     (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2168     compiled pattern from the given file. There must not be any
2169     occurrences of "<" in the file name (pretty unlikely); if there are,
2170     pcretest will instead treat the initial "<" as a pattern delimiter.
2171     After reading in the pattern, pcretest goes on to read data lines as
2172     usual.
2174     (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2175     and 16-bit fields in a compiled pattern, to simulate a pattern that
2176     was compiled on a host of opposite endianness.
2178     21. The pcre-exec() function can now cope with patterns that were compiled on
2179     hosts of opposite endianness, with this restriction:
2181     As for any compiled expression that is saved and used later, the tables
2182     pointer field cannot be preserved; the extra_data field in the arguments
2183     to pcre_exec() should be used to pass in a tables address if a value
2184     other than the default internal tables were used at compile time.
2186     22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2187     now diagnosed as an error. Previously, most of the time, a negative number
2188     would have been treated as zero, but if in addition "ovector" was passed as
2189     NULL, a crash could occur.
2191     23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2192     new versions from the libtool 1.5 distribution (the last one is a copy of
2193     a file called libtool.m4). This seems to have fixed the need to patch
2194     "configure" to support Darwin 1.3 (which I used to do). However, I still
2195     had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2196     workstation).
2198     24. Changed the PCRE licence to be the more standard "BSD" licence.
2201 nigel 73 Version 4.5 01-Dec-03
2202     ---------------------
2204     1. There has been some re-arrangement of the code for the match() function so
2205     that it can be compiled in a version that does not call itself recursively.
2206     Instead, it keeps those local variables that need separate instances for
2207     each "recursion" in a frame on the heap, and gets/frees frames whenever it
2208     needs to "recurse". Keeping track of where control must go is done by means
2209     of setjmp/longjmp. The whole thing is implemented by a set of macros that
2210     hide most of the details from the main code, and operates only if
2211     NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2212     "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2213     operating.
2215     To make it easier for callers to provide specially tailored get/free
2216     functions for this usage, two new functions, pcre_stack_malloc, and
2217     pcre_stack_free, are used. They are always called in strict stacking order,
2218     and the size of block requested is always the same.
2220     The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2221     PCRE has been compiled to use the stack or the heap for recursion. The
2222     -C option of pcretest uses this to show which version is compiled.
2224     A new data escape \S, is added to pcretest; it causes the amounts of store
2225     obtained and freed by both kinds of malloc/free at match time to be added
2226     to the output.
2228     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2229     what's available on my current Linux desktop machine.
2231     3. When matching a UTF-8 string, the test for a valid string at the start has
2232     been extended. If start_offset is not zero, PCRE now checks that it points
2233     to a byte that is the start of a UTF-8 character. If not, it returns
2234     PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2235     this is necessary because there may be backward assertions in the pattern.
2236     When matching the same subject several times, it may save resources to use
2237     PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2239     4. The code for checking the validity of UTF-8 strings has been tightened so
2240     that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2241     containing "overlong sequences".
2243     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2244     I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2245     should have been "&", but it just so happened that all the cases this let
2246     through by mistake were picked up later in the function.
2248     6. I had used a variable called "isblank" - this is a C99 function, causing
2249     some compilers to warn. To avoid this, I renamed it (as "blankclass").
2251     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2252     prompting; (b) run "./pcretest /dev/null" at the start of the test script
2253     so the version is shown; (c) stop "make test" echoing "./RunTest".
2255     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2257     9. The prototype for memmove() for systems that don't have it was using
2258     size_t, but the inclusion of the header that defines size_t was later. I've
2259     moved the #includes for the C headers earlier to avoid this.
2261     10. Added some adjustments to the code to make it easier to compiler on certain
2262     special systems:
2264     (a) Some "const" qualifiers were missing.
2265     (b) Added the macro EXPORT before all exported functions; by default this
2266     is defined to be empty.
2267     (c) Changed the dftables auxiliary program (that builds chartables.c) so
2268     that it reads its output file name as an argument instead of writing
2269     to the standard output and assuming this can be redirected.
2271     11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2272     class containing characters with values greater than 255, PCRE compilation
2273     went into a loop.
2275     12. A recursive reference to a subpattern that was within another subpattern
2276     that had a minimum quantifier of zero caused PCRE to crash. For example,
2277     (x(y(?2))z)? provoked this bug with a subject that got as far as the
2278     recursion. If the recursively-called subpattern itself had a zero repeat,
2279     that was OK.
2281     13. In pcretest, the buffer for reading a data line was set at 30K, but the
2282     buffer into which it was copied (for escape processing) was still set at
2283     1024, so long lines caused crashes.
2285     14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2286     "internal error: code overflow...". This applied to any character class
2287     that was followed by a possessive quantifier.
2289     15. Modified the Makefile to add libpcre.la as a prerequisite for
2290     libpcreposix.la because I was told this is needed for a parallel build to
2291     work.
2293     16. If a pattern that contained .* following optional items at the start was
2294     studied, the wrong optimizing data was generated, leading to matching
2295     errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2296     matching string must start with a or b or c. The correct conclusion for
2297     this pattern is that a match can start with any character.
2300 nigel 71 Version 4.4 13-Aug-03
2301     ---------------------
2303     1. In UTF-8 mode, a character class containing characters with values between
2304     127 and 255 was not handled correctly if the compiled pattern was studied.
2305     In fixing this, I have also improved the studying algorithm for such
2306     classes (slightly).
2308     2. Three internal functions had redundant arguments passed to them. Removal
2309     might give a very teeny performance improvement.
2311     3. Documentation bug: the value of the capture_top field in a callout is *one
2312     more than* the number of the hightest numbered captured substring.
2314     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2315     in incorrectly linking with a previously installed version. They now link
2316     explicitly with libpcre.la.
2318     5. configure.in no longer needs to recognize Cygwin specially.
2320     6. A problem in pcre.in for Windows platforms is fixed.
2322     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2323     pcretest, it used to include the size of the study block as part of its
2324     output. Unfortunately, the structure contains a field that has a different
2325     size on different hardware architectures. This meant that the tests that
2326     showed this size failed. As the block is currently always of a fixed size,
2327     this information isn't actually particularly useful in pcretest output, so
2328     I have just removed it.
2330     8. Three pre-processor statements accidentally did not start in column 1.
2331     Sadly, there are *still* compilers around that complain, even though
2332     standard C has not required this for well over a decade. Sigh.
2334     9. In pcretest, the code for checking callouts passed small integers in the
2335     callout_data field, which is a void * field. However, some picky compilers
2336     complained about the casts involved for this on 64-bit systems. Now
2337     pcretest passes the address of the small integer instead, which should get
2338     rid of the warnings.
2340     10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2341     both compile and run time, and gives an error if an invalid UTF-8 sequence
2342     is found. There is a option for disabling this check in cases where the
2343     string is known to be correct and/or the maximum performance is wanted.
2345     11. In response to a bug report, I changed one line in Makefile.in from
2347     -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2348     to
2349     -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2351     to look similar to other lines, but I have no way of telling whether this
2352     is the right thing to do, as I do not use Windows. No doubt I'll get told
2353     if it's wrong...
2356 nigel 69 Version 4.3 21-May-03
2357     ---------------------
2359     1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2360     Makefile.
2362     2. Some refactoring to improve the quality of the code:
2364     (i) The utf8_table... variables are now declared "const".
2366     (ii) The code for \cx, which used the "case flipping" table to upper case
2367     lower case letters, now just substracts 32. This is ASCII-specific,
2368     but the whole concept of \cx is ASCII-specific, so it seems
2369     reasonable.
2371     (iii) PCRE was using its character types table to recognize decimal and
2372     hexadecimal digits in the pattern. This is silly, because it handles
2373     only 0-9, a-f, and A-F, but the character types table is locale-
2374     specific, which means strange things might happen. A private
2375     table is now used for this - though it costs 256 bytes, a table is
2376     much faster than multiple explicit tests. Of course, the standard
2377     character types table is still used for matching digits in subject
2378     strings against \d.
2380     (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2381     ending in _t are). So I've renamed it as ESC_tee.
2383     3. The first argument for regexec() in the POSIX wrapper should have been
2384     defined as "const".
2386     4. Changed pcretest to use malloc() for its buffers so that they can be
2387     Electric Fenced for debugging.
2389     5. There were several places in the code where, in UTF-8 mode, PCRE would try
2390     to read one or more bytes before the start of the subject string. Often this
2391     had no effect on PCRE's behaviour, but in some circumstances it could
2392     provoke a segmentation fault.
2394     6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2395     to try to read one or more bytes before the start of the subject string.
2397     7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2398     UTF-8 support could misbehave in various ways if the subject string
2399     contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2400     area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2401     back over UTF-8 characters.)
2404 nigel 67 Version 4.2 14-Apr-03
2405     ---------------------
2407     1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2409     2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2410     [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2411     [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2412     [WIN_PREFIX]: new variable, "cyg" for Cygwin
2413     * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2414     and BUILD_EXEEXT
2415     Note: automatic setting of the BUILD variables is not yet working
2416     set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2417     compile-time but not at link-time
2418     [LINK]: use for linking executables only
2419     make different versions for Windows and non-Windows
2420     [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2421     libraries
2422     [LINK_FOR_BUILD]: new variable
2423     [OBJEXT]: use throughout
2424     [EXEEXT]: use throughout
2425     <winshared>: new target
2426     <wininstall>: new target
2427     <dftables.o>: use native compiler
2428     <dftables>: use native linker
2429     <install>: handle Windows platform correctly
2430     <clean>: ditto
2431     <check>: ditto
2432     copy DLL to top builddir before testing
2434     As part of these changes, -no-undefined was removed again. This was reported
2435     to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2436     in any case.
2438     3. Some tidies to get rid of compiler warnings:
2440     . In the match_data structure, match_limit was an unsigned long int, whereas
2441     match_call_count was an int. I've made them both unsigned long ints.
2443     . In pcretest the fact that a const uschar * doesn't automatically cast to
2444     a void * provoked a warning.
2446     . Turning on some more compiler warnings threw up some "shadow" variables
2447     and a few more missing casts.
2449     4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2450     option, a class that contained a single character with a value between 128
2451     and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2453     5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2454     option, a class that contained several characters, but with at least one
2455     whose value was between 128 and 255 caused PCRE to crash.
2458 nigel 65 Version 4.1 12-Mar-03
2459     ---------------------
2460 nigel 63
2461 nigel 65 1. Compiling with gcc -pedantic found a couple of places where casts were
2462     needed, and a string in dftables.c that was longer than standard compilers are
2463     required to support.
2465     2. Compiling with Sun's compiler found a few more places where the code could
2466     be tidied up in order to avoid warnings.
2468     3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2469     first of these names is deprecated in the latest Autoconf in favour of the name
2470     CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2471     compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2472     analogy I have changed it to CFLAGS_FOR_BUILD.
2474     4. Added -no-undefined to the linking command in the Makefile, because this is
2475     apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2476     linking step for the pcreposix library.
2478     5. PCRE was failing to diagnose the case of two named groups with the same
2479     name.
2481     6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2482     literal character that is needed in the subject for a match, and scans along to
2483     ensure that it is present before embarking on the full matching process. This
2484     saves time in cases of nested unlimited repeats that are never going to match.
2485     Problem: the scan can take a lot of time if the subject is very long (e.g.
2486     megabytes), thus penalizing straightforward matches. It is now done only if the
2487     amount of subject to be scanned is less than 1000 bytes.
2489     7. A lesser problem with the same optimization is that it was recording the
2490     first character of an anchored pattern as "needed", thus provoking a search
2491     right along the subject, even when the first match of the pattern was going to
2492     fail. The "needed" character is now not set for anchored patterns, unless it
2493     follows something in the pattern that is of non-fixed length. Thus, it still
2494     fulfils its original purpose of finding quick non-matches in cases of nested
2495     unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2498     Version 4.0 17-Feb-03
2499     ---------------------
2501 nigel 63 1. If a comment in an extended regex that started immediately after a meta-item
2502     extended to the end of string, PCRE compiled incorrect data. This could lead to
2503     all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2505     2. Moved to autoconf 2.53 and libtool 1.4.2.
2507     3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2508     the special perltest8 script is no longer needed - all the tests can be run
2509     from a single perltest script.
2511     4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2512     by \s. It has now been removed in PCRE. This means it isn't recognized as
2513     whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2514     class [:space:] *does* include VT, thereby creating a mess.
2516     5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2517     space and tab.
2519     6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2520     its new features into the main test script, reducing the number of scripts.
2522     7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2523     were backward compatible, and made the (?i) apply to the whole pattern, as if
2524     /i were given. Now it behaves more logically, and applies the option setting
2525     only to what follows. PCRE has been changed to follow suit. However, if it
2526     finds options settings right at the start of the pattern, it extracts them into
2527     the global options, as before. Thus, they show up in the info data.
2529     8. Added support for the \Q...\E escape sequence. Characters in between are
2530     treated as literals. This is slightly different from Perl in that $ and @ are
2531     also handled as literals inside the quotes. In Perl, they will cause variable
2532     interpolation. Note the following examples:
2534     Pattern PCRE matches Perl matches
2536     \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
2537     \Qabc\$xyz\E abc\$xyz abc\$xyz
2538     \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
2540     For compatibility with Perl, \Q...\E sequences are recognized inside character
2541     classes as well as outside them.
2543     9. Re-organized 3 code statements in pcretest to avoid "overflow in
2544     floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2545     (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2546     signed/unsigned warnings.
2548     10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2549     option for pcretest, so I've replaced it by a simple function that does just
2550     that job.
2552     11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2553     "pcregrep -".
2555     12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2556     Java package. This provides some syntactic sugar for simple cases of what my
2557     documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2558     as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2559     item, you can use this simplified notation. Note that only makes sense with
2560     greedy quantifiers. Consequently, the use of the possessive quantifier forces
2561     greediness, whatever the setting of the PCRE_UNGREEDY option.
2563     13. A change of greediness default within a pattern was not taking effect at
2564     the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2565     subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2566     was abstracted outside.
2568     14. PCRE now supports the \G assertion. It is true when the current matching
2569     position is at the start point of the match. This differs from \A when the
2570     starting offset is non-zero. Used with the /g option of pcretest (or similar
2571     code), it works in the same way as it does for Perl's /g option. If all
2572     alternatives of a regex begin with \G, the expression is anchored to the start
2573     match position, and the "anchored" flag is set in the compiled expression.
2575     15. Some bugs concerning the handling of certain option changes within patterns
2576     have been fixed. These applied to options other than (?ims). For example,
2577     "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2578     been the other way round. Some of this was related to change 7 above.
2580     16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2581     features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2582     and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2583     POSIX classes only within a class (e.g. /[[:alpha:]]/).
2585     17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2586     mode. Unlike ".", it always matches newline, whatever the setting of
2587     PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2588     assertions. Perl allows it, but it doesn't (in general) work because it can't
2589     calculate the length of the lookbehind. At least, that's the case for Perl
2590     5.8.0 - I've been told they are going to document that it doesn't work in
2591     future.
2593     18. Added an error diagnosis for escapes that PCRE does not support: these are
2594     \L, \l, \N, \P, \p, \U, \u, and \X.
2596     19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2597     reading past the end of the pattern in cases such as /[abcd/.
2599     20. PCRE was getting more memory than necessary for patterns with classes that
2600     contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2602     21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2603     compiling PCRE for use with Virtual Pascal.
2605     22. Small fix to the Makefile to make it work properly if the build is done
2606     outside the source tree.
2608     23. Added a new extension: a condition to go with recursion. If a conditional
2609     subpattern starts with (?(R) the "true" branch is used if recursion has
2610     happened, whereas the "false" branch is used only at the top level.
2612     24. When there was a very long string of literal characters (over 255 bytes
2613     without UTF support, over 250 bytes with UTF support), the computation of how
2614     much memory was required could be incorrect, leading to segfaults or other
2615     strange effects.
2617     25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2618     start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2619     there was a subsequent back reference to those brackets. This meant that, for
2620     example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2621     possible to check for precisely this case. All we can do is abandon the
2622     optimization if .* occurs inside capturing brackets when there are any back
2623     references whatsoever. (See below for a better fix that came later.)
2625     26. The handling of the optimization for finding the first character of a
2626     non-anchored pattern, and for finding a character that is required later in the
2627     match were failing in some cases. This didn't break the matching; it just
2628     failed to optimize when it could. The way this is done has been re-implemented.
2630     27. Fixed typo in error message for invalid (?R item (it said "(?p").
2632     28. Added a new feature that provides some of the functionality that Perl
2633     provides with (?{...}). The facility is termed a "callout". The way it is done
2634     in PCRE is for the caller to provide an optional function, by setting
2635     pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2636     global variable. By default it is unset, which disables all calling out. To get
2637     the function called, the regex must include (?C) at appropriate points. This
2638     is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2639     This provides a means of identifying different callout points. When PCRE
2640     reaches such a point in the regex, if pcre_callout has been set, the external
2641     function is called. It is provided with data in a structure called
2642     pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2643     matching continues; if it returns a non-zero value, the match at the current
2644     point fails. However, backtracking will occur if possible. [This was changed
2645     later and other features added - see item 49 below.]
2647     29. pcretest is upgraded to test the callout functionality. It provides a
2648     callout function that displays information. By default, it shows the start of
2649     the match and the current position in the text. There are some new data escapes
2650     to vary what happens:
2652     \C+ in addition, show current contents of captured substrings
2653     \C- do not supply a callout function
2654     \C!n return 1 when callout number n is reached
2655     \C!n!m return 1 when callout number n is reached for the mth time
2657     30. If pcregrep was called with the -l option and just a single file name, it
2658     output "<stdin>" if a match was found, instead of the file name.
2660     31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2661     slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2662     pcre_exec(). This saves a malloc/free per call. The default value of
2663     POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2664     when configuring.
2666     32. The default maximum size of a compiled pattern is 64K. There have been a
2667     few cases of people hitting this limit. The code now uses macros to handle the
2668     storing of links as offsets within the compiled pattern. It defaults to 2-byte
2669     links, but this can be changed to 3 or 4 bytes by --with-link-size when
2670     configuring. Tests 2 and 5 work only with 2-byte links because they output
2671     debugging information about compiled patterns.
2673     33. Internal code re-arrangements:
2675     (a) Moved the debugging function for printing out a compiled regex into
2676     its own source file (printint.c) and used #include to pull it into
2677     pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2678     separate copies.
2680     (b) Defined the list of op-code names for debugging as a macro in
2681     internal.h so that it is next to the definition of the opcodes.
2683     (c) Defined a table of op-code lengths for simpler skipping along compiled
2684     code. This is again a macro in internal.h so that it is next to the
2685     definition of the opcodes.
2687     34. Added support for recursive calls to individual subpatterns, along the
2688     lines of Robin Houston's patch (but implemented somewhat differently).
2690     35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2691     allow it to read and process whole directories in Win32. This code was
2692     contributed by Lionel Fourquaux; it has not been tested by me.
2694     36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2695     used to name a group. Names consist of alphanumerics and underscores, and must
2696     be unique. Back references use the syntax (?P=name) and recursive calls use
2697     (?P>name) which is a PCRE extension to the Python extension. Groups still have
2698     numbers. The function pcre_fullinfo() can be used after compilation to extract
2699     a name/number map. There are three relevant calls:
2701     PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
2702     PCRE_INFO_NAMECOUNT yields the number of entries
2703     PCRE_INFO_NAMETABLE yields a pointer to the map.
2705     The map is a vector of fixed-size entries. The size of each entry depends on
2706     the length of the longest name used. The first two bytes of each entry are the
2707     group number, most significant byte first. There follows the corresponding
2708     name, zero terminated. The names are in alphabetical order.
2710     37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2711     case instead of 255. Making it the same both with and without UTF-8 support
2712     means that the same test output works with both.
2714     38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2715     calling malloc() with a zero argument.
2717     39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2718     optimization. I've improved things by keeping a bitmap of backreferences with
2719     numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2720     fact referenced, the optimization can be applied. It is unlikely that a
2721     relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2722     the match to follow \n) will appear inside brackets with a number greater than
2723     31, but if it does, any back reference > 31 suppresses the optimization.
2725     40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2726     of disabling numbered capturing parentheses. Any opening parenthesis that is
2727     not followed by ? behaves as if it were followed by ?: but named parentheses
2728     can still be used for capturing (and they will acquire numbers in the usual
2729     way).
2731     41. Redesigned the return codes from the match() function into yes/no/error so
2732     that errors can be passed back from deep inside the nested calls. A malloc
2733     failure while inside a recursive subpattern call now causes the
2734     PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2736     42. It is now possible to set a limit on the number of times the match()
2737     function is called in a call to pcre_exec(). This facility makes it possible to
2738     limit the amount of recursion and backtracking, though not in a directly
2739     obvious way, because the match() function is used in a number of different
2740     circumstances. The count starts from zero for each position in the subject
2741     string (for non-anchored patterns). The default limit is, for compatibility, a
2742     large number, namely 10 000 000. You can change this in two ways:
2744     (a) When configuring PCRE before making, you can use --with-match-limit=n
2745     to set a default value for the compiled library.
2747     (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2748     a different value is set. See 45 below.
2750     If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2752     43. Added a new function pcre_config(int, void *) to enable run-time extraction
2753     of things that can be changed at compile time. The first argument specifies
2754     what is wanted and the second points to where the information is to be placed.
2755     The current list of available information is:
2759     The output is an integer that is set to one if UTF-8 support is available;
2760     otherwise it is set to zero.
2764     The output is an integer that it set to the value of the code that is used for
2765     newline. It is either LF (10) or CR (13).
2769     The output is an integer that contains the number of bytes used for internal
2770     linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2774     The output is an integer that contains the threshold above which the POSIX
2775     interface uses malloc() for output vectors. See item 31 above.
2779     The output is an unsigned integer that contains the default limit of the number
2780     of match() calls in a pcre_exec() execution. See 42 above.
2782     44. pcretest has been upgraded by the addition of the -C option. This causes it
2783     to extract all the available output from the new pcre_config() function, and to
2784     output it. The program then exits immediately.
2786     45. A need has arisen to pass over additional data with calls to pcre_exec() in
2787     order to support additional features. One way would have been to define
2788     pcre_exec2() (for example) with extra arguments, but this would not have been
2789     extensible, and would also have required all calls to the original function to
2790     be mapped to the new one. Instead, I have chosen to extend the mechanism that
2791     is used for passing in "extra" data from pcre_study().
2793     The pcre_extra structure is now exposed and defined in pcre.h. It currently
2794     contains the following fields:
2796     flags a bitmap indicating which of the following fields are set
2797     study_data opaque data from pcre_study()
2798     match_limit a way of specifying a limit on match() calls for a specific
2799     call to pcre_exec()
2800     callout_data data for callouts (see 49 below)
2802     The flag bits are also defined in pcre.h, and are
2808     The pcre_study() function now returns one of these new pcre_extra blocks, with
2809     the actual study data pointed to by the study_data field, and the
2810     PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2811     before. That is, this change is entirely upwards-compatible and requires no
2812     change to existing code.
2814     If you want to pass in additional data to pcre_exec(), you can either place it
2815     in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2816     block.
2818     46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2819     data string contains the escape sequence \M, pcretest calls pcre_exec() several
2820     times with different match limits, until it finds the minimum value needed for
2821     pcre_exec() to complete. The value is then output. This can be instructive; for
2822     most simple matches the number is quite small, but for pathological cases it
2823     gets very large very quickly.
2825     47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2826     returns the size of the data block pointed to by the study_data field in a
2827     pcre_extra block, that is, the value that was passed as the argument to
2828     pcre_malloc() when PCRE was getting memory in which to place the information
2829     created by pcre_study(). The fourth argument should point to a size_t variable.
2830     pcretest has been extended so that this information is shown after a successful
2831     pcre_study() call when information about the compiled regex is being displayed.
2833     48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2834     because what follows is always an absolute path. (Later: it turns out that this
2835     is more than cosmetic for MinGW, because it doesn't like empty path
2836     components.)
2838     49. Some changes have been made to the callout feature (see 28 above):
2840     (i) A callout function now has three choices for what it returns:
2842     0 => success, carry on matching
2843     > 0 => failure at this point, but backtrack if possible
2844     < 0 => serious error, return this value from pcre_exec()
2846     Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2847     values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2848     "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2849     use by callout functions. It will never be used by PCRE itself.
2851     (ii) The pcre_extra structure (see 45 above) has a void * field called
2852     callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2853     pcre_callout_block structure has a field of the same name. The contents of
2854     the field passed in the pcre_extra structure are passed to the callout
2855     function in the corresponding field in the callout block. This makes it
2856     easier to use the same callout-containing regex from multiple threads. For
2857     testing, the pcretest program has a new data escape
2859     \C*n pass the number n (may be negative) as callout_data
2861     If the callout function in pcretest receives a non-zero value as
2862     callout_data, it returns that value.
2864     50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2865     there were some redundant $(CFLAGS) in commands that are now specified as
2866     $(LINK), which already includes $(CFLAGS).
2868     51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2869     has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2870     with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2871     one-byte characters throughout. Note that case-insensitive matching applies
2872     only to characters whose values are less than 256. PCRE doesn't support the
2873     notion of cases for higher-valued characters.
2875     (i) A character class whose characters are all within 0-255 is handled as
2876     a bit map, and the map is inverted for negative classes. Previously, a
2877     character > 255 always failed to match such a class; however it should
2878     match if the class was a negative one (e.g. [^ab]). This has been fixed.
2880     (ii) A negated character class with a single character < 255 is coded as
2881     "not this character" (OP_NOT). This wasn't working properly when the test
2882     character was multibyte, either singly or repeated.
2884     (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2885     mode, for example: \x{100}{2,3}.
2887     (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2888     singly or repeated) now correctly test multibyte characters. However,
2889     PCRE doesn't recognize any characters with values greater than 255 as
2890     digits, spaces, or word characters. Such characters always match \D, \S,
2891     and \W, and never match \d, \s, or \w.
2893     (v) Classes may now contain characters and character ranges with values
2894     greater than 255. For example: [ab\x{100}-\x{400}].
2896     (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
2897     PCRE in UTF-8 mode.
2899     52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2900     PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2901     retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2902     value.)
2904     53. The single man page has become too large. I have therefore split it up into
2905     a number of separate man pages. These also give rise to individual HTML pages;
2906     these are now put in a separate directory, and there is an index.html page that
2907     lists them all. Some hyperlinking between the pages has been installed.
2909     54. Added convenience functions for handling named capturing parentheses.
2911     55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2912     aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2913     true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2914     are faulted.
2916     56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2917     calling configure. These values are used when compiling the dftables.c program
2918     which is run to generate the source of the default character tables. They
2919     default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2920     you will need to set these values.
2922     57. Updated the building process for Windows DLL, as provided by Fred Cox.
2925     Version 3.9 02-Jan-02
2926 nigel 61 ---------------------
2928     1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2930     2. If --disable-static was given, the building process failed when trying to
2931     build pcretest and pcregrep. (For some reason it was using libtool to compile
2932     them, which is not right, as they aren't part of the library.)
2935 nigel 59 Version 3.8 18-Dec-01
2936     ---------------------
2938     1. The experimental UTF-8 code was completely screwed up. It was packing the
2939     bytes in the wrong order. How dumb can you get?
2942 nigel 57 Version 3.7 29-Oct-01
2943     ---------------------
2945     1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2946     This caused pcretest, when used on the test data, to segfault. Unfortunately,
2947     this didn't happen under Solaris 8, where I normally test things.
2949 nigel 59 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2950     doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2951     isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2952     this fix an hour or so after the initial 3.7 release.)
2953 nigel 57
2954 nigel 59
2955 nigel 55 Version 3.6 23-Oct-01
2956     ---------------------
2958     1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2959     offsets passed as NULL with zero offset count.
2961     2. The config.guess and config.sub files had not been updated when I moved to
2962     the latest autoconf.
2965 nigel 53 Version 3.5 15-Aug-01
2966     ---------------------
2967 nigel 5
2968 nigel 53 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2969     had been forgotten.
2971     2. By using declared but undefined structures, we can avoid using "void"
2972     definitions in pcre.h while keeping the internal definitions of the structures
2973     private.
2975     3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2976     user point of view, this means that both static and shared libraries are built
2977     by default, but this can be individually controlled. More of the work of
2978     handling this static/shared cases is now inside libtool instead of PCRE's make
2979     file.
2981     4. The pcretest utility is now installed along with pcregrep because it is
2982     useful for users (to test regexs) and by doing this, it automatically gets
2983     relinked by libtool. The documentation has been turned into a man page, so
2984     there are now .1, .txt, and .html versions in /doc.
2986     5. Upgrades to pcregrep:
2987     (i) Added long-form option names like gnu grep.
2988     (ii) Added --help to list all options with an explanatory phrase.
2989     (iii) Added -r, --recursive to recurse into sub-directories.
2990     (iv) Added -f, --file to read patterns from a file.
2992     6. pcre_exec() was referring to its "code" argument before testing that
2993     argument for NULL (and giving an error if it was NULL).
2995     7. Upgraded Makefile.in to allow for compiling in a different directory from
2996     the source directory.
2998     8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2999     options bits, the pointer it was passed was to an int instead of to an unsigned
3000     long int. This mattered only on 64-bit systems.
3002     9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3003     generated) instead of pcre.in, which it its source. Also made the same change
3004     in several of the .c files.
3006     10. A new release of gcc defines printf() as a macro, which broke pcretest
3007     because it had an ifdef in the middle of a string argument for printf(). Fixed
3008     by using separate calls to printf().
3010     11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
3011     script, to force use of CR or LF instead of \n in the source. On non-Unix
3012     systems, the value can be set in config.h.
3014     12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
3015     absolute limit. Changed the text of the error message to make this clear, and
3016     likewise updated the man page.
3018     13. The limit of 99 on the number of capturing subpatterns has been removed.
3019     The new limit is 65535, which I hope will not be a "real" limit.
3022 nigel 51 Version 3.4 22-Aug-00
3023     ---------------------
3025     1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
3027     2. Diagnose condition (?(0) as an error instead of crashing on matching.
3030 nigel 49 Version 3.3 01-Aug-00
3031     ---------------------
3033     1. If an octal character was given, but the value was greater than \377, it
3034     was not getting masked to the least significant bits, as documented. This could
3035     lead to crashes in some systems.
3037     2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
3038     the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
3040     3. Added the functions pcre_free_substring() and pcre_free_substring_list().
3041     These just pass their arguments on to (pcre_free)(), but they are provided
3042     because some uses of PCRE bind it to non-C systems that can call its functions,
3043     but cannot call free() or pcre_free() directly.
3045     4. Add "make test" as a synonym for "make check". Corrected some comments in
3046     the Makefile.
3048     5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
3049     Makefile.
3051     6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
3052     command called pgrep for grepping around the active processes.
3054     7. Added the beginnings of support for UTF-8 character strings.
3056     8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
3057     RANLIB to ./ltconfig so that they are used by libtool. I think these are all
3058     the relevant ones. (AR is not passed because ./ltconfig does its own figuring
3059     out for the ar command.)
3062 nigel 47 Version 3.2 12-May-00
3063     ---------------------
3065     This is purely a bug fixing release.
3067     1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
3068     of ZA. This was just one example of several cases that could provoke this bug,
3069     which was introduced by change 9 of version 2.00. The code for breaking
3070     infinite loops after an iteration that matches an empty string was't working
3071     correctly.
3073     2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3074     when matched against abbab (for example). After matching an empty string, it
3075     wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3076     caused it to match further down the string than it should.
3078     3. The code contained an inclusion of sys/types.h. It isn't clear why this
3079     was there because it doesn't seem to be needed, and it causes trouble on some
3080     systems, as it is not a Standard C header. It has been removed.
3082     4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3083     were reported on the Macintosh. The changes were from
3085     while ((c = *(++ptr)) != 0 && c != '\n');
3086     to
3087     while ((c = *(++ptr)) != 0 && c != '\n') ;
3089     Totally extraordinary, but if that's what it takes...
3091     5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3092     available. Added HAVE_BCOPY and an autoconf test for it; if neither
3093     HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3094     assumes the way PCRE uses memmove() (always moving upwards).
3096     6. PCRE is being used in one environment where strchr() is not available. There
3097     was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3098     faster code anyway.
3101 nigel 45 Version 3.1 09-Feb-00
3102     ---------------------
3104     The only change in this release is the fixing of some bugs in Makefile.in for
3105     the "install" target:
3107     (1) It was failing to install pcreposix.h.
3109     (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3112 nigel 43 Version 3.0 01-Feb-00
3113     ---------------------
3114 nigel 41
3115     1. Add support for the /+ modifier to perltest (to output $` like it does in
3116     pcretest).
3118     2. Add support for the /g modifier to perltest.
3120     3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3121     matches null strings.
3123     4. Fix perltest so that it doesn't do unwanted things when fed an empty
3124     pattern. Perl treats empty patterns specially - it reuses the most recent
3125     pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3126     effect.
3128     5. The POSIX interface was broken in that it was just handing over the POSIX
3129     captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3130     required a bigger vector, with some working space on the end. This means that
3131     the POSIX wrapper now has to get and free some memory, and copy the results.
3133 nigel 43 6. Added some simple autoconf support, placing the test data and the
3134     documentation in separate directories, re-organizing some of the
3135     information files, and making it build pcre-config (a GNU standard). Also added
3136     libtool support for building PCRE as a shared library, which is now the
3137     default.
3138 nigel 41
3139 nigel 43 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3140     09 are not valid octal constants. Single digits will be used for minor values
3141     less than 10.
3143     8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3144     existing programs that set these in the POSIX interface can use PCRE without
3145     modification.
3147     9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3148     return all that pcre_info() returns, plus additional data. The pcre_info()
3149     function is retained for compatibility, but is considered to be obsolete.
3151     10. Added experimental recursion feature (?R) to handle one common case that
3152     Perl 5.6 will be able to do with (?p{...}).
3154     11. Added support for POSIX character classes like [:alpha:], which Perl is
3155     adopting.
3158 nigel 39 Version 2.08 31-Aug-99
3159     ----------------------
3161     1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3162     trying to match at the startoffset position, but instead was moving forward to
3163     the next newline as if a previous match had failed.
3165     2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3166     and could get into a loop if a null string was matched other than at the start
3167     of the subject.
3169     3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3170     be distinguished at compile time, and for completeness also added PCRE_DATE.
3172     5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3173     in GnuWin32 environments.
3176 nigel 37 Version 2.07 29-Jul-99
3177     ----------------------
3179     1. The documentation is now supplied in plain text form and HTML as well as in
3180     the form of man page sources.
3182     2. C++ compilers don't like assigning (void *) values to other pointer types.
3183     In particular this affects malloc(). Although there is no problem in Standard
3184     C, I've put in casts to keep C++ compilers happy.
3186     3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3187     should be (const char *).
3189     4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3190     be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3191     However, I haven't made this a standard facility. The documentation doesn't
3192     mention it, and the Makefile doesn't support it.
3194     5. The Makefile now contains an "install" target, with editable destinations at
3195     the top of the file. The pcretest program is not installed.
3197     6. pgrep -V now gives the PCRE version number and date.
3199     7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3200     causing the entire string to be ignored, instead of just the last character.
3202     8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3203     non-matching string, it can take a very, very long time, even for strings of
3204     quite modest length, because of the nested recursion. PCRE now does better in
3205     some of these cases. It does this by remembering the last required literal
3206     character in the pattern, and pre-searching the subject to ensure it is present
3207     before running the real match. In other words, it applies a heuristic to detect
3208     some types of certain failure quickly, and in the above example, if presented
3209     with a string that has no trailing " it gives "no match" very quickly.
3211     9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3212     other alternatives are tried instead.
3215 nigel 35 Version 2.06 09-Jun-99
3216     ----------------------
3218     1. Change pcretest's output for amount of store used to show just the code
3219     space, because the remainder (the data block) varies in size between 32-bit and
3220     64-bit systems.
3222     2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3223     start matching at. This allows lookbehinds to work when searching for multiple
3224     occurrences in a string.
3226     3. Added additional options to pcretest for testing multiple occurrences:
3228     /+ outputs the rest of the string that follows a match
3229     /g loops for multiple occurrences, using the new startoffset argument
3230     /G loops for multiple occurrences by passing an incremented pointer
3232     4. PCRE wasn't doing the "first character" optimization for patterns starting
3233     with \b or \B, though it was doing it for other lookbehind assertions. That is,
3234     it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3235     the letter 'x'. On long subject strings, this gives a significant speed-up.
3238 nigel 33 Version 2.05 21-Apr-99
3239     ----------------------
3241     1. Changed the type of magic_number from int to long int so that it works
3242     properly on 16-bit systems.
3244     2. Fixed a bug which caused patterns starting with .* not to work correctly
3245     when the subject string contained newline characters. PCRE was assuming
3246     anchoring for such patterns in all cases, which is not correct because .* will
3247     not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3248     DOTALL is set at top level; otherwise it knows that patterns starting with .*
3249     must be retried after every newline in the subject.
3252 nigel 31 Version 2.04 18-Feb-99
3253     ----------------------
3255     1. For parenthesized subpatterns with repeats whose minimum was zero, the
3256     computation of the store needed to hold the pattern was incorrect (too large).
3257     If such patterns were nested a few deep, this could multiply and become a real
3258     problem.
3260     2. Added /M option to pcretest to show the memory requirement of a specific
3261     pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3263     3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3264     compiled in such a way that the backtracking after subsequent failure was
3265     pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3266     ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3269 nigel 29 Version 2.03 02-Feb-99
3270     ----------------------
3272     1. Fixed typo and small mistake in man page.
3274 nigel 31 2. Added 4th condition (GPL supersedes if conflict) and created separate
3275     LICENCE file containing the conditions.
3276 nigel 29
3277     3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3278     Perl, that is the internal \ allows the delimiter to be included in the
3279     pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3280     the final delimiter, add \ to the end of the pattern (to test the error).
3282     4. Added the convenience functions for extracting substrings after a successful
3283     match. Updated pcretest to make it able to test these functions.
3286 nigel 27 Version 2.02 14-Jan-99
3287     ----------------------
3289     1. Initialized the working variables associated with each extraction so that
3290     their saving and restoring doesn't refer to uninitialized store.
3292     2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3293     compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3294     fix the problem.
3296     3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3297     calls, and wasn't printing the correct value for compiling calls. Increased the
3298     default value of LOOPREPEAT, and the number of significant figures in the
3299     times.
3301     4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3303     5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3304     a building problem on Windows NT with a FAT file system.
3307 nigel 25 Version 2.01 21-Oct-98
3308     ----------------------
3310     1. Changed the API for pcre_compile() to allow for the provision of a pointer
3311     to character tables built by pcre_maketables() in the current locale. If NULL
3312     is passed, the default tables are used.
3315 nigel 23 Version 2.00 24-Sep-98
3316 nigel 21 ----------------------
3318 nigel 23 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3319     it any more.
3321     2. Allow quantification of (?>) groups, and make it work correctly.
3323     3. The first character computation wasn't working for (?>) groups.
3325     4. Correct the implementation of \Z (it is permitted to match on the \n at the
3326     end of the subject) and add 5.005's \z, which really does match only at the
3327     very end of the subject.
3329     5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3331     6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3332     DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3333     localized options. All options to pcre_study() were also removed.
3335     7. Add other new features from 5.005:
3337     $(?<= positive lookbehind
3338     $(?<! negative lookbehind
3339     (?imsx-imsx) added the unsetting capability
3340     such a setting is global if at outer level; local otherwise
3341     (?imsx-imsx:) non-capturing groups with option setting
3342     (?(cond)re|re) conditional pattern matching
3344     A backreference to itself in a repeated group matches the previous
3345     captured string.
3347     8. General tidying up of studying (both automatic and via "study")
3348     consequential on the addition of new assertions.
3350     9. As in 5.005, unlimited repeated groups that could match an empty substring
3351     are no longer faulted at compile time. Instead, the loop is forcibly broken at
3352     runtime if any iteration does actually match an empty substring.
3354     10. Include the RunTest script in the distribution.
3356     11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3357     discrepancies, some of which were old and were also with respect to 5.004. They
3358     have now been fixed.
3361     Version 1.09 28-Apr-98
3362     ----------------------
3364 nigel 21 1. A negated single character class followed by a quantifier with a minimum
3365     value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
3366     program crashes, or just wrong answers. This did not apply to negated classes
3367     containing more than one character, or to minima other than one.
3370 nigel 19 Version 1.08 27-Mar-98
3371     ----------------------
3373     1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
3375     2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
3376     latter must appear before anything that relies on it in the pattern.
3379 nigel 17 Version 1.07 16-Feb-98
3380     ----------------------
3382     1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
3383     repeat of a potentially empty string).
3386 nigel 15 Version 1.06 23-Jan-98
3387     ----------------------
3389     1. Added Markus Oberhumer's little patches for C++.
3391     2. Literal strings longer than 255 characters were broken.
3394 nigel 13 Version 1.05 23-Dec-97
3395     ----------------------
3397     1. Negated character classes containing more than one character were failing if
3398     PCRE_CASELESS was set at run time.
3401 nigel 11 Version 1.04 19-Dec-97
3402     ----------------------
3404     1. Corrected the man page, where some "const" qualifiers had been omitted.
3406     2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
3407     input syntax.
3409     3. Fixed memory leak which occurred when a regex with back references was
3410     matched with an offsets vector that wasn't big enough. The temporary memory
3411     that is used in this case wasn't being freed if the match failed.
3413     4. Tidied pcretest to ensure it frees memory that it gets.
3415     5. Temporary memory was being obtained in the case where the passed offsets
3416     vector was exactly big enough.
3418     6. Corrected definition of offsetof() from change 5 below.
3420     7. I had screwed up change 6 below and broken the rules for the use of
3421     setjmp(). Now fixed.
3424 nigel 9 Version 1.03 18-Dec-97
3425     ----------------------
3427     1. A erroneous regex with a missing opening parenthesis was correctly
3428     diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
3429     on some systems.
3431     2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
3432     it was reported that one broken compiler failed on the former because "code" is
3433     also an independent variable.
3435     3. The erroneous regex a[]b caused an array overrun reference.
3437     4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
3438     fail on data ending with that character. (It was going on too far, and checking
3439     the next character, typically a binary zero.) This was specific to the
3440     optimized code for single-character negative classes.
3442     5. Added a contributed patch from the TIN world which does the following:
3444     + Add an undef for memmove, in case the the system defines a macro for it.
3446     + Add a definition of offsetof(), in case there isn't one. (I don't know
3447     the reason behind this - offsetof() is part of the ANSI standard - but
3448     it does no harm).
3450     + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
3451     most of the places where whitespace preceded '#'. I have given up and
3452     allowed the remaining 2 cases to be at the margin.
3454     + Rename some variables in pcre to eliminate shadowing. This seems very
3455     pedantic, but does no harm, of course.
3457     6. Moved the call to setjmp() into its own function, to get rid of warnings
3458     from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
3460     7. Constructs such as \d{8,} were compiling into the equivalent of
3461 nigel 11 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
3462 nigel 9 outcome, but in this particular case used more store than had been allocated,
3463     which caused the bug to be discovered because it threw up an internal error.
3465     8. The debugging code in both pcre and pcretest for outputting the compiled
3466     form of a regex was going wrong in the case of back references followed by
3467     curly-bracketed repeats.
3470 nigel 7 Version 1.02 12-Dec-97
3471     ----------------------
3473     1. Typos in pcre.3 and comments in the source fixed.
3475     2. Applied a contributed patch to get rid of places where it used to remove
3476     'const' from variables, and fixed some signed/unsigned and uninitialized
3477     variable warnings.
3479     3. Added the "runtest" target to Makefile.
3481     4. Set default compiler flag to -O2 rather than just -O.
3484 nigel 5 Version 1.01 19-Nov-97
3485     ----------------------
3487     1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
3488     like /([ab]*)*/, that is, for classes with more than one character in them.
3490     2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
3491     as /((?>a*))*/ (a PCRE_EXTRA facility).
3494     Version 1.00 18-Nov-97
3495     ----------------------
3497     1. Added compile-time macros to support systems such as SunOS4 which don't have
3498     memmove() or strerror() but have other things that can be used instead.
3500     2. Arranged that "make clean" removes the executables.
3503 nigel 3 Version 0.99 27-Oct-97
3504     ----------------------
3506     1. Fixed bug in code for optimizing classes with only one character. It was
3507     initializing a 32-byte map regardless, which could cause it to run off the end
3508     of the memory it had got.
3510     2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
3513     Version 0.98 22-Oct-97
3514     ----------------------
3516     1. Fixed bug in code for handling temporary memory usage when there are more
3517     back references than supplied space in the ovector. This could cause segfaults.
3520     Version 0.97 21-Oct-97
3521     ----------------------
3523     1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
3525     2. Optimized negated single characters not to use a bit map.
3527     3. Brought error texts together as macro definitions; clarified some of them;
3528     fixed one that was wrong - it said "range out of order" when it meant "invalid
3529     escape sequence".
3531     4. Changed some char * arguments to const char *.
3533     5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
3535     6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
3536     pcretest.
3539     Version 0.96 16-Oct-97
3540     ----------------------
3542     1. Added a simple "pgrep" utility to the distribution.
3544     2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
3545     unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
3546     where "ddd" means "one or more decimal digits".
3548     3. Fixed serious bug. If a pattern had a back reference, but the call to
3549     pcre_exec() didn't supply a large enough ovector to record the related
3550     identifying subpattern, the match always failed. PCRE now remembers the number
3551     of the largest back reference, and gets some temporary memory in which to save
3552     the offsets during matching if necessary, in order to ensure that
3553     backreferences always work.
3555     4. Increased the compatibility with Perl in a number of ways:
3557     (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
3558     to request this handling. The option can be set at compile or exec time.
3560     (b) $ matches before a terminating newline by default; an option
3561     PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
3562     mode). The option can be set at compile or exec time.
3564     (c) The handling of \ followed by a digit other than 0 is now supposed to be
3565     the same as Perl's. If the decimal number it represents is less than 10
3566     or there aren't that many previous left capturing parentheses, an octal
3567     escape is read. Inside a character class, it's always an octal escape,
3568     even if it is a single digit.
3570     (d) An escaped but undefined alphabetic character is taken as a literal,
3571     unless PCRE_EXTRA is set. Currently this just reserves the remaining
3572     escapes.
3574     (e) {0} is now permitted. (The previous item is removed from the compiled
3575     pattern).
3577     5. Changed all the names of code files so that the basic parts are no longer
3578     than 10 characters, and abolished the teeny "globals.c" file.
3580     6. Changed the handling of character classes; they are now done with a 32-byte
3581     bit map always.
3583     7. Added the -d and /D options to pcretest to make it possible to look at the
3584     internals of compilation without having to recompile pcre.
3587     Version 0.95 23-Sep-97
3588     ----------------------
3590     1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
3591     \x20 at the start of a run of normal characters. These were being treated as
3592     real characters, instead of the source characters being re-checked.
3595     Version 0.94 18-Sep-97
3596     ----------------------
3598     1. The functions are now thread-safe, with the caveat that the global variables
3599     containing pointers to malloc() and free() or alternative functions are the
3600     same for all threads.
3602     2. Get pcre_study() to generate a bitmap of initial characters for non-
3603     anchored patterns when this is possible, and use it if passed to pcre_exec().
3606     Version 0.93 15-Sep-97
3607     ----------------------
3609     1. /(b)|(:+)/ was computing an incorrect first character.
3611     2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
3612     but not actually doing anything yet.
3614     3. Treat "-" characters in classes that cannot be part of ranges as literals,
3615     as Perl does (e.g. [-az] or [az-]).
3617     4. Set the anchored flag if a branch starts with .* or .*? because that tests
3618     all possible positions.
3620     5. Split up into different modules to avoid including unneeded functions in a
3621     compiled binary. However, compile and exec are still in one module. The "study"
3622     function is split off.
3624     6. The character tables are now in a separate module whose source is generated
3625     by an auxiliary program - but can then be edited by hand if required. There are
3626     now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
3627     toupper() in the code.
3629     7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
3630     make them global. Abolish the function for setting them, as the caller can now
3631     set them directly.
3634     Version 0.92 11-Sep-97
3635     ----------------------
3637     1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
3638     (e.g. /a{1,3}/) was broken (I mis-optimized it).
3640     2. Caseless matching was not working in character classes if the characters in
3641     the pattern were in upper case.
3643     3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
3645     4. Make PCRE_ANCHORED public and accept as a compile option.
3647     5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
3648     PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
3649     pass them.
3651     6. Give an error if bad option bits passed at compile or run time.
3653     7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
3654     pcretest to cause it to pass that flag.
3656     8. Add pcre_info(), to get the number of identifying subpatterns, the stored
3657     options, and the first character, if set.
3659     9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
3662     Version 0.91 10-Sep-97
3663     ----------------------
3665     1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
3666     match the empty string as in /(a*)*/. It was looping and ultimately crashing.
3668     2. PCRE was looping on encountering an indefinitely repeated back reference to
3669     a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
3670     Perl does - treats the match as successful.
3672     ****


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

ViewVC Help
Powered by ViewVC 1.1.12