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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 644 - (show annotations) (download)
Sat Jul 30 17:13:00 2011 UTC (3 years, 4 months ago) by ph10
File size: 210970 byte(s)
Changes to pcregrep concerned with very long lines.

1 ChangeLog for PCRE
2 ------------------
3
4 Version 8.13 30-Apr-2011
5 ------------------------
6
7 1. The Unicode data tables have been updated to Unicode 6.0.0.
8
9 2. Two minor typos in pcre_internal.h have been fixed.
10
11 3. Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
12 pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
13 in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
14
15 4. There were a number of related bugs in the code for matching backrefences
16 caselessly in UTF-8 mode when codes for the characters concerned were
17 different numbers of bytes. For example, U+023A and U+2C65 are an upper
18 and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
19 (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
20 code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
21 2-byte code at the end of the subject (it thought there wasn't enough data
22 left).
23
24 5. Comprehensive information about what went wrong is now returned by
25 pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
26 as the output vector has at least 2 elements. The offset of the start of
27 the failing character and a reason code are placed in the vector.
28
29 6. When the UTF-8 string check fails for pcre_compile(), the offset that is
30 now returned is for the first byte of the failing character, instead of the
31 last byte inspected. This is an incompatible change, but I hope it is small
32 enough not to be a problem. It makes the returned offset consistent with
33 pcre_exec() and pcre_dfa_exec().
34
35 7. pcretest now gives a text phrase as well as the error number when
36 pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
37 failure, the offset and reason code are output.
38
39 8. When \R was used with a maximizing quantifier it failed to skip backwards
40 over a \r\n pair if the subsequent match failed. Instead, it just skipped
41 back over a single character (\n). This seems wrong (because it treated the
42 two characters as a single entity when going forwards), conflicts with the
43 documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
44 behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
45 has been changed.
46
47 9. Some internal refactoring has changed the processing so that the handling
48 of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
49 time (the PCRE_DOTALL option was changed this way some time ago: version
50 7.7 change 16). This has made it possible to abolish the OP_OPT op code,
51 which was always a bit of a fudge. It also means that there is one less
52 argument for the match() function, which reduces its stack requirements
53 slightly. This change also fixes an incompatibility with Perl: the pattern
54 (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
55
56 10. More internal refactoring has drastically reduced the number of recursive
57 calls to match() for possessively repeated groups such as (abc)++ when
58 using pcre_exec().
59
60 11. While implementing 10, a number of bugs in the handling of groups were
61 discovered and fixed:
62
63 (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
64 (a|)*(?1) gave a compile-time internal error.
65 ((a|)+)+ did not notice that the outer group could match an empty string.
66 (^a|^)+ was not marked as anchored.
67 (.*a|.*)+ was not marked as matching at start or after a newline.
68
69 12. Yet more internal refactoring has removed another argument from the match()
70 function. Special calls to this function are now indicated by setting a
71 value in a variable in the "match data" data block.
72
73 13. Be more explicit in pcre_study() instead of relying on "default" for
74 opcodes that mean there is no starting character; this means that when new
75 ones are added and accidentally left out of pcre_study(), testing should
76 pick them up.
77
78 14. The -s option of pcretest has been documented for ages as being an old
79 synonym of -m (show memory usage). I have changed it to mean "force study
80 for every regex", that is, assume /S for every regex. This is similar to -i
81 and -d etc. It's slightly incompatible, but I'm hoping nobody is still
82 using it. It makes it easier to run collections of tests with and without
83 study enabled, and thereby test pcre_study() more easily. All the standard
84 tests are now run with and without -s (but some patterns can be marked as
85 "never study" - see 20 below).
86
87 15. When (*ACCEPT) was used in a subpattern that was called recursively, the
88 restoration of the capturing data to the outer values was not happening
89 correctly.
90
91 16. If a recursively called subpattern ended with (*ACCEPT) and matched an
92 empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
93 pattern had matched an empty string, and so incorrectly returned a no
94 match.
95
96 17. There was optimizing code for the last branch of non-capturing parentheses,
97 and also for the obeyed branch of a conditional subexpression, which used
98 tail recursion to cut down on stack usage. Unfortunately, not that there is
99 the possibility of (*THEN) occurring in these branches, tail recursion is
100 no longer possible because the return has to be checked for (*THEN). These
101 two optimizations have therefore been removed.
102
103 18. If a pattern containing \R was studied, it was assumed that \R always
104 matched two bytes, thus causing the minimum subject length to be
105 incorrectly computed because \R can also match just one byte.
106
107 19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
108 was incorrectly computed.
109
110 20. If /S is present twice on a test pattern in pcretest input, it *disables*
111 studying, thereby overriding the use of -s on the command line. This is
112 necessary for one or two tests to keep the output identical in both cases.
113
114 21. When (*ACCEPT) was used in an assertion that matched an empty string and
115 PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
116
117 22. When an atomic group that contained a capturing parenthesis was
118 successfully matched, but the branch in which it appeared failed, the
119 capturing was not being forgotten if a higher numbered group was later
120 captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
121 group 1 to "a", when in fact it should be unset. This applied to multi-
122 branched capturing and non-capturing groups, repeated or not, and also to
123 positive assertions (capturing in negative assertions is not well defined
124 in PCRE) and also to nested atomic groups.
125
126 23. Add the ++ qualifier feature to pcretest, to show the remainder of the
127 subject after a captured substring (to make it easier to tell which of a
128 number of identical substrings has been captured).
129
130 24. The way atomic groups are processed by pcre_exec() has been changed so that
131 if they are repeated, backtracking one repetition now resets captured
132 values correctly. For example, if ((?>(a+)b)+aabab) is matched against
133 "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
134 "aaa". Previously, it would have been "a". As part of this code
135 refactoring, the way recursive calls are handled has also been changed.
136
137 24. If an assertion condition captured any substrings, they were not passed
138 back unless some other capturing happened later. For example, if
139 (?(?=(a))a) was matched against "a", no capturing was returned.
140
141 25. When studying a pattern that contained subroutine calls or assertions,
142 the code for finding the minimum length of a possible match was handling
143 direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
144 group 1 called group 2 while simultaneously a separate group 2 called group
145 1). A stack overflow occurred in this case. I have fixed this by limiting
146 the recursion depth to 10.
147
148 26. Updated RunTest.bat in the distribution to the version supplied by Tom
149 Fortmann. This supports explicit test numbers on the command line, and has
150 argument validation and error reporting.
151
152 27. An instance of \X with an unlimited repeat could fail if at any point the
153 first character it looked at was a mark character.
154
155 28. Some minor code refactoring concerning Unicode properties and scripts
156 should reduce the stack requirement of match() slightly.
157
158 29. Added the '=' option to pcretest to check the setting of unused capturing
159 slots at the end of the pattern, which are documented as being -1, but are
160 not included in the return count.
161
162 30. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
163 compiled something random. Now it gives a compile-time error (as does
164 Perl).
165
166 31. A *MARK encountered during the processing of a positive assertion is now
167 recorded and passed back (compatible with Perl).
168
169 32. If --only-matching or --colour was set on a pcregrep call whose pattern
170 had alternative anchored branches, the search for a second match in a line
171 was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
172 matched the line "0102" twice. The same bug affected patterns that started
173 with a backwards assertion. For example /\b01|\b02/ also matched "0102"
174 twice.
175
176 33. Previously, PCRE did not allow quantification of assertions. However, Perl
177 does, and because of capturing effects, quantifying parenthesized
178 assertions may at times be useful. Quantifiers are now allowed for
179 parenthesized assertions.
180
181 34. A minor code tidy in pcre_compile() when checking options for \R usage.
182
183 35. \g was being checked for fancy things in a character class, when it should
184 just be a literal "g".
185
186 36. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
187 appearance of a nested POSIX class supersedes an apparent external class.
188 For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
189 unescaped square brackets may also appear as part of class names. For
190 example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
191 more like Perl.
192
193 37. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
194 was because it thought it was \N{name}, which is not supported).
195
196 38. Add minix to OS list not supporting the -S option in pcretest.
197
198 39. PCRE tries to detect cases of infinite recursion at compile time, but it
199 cannot analyze patterns in sufficient detail to catch mutual recursions
200 such as ((?1))((?2)). There is now a runtime test that gives an error if a
201 subgroup is called recursively as a subpattern for a second time at the
202 same position in the subject string. In previous releases this might have
203 been caught by the recursion limit, or it might have run out of stack.
204
205 40. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
206 happen only once. PCRE was, however incorrectly giving a compile time error
207 "recursive call could loop indefinitely" because it cannot analyze the
208 pattern in sufficient detail. The compile time test no longer happens when
209 PCRE is compiling a conditional subpattern, but actual runaway loops are
210 now caught at runtime (see 39 above).
211
212 41. It seems that Perl allows any characters other than a closing parenthesis
213 to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
214 has been changed to be the same.
215
216 42. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
217 as not to get warnings when autogen.sh is called. Also changed
218 AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
219
220 43. To help people who use pcregrep to scan files containing exceedingly long
221 lines, the following changes have been made:
222
223 (a) The default value of the buffer size parameter has been increased from
224 8K to 20K. (A buffer three times this size is actually used.)
225
226 (b) The default can be changed by ./configure --with-pcregrep-bufsiz when
227 PCRE is built.
228
229 (c) A --buffer-size=n option has been added to pcregrep, to allow the size
230 to be set at run time.
231
232 (d) Numerical values in pcregrep options can be followed by K or M, for
233 example --buffer-size=50K.
234
235 (e) If a line being scanned overflows pcregrep's buffer, an error is now
236 given and the return code is set to 2.
237
238
239 Version 8.12 15-Jan-2011
240 ------------------------
241
242 1. Fixed some typos in the markup of the man pages, and wrote a script that
243 checks for such things as part of the documentation building process.
244
245 2. On a big-endian 64-bit system, pcregrep did not correctly process the
246 --match-limit and --recursion-limit options (added for 8.11). In
247 particular, this made one of the standard tests fail. (The integer value
248 went into the wrong half of a long int.)
249
250 3. If the --colour option was given to pcregrep with -v (invert match), it
251 did strange things, either producing crazy output, or crashing. It should,
252 of course, ignore a request for colour when reporting lines that do not
253 match.
254
255 4. Another pcregrep bug caused similar problems if --colour was specified with
256 -M (multiline) and the pattern match finished with a line ending.
257
258 5. In pcregrep, when a pattern that ended with a literal newline sequence was
259 matched in multiline mode, the following line was shown as part of the
260 match. This seems wrong, so I have changed it.
261
262 6. Another pcregrep bug in multiline mode, when --colour was specified, caused
263 the check for further matches in the same line (so they could be coloured)
264 to overrun the end of the current line. If another match was found, it was
265 incorrectly shown (and then shown again when found in the next line).
266
267 7. If pcregrep was compiled under Windows, there was a reference to the
268 function pcregrep_exit() before it was defined. I am assuming this was
269 the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
270 reported by a user. I've moved the definition above the reference.
271
272
273 Version 8.11 10-Dec-2010
274 ------------------------
275
276 1. (*THEN) was not working properly if there were untried alternatives prior
277 to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
278 backtracked to try for "b" instead of moving to the next alternative branch
279 at the same level (in this case, to look for "c"). The Perl documentation
280 is clear that when (*THEN) is backtracked onto, it goes to the "next
281 alternative in the innermost enclosing group".
282
283 2. (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
284 such as (A(*COMMIT)B(*THEN)C|D) any failure after matching A should
285 result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
286 (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
287 (*THEN).
288
289 3. If \s appeared in a character class, it removed the VT character from
290 the class, even if it had been included by some previous item, for example
291 in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
292 of \s, but is part of the POSIX "space" class.)
293
294 4. A partial match never returns an empty string (because you can always
295 match an empty string at the end of the subject); however the checking for
296 an empty string was starting at the "start of match" point. This has been
297 changed to the "earliest inspected character" point, because the returned
298 data for a partial match starts at this character. This means that, for
299 example, /(?<=abc)def/ gives a partial match for the subject "abc"
300 (previously it gave "no match").
301
302 5. Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
303 of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
304 previously a full match would be given. However, setting PCRE_PARTIAL_HARD
305 has an implication that the given string is incomplete (because a partial
306 match is preferred over a full match). For this reason, these items now
307 give a partial match in this situation. [Aside: previously, the one case
308 /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
309 match rather than a full match, which was wrong by the old rules, but is
310 now correct.]
311
312 6. There was a bug in the handling of #-introduced comments, recognized when
313 PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
314 If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
315 UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
316 scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
317 but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
318 places in pcre_compile().
319
320 7. Related to (6) above, when pcre_compile() was skipping #-introduced
321 comments when looking ahead for named forward references to subpatterns,
322 the only newline sequence it recognized was NL. It now handles newlines
323 according to the set newline convention.
324
325 8. SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
326 former, but used strtoul(), whereas pcretest avoided strtoul() but did not
327 cater for a lack of strerror(). These oversights have been fixed.
328
329 9. Added --match-limit and --recursion-limit to pcregrep.
330
331 10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
332
333 11. When the -o option was used, pcregrep was setting a return code of 1, even
334 when matches were found, and --line-buffered was not being honoured.
335
336 12. Added an optional parentheses number to the -o and --only-matching options
337 of pcregrep.
338
339 13. Imitating Perl's /g action for multiple matches is tricky when the pattern
340 can match an empty string. The code to do it in pcretest and pcredemo
341 needed fixing:
342
343 (a) When the newline convention was "crlf", pcretest got it wrong, skipping
344 only one byte after an empty string match just before CRLF (this case
345 just got forgotten; "any" and "anycrlf" were OK).
346
347 (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
348 mode when an empty string match preceded an ASCII character followed by
349 a non-ASCII character. (The code for advancing by one character rather
350 than one byte was nonsense.)
351
352 (c) The pcredemo.c sample program did not have any code at all to handle
353 the cases when CRLF is a valid newline sequence.
354
355 14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
356 as a starting offset was within the subject string. There is now a new
357 error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
358 negative or greater than the length of the string. In order to test this,
359 pcretest is extended to allow the setting of negative starting offsets.
360
361 15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
362 starting offset points to the beginning of a UTF-8 character was
363 unnecessarily clumsy. I tidied it up.
364
365 16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
366 bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
367
368 17. Nobody had reported that the --include_dir option, which was added in
369 release 7.7 should have been called --include-dir (hyphen, not underscore)
370 for compatibility with GNU grep. I have changed it to --include-dir, but
371 left --include_dir as an undocumented synonym, and the same for
372 --exclude-dir, though that is not available in GNU grep, at least as of
373 release 2.5.4.
374
375 18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
376 characters from a string of bytes) have been redefined so as not to use
377 loops, in order to improve performance in some environments. At the same
378 time, I abstracted some of the common code into auxiliary macros to save
379 repetition (this should not affect the compiled code).
380
381 19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
382 compile-time error is now given if \c is not followed by an ASCII
383 character, that is, a byte less than 128. (In EBCDIC mode, the code is
384 different, and any byte value is allowed.)
385
386 20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
387 START_OPTIMIZE option, which is now allowed at compile time - but just
388 passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
389 to pcregrep and other applications that have no direct access to PCRE
390 options. The new /Y option in pcretest sets this option when calling
391 pcre_compile().
392
393 21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
394 back references. Groups containing recursive back references were forced to
395 be atomic by that change, but in the case of named groups, the amount of
396 memory required was incorrectly computed, leading to "Failed: internal
397 error: code overflow". This has been fixed.
398
399 22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
400 pcretest.c, to avoid build problems in some Borland environments.
401
402
403 Version 8.10 25-Jun-2010
404 ------------------------
405
406 1. Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
407 THEN.
408
409 2. (*ACCEPT) was not working when inside an atomic group.
410
411 3. Inside a character class, \B is treated as a literal by default, but
412 faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
413 causes the error). The code is unchanged, but I tidied the documentation.
414
415 4. Inside a character class, PCRE always treated \R and \X as literals,
416 whereas Perl faults them if its -w option is set. I have changed PCRE so
417 that it faults them when PCRE_EXTRA is set.
418
419 5. Added support for \N, which always matches any character other than
420 newline. (It is the same as "." when PCRE_DOTALL is not set.)
421
422 6. When compiling pcregrep with newer versions of gcc which may have
423 FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
424 declared with attribute warn_unused_result" were given. Just casting the
425 result to (void) does not stop the warnings; a more elaborate fudge is
426 needed. I've used a macro to implement this.
427
428 7. Minor change to pcretest.c to avoid a compiler warning.
429
430 8. Added four artifical Unicode properties to help with an option to make
431 \s etc use properties (see next item). The new properties are: Xan
432 (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
433
434 9. Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
435 use Unicode properties. (*UCP) at the start of a pattern can be used to set
436 this option. Modified pcretest to add /W to test this facility. Added
437 REG_UCP to make it available via the POSIX interface.
438
439 10. Added --line-buffered to pcregrep.
440
441 11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
442 studied, and the match started with a letter with a code point greater than
443 127 whose first byte was different to the first byte of the other case of
444 the letter, the other case of this starting letter was not recognized
445 (#976).
446
447 12. If a pattern that was studied started with a repeated Unicode property
448 test, for example, \p{Nd}+, there was the theoretical possibility of
449 setting up an incorrect bitmap of starting bytes, but fortunately it could
450 not have actually happened in practice until change 8 above was made (it
451 added property types that matched character-matching opcodes).
452
453 13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
454 possible starting bytes for non-anchored patterns.
455
456 14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
457 \R, and also a number of cases that involve Unicode properties, both
458 explicit and implicit when PCRE_UCP is set.
459
460 15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
461 input, it could crash or give wrong results if characters with values
462 greater than 0xc0 were present in the subject string. (Detail: it assumed
463 UTF-8 input when processing these items.)
464
465 16. Added a lot of (int) casts to avoid compiler warnings in systems where
466 size_t is 64-bit (#991).
467
468 17. Added a check for running out of memory when PCRE is compiled with
469 --disable-stack-for-recursion (#990).
470
471 18. If the last data line in a file for pcretest does not have a newline on
472 the end, a newline was missing in the output.
473
474 19. The default pcre_chartables.c file recognizes only ASCII characters (values
475 less than 128) in its various bitmaps. However, there is a facility for
476 generating tables according to the current locale when PCRE is compiled. It
477 turns out that in some environments, 0x85 and 0xa0, which are Unicode space
478 characters, are recognized by isspace() and therefore were getting set in
479 these tables, and indeed these tables seem to approximate to ISO 8859. This
480 caused a problem in UTF-8 mode when pcre_study() was used to create a list
481 of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
482 which of course cannot start UTF-8 characters. I have changed the code so
483 that only real ASCII characters (less than 128) and the correct starting
484 bytes for UTF-8 encodings are set for characters greater than 127 when in
485 UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
486 altogether.)
487
488 20. Added the /T option to pcretest so as to be able to run tests with non-
489 standard character tables, thus making it possible to include the tests
490 used for 19 above in the standard set of tests.
491
492 21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
493 reference to a subpattern the other side of a comment that contains an
494 opening parenthesis caused either an internal compiling error, or a
495 reference to the wrong subpattern.
496
497
498 Version 8.02 19-Mar-2010
499 ------------------------
500
501 1. The Unicode data tables have been updated to Unicode 5.2.0.
502
503 2. Added the option --libs-cpp to pcre-config, but only when C++ support is
504 configured.
505
506 3. Updated the licensing terms in the pcregexp.pas file, as agreed with the
507 original author of that file, following a query about its status.
508
509 4. On systems that do not have stdint.h (e.g. Solaris), check for and include
510 inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
511
512 5. A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
513 quantifier applied to a forward-referencing subroutine call, could compile
514 incorrect code or give the error "internal error: previously-checked
515 referenced subpattern not found".
516
517 6. Both MS Visual Studio and Symbian OS have problems with initializing
518 variables to point to external functions. For these systems, therefore,
519 pcre_malloc etc. are now initialized to local functions that call the
520 relevant global functions.
521
522 7. There were two entries missing in the vectors called coptable and poptable
523 in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
524 I've fixed the data, and added a kludgy way of testing at compile time that
525 the lengths are correct (equal to the number of opcodes).
526
527 8. Following on from 7, I added a similar kludge to check the length of the
528 eint vector in pcreposix.c.
529
530 9. Error texts for pcre_compile() are held as one long string to avoid too
531 much relocation at load time. To find a text, the string is searched,
532 counting zeros. There was no check for running off the end of the string,
533 which could happen if a new error number was added without updating the
534 string.
535
536 10. \K gave a compile-time error if it appeared in a lookbehind assersion.
537
538 11. \K was not working if it appeared in an atomic group or in a group that
539 was called as a "subroutine", or in an assertion. Perl 5.11 documents that
540 \K is "not well defined" if used in an assertion. PCRE now accepts it if
541 the assertion is positive, but not if it is negative.
542
543 12. Change 11 fortuitously reduced the size of the stack frame used in the
544 "match()" function of pcre_exec.c by one pointer. Forthcoming
545 implementation of support for (*MARK) will need an extra pointer on the
546 stack; I have reserved it now, so that the stack frame size does not
547 decrease.
548
549 13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
550 item in branch that calls a recursion is a subroutine call - as in the
551 second branch in the above example - was incorrectly given the compile-
552 time error "recursive call could loop indefinitely" because pcre_compile()
553 was not correctly checking the subroutine for matching a non-empty string.
554
555 14. The checks for overrunning compiling workspace could trigger after an
556 overrun had occurred. This is a "should never occur" error, but it can be
557 triggered by pathological patterns such as hundreds of nested parentheses.
558 The checks now trigger 100 bytes before the end of the workspace.
559
560 15. Fix typo in configure.ac: "srtoq" should be "strtoq".
561
562
563 Version 8.01 19-Jan-2010
564 ------------------------
565
566 1. If a pattern contained a conditional subpattern with only one branch (in
567 particular, this includes all (*DEFINE) patterns), a call to pcre_study()
568 computed the wrong minimum data length (which is of course zero for such
569 subpatterns). This could cause incorrect "no match" results.
570
571 2. For patterns such as (?i)a(?-i)b|c where an option setting at the start of
572 the pattern is reset in the first branch, pcre_compile() failed with
573 "internal error: code overflow at offset...". This happened only when
574 the reset was to the original external option setting. (An optimization
575 abstracts leading options settings into an external setting, which was the
576 cause of this.)
577
578 3. A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
579 of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
580 assertion pattern did not match (meaning that the assertion was true), it
581 was incorrectly treated as false if the SKIP had been reached during the
582 matching. This also applied to assertions used as conditions.
583
584 4. If an item that is not supported by pcre_dfa_exec() was encountered in an
585 assertion subpattern, including such a pattern used as a condition,
586 unpredictable results occurred, instead of the error return
587 PCRE_ERROR_DFA_UITEM.
588
589 5. The C++ GlobalReplace function was not working like Perl for the special
590 situation when an empty string is matched. It now does the fancy magic
591 stuff that is necessary.
592
593 6. In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
594 removed. (These were left over from very, very early versions of PCRE.)
595
596 7. Some cosmetic changes to the code to make life easier when compiling it
597 as part of something else:
598
599 (a) Change DEBUG to PCRE_DEBUG.
600
601 (b) In pcre_compile(), rename the member of the "branch_chain" structure
602 called "current" as "current_branch", to prevent a collision with the
603 Linux macro when compiled as a kernel module.
604
605 (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
606 prevent a collision with the Linux macro when compiled as a kernel
607 module.
608
609 8. In pcre_compile() there are some checks for integer overflows that used to
610 cast potentially large values to (double). This has been changed to that
611 when building, a check for int64_t is made, and if it is found, it is used
612 instead, thus avoiding the use of floating point arithmetic. (There is no
613 other use of FP in PCRE.) If int64_t is not found, the fallback is to
614 double.
615
616 9. Added two casts to avoid signed/unsigned warnings from VS Studio Express
617 2005 (difference between two addresses compared to an unsigned value).
618
619 10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
620 custom one, because of the following reported problem in Windows:
621
622 - libbz2 uses the Pascal calling convention (WINAPI) for the functions
623 under Win32.
624 - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
625 therefore missing the function definition.
626 - The compiler thus generates a "C" signature for the test function.
627 - The linker fails to find the "C" function.
628 - PCRE fails to configure if asked to do so against libbz2.
629
630 11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
631 messages were output:
632
633 Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
634 rerunning libtoolize, to keep the correct libtool macros in-tree.
635 Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
636
637 I have done both of these things.
638
639 12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
640 most of the time, it *can* run out if it is given a pattern that contains a
641 runaway infinite recursion. I updated the discussion in the pcrestack man
642 page.
643
644 13. Now that we have gone to the x.xx style of version numbers, the minor
645 version may start with zero. Using 08 or 09 is a bad idea because users
646 might check the value of PCRE_MINOR in their code, and 08 or 09 may be
647 interpreted as invalid octal numbers. I've updated the previous comment in
648 configure.ac, and also added a check that gives an error if 08 or 09 are
649 used.
650
651 14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
652 causing partial matching to fail when the end of the subject matched \W
653 in a UTF-8 pattern where \W was quantified with a minimum of 3.
654
655 15. There were some discrepancies between the declarations in pcre_internal.h
656 of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
657 their definitions. The declarations used "const uschar *" and the
658 definitions used USPTR. Even though USPTR is normally defined as "const
659 unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
660 reported that: "This difference in casting confuses some C++ compilers, for
661 example, SunCC recognizes above declarations as different functions and
662 generates broken code for hbpcre." I have changed the declarations to use
663 USPTR.
664
665 16. GNU libtool is named differently on some systems. The autogen.sh script now
666 tries several variants such as glibtoolize (MacOSX) and libtoolize1x
667 (FreeBSD).
668
669 17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
670 (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
671 comment: "Figure out how to create a longlong from a string: strtoll and
672 equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
673 instance, but it only takes 2 args instead of 3!"
674
675 18. A subtle bug concerned with back references has been fixed by a change of
676 specification, with a corresponding code fix. A pattern such as
677 ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
678 refers, was giving matches when it shouldn't. For example, xa=xaaa would
679 match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
680 same bug. Such groups have to be quantified to be useful, or contained
681 inside another quantified group. (If there's no repetition, the reference
682 can never match.) The problem arises because, having left the group and
683 moved on to the rest of the pattern, a later failure that backtracks into
684 the group uses the captured value from the final iteration of the group
685 rather than the correct earlier one. I have fixed this in PCRE by forcing
686 any group that contains a reference to itself to be an atomic group; that
687 is, there cannot be any backtracking into it once it has completed. This is
688 similar to recursive and subroutine calls.
689
690
691 Version 8.00 19-Oct-09
692 ----------------------
693
694 1. The table for translating pcre_compile() error codes into POSIX error codes
695 was out-of-date, and there was no check on the pcre_compile() error code
696 being within the table. This could lead to an OK return being given in
697 error.
698
699 2. Changed the call to open a subject file in pcregrep from fopen(pathname,
700 "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
701 in a Windows environment.
702
703 3. The pcregrep --count option prints the count for each file even when it is
704 zero, as does GNU grep. However, pcregrep was also printing all files when
705 --files-with-matches was added. Now, when both options are given, it prints
706 counts only for those files that have at least one match. (GNU grep just
707 prints the file name in this circumstance, but including the count seems
708 more useful - otherwise, why use --count?) Also ensured that the
709 combination -clh just lists non-zero counts, with no names.
710
711 4. The long form of the pcregrep -F option was incorrectly implemented as
712 --fixed_strings instead of --fixed-strings. This is an incompatible change,
713 but it seems right to fix it, and I didn't think it was worth preserving
714 the old behaviour.
715
716 5. The command line items --regex=pattern and --regexp=pattern were not
717 recognized by pcregrep, which required --regex pattern or --regexp pattern
718 (with a space rather than an '='). The man page documented the '=' forms,
719 which are compatible with GNU grep; these now work.
720
721 6. No libpcreposix.pc file was created for pkg-config; there was just
722 libpcre.pc and libpcrecpp.pc. The omission has been rectified.
723
724 7. Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
725 when UCP support is not needed, by modifying the Python script that
726 generates it from Unicode data files. This should not matter if the module
727 is correctly used as a library, but I received one complaint about 50K of
728 unwanted data. My guess is that the person linked everything into his
729 program rather than using a library. Anyway, it does no harm.
730
731 8. A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
732 was a minimum greater than 1 for a wide character in a possessive
733 repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
734 which had an unlimited repeat of a nested, fixed maximum repeat of a wide
735 character. Chaos in the form of incorrect output or a compiling loop could
736 result.
737
738 9. The restrictions on what a pattern can contain when partial matching is
739 requested for pcre_exec() have been removed. All patterns can now be
740 partially matched by this function. In addition, if there are at least two
741 slots in the offset vector, the offset of the earliest inspected character
742 for the match and the offset of the end of the subject are set in them when
743 PCRE_ERROR_PARTIAL is returned.
744
745 10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
746 synonymous with PCRE_PARTIAL, for backwards compatibility, and
747 PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
748 and may be more useful for multi-segment matching.
749
750 11. Partial matching with pcre_exec() is now more intuitive. A partial match
751 used to be given if ever the end of the subject was reached; now it is
752 given only if matching could not proceed because another character was
753 needed. This makes a difference in some odd cases such as Z(*FAIL) with the
754 string "Z", which now yields "no match" instead of "partial match". In the
755 case of pcre_dfa_exec(), "no match" is given if every matching path for the
756 final character ended with (*FAIL).
757
758 12. Restarting a match using pcre_dfa_exec() after a partial match did not work
759 if the pattern had a "must contain" character that was already found in the
760 earlier partial match, unless partial matching was again requested. For
761 example, with the pattern /dog.(body)?/, the "must contain" character is
762 "g". If the first part-match was for the string "dog", restarting with
763 "sbody" failed. This bug has been fixed.
764
765 13. The string returned by pcre_dfa_exec() after a partial match has been
766 changed so that it starts at the first inspected character rather than the
767 first character of the match. This makes a difference only if the pattern
768 starts with a lookbehind assertion or \b or \B (\K is not supported by
769 pcre_dfa_exec()). It's an incompatible change, but it makes the two
770 matching functions compatible, and I think it's the right thing to do.
771
772 14. Added a pcredemo man page, created automatically from the pcredemo.c file,
773 so that the demonstration program is easily available in environments where
774 PCRE has not been installed from source.
775
776 15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
777 libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
778 library.
779
780 16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
781 It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
782 is not the first non-POSIX option to be added. Clearly some people find
783 these options useful.
784
785 17. If a caller to the POSIX matching function regexec() passes a non-zero
786 value for nmatch with a NULL value for pmatch, the value of
787 nmatch is forced to zero.
788
789 18. RunGrepTest did not have a test for the availability of the -u option of
790 the diff command, as RunTest does. It now checks in the same way as
791 RunTest, and also checks for the -b option.
792
793 19. If an odd number of negated classes containing just a single character
794 interposed, within parentheses, between a forward reference to a named
795 subpattern and the definition of the subpattern, compilation crashed with
796 an internal error, complaining that it could not find the referenced
797 subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
798 [The bug was that it was starting one character too far in when skipping
799 over the character class, thus treating the ] as data rather than
800 terminating the class. This meant it could skip too much.]
801
802 20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
803 /g option in pcretest when the pattern contains \K, which makes it possible
804 to have an empty string match not at the start, even when the pattern is
805 anchored. Updated pcretest and pcredemo to use this option.
806
807 21. If the maximum number of capturing subpatterns in a recursion was greater
808 than the maximum at the outer level, the higher number was returned, but
809 with unset values at the outer level. The correct (outer level) value is
810 now given.
811
812 22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
813 PCRE did not set those parentheses (unlike Perl). I have now found a way to
814 make it do so. The string so far is captured, making this feature
815 compatible with Perl.
816
817 23. The tests have been re-organized, adding tests 11 and 12, to make it
818 possible to check the Perl 5.10 features against Perl 5.10.
819
820 24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
821 pattern matches a fixed length string. PCRE did not allow this; now it
822 does. Neither allows recursion.
823
824 25. I finally figured out how to implement a request to provide the minimum
825 length of subject string that was needed in order to match a given pattern.
826 (It was back references and recursion that I had previously got hung up
827 on.) This code has now been added to pcre_study(); it finds a lower bound
828 to the length of subject needed. It is not necessarily the greatest lower
829 bound, but using it to avoid searching strings that are too short does give
830 some useful speed-ups. The value is available to calling programs via
831 pcre_fullinfo().
832
833 26. While implementing 25, I discovered to my embarrassment that pcretest had
834 not been passing the result of pcre_study() to pcre_dfa_exec(), so the
835 study optimizations had never been tested with that matching function.
836 Oops. What is worse, even when it was passed study data, there was a bug in
837 pcre_dfa_exec() that meant it never actually used it. Double oops. There
838 were also very few tests of studied patterns with pcre_dfa_exec().
839
840 27. If (?| is used to create subpatterns with duplicate numbers, they are now
841 allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
842 on the other side of the coin, they are no longer allowed to have different
843 names, because these cannot be distinguished in PCRE, and this has caused
844 confusion. (This is a difference from Perl.)
845
846 28. When duplicate subpattern names are present (necessarily with different
847 numbers, as required by 27 above), and a test is made by name in a
848 conditional pattern, either for a subpattern having been matched, or for
849 recursion in such a pattern, all the associated numbered subpatterns are
850 tested, and the overall condition is true if the condition is true for any
851 one of them. This is the way Perl works, and is also more like the way
852 testing by number works.
853
854
855 Version 7.9 11-Apr-09
856 ---------------------
857
858 1. When building with support for bzlib/zlib (pcregrep) and/or readline
859 (pcretest), all targets were linked against these libraries. This included
860 libpcre, libpcreposix, and libpcrecpp, even though they do not use these
861 libraries. This caused unwanted dependencies to be created. This problem
862 has been fixed, and now only pcregrep is linked with bzlib/zlib and only
863 pcretest is linked with readline.
864
865 2. The "typedef int BOOL" in pcre_internal.h that was included inside the
866 "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
867 moved outside it again, because FALSE and TRUE are already defined in AIX,
868 but BOOL is not.
869
870 3. The pcre_config() function was treating the PCRE_MATCH_LIMIT and
871 PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
872
873 4. The pcregrep documentation said spaces were inserted as well as colons (or
874 hyphens) following file names and line numbers when outputting matching
875 lines. This is not true; no spaces are inserted. I have also clarified the
876 wording for the --colour (or --color) option.
877
878 5. In pcregrep, when --colour was used with -o, the list of matching strings
879 was not coloured; this is different to GNU grep, so I have changed it to be
880 the same.
881
882 6. When --colo(u)r was used in pcregrep, only the first matching substring in
883 each matching line was coloured. Now it goes on to look for further matches
884 of any of the test patterns, which is the same behaviour as GNU grep.
885
886 7. A pattern that could match an empty string could cause pcregrep to loop; it
887 doesn't make sense to accept an empty string match in pcregrep, so I have
888 locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
889 seems to be how GNU grep behaves.
890
891 8. The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
892 start or after a newline", because the conditional assertion was not being
893 correctly handled. The rule now is that both the assertion and what follows
894 in the first alternative must satisfy the test.
895
896 9. If auto-callout was enabled in a pattern with a conditional group whose
897 condition was an assertion, PCRE could crash during matching, both with
898 pcre_exec() and pcre_dfa_exec().
899
900 10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
901 used for matching.
902
903 11. Unicode property support in character classes was not working for
904 characters (bytes) greater than 127 when not in UTF-8 mode.
905
906 12. Added the -M command line option to pcretest.
907
908 14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
909
910 15. Added the PCRE_NO_START_OPTIMIZE match-time option.
911
912 16. Added comments and documentation about mis-use of no_arg in the C++
913 wrapper.
914
915 17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
916 from Martin Jerabek that uses macro names for all relevant character and
917 string constants.
918
919 18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
920 SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
921 SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
922 these, but not everybody uses configure.
923
924 19. A conditional group that had only one branch was not being correctly
925 recognized as an item that could match an empty string. This meant that an
926 enclosing group might also not be so recognized, causing infinite looping
927 (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
928 with the subject "ab", where knowledge that the repeated group can match
929 nothing is needed in order to break the loop.
930
931 20. If a pattern that was compiled with callouts was matched using pcre_dfa_
932 exec(), but without supplying a callout function, matching went wrong.
933
934 21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
935 leak if the size of the offset vector was greater than 30. When the vector
936 is smaller, the saved offsets during recursion go onto a local stack
937 vector, but for larger vectors malloc() is used. It was failing to free
938 when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
939 error, in fact).
940
941 22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
942 heapframe that is used only when UTF-8 support is enabled. This caused no
943 problem, but was untidy.
944
945 23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
946 CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
947 included within another project.
948
949 24. Steven Van Ingelgem's patches to add more options to the CMake support,
950 slightly modified by me:
951
952 (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
953 not building pcregrep.
954
955 (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
956 if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
957
958 25. Forward references, both numeric and by name, in patterns that made use of
959 duplicate group numbers, could behave incorrectly or give incorrect errors,
960 because when scanning forward to find the reference group, PCRE was not
961 taking into account the duplicate group numbers. A pattern such as
962 ^X(?3)(a)(?|(b)|(q))(Y) is an example.
963
964 26. Changed a few more instances of "const unsigned char *" to USPTR, making
965 the feature of a custom pointer more persuasive (as requested by a user).
966
967 27. Wrapped the definitions of fileno and isatty for Windows, which appear in
968 pcretest.c, inside #ifndefs, because it seems they are sometimes already
969 pre-defined.
970
971 28. Added support for (*UTF8) at the start of a pattern.
972
973 29. Arrange for flags added by the "release type" setting in CMake to be shown
974 in the configuration summary.
975
976
977 Version 7.8 05-Sep-08
978 ---------------------
979
980 1. Replaced UCP searching code with optimized version as implemented for Ad
981 Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
982 stage table and inline lookup instead of a function, giving speed ups of 2
983 to 5 times on some simple patterns that I tested. Permission was given to
984 distribute the MultiStage2.py script that generates the tables (it's not in
985 the tarball, but is in the Subversion repository).
986
987 2. Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
988 scripts.
989
990 3. Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
991 a group with a zero qualifier. The result of the study could be incorrect,
992 or the function might crash, depending on the pattern.
993
994 4. Caseless matching was not working for non-ASCII characters in back
995 references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
996 It now works when Unicode Property Support is available.
997
998 5. In pcretest, an escape such as \x{de} in the data was always generating
999 a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1000 non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1001 truncation.
1002
1003 6. Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1004
1005 7. Added two (int) casts to pcregrep when printing the difference of two
1006 pointers, in case they are 64-bit values.
1007
1008 8. Added comments about Mac OS X stack usage to the pcrestack man page and to
1009 test 2 if it fails.
1010
1011 9. Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1012 and a #define of that name to empty if it is not externally set. This is to
1013 allow users of MSVC to set it if necessary.
1014
1015 10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1016 the convenience functions in the pcre_get.c source file.
1017
1018 11. An option change at the start of a pattern that had top-level alternatives
1019 could cause overwriting and/or a crash. This command provoked a crash in
1020 some environments:
1021
1022 printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1023
1024 This potential security problem was recorded as CVE-2008-2371.
1025
1026 12. For a pattern where the match had to start at the beginning or immediately
1027 after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1028 pcre_dfa_exec() could read past the end of the passed subject if there was
1029 no match. To help with detecting such bugs (e.g. with valgrind), I modified
1030 pcretest so that it places the subject at the end of its malloc-ed buffer.
1031
1032 13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1033 exec() might read past the end of the data buffer in UTF-8 mode.
1034
1035 14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1036 the data contained the byte 0x85 as part of a UTF-8 character within its
1037 first line. This applied both to normal and DFA matching.
1038
1039 15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1040 /^[^d]*?$/8 failed to match "abc".
1041
1042 16. Added a missing copyright notice to pcrecpp_internal.h.
1043
1044 17. Make it more clear in the documentation that values returned from
1045 pcre_exec() in ovector are byte offsets, not character counts.
1046
1047 18. Tidied a few places to stop certain compilers from issuing warnings.
1048
1049 19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1050 supplied by Stefan Weber. I made a further small update for 7.8 because
1051 there is a change of source arrangements: the pcre_searchfuncs.c module is
1052 replaced by pcre_ucd.c.
1053
1054
1055 Version 7.7 07-May-08
1056 ---------------------
1057
1058 1. Applied Craig's patch to sort out a long long problem: "If we can't convert
1059 a string to a long long, pretend we don't even have a long long." This is
1060 done by checking for the strtoq, strtoll, and _strtoi64 functions.
1061
1062 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1063 pre-7.6 versions, which defined a global no_arg variable instead of putting
1064 it in the RE class. (See also #8 below.)
1065
1066 3. Remove a line of dead code, identified by coverity and reported by Nuno
1067 Lopes.
1068
1069 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
1070
1071 (1) The include/exclude patterns were being applied to the whole pathnames
1072 of files, instead of just to the final components.
1073
1074 (2) If there was more than one level of directory, the subdirectories were
1075 skipped unless they satisfied the include/exclude conditions. This is
1076 inconsistent with GNU grep (and could even be seen as contrary to the
1077 pcregrep specification - which I improved to make it absolutely clear).
1078 The action now is always to scan all levels of directory, and just
1079 apply the include/exclude patterns to regular files.
1080
1081 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1082 --exclude_dir in the tests to avoid scanning .svn directories.
1083
1084 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
1085 NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1086 doesn't support NULs in patterns.
1087
1088 7. Added some missing "const"s to declarations of static tables in
1089 pcre_compile.c and pcre_dfa_exec.c.
1090
1091 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1092 caused by fix #2 above. (Subsequently also a second patch to fix the
1093 first patch. And a third patch - this was a messy problem.)
1094
1095 9. Applied Craig's patch to remove the use of push_back().
1096
1097 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1098 matching function regexec().
1099
1100 11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1101 which, however, unlike Perl's \g{...}, are subroutine calls, not back
1102 references. PCRE supports relative numbers with this syntax (I don't think
1103 Oniguruma does).
1104
1105 12. Previously, a group with a zero repeat such as (...){0} was completely
1106 omitted from the compiled regex. However, this means that if the group
1107 was called as a subroutine from elsewhere in the pattern, things went wrong
1108 (an internal error was given). Such groups are now left in the compiled
1109 pattern, with a new opcode that causes them to be skipped at execution
1110 time.
1111
1112 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1113 to the way PCRE behaves:
1114
1115 (a) A lone ] character is dis-allowed (Perl treats it as data).
1116
1117 (b) A back reference to an unmatched subpattern matches an empty string
1118 (Perl fails the current match path).
1119
1120 (c) A data ] in a character class must be notated as \] because if the
1121 first data character in a class is ], it defines an empty class. (In
1122 Perl it is not possible to have an empty class.) The empty class []
1123 never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1124 The negative empty class [^] matches any one character, independently
1125 of the DOTALL setting.
1126
1127 14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1128 non-existent subpattern following a character class starting with ']' and
1129 containing () gave an internal compiling error instead of "reference to
1130 non-existent subpattern". Fortunately, when the pattern did exist, the
1131 compiled code was correct. (When scanning forwards to check for the
1132 existencd of the subpattern, it was treating the data ']' as terminating
1133 the class, so got the count wrong. When actually compiling, the reference
1134 was subsequently set up correctly.)
1135
1136 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1137 it was being rejected as not supported by pcre_dfa_exec(), even though
1138 other assertions are supported. I have made pcre_dfa_exec() support
1139 (*FAIL).
1140
1141 16. The implementation of 13c above involved the invention of a new opcode,
1142 OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1143 cannot be changed at match time, I realized I could make a small
1144 improvement to matching performance by compiling OP_ALLANY instead of
1145 OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1146 on the OP_ANY path.
1147
1148 17. Compiling pcretest on Windows with readline support failed without the
1149 following two fixes: (1) Make the unistd.h include conditional on
1150 HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1151
1152 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1153 ncurses library to be included for pcretest when ReadLine support is
1154 requested, but also to allow for it to be overridden. This patch came from
1155 Daniel Bergström.
1156
1157 19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1158 as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1159 any errors with the current Unicode tables. Thanks to Peter Kankowski for
1160 spotting this.
1161
1162
1163 Version 7.6 28-Jan-08
1164 ---------------------
1165
1166 1. A character class containing a very large number of characters with
1167 codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1168 overflow.
1169
1170 2. Patch to cut out the "long long" test in pcrecpp_unittest when
1171 HAVE_LONG_LONG is not defined.
1172
1173 3. Applied Christian Ehrlicher's patch to update the CMake build files to
1174 bring them up to date and include new features. This patch includes:
1175
1176 - Fixed PH's badly added libz and libbz2 support.
1177 - Fixed a problem with static linking.
1178 - Added pcredemo. [But later removed - see 7 below.]
1179 - Fixed dftables problem and added an option.
1180 - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1181 HAVE_LONG_LONG.
1182 - Added readline support for pcretest.
1183 - Added an listing of the option settings after cmake has run.
1184
1185 4. A user submitted a patch to Makefile that makes it easy to create
1186 "pcre.dll" under mingw when using Configure/Make. I added stuff to
1187 Makefile.am that cause it to include this special target, without
1188 affecting anything else. Note that the same mingw target plus all
1189 the other distribution libraries and programs are now supported
1190 when configuring with CMake (see 6 below) instead of with
1191 Configure/Make.
1192
1193 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1194 This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1195 exported in the Windows port". It has not yet been confirmed that the patch
1196 solves the problem, but it does no harm.
1197
1198 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1199 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1200 with CMake, and also correct the comment about stack recursion.
1201
1202 7. Remove the automatic building of pcredemo from the ./configure system and
1203 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1204 of a program that users should build themselves after PCRE is installed, so
1205 building it automatically is not really right. What is more, it gave
1206 trouble in some build environments.
1207
1208 8. Further tidies to CMakeLists.txt from Sheri and Christian.
1209
1210
1211 Version 7.5 10-Jan-08
1212 ---------------------
1213
1214 1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
1215 values in parens when parsing an RE using the C++ wrapper."
1216
1217 2. Negative specials like \S did not work in character classes in UTF-8 mode.
1218 Characters greater than 255 were excluded from the class instead of being
1219 included.
1220
1221 3. The same bug as (2) above applied to negated POSIX classes such as
1222 [:^space:].
1223
1224 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1225 defined or documented. It seems to have been a typo for PCRE_STATIC, so
1226 I have changed it.
1227
1228 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
1229 first named subpattern) and a construct such as (?&a) would reference the
1230 first named subpattern whose name started with "a" (in other words, the
1231 length check was missing). Both these problems are fixed. "Subpattern name
1232 expected" is now given for (?&) (a zero-length name), and this patch also
1233 makes it give the same error for \k'' (previously it complained that that
1234 was a reference to a non-existent subpattern).
1235
1236 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
1237 this is right because (?- can be followed by option settings as well as by
1238 digits. I have, however, made the messages clearer.
1239
1240 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1241 than the number used in the conditional) now cause a compile-time error.
1242 This is actually not compatible with Perl, which accepts such patterns, but
1243 treats the conditional as always being FALSE (as PCRE used to), but it
1244 seems to me that giving a diagnostic is better.
1245
1246 8. Change "alphameric" to the more common word "alphanumeric" in comments
1247 and messages.
1248
1249 9. Fix two occurrences of "backslash" in comments that should have been
1250 "backspace".
1251
1252 10. Remove two redundant lines of code that can never be obeyed (their function
1253 was moved elsewhere).
1254
1255 11. The program that makes PCRE's Unicode character property table had a bug
1256 which caused it to generate incorrect table entries for sequences of
1257 characters that have the same character type, but are in different scripts.
1258 It amalgamated them into a single range, with the script of the first of
1259 them. In other words, some characters were in the wrong script. There were
1260 thirteen such cases, affecting characters in the following ranges:
1261
1262 U+002b0 - U+002c1
1263 U+0060c - U+0060d
1264 U+0061e - U+00612
1265 U+0064b - U+0065e
1266 U+0074d - U+0076d
1267 U+01800 - U+01805
1268 U+01d00 - U+01d77
1269 U+01d9b - U+01dbf
1270 U+0200b - U+0200f
1271 U+030fc - U+030fe
1272 U+03260 - U+0327f
1273 U+0fb46 - U+0fbb1
1274 U+10450 - U+1049d
1275
1276 12. The -o option (show only the matching part of a line) for pcregrep was not
1277 compatible with GNU grep in that, if there was more than one match in a
1278 line, it showed only the first of them. It now behaves in the same way as
1279 GNU grep.
1280
1281 13. If the -o and -v options were combined for pcregrep, it printed a blank
1282 line for every non-matching line. GNU grep prints nothing, and pcregrep now
1283 does the same. The return code can be used to tell if there were any
1284 non-matching lines.
1285
1286 14. Added --file-offsets and --line-offsets to pcregrep.
1287
1288 15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1289 infinitely looping recursion. The bug was that positive lookaheads were not
1290 being skipped when checking for a possible empty match (negative lookaheads
1291 and both kinds of lookbehind were skipped).
1292
1293 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1294 inclusion of <windows.h> to before rather than after the definition of
1295 INVALID_FILE_ATTRIBUTES (patch from David Byron).
1296
1297 17. Specifying a possessive quantifier with a specific limit for a Unicode
1298 character property caused pcre_compile() to compile bad code, which led at
1299 runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1300 are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1301 caused the error; without that there was no problem.
1302
1303 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1304
1305 19. Added --enable-pcretest-libreadline.
1306
1307 20. In pcrecpp.cc, the variable 'count' was incremented twice in
1308 RE::GlobalReplace(). As a result, the number of replacements returned was
1309 double what it should be. I removed one of the increments, but Craig sent a
1310 later patch that removed the other one (the right fix) and added unit tests
1311 that check the return values (which was not done before).
1312
1313 21. Several CMake things:
1314
1315 (1) Arranged that, when cmake is used on Unix, the libraries end up with
1316 the names libpcre and libpcreposix, not just pcre and pcreposix.
1317
1318 (2) The above change means that pcretest and pcregrep are now correctly
1319 linked with the newly-built libraries, not previously installed ones.
1320
1321 (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
1322
1323 22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1324 crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1325 UTF-8 newline character). The key issue is that the pattern starts .*;
1326 this means that the match must be either at the beginning, or after a
1327 newline. The bug was in the code for advancing after a failed match and
1328 checking that the new position followed a newline. It was not taking
1329 account of UTF-8 characters correctly.
1330
1331 23. PCRE was behaving differently from Perl in the way it recognized POSIX
1332 character classes. PCRE was not treating the sequence [:...:] as a
1333 character class unless the ... were all letters. Perl, however, seems to
1334 allow any characters between [: and :], though of course it rejects as
1335 unknown any "names" that contain non-letters, because all the known class
1336 names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1337 for example, whereas PCRE did not - it did not recognize a POSIX character
1338 class. This seemed a bit dangerous, so the code has been changed to be
1339 closer to Perl. The behaviour is not identical to Perl, because PCRE will
1340 diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1341 treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1342 Perl does, and where it didn't before.
1343
1344 24. Rewrite so as to remove the single use of %n from pcregrep because in some
1345 Windows environments %n is disabled by default.
1346
1347
1348 Version 7.4 21-Sep-07
1349 ---------------------
1350
1351 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
1352 means that a class such as [\s] counted as "explicit reference to CR or
1353 LF". That isn't really right - the whole point of the change was to try to
1354 help when there was an actual mention of one of the two characters. So now
1355 the change happens only if \r or \n (or a literal CR or LF) character is
1356 encountered.
1357
1358 2. The 32-bit options word was also used for 6 internal flags, but the numbers
1359 of both had grown to the point where there were only 3 bits left.
1360 Fortunately, there was spare space in the data structure, and so I have
1361 moved the internal flags into a new 16-bit field to free up more option
1362 bits.
1363
1364 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1365 but did not set the internal JCHANGED flag - either of these is enough to
1366 control the way the "get" function works - but the PCRE_INFO_JCHANGED
1367 facility is supposed to tell if (?J) was ever used, so now (?J) at the
1368 start sets both bits.
1369
1370 4. Added options (at build time, compile time, exec time) to change \R from
1371 matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1372
1373 5. doc/pcresyntax.html was missing from the distribution.
1374
1375 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1376 compatibility, even though it is no longer used.
1377
1378 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1379 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1380 windows.h file is present (where different names are used). [This was
1381 reversed later after testing - see 16 below.]
1382
1383 8. Changed all #include <config.h> to #include "config.h". There were also
1384 some further <pcre.h> cases that I changed to "pcre.h".
1385
1386 9. When pcregrep was used with the --colour option, it missed the line ending
1387 sequence off the lines that it output.
1388
1389 10. It was pointed out to me that arrays of string pointers cause lots of
1390 relocations when a shared library is dynamically loaded. A technique of
1391 using a single long string with a table of offsets can drastically reduce
1392 these. I have refactored PCRE in four places to do this. The result is
1393 dramatic:
1394
1395 Originally: 290
1396 After changing UCP table: 187
1397 After changing error message table: 43
1398 After changing table of "verbs" 36
1399 After changing table of Posix names 22
1400
1401 Thanks to the folks working on Gregex for glib for this insight.
1402
1403 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1404 unicode-properties was also set.
1405
1406 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1407
1408 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1409 checked only for CRLF.
1410
1411 14. Added casts to pcretest.c to avoid compiler warnings.
1412
1413 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1414
1415 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1416 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1417 entirely. This removes changes made in 7 above.
1418
1419 17. The CMake files have been updated, and there is now more information about
1420 building with CMake in the NON-UNIX-USE document.
1421
1422
1423 Version 7.3 28-Aug-07
1424 ---------------------
1425
1426 1. In the rejigging of the build system that eventually resulted in 7.1, the
1427 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1428 brackets there is not right, since it causes compilers to look for an
1429 installed pcre.h, not the version that is in the source that is being
1430 compiled (which of course may be different). I have changed it back to:
1431
1432 #include "pcre.h"
1433
1434 I have a vague recollection that the change was concerned with compiling in
1435 different directories, but in the new build system, that is taken care of
1436 by the VPATH setting the Makefile.
1437
1438 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1439 when the subject happened to end in the byte 0x85 (e.g. if the last
1440 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1441 characters but of course it shouldn't be taken as a newline when it is part
1442 of another character. The bug was that, for an unlimited repeat of . in
1443 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1444 characters when looking for a newline.
1445
1446 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1447
1448 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1449 in debug output.
1450
1451 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1452 long printing in the pcrecpp unittest when running under MinGW.
1453
1454 6. ESC_K was left out of the EBCDIC table.
1455
1456 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1457 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1458 limit also applies to "virtual nesting" when a pattern is recursive, and in
1459 this case 1000 isn't so big. I have been able to remove this limit at the
1460 expense of backing off one optimization in certain circumstances. Normally,
1461 when pcre_exec() would call its internal match() function recursively and
1462 immediately return the result unconditionally, it uses a "tail recursion"
1463 feature to save stack. However, when a subpattern that can match an empty
1464 string has an unlimited repetition quantifier, it no longer makes this
1465 optimization. That gives it a stack frame in which to save the data for
1466 checking that an empty string has been matched. Previously this was taken
1467 from the 1000-entry workspace that had been reserved. So now there is no
1468 explicit limit, but more stack is used.
1469
1470 8. Applied Daniel's patches to solve problems with the import/export magic
1471 syntax that is required for Windows, and which was going wrong for the
1472 pcreposix and pcrecpp parts of the library. These were overlooked when this
1473 problem was solved for the main library.
1474
1475 9. There were some crude static tests to avoid integer overflow when computing
1476 the size of patterns that contain repeated groups with explicit upper
1477 limits. As the maximum quantifier is 65535, the maximum group length was
1478 set at 30,000 so that the product of these two numbers did not overflow a
1479 32-bit integer. However, it turns out that people want to use groups that
1480 are longer than 30,000 bytes (though not repeat them that many times).
1481 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1482 made it possible to implement the integer overflow checks in a much more
1483 dynamic way, which I have now done. The artificial limitation on group
1484 length has been removed - we now have only the limit on the total length of
1485 the compiled pattern, which depends on the LINK_SIZE setting.
1486
1487 10. Fixed a bug in the documentation for get/copy named substring when
1488 duplicate names are permitted. If none of the named substrings are set, the
1489 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1490 empty string.
1491
1492 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1493 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1494 because the ] is interpreted as the first data character and the
1495 terminating ] is not found. PCRE has been made compatible with Perl in this
1496 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1497 cause memory overwriting.
1498
1499 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1500 string has been matched (to stop an infinite loop). It was not recognizing
1501 a conditional subpattern that could match an empty string if that
1502 subpattern was within another subpattern. For example, it looped when
1503 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
1504 condition was not nested. This bug has been fixed.
1505
1506 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1507 past the start of the subject in the presence of bytes with the top bit
1508 set, for example "\x8aBCD".
1509
1510 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1511 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1512
1513 14. Optimized (?!) to (*FAIL).
1514
1515 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1516 This restricts code points to be within the range 0 to 0x10FFFF, excluding
1517 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1518 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1519 does: it's just the validity check that is more restrictive.
1520
1521 16. Inserted checks for integer overflows during escape sequence (backslash)
1522 processing, and also fixed erroneous offset values for syntax errors during
1523 backslash processing.
1524
1525 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1526 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1527
1528 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1529 caused an overrun.
1530
1531 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1532 something other than just ASCII characters) inside a group that had an
1533 unlimited repeat caused a loop at compile time (while checking to see
1534 whether the group could match an empty string).
1535
1536 20. Debugging a pattern containing \p or \P could cause a crash. For example,
1537 [\P{Any}] did so. (Error in the code for printing property names.)
1538
1539 21. An orphan \E inside a character class could cause a crash.
1540
1541 22. A repeated capturing bracket such as (A)? could cause a wild memory
1542 reference during compilation.
1543
1544 23. There are several functions in pcre_compile() that scan along a compiled
1545 expression for various reasons (e.g. to see if it's fixed length for look
1546 behind). There were bugs in these functions when a repeated \p or \P was
1547 present in the pattern. These operators have additional parameters compared
1548 with \d, etc, and these were not being taken into account when moving along
1549 the compiled data. Specifically:
1550
1551 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1552 length.
1553
1554 (b) An item such as \pL+ within a repeated group could cause crashes or
1555 loops.
1556
1557 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1558 "reference to non-existent subpattern" error.
1559
1560 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1561
1562 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1563 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1564
1565 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1566
1567 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1568 character were causing crashes (broken optimization).
1569
1570 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1571 \p or \P) caused a compile-time loop.
1572
1573 28. More problems have arisen in unanchored patterns when CRLF is a valid line
1574 break. For example, the unstudied pattern [\r\n]A does not match the string
1575 "\r\nA" because change 7.0/46 below moves the current point on by two
1576 characters after failing to match at the start. However, the pattern \nA
1577 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1578 the same is true. There doesn't seem any very clean way out of this, but
1579 what I have chosen to do makes the common cases work: PCRE now takes note
1580 of whether there can be an explicit match for \r or \n anywhere in the
1581 pattern, and if so, 7.0/46 no longer applies. As part of this change,
1582 there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1583 pattern has explicit CR or LF references.
1584
1585 29. Added (*CR) etc for changing newline setting at start of pattern.
1586
1587
1588 Version 7.2 19-Jun-07
1589 ---------------------
1590
1591 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1592 which is apparently normally available under Windows.
1593
1594 2. Re-jig the pcregrep tests with different newline settings in an attempt
1595 to make them independent of the local environment's newline setting.
1596
1597 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1598
1599 4. Some of the "internals" tests were previously cut out when the link size
1600 was not 2, because the output contained actual offsets. The recent new
1601 "Z" feature of pcretest means that these can be cut out, making the tests
1602 usable with all link sizes.
1603
1604 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1605 stack recursion. This gives a massive performance boost under BSD, but just
1606 a small improvement under Linux. However, it saves one field in the frame
1607 in all cases.
1608
1609 6. Added more features from the forthcoming Perl 5.10:
1610
1611 (a) (?-n) (where n is a string of digits) is a relative subroutine or
1612 recursion call. It refers to the nth most recently opened parentheses.
1613
1614 (b) (?+n) is also a relative subroutine call; it refers to the nth next
1615 to be opened parentheses.
1616
1617 (c) Conditions that refer to capturing parentheses can be specified
1618 relatively, for example, (?(-2)... or (?(+3)...
1619
1620 (d) \K resets the start of the current match so that everything before
1621 is not part of it.
1622
1623 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1624
1625 (f) \g{name} is another synonym - part of Perl 5.10's unification of
1626 reference syntax.
1627
1628 (g) (?| introduces a group in which the numbering of parentheses in each
1629 alternative starts with the same number.
1630
1631 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1632
1633 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1634 PCRE_INFO_JCHANGED.
1635
1636 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
1637 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1638 for detecting groups that can match an empty string.
1639
1640 9. A pattern with a very large number of alternatives (more than several
1641 hundred) was running out of internal workspace during the pre-compile
1642 phase, where pcre_compile() figures out how much memory will be needed. A
1643 bit of new cunning has reduced the workspace needed for groups with
1644 alternatives. The 1000-alternative test pattern now uses 12 bytes of
1645 workspace instead of running out of the 4096 that are available.
1646
1647 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1648
1649 11. Applied patch from Google to remove an optimization that didn't quite work.
1650 The report of the bug said:
1651
1652 pcrecpp::RE("a*").FullMatch("aaa") matches, while
1653 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1654 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1655
1656 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1657 it matched the wrong number of bytes.
1658
1659
1660 Version 7.1 24-Apr-07
1661 ---------------------
1662
1663 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1664 that is more "standard", making use of automake and other Autotools. There
1665 is some re-arrangement of the files and adjustment of comments consequent
1666 on this.
1667
1668 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1669 for recursive directory scanning broke on some systems because the files
1670 are not scanned in any specific order and on different systems the order
1671 was different. A call to "sort" has been inserted into RunGrepTest for the
1672 approprate test as a short-term fix. In the longer term there may be an
1673 alternative.
1674
1675 3. I had an email from Eric Raymond about problems translating some of PCRE's
1676 man pages to HTML (despite the fact that I distribute HTML pages, some
1677 people do their own conversions for various reasons). The problems
1678 concerned the use of low-level troff macros .br and .in. I have therefore
1679 removed all such uses from the man pages (some were redundant, some could
1680 be replaced by .nf/.fi pairs). The 132html script that I use to generate
1681 HTML has been updated to handle .nf/.fi and to complain if it encounters
1682 .br or .in.
1683
1684 4. Updated comments in configure.ac that get placed in config.h.in and also
1685 arranged for config.h to be included in the distribution, with the name
1686 config.h.generic, for the benefit of those who have to compile without
1687 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1688
1689 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1690 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1691 makevp.bat for the current PCRE, using the additional files
1692 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1693
1694 6. A Windows user reported a minor discrepancy with test 2, which turned out
1695 to be caused by a trailing space on an input line that had got lost in his
1696 copy. The trailing space was an accident, so I've just removed it.
1697
1698 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1699 that is needed.
1700
1701 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1702 as "const" (a) because they are and (b) because it helps the PHP
1703 maintainers who have recently made a script to detect big data structures
1704 in the php code that should be moved to the .rodata section. I remembered
1705 to update Builducptable as well, so it won't revert if ucptable.h is ever
1706 re-created.
1707
1708 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1709 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1710 order to be able to cut out the UTF-8 tables in the latter when UTF-8
1711 support is not required. This saves 1.5-2K of code, which is important in
1712 some applications.
1713
1714 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1715 so as not to refer to the tables, even though these functions will never be
1716 called when UTF-8 support is disabled. Otherwise there are problems with a
1717 shared library.
1718
1719 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1720
1721 (a) It was defining its arguments as char * instead of void *.
1722
1723 (b) It was assuming that all moves were upwards in memory; this was true
1724 a long time ago when I wrote it, but is no longer the case.
1725
1726 The emulated memove() is provided for those environments that have neither
1727 memmove() nor bcopy(). I didn't think anyone used it these days, but that
1728 is clearly not the case, as these two bugs were recently reported.
1729
1730 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1731 and Detrail to create the HTML documentation, the .txt form of the man
1732 pages, and it removes trailing spaces from listed files. It also creates
1733 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1734 case, it wraps all the #defines with #ifndefs. This script should be run
1735 before "make dist".
1736
1737 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1738 with Unicode property support.
1739
1740 (a) For a maximizing quantifier, if the two different cases of the
1741 character were of different lengths in their UTF-8 codings (there are
1742 some cases like this - I found 11), and the matching function had to
1743 back up over a mixture of the two cases, it incorrectly assumed they
1744 were both the same length.
1745
1746 (b) When PCRE was configured to use the heap rather than the stack for
1747 recursion during matching, it was not correctly preserving the data for
1748 the other case of a UTF-8 character when checking ahead for a match
1749 while processing a minimizing repeat. If the check also involved
1750 matching a wide character, but failed, corruption could cause an
1751 erroneous result when trying to check for a repeat of the original
1752 character.
1753
1754 13. Some tidying changes to the testing mechanism:
1755
1756 (a) The RunTest script now detects the internal link size and whether there
1757 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1758 values substituted by "configure". (The RunGrepTest script already did
1759 this for UTF-8.) The configure.ac script no longer substitutes the
1760 relevant variables.
1761
1762 (b) The debugging options /B and /D in pcretest show the compiled bytecode
1763 with length and offset values. This means that the output is different
1764 for different internal link sizes. Test 2 is skipped for link sizes
1765 other than 2 because of this, bypassing the problem. Unfortunately,
1766 there was also a test in test 3 (the locale tests) that used /B and
1767 failed for link sizes other than 2. Rather than cut the whole test out,
1768 I have added a new /Z option to pcretest that replaces the length and
1769 offset values with spaces. This is now used to make test 3 independent
1770 of link size. (Test 2 will be tidied up later.)
1771
1772 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1773 segmentation fault instead of returning the appropriate error message.
1774
1775 15. In multiline mode when the newline sequence was set to "any", the pattern
1776 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1777 This doesn't seem right; it now treats the CRLF combination as the line
1778 ending, and so does not match in that case. It's only a pattern such as ^$
1779 that would hit this one: something like ^ABC$ would have failed after \r
1780 and then tried again after \r\n.
1781
1782 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1783 in an attempt to make files that differ only in their line terminators
1784 compare equal. This works on Linux.
1785
1786 17. Under certain error circumstances pcregrep might try to free random memory
1787 as it exited. This is now fixed, thanks to valgrind.
1788
1789 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1790 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1791 was because its rules for how to advance for /g after matching an empty
1792 string at the end of a line did not allow for this case. They now check for
1793 it specially.
1794
1795 20. pcretest is supposed to handle patterns and data of any length, by
1796 extending its buffers when necessary. It was getting this wrong when the
1797 buffer for a data line had to be extended.
1798
1799 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1800 CRLF as a newline sequence.
1801
1802 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1803 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1804 I have nevertheless tidied it up.
1805
1806 23. Added some casts to kill warnings from HP-UX ia64 compiler.
1807
1808 24. Added a man page for pcre-config.
1809
1810
1811 Version 7.0 19-Dec-06
1812 ---------------------
1813
1814 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1815 moving to gcc 4.1.1.
1816
1817 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1818 sys/time.h, which is documented as needed for this function. It doesn't
1819 seem to matter on Linux, but it showed up on some releases of OS X.
1820
1821 3. It seems that there are systems where bytes whose values are greater than
1822 127 match isprint() in the "C" locale. The "C" locale should be the
1823 default when a C program starts up. In most systems, only ASCII printing
1824 characters match isprint(). This difference caused the output from pcretest
1825 to vary, making some of the tests fail. I have changed pcretest so that:
1826
1827 (a) When it is outputting text in the compiled version of a pattern, bytes
1828 other than 32-126 are always shown as hex escapes.
1829
1830 (b) When it is outputting text that is a matched part of a subject string,
1831 it does the same, unless a different locale has been set for the match
1832 (using the /L modifier). In this case, it uses isprint() to decide.
1833
1834 4. Fixed a major bug that caused incorrect computation of the amount of memory
1835 required for a compiled pattern when options that changed within the
1836 pattern affected the logic of the preliminary scan that determines the
1837 length. The relevant options are -x, and -i in UTF-8 mode. The result was
1838 that the computed length was too small. The symptoms of this bug were
1839 either the PCRE error "internal error: code overflow" from pcre_compile(),
1840 or a glibc crash with a message such as "pcretest: free(): invalid next
1841 size (fast)". Examples of patterns that provoked this bug (shown in
1842 pcretest format) are:
1843
1844 /(?-x: )/x
1845 /(?x)(?-x: \s*#\s*)/
1846 /((?i)[\x{c0}])/8
1847 /(?i:[\x{c0}])/8
1848
1849 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1850 is now done differently.
1851
1852 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1853 wrapper classes; (b) implement a new function in the C++ scanner that is
1854 more efficient than the old way of doing things because it avoids levels of
1855 recursion in the regex matching; (c) add a paragraph to the documentation
1856 for the FullMatch() function.
1857
1858 6. The escape sequence \n was being treated as whatever was defined as
1859 "newline". Not only was this contrary to the documentation, which states
1860 that \n is character 10 (hex 0A), but it also went horribly wrong when
1861 "newline" was defined as CRLF. This has been fixed.
1862
1863 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1864 was being set to -1 for the "end of line" case (supposedly a value that no
1865 character can have). Though this value is never used (the check for end of
1866 line is "zero bytes in current character"), it caused compiler complaints.
1867 I've changed it to 0xffffffff.
1868
1869 8. In pcre_version.c, the version string was being built by a sequence of
1870 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1871 string (as it is for production releases) called a macro with an empty
1872 argument. The C standard says the result of this is undefined. The gcc
1873 compiler treats it as an empty string (which was what was wanted) but it is
1874 reported that Visual C gives an error. The source has been hacked around to
1875 avoid this problem.
1876
1877 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1878 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1879 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1880 of them did).
1881
1882 10. Originally, pcretest opened its input and output without "b"; then I was
1883 told that "b" was needed in some environments, so it was added for release
1884 5.0 to both the input and output. (It makes no difference on Unix-like
1885 systems.) Later I was told that it is wrong for the input on Windows. I've
1886 now abstracted the modes into two macros, to make it easier to fiddle with
1887 them, and removed "b" from the input mode under Windows.
1888
1889 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1890
1891 12. Added -help and --help to pcretest as an official way of being reminded
1892 of the options.
1893
1894 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1895 and pcrecpp.cc because they annoy compilers at high warning levels.
1896
1897 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1898
1899 15. Fixed an occurrence of == in configure.ac that should have been = (shell
1900 scripts are not C programs :-) and which was not noticed because it works
1901 on Linux.
1902
1903 16. pcretest is supposed to handle any length of pattern and data line (as one
1904 line or as a continued sequence of lines) by extending its input buffer if
1905 necessary. This feature was broken for very long pattern lines, leading to
1906 a string of junk being passed to pcre_compile() if the pattern was longer
1907 than about 50K.
1908
1909 17. I have done a major re-factoring of the way pcre_compile() computes the
1910 amount of memory needed for a compiled pattern. Previously, there was code
1911 that made a preliminary scan of the pattern in order to do this. That was
1912 OK when PCRE was new, but as the facilities have expanded, it has become
1913 harder and harder to keep it in step with the real compile phase, and there
1914 have been a number of bugs (see for example, 4 above). I have now found a
1915 cunning way of running the real compile function in a "fake" mode that
1916 enables it to compute how much memory it would need, while actually only
1917 ever using a few hundred bytes of working memory and without too many
1918 tests of the mode. This should make future maintenance and development
1919 easier. A side effect of this work is that the limit of 200 on the nesting
1920 depth of parentheses has been removed (though this was never a serious
1921 limitation, I suspect). However, there is a downside: pcre_compile() now
1922 runs more slowly than before (30% or more, depending on the pattern). I
1923 hope this isn't a big issue. There is no effect on runtime performance.
1924
1925 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1926 newline (only possible for the last line of a file) and it was a
1927 pattern that set a locale (followed by /Lsomething), pcretest crashed.
1928
1929 19. Added additional timing features to pcretest. (1) The -tm option now times
1930 matching only, not compiling. (2) Both -t and -tm can be followed, as a
1931 separate command line item, by a number that specifies the number of
1932 repeats to use when timing. The default is 50000; this gives better
1933 precision, but takes uncomfortably long for very large patterns.
1934
1935 20. Extended pcre_study() to be more clever in cases where a branch of a
1936 subpattern has no definite first character. For example, (a*|b*)[cd] would
1937 previously give no result from pcre_study(). Now it recognizes that the
1938 first character must be a, b, c, or d.
1939
1940 21. There was an incorrect error "recursive call could loop indefinitely" if
1941 a subpattern (or the entire pattern) that was being tested for matching an
1942 empty string contained only one non-empty item after a nested subpattern.
1943 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1944 incorrectly, because the \d was being skipped in the check.
1945
1946 22. The pcretest program now has a new pattern option /B and a command line
1947 option -b, which is equivalent to adding /B to every pattern. This causes
1948 it to show the compiled bytecode, without the additional information that
1949 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1950 is the same as /B/I).
1951
1952 23. A new optimization is now able automatically to treat some sequences such
1953 as a*b as a*+b. More specifically, if something simple (such as a character
1954 or a simple class like \d) has an unlimited quantifier, and is followed by
1955 something that cannot possibly match the quantified thing, the quantifier
1956 is automatically "possessified".
1957
1958 24. A recursive reference to a subpattern whose number was greater than 39
1959 went wrong under certain circumstances in UTF-8 mode. This bug could also
1960 have affected the operation of pcre_study().
1961
1962 25. Realized that a little bit of performance could be had by replacing
1963 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1964
1965 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1966
1967 27. Possessive quantifiers such as a++ were previously implemented by turning
1968 them into atomic groups such as ($>a+). Now they have their own opcodes,
1969 which improves performance. This includes the automatically created ones
1970 from 23 above.
1971
1972 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1973 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1974 the first matched character to be a colon. This applied both to named and
1975 numbered groups.
1976
1977 29. The ucpinternal.h header file was missing its idempotency #ifdef.
1978
1979 30. I was sent a "project" file called libpcre.a.dev which I understand makes
1980 building PCRE on Windows easier, so I have included it in the distribution.
1981
1982 31. There is now a check in pcretest against a ridiculously large number being
1983 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1984 loop, the loop is abandoned.
1985
1986 32. Forward references to subpatterns in conditions such as (?(2)...) where
1987 subpattern 2 is defined later cause pcre_compile() to search forwards in
1988 the pattern for the relevant set of parentheses. This search went wrong
1989 when there were unescaped parentheses in a character class, parentheses
1990 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1991
1992 33. "Subroutine" calls and backreferences were previously restricted to
1993 referencing subpatterns earlier in the regex. This restriction has now
1994 been removed.
1995
1996 34. Added a number of extra features that are going to be in Perl 5.10. On the
1997 whole, these are just syntactic alternatives for features that PCRE had
1998 previously implemented using the Python syntax or my own invention. The
1999 other formats are all retained for compatibility.
2000
2001 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
2002 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
2003 also .NET compatible.
2004
2005 (b) A recursion or subroutine call to a named group can now be defined as
2006 (?&name) as well as (?P>name).
2007
2008 (c) A backreference to a named group can now be defined as \k<name> or
2009 \k'name' as well as (?P=name). The new forms, as well as being in Perl
2010 5.10, are also .NET compatible.
2011
2012 (d) A conditional reference to a named group can now use the syntax
2013 (?(<name>) or (?('name') as well as (?(name).
2014
2015 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
2016 groups (named and numbered) that are never evaluated inline, but can be
2017 called as "subroutines" from elsewhere. In effect, the DEFINE condition
2018 is always false. There may be only one alternative in such a group.
2019
2020 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
2021 as the simple (?(R). The condition is true only if the most recent
2022 recursion is that of the given number or name. It does not search out
2023 through the entire recursion stack.
2024
2025 (g) The escape \gN or \g{N} has been added, where N is a positive or
2026 negative number, specifying an absolute or relative reference.
2027
2028 35. Tidied to get rid of some further signed/unsigned compiler warnings and
2029 some "unreachable code" warnings.
2030
2031 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
2032 things, this adds five new scripts.
2033
2034 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
2035 There were also incompatibilities regarding the handling of \Q..\E inside
2036 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
2037 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
2038
2039 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
2040 matches an empty string, and forcibly breaks the loop. There were bugs in
2041 this code in non-simple cases. For a pattern such as ^(a()*)* matched
2042 against aaaa the result was just "a" rather than "aaaa", for example. Two
2043 separate and independent bugs (that affected different cases) have been
2044 fixed.
2045
2046 39. Refactored the code to abolish the use of different opcodes for small
2047 capturing bracket numbers. This is a tidy that I avoided doing when I
2048 removed the limit on the number of capturing brackets for 3.5 back in 2001.
2049 The new approach is not only tidier, it makes it possible to reduce the
2050 memory needed to fix the previous bug (38).
2051
2052 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
2053 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
2054 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
2055 mode.
2056
2057 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
2058 report.
2059
2060 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
2061 copy construction and assignment in the C++ wrapper.
2062
2063 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2064 couple of bugs that could have given wrong results in the "--newline=crlf"
2065 case.
2066
2067 44. Added a number of casts and did some reorganization of signed/unsigned int
2068 variables following suggestions from Dair Grant. Also renamed the variable
2069 "this" as "item" because it is a C++ keyword.
2070
2071 45. Arranged for dftables to add
2072
2073 #include "pcre_internal.h"
2074
2075 to pcre_chartables.c because without it, gcc 4.x may remove the array
2076 definition from the final binary if PCRE is built into a static library and
2077 dead code stripping is activated.
2078
2079 46. For an unanchored pattern, if a match attempt fails at the start of a
2080 newline sequence, and the newline setting is CRLF or ANY, and the next two
2081 characters are CRLF, advance by two characters instead of one.
2082
2083
2084 Version 6.7 04-Jul-06
2085 ---------------------
2086
2087 1. In order to handle tests when input lines are enormously long, pcretest has
2088 been re-factored so that it automatically extends its buffers when
2089 necessary. The code is crude, but this _is_ just a test program. The
2090 default size has been increased from 32K to 50K.
2091
2092 2. The code in pcre_study() was using the value of the re argument before
2093 testing it for NULL. (Of course, in any sensible call of the function, it
2094 won't be NULL.)
2095
2096 3. The memmove() emulation function in pcre_internal.h, which is used on
2097 systems that lack both memmove() and bcopy() - that is, hardly ever -
2098 was missing a "static" storage class specifier.
2099
2100 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
2101 containing an extended class (one that cannot be represented by a bitmap
2102 because it contains high-valued characters or Unicode property items, e.g.
2103 [\pZ]). Almost always one would set UTF-8 mode when processing such a
2104 pattern, but PCRE should not loop if you do not (it no longer does).
2105 [Detail: two cases were found: (a) a repeated subpattern containing an
2106 extended class; (b) a recursive reference to a subpattern that followed a
2107 previous extended class. It wasn't skipping over the extended class
2108 correctly when UTF-8 mode was not set.]
2109
2110 5. A negated single-character class was not being recognized as fixed-length
2111 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
2112 compile error "lookbehind assertion is not fixed length".
2113
2114 6. The RunPerlTest auxiliary script was showing an unexpected difference
2115 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
2116 write a Perl script that can interpret lines of an input file either as
2117 byte characters or as UTF-8, which is what "perltest" was being required to
2118 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
2119 can't do is switch easily at run time between having the "use utf8;" pragma
2120 or not. In the end, I fudged it by using the RunPerlTest script to insert
2121 "use utf8;" explicitly for the UTF-8 tests.
2122
2123 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
2124 the end of the subject string, contrary to the documentation and to what
2125 Perl does. This was true of both matching functions. Now it matches only at
2126 the start of the subject and immediately after *internal* newlines.
2127
2128 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
2129 a pointer to an int instead of a pointer to an unsigned long int. This
2130 caused problems on 64-bit systems.
2131
2132 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
2133 instance of the 'standard' template library not being so standard".
2134
2135 10. There was no check on the number of named subpatterns nor the maximum
2136 length of a subpattern name. The product of these values is used to compute
2137 the size of the memory block for a compiled pattern. By supplying a very
2138 long subpattern name and a large number of named subpatterns, the size
2139 computation could be caused to overflow. This is now prevented by limiting
2140 the length of names to 32 characters, and the number of named subpatterns
2141 to 10,000.
2142
2143 11. Subpatterns that are repeated with specific counts have to be replicated in
2144 the compiled pattern. The size of memory for this was computed from the
2145 length of the subpattern and the repeat count. The latter is limited to
2146 65535, but there was no limit on the former, meaning that integer overflow
2147 could in principle occur. The compiled length of a repeated subpattern is
2148 now limited to 30,000 bytes in order to prevent this.
2149
2150 12. Added the optional facility to have named substrings with the same name.
2151
2152 13. Added the ability to use a named substring as a condition, using the
2153 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2154 are numbers (not recommended). Forward references are permitted.
2155
2156 14. Added forward references in named backreferences (if you see what I mean).
2157
2158 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2159 pattern could run off the end of the subject. For example, the pattern
2160 "(?s)(.{1,5})"8 did this with the subject "ab".
2161
2162 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2163 PCRE_CASELESS was set when matching characters that were quantified with ?
2164 or *.
2165
2166 17. A character class other than a single negated character that had a minimum
2167 but no maximum quantifier - for example [ab]{6,} - was not handled
2168 correctly by pce_dfa_exec(). It would match only one character.
2169
2170 18. A valid (though odd) pattern that looked like a POSIX character
2171 class but used an invalid character after [ (for example [[,abc,]]) caused
2172 pcre_compile() to give the error "Failed: internal error: code overflow" or
2173 in some cases to crash with a glibc free() error. This could even happen if
2174 the pattern terminated after [[ but there just happened to be a sequence of
2175 letters, a binary zero, and a closing ] in the memory that followed.
2176
2177 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2178 over the years. Originally (before any Unicode support), just the bottom 8
2179 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2180 output from "man perlunicode" includes this:
2181
2182 The regular expression compiler produces polymorphic opcodes. That
2183 is, the pattern adapts to the data and automatically switches to
2184 the Unicode character scheme when presented with Unicode data--or
2185 instead uses a traditional byte scheme when presented with byte
2186 data.
2187
2188 Sadly, a wide octal escape does not cause a switch, and in a string with
2189 no other multibyte characters, these octal escapes are treated as before.
2190 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
2191 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2192 Unicode string.
2193
2194 I have not perpetrated such confusion in PCRE. Up till now, it took just
2195 the bottom 8 bits, as in old Perl. I have now made octal escapes with
2196 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2197 translate to the appropriate multibyte character.
2198
2199 29. Applied some refactoring to reduce the number of warnings from Microsoft
2200 and Borland compilers. This has included removing the fudge introduced
2201 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2202 a warning about an unused variable.
2203
2204 21. PCRE has not included VT (character 0x0b) in the set of whitespace
2205 characters since release 4.0, because Perl (from release 5.004) does not.
2206 [Or at least, is documented not to: some releases seem to be in conflict
2207 with the documentation.] However, when a pattern was studied with
2208 pcre_study() and all its branches started with \s, PCRE still included VT
2209 as a possible starting character. Of course, this did no harm; it just
2210 caused an unnecessary match attempt.
2211
2212 22. Removed a now-redundant internal flag bit that recorded the fact that case
2213 dependency changed within the pattern. This was once needed for "required
2214 byte" processing, but is no longer used. This recovers a now-scarce options
2215 bit. Also moved the least significant internal flag bit to the most-
2216 significant bit of the word, which was not previously used (hangover from
2217 the days when it was an int rather than a uint) to free up another bit for
2218 the future.
2219
2220 23. Added support for CRLF line endings as well as CR and LF. As well as the
2221 default being selectable at build time, it can now be changed at runtime
2222 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2223 specify that it is scanning data with non-default line endings.
2224
2225 24. Changed the definition of CXXLINK to make it agree with the definition of
2226 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2227
2228 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2229 recursions. This makes a big different to stack usage for some patterns.
2230
2231 26. If a subpattern containing a named recursion or subroutine reference such
2232 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2233 the space required for the compiled pattern went wrong and gave too small a
2234 value. Depending on the environment, this could lead to "Failed: internal
2235 error: code overflow at offset 49" or "glibc detected double free or
2236 corruption" errors.
2237
2238 27. Applied patches from Google (a) to support the new newline modes and (b) to
2239 advance over multibyte UTF-8 characters in GlobalReplace.
2240
2241 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2242 difference for some implementation of PCRE in some Windows version.
2243
2244 29. Added some extra testing facilities to pcretest:
2245
2246 \q<number> in a data line sets the "match limit" value
2247 \Q<number> in a data line sets the "match recursion limt" value
2248 -S <number> sets the stack size, where <number> is in megabytes
2249
2250 The -S option isn't available for Windows.
2251
2252
2253 Version 6.6 06-Feb-06
2254 ---------------------
2255
2256 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2257 in pcreposix.h. I have copied the definition from pcre.h.
2258
2259 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2260 because pcre.h is no longer a built file.
2261
2262 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2263 not normally included in the compiled code.
2264
2265
2266 Version 6.5 01-Feb-06
2267 ---------------------
2268
2269 1. When using the partial match feature with pcre_dfa_exec(), it was not
2270 anchoring the second and subsequent partial matches at the new starting
2271 point. This could lead to incorrect results. For example, with the pattern
2272 /1234/, partially matching against "123" and then "a4" gave a match.
2273
2274 2. Changes to pcregrep:
2275
2276 (a) All non-match returns from pcre_exec() were being treated as failures
2277 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2278 error message is output. Some extra information is given for the
2279 PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2280 probably the only errors that are likely to be caused by users (by
2281 specifying a regex that has nested indefinite repeats, for instance).
2282 If there are more than 20 of these errors, pcregrep is abandoned.
2283
2284 (b) A binary zero was treated as data while matching, but terminated the
2285 output line if it was written out. This has been fixed: binary zeroes
2286 are now no different to any other data bytes.
2287
2288 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2289 used to set a locale for matching. The --locale=xxxx long option has
2290 been added (no short equivalent) to specify a locale explicitly on the
2291 pcregrep command, overriding the environment variables.
2292
2293 (d) When -B was used with -n, some line numbers in the output were one less
2294 than they should have been.
2295
2296 (e) Added the -o (--only-matching) option.
2297
2298 (f) If -A or -C was used with -c (count only), some lines of context were
2299 accidentally printed for the final match.
2300
2301 (g) Added the -H (--with-filename) option.
2302
2303 (h) The combination of options -rh failed to suppress file names for files
2304 that were found from directory arguments.
2305
2306 (i) Added the -D (--devices) and -d (--directories) options.
2307
2308 (j) Added the -F (--fixed-strings) option.
2309
2310 (k) Allow "-" to be used as a file name for -f as well as for a data file.
2311
2312 (l) Added the --colo(u)r option.
2313
2314 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2315 is not present by default.
2316
2317 3. A nasty bug was discovered in the handling of recursive patterns, that is,
2318 items such as (?R) or (?1), when the recursion could match a number of
2319 alternatives. If it matched one of the alternatives, but subsequently,
2320 outside the recursion, there was a failure, the code tried to back up into
2321 the recursion. However, because of the way PCRE is implemented, this is not
2322 possible, and the result was an incorrect result from the match.
2323
2324 In order to prevent this happening, the specification of recursion has
2325 been changed so that all such subpatterns are automatically treated as
2326 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2327
2328 4. I had overlooked the fact that, in some locales, there are characters for
2329 which isalpha() is true but neither isupper() nor islower() are true. In
2330 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2331 and ordfeminine) are like this. This affected the treatment of \w and \W
2332 when they appeared in character classes, but not when they appeared outside
2333 a character class. The bit map for "word" characters is now created
2334 separately from the results of isalnum() instead of just taking it from the
2335 upper, lower, and digit maps. (Plus the underscore character, of course.)
2336
2337 5. The above bug also affected the handling of POSIX character classes such as
2338 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2339 permanent tables. Instead, the bit maps for such a class were previously
2340 created as the appropriate unions of the upper, lower, and digit bitmaps.
2341 Now they are created by subtraction from the [[:word:]] class, which has
2342 its own bitmap.
2343
2344 6. The [[:blank:]] character class matches horizontal, but not vertical space.
2345 It is created by subtracting the vertical space characters (\x09, \x0a,
2346 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2347 subtraction was done in the overall bitmap for a character class, meaning
2348 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2349 be recognized. This bug has been fixed.
2350
2351 7. Patches from the folks at Google:
2352
2353 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2354 real life, but is still worth protecting against".
2355
2356 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2357 regular expressions".
2358
2359 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2360 have it.
2361
2362 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2363 "configure" and the latter not, in order to fix a problem somebody had
2364 with compiling the Arg class on HP-UX.
2365
2366 (e) Improve the error-handling of the C++ wrapper a little bit.
2367
2368 (f) New tests for checking recursion limiting.
2369
2370 8. The pcre_memmove() function, which is used only if the environment does not
2371 have a standard memmove() function (and is therefore rarely compiled),
2372 contained two bugs: (a) use of int instead of size_t, and (b) it was not
2373 returning a result (though PCRE never actually uses the result).
2374
2375 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2376 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2377 returned instead of calling malloc() with an overflowing number that would
2378 most likely cause subsequent chaos.
2379
2380 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2381
2382 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2383 with this option is matched, the nmatch and pmatch options of regexec() are
2384 ignored.
2385
2386 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2387 provided in case anyone wants to the the POSIX interface with UTF-8
2388 strings.
2389
2390 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2391 C++ linking (needed for some HP-UX environments).
2392
2393 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2394 (unused parameter) and in the pcre_printint() function (omitted "default"
2395 switch label when the default is to do nothing).
2396
2397 15. Added some code to make it possible, when PCRE is compiled as a C++
2398 library, to replace subject pointers for pcre_exec() with a smart pointer
2399 class, thus making it possible to process discontinuous strings.
2400
2401 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2402 much the same function. They were added by different people who were trying
2403 to make PCRE easy to compile on non-Unix systems. It has been suggested
2404 that PCRE_EXPORT be abolished now that there is more automatic apparatus
2405 for compiling on Windows systems. I have therefore replaced it with
2406 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2407 defaults to "extern" for C or "extern C" for C++, which works fine on
2408 Unix-like systems. It is now possible to override the value of PCRE_DATA_
2409 SCOPE with something explicit in config.h. In addition:
2410
2411 (a) pcreposix.h still had just "extern" instead of either of these macros;
2412 I have replaced it with PCRE_DATA_SCOPE.
2413
2414 (b) Functions such as _pcre_xclass(), which are internal to the library,
2415 but external in the C sense, all had PCRE_EXPORT in their definitions.
2416 This is apparently wrong for the Windows case, so I have removed it.
2417 (It makes no difference on Unix-like systems.)
2418
2419 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2420 of recursive calls to match(). This is different to MATCH_LIMIT because
2421 that limits the total number of calls to match(), not all of which increase
2422 the depth of recursion. Limiting the recursion depth limits the amount of
2423 stack (or heap if NO_RECURSE is set) that is used. The default can be set
2424 when PCRE is compiled, and changed at run time. A patch from Google adds
2425 this functionality to the C++ interface.
2426
2427 18. Changes to the handling of Unicode character properties:
2428
2429 (a) Updated the table to Unicode 4.1.0.
2430
2431 (b) Recognize characters that are not in the table as "Cn" (undefined).
2432
2433 (c) I revised the way the table is implemented to a much improved format
2434 which includes recognition of ranges. It now supports the ranges that
2435 are defined in UnicodeData.txt, and it also amalgamates other
2436 characters into ranges. This has reduced the number of entries in the
2437 table from around 16,000 to around 3,000, thus reducing its size
2438 considerably. I realized I did not need to use a tree structure after
2439 all - a binary chop search is just as efficient. Having reduced the
2440 number of entries, I extended their size from 6 bytes to 8 bytes to
2441 allow for more data.
2442
2443 (d) Added support for Unicode script names via properties such as \p{Han}.
2444
2445 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2446 matching that character.
2447
2448 20. When matching a repeated Unicode property with a minimum greater than zero,
2449 (for example \pL{2,}), PCRE could look past the end of the subject if it
2450 reached it while seeking the minimum number of characters. This could
2451 happen only if some of the characters were more than one byte long, because
2452 there is a check for at least the minimum number of bytes.
2453
2454 21. Refactored the implementation of \p and \P so as to be more general, to
2455 allow for more different types of property in future. This has changed the
2456 compiled form incompatibly. Anybody with saved compiled patterns that use
2457 \p or \P will have to recompile them.
2458
2459 22. Added "Any" and "L&" to the supported property types.
2460
2461 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2462 but give a compile time error if the value is greater than 0xff.
2463
2464 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2465 accidentally not being installed or uninstalled.
2466
2467 25. The pcre.h file was built from pcre.h.in, but the only changes that were
2468 made were to insert the current release number. This seemed silly, because
2469 it made things harder for people building PCRE on systems that don't run
2470 "configure". I have turned pcre.h into a distributed file, no longer built
2471 by "configure", with the version identification directly included. There is
2472 no longer a pcre.h.in file.
2473
2474 However, this change necessitated a change to the pcre-config script as
2475 well. It is built from pcre-config.in, and one of the substitutions was the
2476 release number. I have updated configure.ac so that ./configure now finds
2477 the release number by grepping pcre.h.
2478
2479 26. Added the ability to run the tests under valgrind.
2480
2481
2482 Version 6.4 05-Sep-05
2483 ---------------------
2484
2485 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2486 "--" to be printed when multiple files were scanned, even when none of the
2487 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2488 consider it to be a bug, and have restored the previous behaviour.
2489
2490 2. A couple of code tidies to get rid of compiler warnings.
2491
2492 3. The pcretest program used to cheat by referring to symbols in the library
2493 whose names begin with _pcre_. These are internal symbols that are not
2494 really supposed to be visible externally, and in some environments it is
2495 possible to suppress them. The cheating is now confined to including
2496 certain files from the library's source, which is a bit cleaner.
2497
2498 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2499 file's purpose clearer.
2500
2501 5. Reorganized pcre_ucp_findchar().
2502
2503
2504 Version 6.3 15-Aug-05
2505 ---------------------
2506
2507 1. The file libpcre.pc.in did not have general read permission in the tarball.
2508
2509 2. There were some problems when building without C++ support:
2510
2511 (a) If C++ support was not built, "make install" and "make test" still
2512 tried to test it.
2513
2514 (b) There were problems when the value of CXX was explicitly set. Some
2515 changes have been made to try to fix these, and ...
2516
2517 (c) --disable-cpp can now be used to explicitly disable C++ support.
2518
2519 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2520 backslash in a target when C++ was disabled. This confuses some
2521 versions of "make", apparently. Using an intermediate variable solves
2522 this. (Same for CPP_LOBJ.)
2523
2524 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2525 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2526 necessary on certain architectures.
2527
2528 4. Added a setting of -export-symbols-regex to the link command to remove
2529 those symbols that are exported in the C sense, but actually are local
2530 within the library, and not documented. Their names all begin with
2531 "_pcre_". This is not a perfect job, because (a) we have to except some
2532 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2533 available (and never for static libraries). I have made a note to try to
2534 find a way round (a) in the future.
2535
2536
2537 Version 6.2 01-Aug-05
2538 ---------------------
2539
2540 1. There was no test for integer overflow of quantifier values. A construction
2541 such as {1111111111111111} would give undefined results. What is worse, if
2542 a minimum quantifier for a parenthesized subpattern overflowed and became
2543 negative, the calculation of the memory size went wrong. This could have
2544 led to memory overwriting.
2545
2546 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2547
2548 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2549 operating environments where this matters.
2550
2551 4. Applied Giuseppe Maxia's patch to add additional features for controlling
2552 PCRE options from within the C++ wrapper.
2553
2554 5. Named capturing subpatterns were not being correctly counted when a pattern
2555 was compiled. This caused two problems: (a) If there were more than 100
2556 such subpatterns, the calculation of the memory needed for the whole
2557 compiled pattern went wrong, leading to an overflow error. (b) Numerical
2558 back references of the form \12, where the number was greater than 9, were
2559 not recognized as back references, even though there were sufficient
2560 previous subpatterns.
2561
2562 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2563 versions of gcc, e.g. 2.95.4.
2564
2565
2566 Version 6.1 21-Jun-05
2567 ---------------------
2568
2569 1. There was one reference to the variable "posix" in pcretest.c that was not
2570 surrounded by "#if !defined NOPOSIX".
2571
2572 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2573 the cross-check on the old pcre_info() function, for the benefit of the
2574 cut-down version of PCRE that is currently imported into Exim.
2575
2576 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2577 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2578 patterns that start (?i) but I don't think that matters. The use of (?i) is
2579 just an example; this all applies to the other options as well.
2580
2581 4. Since libtool seems to echo the compile commands it is issuing, the output
2582 from "make" can be reduced a bit by putting "@" in front of each libtool
2583 compile command.
2584
2585 5. Patch from the folks at Google for configure.in to be a bit more thorough
2586 in checking for a suitable C++ installation before trying to compile the
2587 C++ stuff. This should fix a reported problem when a compiler was present,
2588 but no suitable headers.
2589
2590 6. The man pages all had just "PCRE" as their title. I have changed them to
2591 be the relevant file name. I have also arranged that these names are
2592 retained in the file doc/pcre.txt, which is a concatenation in text format
2593 of all the man pages except the little individual ones for each function.
2594
2595 7. The NON-UNIX-USE file had not been updated for the different set of source
2596 files that come with release 6. I also added a few comments about the C++
2597 wrapper.
2598
2599
2600 Version 6.0 07-Jun-05
2601 ---------------------
2602
2603 1. Some minor internal re-organization to help with my DFA experiments.
2604
2605 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2606 didn't matter for the library itself when fully configured, but did matter
2607 when compiling without UCP support, or within Exim, where the ucp files are
2608 not imported.
2609
2610 3. Refactoring of the library code to split up the various functions into
2611 different source modules. The addition of the new DFA matching code (see
2612 below) to a single monolithic source would have made it really too
2613 unwieldy, quite apart from causing all the code to be include in a
2614 statically linked application, when only some functions are used. This is
2615 relevant even without the DFA addition now that patterns can be compiled in
2616 one application and matched in another.
2617
2618 The downside of splitting up is that there have to be some external
2619 functions and data tables that are used internally in different modules of
2620 the library but which are not part of the API. These have all had their
2621 names changed to start with "_pcre_" so that they are unlikely to clash
2622 with other external names.
2623
2624 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2625 a different (DFA) algorithm. Although it is slower than the original
2626 function, it does have some advantages for certain types of matching
2627 problem.
2628
2629 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2630 including restarting after a partial match.
2631
2632 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2633 defined when compiling for Windows was sent to me. I have put it into the
2634 code, though I have no means of testing or verifying it.
2635
2636 7. Added the pcre_refcount() auxiliary function.
2637
2638 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2639 match before or at the first newline in the subject string. In pcretest,
2640 the /f option on a pattern can be used to set this.
2641
2642 9. A repeated \w when used in UTF-8 mode with characters greater than 256
2643 would behave wrongly. This has been present in PCRE since release 4.0.
2644
2645 10. A number of changes to the pcregrep command:
2646
2647 (a) Refactored how -x works; insert ^(...)$ instead of setting
2648 PCRE_ANCHORED and checking the length, in preparation for adding
2649 something similar for -w.
2650
2651 (b) Added the -w (match as a word) option.
2652
2653 (c) Refactored the way lines are read and buffered so as to have more
2654 than one at a time available.
2655
2656 (d) Implemented a pcregrep test script.
2657
2658 (e) Added the -M (multiline match) option. This allows patterns to match
2659 over several lines of the subject. The buffering ensures that at least
2660 8K, or the rest of the document (whichever is the shorter) is available
2661 for matching (and similarly the previous 8K for lookbehind assertions).
2662
2663 (f) Changed the --help output so that it now says
2664
2665 -w, --word-regex(p)
2666
2667 instead of two lines, one with "regex" and the other with "regexp"
2668 because that confused at least one person since the short forms are the
2669 same. (This required a bit of code, as the output is generated
2670 automatically from a table. It wasn't just a text change.)
2671
2672 (g) -- can be used to terminate pcregrep options if the next thing isn't an
2673 option but starts with a hyphen. Could be a pattern or a path name
2674 starting with a hyphen, for instance.
2675
2676 (h) "-" can be given as a file name to represent stdin.
2677
2678 (i) When file names are being printed, "(standard input)" is used for
2679 the standard input, for compatibility with GNU grep. Previously
2680 "<stdin>" was used.
2681
2682 (j) The option --label=xxx can be used to supply a name to be used for
2683 stdin when file names are being printed. There is no short form.
2684
2685 (k) Re-factored the options decoding logic because we are going to add
2686 two more options that take data. Such options can now be given in four
2687 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2688
2689 (l) Added the -A, -B, and -C options for requesting that lines of context
2690 around matches be printed.
2691
2692 (m) Added the -L option to print the names of files that do not contain
2693 any matching lines, that is, the complement of -l.
2694
2695 (n) The return code is 2 if any file cannot be opened, but pcregrep does
2696 continue to scan other files.
2697
2698 (o) The -s option was incorrectly implemented. For compatibility with other
2699 greps, it now suppresses the error message for a non-existent or non-
2700 accessible file (but not the return code). There is a new option called
2701 -q that suppresses the output of matching lines, which was what -s was
2702 previously doing.
2703
2704 (p) Added --include and --exclude options to specify files for inclusion
2705 and exclusion when recursing.
2706
2707 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2708 Hopefully, it now does.
2709
2710 12. Missing cast in pcre_study().
2711
2712 13. Added an "uninstall" target to the makefile.
2713
2714 14. Replaced "extern" in the function prototypes in Makefile.in with
2715 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2716 world, but is set differently for Windows.
2717
2718 15. Added a second compiling function called pcre_compile2(). The only
2719 difference is that it has an extra argument, which is a pointer to an
2720 integer error code. When there is a compile-time failure, this is set
2721 non-zero, in addition to the error test pointer being set to point to an
2722 error message. The new argument may be NULL if no error number is required
2723 (but then you may as well call pcre_compile(), which is now just a
2724 wrapper). This facility is provided because some applications need a
2725 numeric error indication, but it has also enabled me to tidy up the way
2726 compile-time errors are handled in the POSIX wrapper.
2727
2728 16. Added VPATH=.libs to the makefile; this should help when building with one
2729 prefix path and installing with another. (Or so I'm told by someone who
2730 knows more about this stuff than I do.)
2731
2732 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2733 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2734 match everything, including newlines. This is not POSIX-compatible, but
2735 somebody wanted the feature. From pcretest it can be activated by using
2736 both the P and the s flags.
2737
2738 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2739
2740 19. libpcre.pc was being incorrectly installed as executable.
2741
2742 20. A couple of places in pcretest check for end-of-line by looking for '\n';
2743 it now also looks for '\r' so that it will work unmodified on Windows.
2744
2745 21. Added Google's contributed C++ wrapper to the distribution.
2746
2747 22. Added some untidy missing memory free() calls in pcretest, to keep
2748 Electric Fence happy when testing.
2749
2750
2751
2752 Version 5.0 13-Sep-04
2753 ---------------------
2754
2755 1. Internal change: literal characters are no longer packed up into items
2756 containing multiple characters in a single byte-string. Each character
2757 is now matched using a separate opcode. However, there may be more than one
2758 byte in the character in UTF-8 mode.
2759
2760 2. The pcre_callout_block structure has two new fields: pattern_position and
2761 next_item_length. These contain the offset in the pattern to the next match
2762 item, and its length, respectively.
2763
2764 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2765 insertion of callouts before each pattern item. Added the /C option to
2766 pcretest to make use of this.
2767
2768 4. On the advice of a Windows user, the lines
2769
2770 #if defined(_WIN32) || defined(WIN32)
2771 _setmode( _fileno( stdout ), 0x8000 );
2772 #endif /* defined(_WIN32) || defined(WIN32) */
2773
2774 have been added to the source of pcretest. This apparently does useful
2775 magic in relation to line terminators.
2776
2777 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2778 for the benefit of those environments where the "b" makes a difference.
2779
2780 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2781 to know about it. I have put a hack into configure.in that adds in code
2782 to set GCC=yes if CC=icc. This seems to end up at a point in the
2783 generated configure script that is early enough to affect the setting of
2784 compiler options, which is what is needed, but I have no means of testing
2785 whether it really works. (The user who reported this had patched the
2786 generated configure script, which of course I cannot do.)
2787
2788 LATER: After change 22 below (new libtool files), the configure script
2789 seems to know about icc (and also ecc). Therefore, I have commented out
2790 this hack in configure.in.
2791
2792 7. Added support for pkg-config (2 patches were sent in).
2793
2794 8. Negated POSIX character classes that used a combination of internal tables
2795 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2796 [[:^ascii]]. Typically, they would match almost any characters. The other
2797 POSIX classes were not broken in this way.
2798
2799 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2800 to find the match, as PCRE was deluded into thinking that the match had to
2801 start at the start point or following a newline. The same bug applied to
2802 patterns with negative forward assertions or any backward assertions
2803 preceding ".*" at the start, unless the pattern required a fixed first
2804 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2805
2806 10. In UTF-8 mode, when moving forwards in the subject after a failed match
2807 starting at the last subject character, bytes beyond the end of the subject
2808 string were read.
2809
2810 11. Renamed the variable "class" as "classbits" to make life easier for C++
2811 users. (Previously there was a macro definition, but it apparently wasn't
2812 enough.)
2813
2814 12. Added the new field "tables" to the extra data so that tables can be passed
2815 in at exec time, or the internal tables can be re-selected. This allows
2816 a compiled regex to be saved and re-used at a later time by a different
2817 program that might have everything at different addresses.
2818
2819 13. Modified the pcre-config script so that, when run on Solaris, it shows a
2820 -R library as well as a -L library.
2821
2822 14. The debugging options of pcretest (-d on the command line or D on a
2823 pattern) showed incorrect output for anything following an extended class
2824 that contained multibyte characters and which was followed by a quantifier.
2825
2826 15. Added optional support for general category Unicode character properties
2827 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2828 support. It adds about 90K to the size of the library. The meanings of the
2829 inbuilt class escapes such as \d and \s have NOT been changed.
2830
2831 16. Updated pcredemo.c to include calls to free() to release the memory for the
2832 compiled pattern.
2833
2834 17. The generated file chartables.c was being created in the source directory
2835 instead of in the building directory. This caused the build to fail if the
2836 source directory was different from the building directory, and was
2837 read-only.
2838
2839 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2840 file. No doubt somebody will tell me if they don't make sense... Also added
2841 Dan Mooney's comments about building on OpenVMS.
2842
2843 19. Added support for partial matching via the PCRE_PARTIAL option for
2844 pcre_exec() and the \P data escape in pcretest.
2845
2846 20. Extended pcretest with 3 new pattern features:
2847
2848 (i) A pattern option of the form ">rest-of-line" causes pcretest to
2849 write the compiled pattern to the file whose name is "rest-of-line".
2850 This is a straight binary dump of the data, with the saved pointer to
2851 the character tables forced to be NULL. The study data, if any, is
2852 written too. After writing, pcretest reads a new pattern.
2853
2854 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2855 compiled pattern from the given file. There must not be any
2856 occurrences of "<" in the file name (pretty unlikely); if there are,
2857 pcretest will instead treat the initial "<" as a pattern delimiter.
2858 After reading in the pattern, pcretest goes on to read data lines as
2859 usual.
2860
2861 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2862 and 16-bit fields in a compiled pattern, to simulate a pattern that
2863 was compiled on a host of opposite endianness.
2864
2865 21. The pcre-exec() function can now cope with patterns that were compiled on
2866 hosts of opposite endianness, with this restriction:
2867
2868 As for any compiled expression that is saved and used later, the tables
2869 pointer field cannot be preserved; the extra_data field in the arguments
2870 to pcre_exec() should be used to pass in a tables address if a value
2871 other than the default internal tables were used at compile time.
2872
2873 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2874 now diagnosed as an error. Previously, most of the time, a negative number
2875 would have been treated as zero, but if in addition "ovector" was passed as
2876 NULL, a crash could occur.
2877
2878 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2879 new versions from the libtool 1.5 distribution (the last one is a copy of
2880 a file called libtool.m4). This seems to have fixed the need to patch
2881 "configure" to support Darwin 1.3 (which I used to do). However, I still
2882 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2883 workstation).
2884
2885 24. Changed the PCRE licence to be the more standard "BSD" licence.
2886
2887
2888 Version 4.5 01-Dec-03
2889 ---------------------
2890
2891 1. There has been some re-arrangement of the code for the match() function so
2892 that it can be compiled in a version that does not call itself recursively.
2893 Instead, it keeps those local variables that need separate instances for
2894 each "recursion" in a frame on the heap, and gets/frees frames whenever it
2895 needs to "recurse". Keeping track of where control must go is done by means
2896 of setjmp/longjmp. The whole thing is implemented by a set of macros that
2897 hide most of the details from the main code, and operates only if
2898 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2899 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2900 operating.
2901
2902 To make it easier for callers to provide specially tailored get/free
2903 functions for this usage, two new functions, pcre_stack_malloc, and
2904 pcre_stack_free, are used. They are always called in strict stacking order,
2905 and the size of block requested is always the same.
2906
2907 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2908 PCRE has been compiled to use the stack or the heap for recursion. The
2909 -C option of pcretest uses this to show which version is compiled.
2910
2911 A new data escape \S, is added to pcretest; it causes the amounts of store
2912 obtained and freed by both kinds of malloc/free at match time to be added
2913 to the output.
2914
2915 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2916 what's available on my current Linux desktop machine.
2917
2918 3. When matching a UTF-8 string, the test for a valid string at the start has
2919 been extended. If start_offset is not zero, PCRE now checks that it points
2920 to a byte that is the start of a UTF-8 character. If not, it returns
2921 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2922 this is necessary because there may be backward assertions in the pattern.
2923 When matching the same subject several times, it may save resources to use
2924 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2925
2926 4. The code for checking the validity of UTF-8 strings has been tightened so
2927 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2928 containing "overlong sequences".
2929
2930 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2931 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2932 should have been "&", but it just so happened that all the cases this let
2933 through by mistake were picked up later in the function.
2934
2935 6. I had used a variable called "isblank" - this is a C99 function, causing
2936 some compilers to warn. To avoid this, I renamed it (as "blankclass").
2937
2938 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2939 prompting; (b) run "./pcretest /dev/null" at the start of the test script
2940 so the version is shown; (c) stop "make test" echoing "./RunTest".
2941
2942 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2943
2944 9. The prototype for memmove() for systems that don't have it was using
2945 size_t, but the inclusion of the header that defines size_t was later. I've
2946 moved the #includes for the C headers earlier to avoid this.
2947
2948 10. Added some adjustments to the code to make it easier to compiler on certain
2949 special systems:
2950
2951 (a) Some "const" qualifiers were missing.
2952 (b) Added the macro EXPORT before all exported functions; by default this
2953 is defined to be empty.
2954 (c) Changed the dftables auxiliary program (that builds chartables.c) so
2955 that it reads its output file name as an argument instead of writing
2956 to the standard output and assuming this can be redirected.
2957
2958 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2959 class containing characters with values greater than 255, PCRE compilation
2960 went into a loop.
2961
2962 12. A recursive reference to a subpattern that was within another subpattern
2963 that had a minimum quantifier of zero caused PCRE to crash. For example,
2964 (x(y(?2))z)? provoked this bug with a subject that got as far as the
2965 recursion. If the recursively-called subpattern itself had a zero repeat,
2966 that was OK.
2967
2968 13. In pcretest, the buffer for reading a data line was set at 30K, but the
2969 buffer into which it was copied (for escape processing) was still set at
2970 1024, so long lines caused crashes.
2971
2972 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2973 "internal error: code overflow...". This applied to any character class
2974 that was followed by a possessive quantifier.
2975
2976 15. Modified the Makefile to add libpcre.la as a prerequisite for
2977 libpcreposix.la because I was told this is needed for a parallel build to
2978 work.
2979
2980 16. If a pattern that contained .* following optional items at the start was
2981 studied, the wrong optimizing data was generated, leading to matching
2982 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2983 matching string must start with a or b or c. The correct conclusion for
2984 this pattern is that a match can start with any character.
2985
2986
2987 Version 4.4 13-Aug-03
2988 ---------------------
2989
2990 1. In UTF-8 mode, a character class containing characters with values between
2991 127 and 255 was not handled correctly if the compiled pattern was studied.
2992 In fixing this, I have also improved the studying algorithm for such
2993 classes (slightly).
2994
2995 2. Three internal functions had redundant arguments passed to them. Removal
2996 might give a very teeny performance improvement.
2997
2998 3. Documentation bug: the value of the capture_top field in a callout is *one
2999 more than* the number of the hightest numbered captured substring.
3000
3001 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
3002 in incorrectly linking with a previously installed version. They now link
3003 explicitly with libpcre.la.
3004
3005 5. configure.in no longer needs to recognize Cygwin specially.
3006
3007 6. A problem in pcre.in for Windows platforms is fixed.
3008
3009 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
3010 pcretest, it used to include the size of the study block as part of its
3011 output. Unfortunately, the structure contains a field that has a different
3012 size on different hardware architectures. This meant that the tests that
3013 showed this size failed. As the block is currently always of a fixed size,
3014 this information isn't actually particularly useful in pcretest output, so
3015 I have just removed it.
3016
3017 8. Three pre-processor statements accidentally did not start in column 1.
3018 Sadly, there are *still* compilers around that complain, even though
3019 standard C has not required this for well over a decade. Sigh.
3020
3021 9. In pcretest, the code for checking callouts passed small integers in the
3022 callout_data field, which is a void * field. However, some picky compilers
3023 complained about the casts involved for this on 64-bit systems. Now
3024 pcretest passes the address of the small integer instead, which should get
3025 rid of the warnings.
3026
3027 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
3028 both compile and run time, and gives an error if an invalid UTF-8 sequence
3029 is found. There is a option for disabling this check in cases where the
3030 string is known to be correct and/or the maximum performance is wanted.
3031
3032 11. In response to a bug report, I changed one line in Makefile.in from
3033
3034 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
3035 to
3036 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
3037
3038 to look similar to other lines, but I have no way of telling whether this
3039 is the right thing to do, as I do not use Windows. No doubt I'll get told
3040 if it's wrong...
3041
3042
3043 Version 4.3 21-May-03
3044 ---------------------
3045
3046 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
3047 Makefile.
3048
3049 2. Some refactoring to improve the quality of the code:
3050
3051 (i) The utf8_table... variables are now declared "const".
3052
3053 (ii) The code for \cx, which used the "case flipping" table to upper case
3054 lower case letters, now just substracts 32. This is ASCII-specific,
3055 but the whole concept of \cx is ASCII-specific, so it seems
3056 reasonable.
3057
3058 (iii) PCRE was using its character types table to recognize decimal and
3059 hexadecimal digits in the pattern. This is silly, because it handles
3060 only 0-9, a-f, and A-F, but the character types table is locale-
3061 specific, which means strange things might happen. A private
3062 table is now used for this - though it costs 256 bytes, a table is
3063 much faster than multiple explicit tests. Of course, the standard
3064 character types table is still used for matching digits in subject
3065 strings against \d.
3066
3067 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
3068 ending in _t are). So I've renamed it as ESC_tee.
3069
3070 3. The first argument for regexec() in the POSIX wrapper should have been
3071 defined as "const".
3072
3073 4. Changed pcretest to use malloc() for its buffers so that they can be
3074 Electric Fenced for debugging.
3075
3076 5. There were several places in the code where, in UTF-8 mode, PCRE would try
3077 to read one or more bytes before the start of the subject string. Often this
3078 had no effect on PCRE's behaviour, but in some circumstances it could
3079 provoke a segmentation fault.
3080
3081 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
3082 to try to read one or more bytes before the start of the subject string.
3083
3084 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
3085 UTF-8 support could misbehave in various ways if the subject string
3086 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
3087 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
3088 back over UTF-8 characters.)
3089
3090
3091 Version 4.2 14-Apr-03
3092 ---------------------
3093
3094 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
3095
3096 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
3097 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
3098 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
3099 [WIN_PREFIX]: new variable, "cyg" for Cygwin
3100 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
3101 and BUILD_EXEEXT
3102 Note: automatic setting of the BUILD variables is not yet working
3103 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
3104 compile-time but not at link-time
3105 [LINK]: use for linking executables only
3106 make different versions for Windows and non-Windows
3107 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
3108 libraries
3109 [LINK_FOR_BUILD]: new variable
3110 [OBJEXT]: use throughout
3111 [EXEEXT]: use throughout
3112 <winshared>: new target
3113 <wininstall>: new target
3114 <dftables.o>: use native compiler
3115 <dftables>: use native linker
3116 <install>: handle Windows platform correctly
3117 <clean>: ditto
3118 <check>: ditto
3119 copy DLL to top builddir before testing
3120
3121 As part of these changes, -no-undefined was removed again. This was reported
3122 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
3123 in any case.
3124
3125 3. Some tidies to get rid of compiler warnings:
3126
3127 . In the match_data structure, match_limit was an unsigned long int, whereas
3128 match_call_count was an int. I've made them both unsigned long ints.
3129
3130 . In pcretest the fact that a const uschar * doesn't automatically cast to
3131 a void * provoked a warning.
3132
3133 . Turning on some more compiler warnings threw up some "shadow" variables
3134 and a few more missing casts.
3135
3136 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
3137 option, a class that contained a single character with a value between 128
3138 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
3139
3140 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
3141 option, a class that contained several characters, but with at least one
3142 whose value was between 128 and 255 caused PCRE to crash.
3143
3144
3145 Version 4.1 12-Mar-03
3146 ---------------------
3147
3148 1. Compiling with gcc -pedantic found a couple of places where casts were
3149 needed, and a string in dftables.c that was longer than standard compilers are
3150 required to support.
3151
3152 2. Compiling with Sun's compiler found a few more places where the code could
3153 be tidied up in order to avoid warnings.
3154
3155 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
3156 first of these names is deprecated in the latest Autoconf in favour of the name
3157 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
3158 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
3159 analogy I have changed it to CFLAGS_FOR_BUILD.
3160
3161 4. Added -no-undefined to the linking command in the Makefile, because this is
3162 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
3163 linking step for the pcreposix library.
3164
3165 5. PCRE was failing to diagnose the case of two named groups with the same
3166 name.
3167
3168 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
3169 literal character that is needed in the subject for a match, and scans along to
3170 ensure that it is present before embarking on the full matching process. This
3171 saves time in cases of nested unlimited repeats that are never going to match.
3172 Problem: the scan can take a lot of time if the subject is very long (e.g.
3173 megabytes), thus penalizing straightforward matches. It is now done only if the
3174 amount of subject to be scanned is less than 1000 bytes.
3175
3176 7. A lesser problem with the same optimization is that it was recording the
3177 first character of an anchored pattern as "needed", thus provoking a search
3178 right along the subject, even when the first match of the pattern was going to
3179 fail. The "needed" character is now not set for anchored patterns, unless it
3180 follows something in the pattern that is of non-fixed length. Thus, it still
3181 fulfils its original purpose of finding quick non-matches in cases of nested
3182 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
3183
3184
3185 Version 4.0 17-Feb-03
3186 ---------------------
3187
3188 1. If a comment in an extended regex that started immediately after a meta-item
3189 extended to the end of string, PCRE compiled incorrect data. This could lead to
3190 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
3191
3192 2. Moved to autoconf 2.53 and libtool 1.4.2.
3193
3194 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
3195 the special perltest8 script is no longer needed - all the tests can be run
3196 from a single perltest script.
3197
3198 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
3199 by \s. It has now been removed in PCRE. This means it isn't recognized as
3200 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
3201 class [:space:] *does* include VT, thereby creating a mess.
3202
3203 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
3204 space and tab.
3205
3206 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
3207 its new features into the main test script, reducing the number of scripts.
3208
3209 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
3210 were backward compatible, and made the (?i) apply to the whole pattern, as if
3211 /i were given. Now it behaves more logically, and applies the option setting
3212 only to what follows. PCRE has been changed to follow suit. However, if it
3213 finds options settings right at the start of the pattern, it extracts them into
3214 the global options, as before. Thus, they show up in the info data.
3215
3216 8. Added support for the \Q...\E escape sequence. Characters in between are
3217 treated as literals. This is slightly different from Perl in that $ and @ are
3218 also handled as literals inside the quotes. In Perl, they will cause variable
3219 interpolation. Note the following examples:
3220
3221 Pattern PCRE matches Perl matches
3222
3223 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
3224 \Qabc\$xyz\E abc\$xyz abc\$xyz
3225 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
3226
3227 For compatibility with Perl, \Q...\E sequences are recognized inside character
3228 classes as well as outside them.
3229
3230 9. Re-organized 3 code statements in pcretest to avoid "overflow in
3231 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
3232 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
3233 signed/unsigned warnings.
3234
3235 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
3236 option for pcretest, so I've replaced it by a simple function that does just
3237 that job.
3238
3239 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
3240 "pcregrep -".
3241
3242 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
3243 Java package. This provides some syntactic sugar for simple cases of what my
3244 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
3245 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
3246 item, you can use this simplified notation. Note that only makes sense with
3247 greedy quantifiers. Consequently, the use of the possessive quantifier forces
3248 greediness, whatever the setting of the PCRE_UNGREEDY option.
3249
3250 13. A change of greediness default within a pattern was not taking effect at
3251 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
3252 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
3253 was abstracted outside.
3254
3255 14. PCRE now supports the \G assertion. It is true when the current matching
3256 position is at the start point of the match. This differs from \A when the
3257 starting offset is non-zero. Used with the /g option of pcretest (or similar
3258 code), it works in the same way as it does for Perl's /g option. If all
3259 alternatives of a regex begin with \G, the expression is anchored to the start
3260 match position, and the "anchored" flag is set in the compiled expression.
3261
3262 15. Some bugs concerning the handling of certain option changes within patterns
3263 have been fixed. These applied to options other than (?ims). For example,
3264 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
3265 been the other way round. Some of this was related to change 7 above.
3266
3267 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
3268 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
3269 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
3270 POSIX classes only within a class (e.g. /[[:alpha:]]/).
3271
3272 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
3273 mode. Unlike ".", it always matches newline, whatever the setting of
3274 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
3275 assertions. Perl allows it, but it doesn't (in general) work because it can't
3276 calculate the length of the lookbehind. At least, that's the case for Perl
3277 5.8.0 - I've been told they are going to document that it doesn't work in
3278 future.
3279
3280 18. Added an error diagnosis for escapes that PCRE does not support: these are
3281 \L, \l, \N, \P, \p, \U, \u, and \X.
3282
3283 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
3284 reading past the end of the pattern in cases such as /[abcd/.
3285
3286 20. PCRE was getting more memory than necessary for patterns with classes that
3287 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
3288
3289 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
3290 compiling PCRE for use with Virtual Pascal.
3291
3292 22. Small fix to the Makefile to make it work properly if the build is done
3293 outside the source tree.
3294
3295 23. Added a new extension: a condition to go with recursion. If a conditional
3296 subpattern starts with (?(R) the "true" branch is used if recursion has
3297 happened, whereas the "false" branch is used only at the top level.
3298
3299 24. When there was a very long string of literal characters (over 255 bytes
3300 without UTF support, over 250 bytes with UTF support), the computation of how
3301 much memory was required could be incorrect, leading to segfaults or other
3302 strange effects.
3303
3304 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
3305 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
3306 there was a subsequent back reference to those brackets. This meant that, for
3307 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
3308 possible to check for precisely this case. All we can do is abandon the
3309 optimization if .* occurs inside capturing brackets when there are any back
3310 references whatsoever. (See below for a better fix that came later.)
3311
3312 26. The handling of the optimization for finding the first character of a
3313 non-anchored pattern, and for finding a character that is required later in the
3314 match were failing in some cases. This didn't break the matching; it just
3315 failed to optimize when it could. The way this is done has been re-implemented.
3316
3317 27. Fixed typo in error message for invalid (?R item (it said "(?p").
3318
3319 28. Added a new feature that provides some of the functionality that Perl
3320 provides with (?{...}). The facility is termed a "callout". The way it is done
3321 in PCRE is for the caller to provide an optional function, by setting
3322 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
3323 global variable. By default it is unset, which disables all calling out. To get
3324 the function called, the regex must include (?C) at appropriate points. This
3325 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
3326 This provides a means of identifying different callout points. When PCRE
3327 reaches such a point in the regex, if pcre_callout has been set, the external
3328 function is called. It is provided with data in a structure called
3329 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
3330 matching continues; if it returns a non-zero value, the match at the current
3331 point fails. However, backtracking will occur if possible. [This was changed
3332 later and other features added - see item 49 below.]
3333
3334 29. pcretest is upgraded to test the callout functionality. It provides a
3335 callout function that displays information. By default, it shows the start of
3336 the match and the current position in the text. There are some new data escapes
3337 to vary what happens:
3338
3339 \C+ in addition, show current contents of captured substrings
3340 \C- do not supply a callout function
3341 \C!n return 1 when callout number n is reached
3342 \C!n!m return 1 when callout number n is reached for the mth time
3343
3344 30. If pcregrep was called with the -l option and just a single file name, it
3345 output "<stdin>" if a match was found, instead of the file name.
3346
3347 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
3348 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
3349 pcre_exec(). This saves a malloc/free per call. The default value of
3350 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
3351 when configuring.
3352
3353 32. The default maximum size of a compiled pattern is 64K. There have been a
3354 few cases of people hitting this limit. The code now uses macros to handle the
3355 storing of links as offsets within the compiled pattern. It defaults to 2-byte
3356 links, but this can be changed to 3 or 4 bytes by --with-link-size when
3357 configuring. Tests 2 and 5 work only with 2-byte links because they output
3358 debugging information about compiled patterns.
3359
3360 33. Internal code re-arrangements:
3361
3362 (a) Moved the debugging function for printing out a compiled regex into
3363 its own source file (printint.c) and used #include to pull it into
3364 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
3365 separate copies.
3366
3367 (b) Defined the list of op-code names for debugging as a macro in
3368 internal.h so that it is next to the definition of the opcodes.
3369
3370 (c) Defined a table of op-code lengths for simpler skipping along compiled
3371 code. This is again a macro in internal.h so that it is next to the
3372 definition of the opcodes.
3373
3374 34. Added support for recursive calls to individual subpatterns, along the
3375 lines of Robin Houston's patch (but implemented somewhat differently).
3376
3377 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
3378 allow it to read and process whole directories in Win32. This code was
3379 contributed by Lionel Fourquaux; it has not been tested by me.
3380
3381 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
3382 used to name a group. Names consist of alphanumerics and underscores, and must
3383 be unique. Back references use the syntax (?P=name) and recursive calls use
3384 (?P>name) which is a PCRE extension to the Python extension. Groups still have
3385 numbers. The function pcre_fullinfo() can be used after compilation to extract
3386 a name/number map. There are three relevant calls:
3387
3388 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
3389 PCRE_INFO_NAMECOUNT yields the number of entries
3390 PCRE_INFO_NAMETABLE yields a pointer to the map.
3391
3392 The map is a vector of fixed-size entries. The size of each entry depends on
3393 the length of the longest name used. The first two bytes of each entry are the
3394 group number, most significant byte first. There follows the corresponding
3395 name, zero terminated. The names are in alphabetical order.
3396
3397 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
3398 case instead of 255. Making it the same both with and without UTF-8 support
3399 means that the same test output works with both.
3400
3401 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
3402 calling malloc() with a zero argument.
3403
3404 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
3405 optimization. I've improved things by keeping a bitmap of backreferences with
3406 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
3407 fact referenced, the optimization can be applied. It is unlikely that a
3408 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
3409 the match to follow \n) will appear inside brackets with a number greater than
3410 31, but if it does, any back reference > 31 suppresses the optimization.
3411
3412 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
3413 of disabling numbered capturing parentheses. Any opening parenthesis that is
3414 not followed by ? behaves as if it were followed by ?: but named parentheses
3415 can still be used for capturing (and they will acquire numbers in the usual
3416 way).
3417
3418 41. Redesigned the return codes from the match() function into yes/no/error so
3419 that errors can be passed back from deep inside the nested calls. A malloc
3420 failure while inside a recursive subpattern call now causes the
3421 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
3422
3423 42. It is now possible to set a limit on the number of times the match()
3424 function is called in a call to pcre_exec(). This facility makes it possible to
3425 limit the amount of recursion and backtracking, though not in a directly
3426 obvious way, because the match() function is used in a number of different
3427 circumstances. The count starts from zero for each position in the subject
3428 string (for non-anchored patterns). The default limit is, for compatibility, a
3429 large number, namely 10 000 000. You can change this in two ways:
3430
3431 (a) When configuring PCRE before making, you can use --with-match-limit=n
3432 to set a default value for the compiled library.
3433
3434 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
3435 a different value is set. See 45 below.
3436
3437 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3438
3439 43. Added a new function pcre_config(int, void *) to enable run-time extraction
3440 of things that can be changed at compile time. The first argument specifies
3441 what is wanted and the second points to where the information is to be placed.
3442 The current list of available information is:
3443
3444 PCRE_CONFIG_UTF8
3445
3446 The output is an integer that is set to one if UTF-8 support is available;
3447 otherwise it is set to zero.
3448
3449 PCRE_CONFIG_NEWLINE
3450
3451 The output is an integer that it set to the value of the code that is used for
3452 newline. It is either LF (10) or CR (13).
3453
3454 PCRE_CONFIG_LINK_SIZE
3455
3456 The output is an integer that contains the number of bytes used for internal
3457 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3458
3459 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
3460
3461 The output is an integer that contains the threshold above which the POSIX
3462 interface uses malloc() for output vectors. See item 31 above.
3463
3464 PCRE_CONFIG_MATCH_LIMIT
3465
3466 The output is an unsigned integer that contains the default limit of the number
3467 of match() calls in a pcre_exec() execution. See 42 above.
3468
3469 44. pcretest has been upgraded by the addition of the -C option. This causes it
3470 to extract all the available output from the new pcre_config() function, and to
3471 output it. The program then exits immediately.
3472
3473 45. A need has arisen to pass over additional data with calls to pcre_exec() in
3474 order to support additional features. One way would have been to define
3475 pcre_exec2() (for example) with extra arguments, but this would not have been
3476 extensible, and would also have required all calls to the original function to
3477 be mapped to the new one. Instead, I have chosen to extend the mechanism that
3478 is used for passing in "extra" data from pcre_study().
3479
3480 The pcre_extra structure is now exposed and defined in pcre.h. It currently
3481 contains the following fields:
3482
3483 flags a bitmap indicating which of the following fields are set
3484 study_data opaque data from pcre_study()
3485 match_limit a way of specifying a limit on match() calls for a specific
3486 call to pcre_exec()
3487 callout_data data for callouts (see 49 below)
3488
3489 The flag bits are also defined in pcre.h, and are
3490
3491 PCRE_EXTRA_STUDY_DATA
3492 PCRE_EXTRA_MATCH_LIMIT
3493 PCRE_EXTRA_CALLOUT_DATA
3494
3495 The pcre_study() function now returns one of these new pcre_extra blocks, with
3496 the actual study data pointed to by the study_data field, and the
3497 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3498 before. That is, this change is entirely upwards-compatible and requires no
3499 change to existing code.
3500
3501 If you want to pass in additional data to pcre_exec(), you can either place it
3502 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3503 block.
3504
3505 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3506 data string contains the escape sequence \M, pcretest calls pcre_exec() several
3507 times with different match limits, until it finds the minimum value needed for
3508 pcre_exec() to complete. The value is then output. This can be instructive; for
3509 most simple matches the number is quite small, but for pathological cases it
3510 gets very large very quickly.
3511
3512 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3513 returns the size of the data block pointed to by the study_data field in a
3514 pcre_extra block, that is, the value that was passed as the argument to
3515 pcre_malloc() when PCRE was getting memory in which to place the information
3516 created by pcre_study(). The fourth argument should point to a size_t variable.
3517 pcretest has been extended so that this information is shown after a successful
3518 pcre_study() call when information about the compiled regex is being displayed.
3519
3520 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3521 because what follows is always an absolute path. (Later: it turns out that this
3522 is more than cosmetic for MinGW, because it doesn't like empty path
3523 components.)
3524
3525 49. Some changes have been made to the callout feature (see 28 above):
3526
3527 (i) A callout function now has three choices for what it returns:
3528
3529 0 => success, carry on matching
3530 > 0 => failure at this point, but backtrack if possible
3531 < 0 => serious error, return this value from pcre_exec()
3532
3533 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3534 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3535 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3536 use by callout functions. It will never be used by PCRE itself.
3537
3538 (ii) The pcre_extra structure (see 45 above) has a void * field called
3539 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3540 pcre_callout_block structure has a field of the same name. The contents of
3541 the field passed in the pcre_extra structure are passed to the callout
3542 function in the corresponding field in the callout block. This makes it
3543 easier to use the same callout-containing regex from multiple threads. For
3544 testing, the pcretest program has a new data escape
3545
3546 \C*n pass the number n (may be negative) as callout_data
3547
3548 If the callout function in pcretest receives a non-zero value as
3549 callout_data, it returns that value.
3550
3551 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3552 there were some redundant $(CFLAGS) in commands that are now specified as
3553 $(LINK), which already includes $(CFLAGS).
3554
3555 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3556 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3557 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3558 one-byte characters throughout. Note that case-insensitive matching applies
3559 only to characters whose values are less than 256. PCRE doesn't support the
3560 notion of cases for higher-valued characters.
3561
3562 (i) A character class whose characters are all within 0-255 is handled as
3563 a bit map, and the map is inverted for negative classes. Previously, a
3564 character > 255 always failed to match such a class; however it should
3565 match if the class was a negative one (e.g. [^ab]). This has been fixed.
3566
3567 (ii) A negated character class with a single character < 255 is coded as
3568 "not this character" (OP_NOT). This wasn't working properly when the test
3569 character was multibyte, either singly or repeated.
3570
3571 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3572 mode, for example: \x{100}{2,3}.
3573
3574 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3575 singly or repeated) now correctly test multibyte characters. However,
3576 PCRE doesn't recognize any characters with values greater than 255 as
3577 digits, spaces, or word characters. Such characters always match \D, \S,
3578 and \W, and never match \d, \s, or \w.
3579
3580 (v) Classes may now contain characters and character ranges with values
3581 greater than 255. For example: [ab\x{100}-\x{400}].
3582
3583 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
3584 PCRE in UTF-8 mode.
3585
3586 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3587 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3588 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3589 value.)
3590
3591 53. The single man page has become too large. I have therefore split it up into
3592 a number of separate man pages. These also give rise to individual HTML pages;
3593 these are now put in a separate directory, and there is an index.html page that
3594 lists them all. Some hyperlinking between the pages has been installed.
3595
3596 54. Added convenience functions for handling named capturing parentheses.
3597
3598 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3599 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3600 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3601 are faulted.
3602
3603 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3604 calling configure. These values are used when compiling the dftables.c program
3605 which is run to generate the source of the default character tables. They
3606 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3607 you will need to set these values.
3608
3609 57. Updated the building process for Windows DLL, as provided by Fred Cox.
3610
3611
3612 Version 3.9 02-Jan-02
3613 ---------------------
3614
3615 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3616
3617 2. If --disable-static was given, the building process failed when trying to
3618 build pcretest and pcregrep. (For some reason it was using libtool to compile
3619 them, which is not right, as they aren't part of the library.)
3620
3621
3622 Version 3.8 18-Dec-01
3623 ---------------------
3624
3625 1. The experimental UTF-8 code was completely screwed up. It was packing the
3626 bytes in the wrong order. How dumb can you get?
3627
3628
3629 Version 3.7 29-Oct-01
3630 ---------------------
3631
3632 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3633 This caused pcretest, when used on the test data, to segfault. Unfortunately,
3634 this didn't happen under Solaris 8, where I normally test things.
3635
3636 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3637 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3638 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3639 this fix an hour or so after the initial 3.7 release.)
3640
3641
3642 Version 3.6 23-Oct-01
3643 ---------------------
3644
3645 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3646 offsets passed as NULL with zero offset count.
3647
3648 2. The config.guess and config.sub files had not been updated when I moved to
3649 the latest autoconf.
3650
3651
3652 Version 3.5 15-Aug-01
3653 ---------------------
3654
3655 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
3656 had been forgotten.
3657
3658 2. By using declared but undefined structures, we can avoid using "void"
3659 definitions in pcre.h while keeping the internal definitions of the structures
3660 private.
3661
3662 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
3663 user point of view, this means that both static and shared libraries are built
3664 by default, but this can be individually controlled. More of the work of
3665 handling this static/shared cases is now inside libtool instead of PCRE's make
3666 file.
3667
3668 4. The pcretest utility is now installed along with pcregrep because it is
3669 useful for users (to test regexs) and by doing this, it automatically gets
3670 relinked by libtool. The documentation has been turned into a man page, so
3671 there are now .1, .txt, and .html versions in /doc.
3672
3673 5. Upgrades to pcregrep:
3674 (i) Added long-form option names like gnu grep.
3675 (ii) Added --help to list all options with an explanatory phrase.
3676 (iii) Added -r, --recursive to recurse into sub-directories.
3677 (iv) Added -f, --file to read patterns from a file.
3678
3679 6. pcre_exec() was referring to its "code" argument before testing that
3680 argument for NULL (and giving an error if it was NULL).
3681
3682 7. Upgraded Makefile.in to allow for compiling in a different directory from
3683 the source directory.
3684
3685 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
3686 options bits, the pointer it was passed was to an int instead of to an unsigned
3687 long int. This mattered only on 64-bit systems.
3688
3689 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3690 generated) instead of pcre.in, which it its source. Also made the same change
3691 in several of the .c files.
3692
3693 10. A new release of gcc defines printf() as a macro, which broke pcretest
3694 because it had an ifdef in the middle of a string argument for printf(). Fixed
3695 by using separate calls to printf().
3696
3697 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
3698 script, to force use of CR or LF instead of \n in the source. On non-Unix
3699 systems, the value can be set in config.h.
3700
3701 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
3702 absolute limit. Changed the text of the error message to make this clear, and
3703 likewise updated the man page.
3704
3705 13. The limit of 99 on the number of capturing subpatterns has been removed.
3706 The new limit is 65535, which I hope will not be a "real" limit.
3707
3708
3709 Version 3.4 22-Aug-00
3710 ---------------------
3711
3712 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
3713
3714 2. Diagnose condition (?(0) as an error instead of crashing on matching.
3715
3716
3717 Version 3.3 01-Aug-00
3718 ---------------------
3719
3720 1. If an octal character was given, but the value was greater than \377, it
3721 was not getting masked to the least significant bits, as documented. This could
3722 lead to crashes in some systems.
3723
3724 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
3725 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
3726
3727 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
3728 These just pass their arguments on to (pcre_free)(), but they are provided
3729 because some uses of PCRE bind it to non-C systems that can call its functions,
3730 but cannot call free() or pcre_free() directly.
3731
3732 4. Add "make test" as a synonym for "make check". Corrected some comments in
3733 the Makefile.
3734
3735 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
3736 Makefile.
3737
3738 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
3739 command called pgrep for grepping around the active processes.
3740
3741 7. Added the beginnings of support for UTF-8 character strings.
3742
3743 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
3744 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
3745 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
3746 out for the ar command.)
3747
3748
3749 Version 3.2 12-May-00
3750 ---------------------
3751
3752 This is purely a bug fixing release.
3753
3754 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
3755 of ZA. This was just one example of several cases that could provoke this bug,
3756 which was introduced by change 9 of version 2.00. The code for breaking
3757 infinite loops after an iteration that matches an empty string was't working
3758 correctly.
3759
3760 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3761 when matched against abbab (for example). After matching an empty string, it
3762 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3763 caused it to match further down the string than it should.
3764
3765 3. The code contained an inclusion of sys/types.h. It isn't clear why this
3766 was there because it doesn't seem to be needed, and it causes trouble on some
3767 systems, as it is not a Standard C header. It has been removed.
3768
3769 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3770 were reported on the Macintosh. The changes were from
3771
3772 while ((c = *(++ptr)) != 0 && c != '\n');
3773 to
3774 while ((c = *(++ptr)) != 0 && c != '\n') ;
3775
3776 Totally extraordinary, but if that's what it takes...
3777
3778 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3779 available. Added HAVE_BCOPY and an autoconf test for it; if neither
3780 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3781 assumes the way PCRE uses memmove() (always moving upwards).
3782
3783 6. PCRE is being used in one environment where strchr() is not available. There
3784 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3785 faster code anyway.
3786
3787
3788 Version 3.1 09-Feb-00
3789 ---------------------
3790
3791 The only change in this release is the fixing of some bugs in Makefile.in for
3792 the "install" target:
3793
3794 (1) It was failing to install pcreposix.h.
3795
3796 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3797
3798
3799 Version 3.0 01-Feb-00
3800 ---------------------
3801
3802 1. Add support for the /+ modifier to perltest (to output $` like it does in
3803 pcretest).
3804
3805 2. Add support for the /g modifier to perltest.
3806
3807 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3808 matches null strings.
3809
3810 4. Fix perltest so that it doesn't do unwanted things when fed an empty
3811 pattern. Perl treats empty patterns specially - it reuses the most recent
3812 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3813 effect.
3814
3815 5. The POSIX interface was broken in that it was just handing over the POSIX
3816 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3817 required a bigger vector, with some working space on the end. This means that
3818 the POSIX wrapper now has to get and free some memory, and copy the results.
3819
3820 6. Added some simple autoconf support, placing the test data and the
3821 documentation in separate directories, re-organizing some of the
3822 information files, and making it build pcre-config (a GNU standard). Also added
3823 libtool support for building PCRE as a shared library, which is now the
3824 default.
3825
3826 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3827 09 are not valid octal constants. Single digits will be used for minor values
3828 less than 10.
3829
3830 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3831 existing programs that set these in the POSIX interface can use PCRE without
3832 modification.
3833
3834 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3835 return all that pcre_info() returns, plus additional data. The pcre_info()
3836 function is retained for compatibility, but is considered to be obsolete.
3837
3838 10. Added experimental recursion feature (?R) to handle one common case that
3839 Perl 5.6 will be able to do with (?p{...}).
3840
3841 11. Added support for POSIX character classes like [:alpha:], which Perl is
3842 adopting.
3843
3844
3845 Version 2.08 31-Aug-99
3846 ----------------------
3847
3848 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3849 trying to match at the startoffset position, but instead was moving forward to
3850 the next newline as if a previous match had failed.
3851
3852 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3853 and could get into a loop if a null string was matched other than at the start
3854 of the subject.
3855
3856 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3857 be distinguished at compile time, and for completeness also added PCRE_DATE.
3858
3859 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3860 in GnuWin32 environments.
3861
3862
3863 Version 2.07 29-Jul-99
3864 ----------------------
3865
3866 1. The documentation is now supplied in plain text form and HTML as well as in
3867 the form of man page sources.
3868
3869 2. C++ compilers don't like assigning (void *) values to other pointer types.
3870 In particular this affects malloc(). Although there is no problem in Standard
3871 C, I've put in casts to keep C++ compilers happy.
3872
3873 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3874 should be (const char *).
3875
3876 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3877 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3878 However, I haven't made this a standard facility. The documentation doesn't
3879 mention it, and the Makefile doesn't support it.
3880
3881 5. The Makefile now contains an "install" target, with editable destinations at
3882 the top of the file. The pcretest program is not installed.
3883
3884 6. pgrep -V now gives the PCRE version number and date.
3885
3886 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3887 causing the entire string to be ignored, instead of just the last character.
3888
3889 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3890 non-matching string, it can take a very, very long time, even for strings of
3891 quite modest length, because of the nested recursion. PCRE now does better in
3892 some of these cases. It does this by remembering the last required literal
3893 character in the pattern, and pre-searching the subject to ensure it is present
3894 before running the real match. In other words, it applies a heuristic to detect
3895 some types of certain failure quickly, and in the above example, if presented
3896 with a string that has no trailing " it gives "no match" very quickly.
3897
3898 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3899 other alternatives are tried instead.
3900
3901
3902 Version 2.06 09-Jun-99
3903 ----------------------
3904
3905 1. Change pcretest's output for amount of store used to show just the code
3906 space, because the remainder (the data block) varies in size between 32-bit and
3907 64-bit systems.
3908
3909 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3910 start matching at. This allows lookbehinds to work when searching for multiple
3911 occurrences in a string.
3912
3913 3. Added additional options to pcretest for testing multiple occurrences:
3914
3915 /+ outputs the rest of the string that follows a match
3916 /g loops for multiple occurrences, using the new startoffset argument
3917 /G loops for multiple occurrences by passing an incremented pointer
3918
3919 4. PCRE wasn't doing the "first character" optimization for patterns starting
3920 with \b or \B, though it was doing it for other lookbehind assertions. That is,
3921 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3922 the letter 'x'. On long subject strings, this gives a significant speed-up.
3923
3924
3925 Version 2.05 21-Apr-99
3926 ----------------------
3927
3928 1. Changed the type of magic_number from int to long int so that it works
3929 properly on 16-bit systems.
3930
3931 2. Fixed a bug which caused patterns starting with .* not to work correctly
3932 when the subject string contained newline characters. PCRE was assuming
3933 anchoring for such patterns in all cases, which is not correct because .* will
3934 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3935 DOTALL is set at top level; otherwise it knows that patterns starting with .*
3936 must be retried after every newline in the subject.
3937
3938
3939 Version 2.04 18-Feb-99
3940 ----------------------
3941
3942 1. For parenthesized subpatterns with repeats whose minimum was zero, the
3943 computation of the store needed to hold the pattern was incorrect (too large).
3944 If such patterns were nested a few deep, this could multiply and become a real
3945 problem.
3946
3947 2. Added /M option to pcretest to show the memory requirement of a specific
3948 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3949
3950 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3951 compiled in such a way that the backtracking after subsequent failure was
3952 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3953 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3954
3955
3956 Version 2.03 02-Feb-99
3957 ----------------------
3958
3959 1. Fixed typo and small mistake in man page.
3960
3961 2. Added 4th condition (GPL supersedes if conflict) and created separate
3962 LICENCE file containing the conditions.
3963
3964 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3965 Perl, that is the internal \ allows the delimiter to be included in the
3966 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3967 the final delimiter, add \ to the end of the pattern (to test the error).
3968
3969 4. Added the convenience functions for extracting substrings after a successful
3970 match. Updated pcretest to make it able to test these functions.
3971
3972
3973 Version 2.02 14-Jan-99
3974 ----------------------
3975
3976 1. Initialized the working variables associated with each extraction so that
3977 their saving and restoring doesn't refer to uninitialized store.
3978
3979 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3980 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3981 fix the problem.
3982
3983 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3984 calls, and wasn't printing the correct value for compiling calls. Increased the
3985 default value of LOOPREPEAT, and the number of significant figures in the
3986 times.
3987
3988 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3989
3990 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3991 a building problem on Windows NT with a FAT file system.
3992
3993
3994 Version 2.01 21-Oct-98
3995 ----------------------
3996
3997 1. Changed the API for pcre_compile() to allow for the provision of a pointer
3998 to character tables built by pcre_maketables() in the current locale. If NULL
3999 is passed, the default tables are used.
4000
4001
4002 Version 2.00 24-Sep-98
4003 ----------------------
4004
4005 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
4006 it any more.
4007
4008 2. Allow quantification of (?>) groups, and make it work correctly.
4009
4010 3. The first character computation wasn't working for (?>) groups.
4011
4012 4. Correct the implementation of \Z (it is permitted to match on the \n at the
4013 end of the subject) and add 5.005's \z, which really does match only at the
4014 very end of the subject.
4015
4016 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
4017
4018 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
4019 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
4020 localized options. All options to pcre_study() were also removed.
4021
4022 7. Add other new features from 5.005:
4023
4024 $(?<= positive lookbehind
4025 $(?<! negative lookbehind
4026 (?imsx-imsx) added the unsetting capability
4027 such a setting is global if at outer level; local otherwise
4028 (?imsx-imsx:) non-capturing groups with option setting
4029 (?(cond)re|re) conditional pattern matching
4030
4031 A backreference to itself in a repeated group matches the previous
4032 captured string.
4033
4034 8. General tidying up of studying (both automatic and via "study")
4035 consequential on the addition of new assertions.
4036
4037 9. As in 5.005, unlimited repeated groups that could match an empty substring
4038 are no longer faulted at compile time. Instead, the loop is forcibly broken at
4039 runtime if any iteration does actually match an empty substring.
4040
4041 10. Include the RunTest script in the distribution.
4042
4043 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
4044 discrepancies, some of which were old and were also with respect to 5.004. They
4045 have now been fixed.
4046
4047
4048 Version 1.09 28-Apr-98
4049 ----------------------
4050
4051 1. A negated single character class followed by a quantifier with a minimum
4052 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
4053 program crashes, or just wrong answers. This did not apply to negated classes
4054 containing more than one character, or to minima other than one.
4055
4056
4057 Version 1.08 27-Mar-98
4058 ----------------------
4059
4060 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
4061
4062 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
4063 latter must appear before anything that relies on it in the pattern.
4064
4065
4066 Version 1.07 16-Feb-98
4067 ----------------------
4068
4069 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
4070 repeat of a potentially empty string).
4071
4072
4073 Version 1.06 23-Jan-98
4074 ----------------------
4075
4076 1. Added Markus Oberhumer's little patches for C++.
4077
4078 2. Literal strings longer than 255 characters were broken.
4079
4080
4081 Version 1.05 23-Dec-97
4082 ----------------------
4083
4084 1. Negated character classes containing more than one character were failing if
4085 PCRE_CASELESS was set at run time.
4086
4087
4088 Version 1.04 19-Dec-97
4089 ----------------------
4090
4091 1. Corrected the man page, where some "const" qualifiers had been omitted.
4092
4093 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
4094 input syntax.
4095
4096 3. Fixed memory leak which occurred when a regex with back references was
4097 matched with an offsets vector that wasn't big enough. The temporary memory
4098 that is used in this case wasn't being freed if the match failed.
4099
4100 4. Tidied pcretest to ensure it frees memory that it gets.
4101
4102 5. Temporary memory was being obtained in the case where the passed offsets
4103 vector was exactly big enough.
4104
4105 6. Corrected definition of offsetof() from change 5 below.
4106
4107 7. I had screwed up change 6 below and broken the rules for the use of
4108 setjmp(). Now fixed.
4109
4110
4111 Version 1.03 18-Dec-97
4112 ----------------------
4113
4114 1. A erroneous regex with a missing opening parenthesis was correctly
4115 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
4116 on some systems.
4117
4118 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
4119 it was reported that one broken compiler failed on the former because "code" is
4120 also an independent variable.
4121
4122 3. The erroneous regex a[]b caused an array overrun reference.
4123
4124 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
4125 fail on data ending with that character. (It was going on too far, and checking
4126 the next character, typically a binary zero.) This was specific to the
4127 optimized code for single-character negative classes.
4128
4129 5. Added a contributed patch from the TIN world which does the following:
4130
4131 + Add an undef for memmove, in case the the system defines a macro for it.
4132
4133 + Add a definition of offsetof(), in case there isn't one. (I don't know
4134 the reason behind this - offsetof() is part of the ANSI standard - but
4135 it does no harm).
4136
4137 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
4138 most of the places where whitespace preceded '#'. I have given up and
4139 allowed the remaining 2 cases to be at the margin.
4140
4141 + Rename some variables in pcre to eliminate shadowing. This seems very
4142 pedantic, but does no harm, of course.
4143
4144 6. Moved the call to setjmp() into its own function, to get rid of warnings
4145 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
4146
4147 7. Constructs such as \d{8,} were compiling into the equivalent of
4148 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
4149 outcome, but in this particular case used more store than had been allocated,
4150 which caused the bug to be discovered because it threw up an internal error.
4151
4152 8. The debugging code in both pcre and pcretest for outputting the compiled
4153 form of a regex was going wrong in the case of back references followed by
4154 curly-bracketed repeats.
4155
4156
4157 Version 1.02 12-Dec-97
4158 ----------------------
4159
4160 1. Typos in pcre.3 and comments in the source fixed.
4161
4162 2. Applied a contributed patch to get rid of places where it used to remove
4163 'const' from variables, and fixed some signed/unsigned and uninitialized
4164 variable warnings.
4165
4166 3. Added the "runtest" target to Makefile.
4167
4168 4. Set default compiler flag to -O2 rather than just -O.
4169
4170
4171 Version 1.01 19-Nov-97
4172 ----------------------
4173
4174 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
4175 like /([ab]*)*/, that is, for classes with more than one character in them.
4176
4177 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
4178 as /((?>a*))*/ (a PCRE_EXTRA facility).
4179
4180
4181 Version 1.00 18-Nov-97
4182 ----------------------
4183
4184 1. Added compile-time macros to support systems such as SunOS4 which don't have
4185 memmove() or strerror() but have other things that can be used instead.
4186
4187 2. Arranged that "make clean" removes the executables.
4188
4189
4190 Version 0.99 27-Oct-97
4191 ----------------------
4192
4193 1. Fixed bug in code for optimizing classes with only one character. It was
4194 initializing a 32-byte map regardless, which could cause it to run off the end
4195 of the memory it had got.
4196
4197 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
4198
4199
4200 Version 0.98 22-Oct-97
4201 ----------------------
4202
4203 1. Fixed bug in code for handling temporary memory usage when there are more
4204 back references than supplied space in the ovector. This could cause segfaults.
4205
4206
4207 Version 0.97 21-Oct-97
4208 ----------------------
4209
4210 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
4211
4212 2. Optimized negated single characters not to use a bit map.
4213
4214 3. Brought error texts together as macro definitions; clarified some of them;
4215 fixed one that was wrong - it said "range out of order" when it meant "invalid
4216 escape sequence".
4217
4218 4. Changed some char * arguments to const char *.
4219
4220 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
4221
4222 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
4223 pcretest.
4224
4225
4226 Version 0.96 16-Oct-97
4227 ----------------------
4228
4229 1. Added a simple "pgrep" utility to the distribution.
4230
4231 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
4232 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
4233 where "ddd" means "one or more decimal digits".
4234
4235 3. Fixed serious bug. If a pattern had a back reference, but the call to
4236 pcre_exec() didn't supply a large enough ovector to record the related
4237 identifying subpattern, the match always failed. PCRE now remembers the number
4238 of the largest back reference, and gets some temporary memory in which to save
4239 the offsets during matching if necessary, in order to ensure that
4240 backreferences always work.
4241
4242 4. Increased the compatibility with Perl in a number of ways:
4243
4244 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
4245 to request this handling. The option can be set at compile or exec time.
4246
4247 (b) $ matches before a terminating newline by default; an option
4248 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
4249 mode). The option can be set at compile or exec time.
4250
4251 (c) The handling of \ followed by a digit other than 0 is now supposed to be
4252 the same as Perl's. If the decimal number it represents is less than 10
4253 or there aren't that many previous left capturing parentheses, an octal
4254 escape is read. Inside a character class, it's always an octal escape,
4255 even if it is a single digit.
4256
4257 (d) An escaped but undefined alphabetic character is taken as a literal,
4258 unless PCRE_EXTRA is set. Currently this just reserves the remaining
4259 escapes.
4260
4261 (e) {0} is now permitted. (The previous item is removed from the compiled
4262 pattern).
4263
4264 5. Changed all the names of code files so that the basic parts are no longer
4265 than 10 characters, and abolished the teeny "globals.c" file.
4266
4267 6. Changed the handling of character classes; they are now done with a 32-byte
4268 bit map always.
4269
4270 7. Added the -d and /D options to pcretest to make it possible to look at the
4271 internals of compilation without having to recompile pcre.
4272
4273
4274 Version 0.95 23-Sep-97
4275 ----------------------
4276
4277 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
4278 \x20 at the start of a run of normal characters. These were being treated as
4279 real characters, instead of the source characters being re-checked.
4280
4281
4282 Version 0.94 18-Sep-97
4283 ----------------------
4284
4285 1. The functions are now thread-safe, with the caveat that the global variables
4286 containing pointers to malloc() and free() or alternative functions are the
4287 same for all threads.
4288
4289 2. Get pcre_study() to generate a bitmap of initial characters for non-
4290 anchored patterns when this is possible, and use it if passed to pcre_exec().
4291
4292
4293 Version 0.93 15-Sep-97
4294 ----------------------
4295
4296 1. /(b)|(:+)/ was computing an incorrect first character.
4297
4298 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
4299 but not actually doing anything yet.
4300
4301 3. Treat "-" characters in classes that cannot be part of ranges as literals,
4302 as Perl does (e.g. [-az] or [az-]).
4303
4304 4. Set the anchored flag if a branch starts with .* or .*? because that tests
4305 all possible positions.
4306
4307 5. Split up into different modules to avoid including unneeded functions in a
4308 compiled binary. However, compile and exec are still in one module. The "study"
4309 function is split off.
4310
4311 6. The character tables are now in a separate module whose source is generated
4312 by an auxiliary program - but can then be edited by hand if required. There are
4313 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
4314 toupper() in the code.
4315
4316 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
4317 make them global. Abolish the function for setting them, as the caller can now
4318 set them directly.
4319
4320
4321 Version 0.92 11-Sep-97
4322 ----------------------
4323
4324 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
4325 (e.g. /a{1,3}/) was broken (I mis-optimized it).
4326
4327 2. Caseless matching was not working in character classes if the characters in
4328 the pattern were in upper case.
4329
4330 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
4331
4332 4. Make PCRE_ANCHORED public and accept as a compile option.
4333
4334 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
4335 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
4336 pass them.
4337
4338 6. Give an error if bad option bits passed at compile or run time.
4339
4340 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
4341 pcretest to cause it to pass that flag.
4342
4343 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
4344 options, and the first character, if set.
4345
4346 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
4347
4348
4349 Version 0.91 10-Sep-97
4350 ----------------------
4351
4352 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
4353 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
4354
4355 2. PCRE was looping on encountering an indefinitely repeated back reference to
4356 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
4357 Perl does - treats the match as successful.
4358
4359 ****

Properties

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

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12