ViewVC logotype

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 623 - (show annotations)
Tue Jul 19 09:58:42 2011 UTC (9 years, 4 months ago) by ph10
File size: 206887 byte(s)
Fix \X* bug when first character has the mark property. Also improve code for 
property and script handling.  
1 ChangeLog for PCRE
2 ------------------
4 Version 8.13 30-Apr-2011
5 ------------------------
7 1. The Unicode data tables have been updated to Unicode 6.0.0.
9 2. Two minor typos in pcre_internal.h have been fixed.
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).
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).
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.
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().
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.
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.
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.
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().
60 11. While implementing 10, a number of bugs in the handling of groups were
61 discovered and fixed:
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.
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.
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.
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).
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.
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.
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.
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.
107 19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
108 was incorrectly computed.
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.
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.
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.
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).
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.
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.
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.
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.
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.
155 28. Some minor code refactoring concerning Unicode properties and scripts
156 should reduce the stack requirement of match() slightly.
160 Version 8.12 15-Jan-2011
161 ------------------------
163 1. Fixed some typos in the markup of the man pages, and wrote a script that
164 checks for such things as part of the documentation building process.
166 2. On a big-endian 64-bit system, pcregrep did not correctly process the
167 --match-limit and --recursion-limit options (added for 8.11). In
168 particular, this made one of the standard tests fail. (The integer value
169 went into the wrong half of a long int.)
171 3. If the --colour option was given to pcregrep with -v (invert match), it
172 did strange things, either producing crazy output, or crashing. It should,
173 of course, ignore a request for colour when reporting lines that do not
174 match.
176 4. Another pcregrep bug caused similar problems if --colour was specified with
177 -M (multiline) and the pattern match finished with a line ending.
179 5. In pcregrep, when a pattern that ended with a literal newline sequence was
180 matched in multiline mode, the following line was shown as part of the
181 match. This seems wrong, so I have changed it.
183 6. Another pcregrep bug in multiline mode, when --colour was specified, caused
184 the check for further matches in the same line (so they could be coloured)
185 to overrun the end of the current line. If another match was found, it was
186 incorrectly shown (and then shown again when found in the next line).
188 7. If pcregrep was compiled under Windows, there was a reference to the
189 function pcregrep_exit() before it was defined. I am assuming this was
190 the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
191 reported by a user. I've moved the definition above the reference.
194 Version 8.11 10-Dec-2010
195 ------------------------
197 1. (*THEN) was not working properly if there were untried alternatives prior
198 to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
199 backtracked to try for "b" instead of moving to the next alternative branch
200 at the same level (in this case, to look for "c"). The Perl documentation
201 is clear that when (*THEN) is backtracked onto, it goes to the "next
202 alternative in the innermost enclosing group".
204 2. (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
205 such as (A(*COMMIT)B(*THEN)C|D) any failure after matching A should
206 result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
207 (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
208 (*THEN).
210 3. If \s appeared in a character class, it removed the VT character from
211 the class, even if it had been included by some previous item, for example
212 in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
213 of \s, but is part of the POSIX "space" class.)
215 4. A partial match never returns an empty string (because you can always
216 match an empty string at the end of the subject); however the checking for
217 an empty string was starting at the "start of match" point. This has been
218 changed to the "earliest inspected character" point, because the returned
219 data for a partial match starts at this character. This means that, for
220 example, /(?<=abc)def/ gives a partial match for the subject "abc"
221 (previously it gave "no match").
223 5. Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
224 of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
225 previously a full match would be given. However, setting PCRE_PARTIAL_HARD
226 has an implication that the given string is incomplete (because a partial
227 match is preferred over a full match). For this reason, these items now
228 give a partial match in this situation. [Aside: previously, the one case
229 /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
230 match rather than a full match, which was wrong by the old rules, but is
231 now correct.]
233 6. There was a bug in the handling of #-introduced comments, recognized when
234 PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
235 If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
236 UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
237 scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
238 but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
239 places in pcre_compile().
241 7. Related to (6) above, when pcre_compile() was skipping #-introduced
242 comments when looking ahead for named forward references to subpatterns,
243 the only newline sequence it recognized was NL. It now handles newlines
244 according to the set newline convention.
246 8. SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
247 former, but used strtoul(), whereas pcretest avoided strtoul() but did not
248 cater for a lack of strerror(). These oversights have been fixed.
250 9. Added --match-limit and --recursion-limit to pcregrep.
252 10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
254 11. When the -o option was used, pcregrep was setting a return code of 1, even
255 when matches were found, and --line-buffered was not being honoured.
257 12. Added an optional parentheses number to the -o and --only-matching options
258 of pcregrep.
260 13. Imitating Perl's /g action for multiple matches is tricky when the pattern
261 can match an empty string. The code to do it in pcretest and pcredemo
262 needed fixing:
264 (a) When the newline convention was "crlf", pcretest got it wrong, skipping
265 only one byte after an empty string match just before CRLF (this case
266 just got forgotten; "any" and "anycrlf" were OK).
268 (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
269 mode when an empty string match preceded an ASCII character followed by
270 a non-ASCII character. (The code for advancing by one character rather
271 than one byte was nonsense.)
273 (c) The pcredemo.c sample program did not have any code at all to handle
274 the cases when CRLF is a valid newline sequence.
276 14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
277 as a starting offset was within the subject string. There is now a new
278 error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
279 negative or greater than the length of the string. In order to test this,
280 pcretest is extended to allow the setting of negative starting offsets.
282 15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
283 starting offset points to the beginning of a UTF-8 character was
284 unnecessarily clumsy. I tidied it up.
286 16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
287 bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
289 17. Nobody had reported that the --include_dir option, which was added in
290 release 7.7 should have been called --include-dir (hyphen, not underscore)
291 for compatibility with GNU grep. I have changed it to --include-dir, but
292 left --include_dir as an undocumented synonym, and the same for
293 --exclude-dir, though that is not available in GNU grep, at least as of
294 release 2.5.4.
296 18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
297 characters from a string of bytes) have been redefined so as not to use
298 loops, in order to improve performance in some environments. At the same
299 time, I abstracted some of the common code into auxiliary macros to save
300 repetition (this should not affect the compiled code).
302 19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
303 compile-time error is now given if \c is not followed by an ASCII
304 character, that is, a byte less than 128. (In EBCDIC mode, the code is
305 different, and any byte value is allowed.)
307 20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
308 START_OPTIMIZE option, which is now allowed at compile time - but just
309 passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
310 to pcregrep and other applications that have no direct access to PCRE
311 options. The new /Y option in pcretest sets this option when calling
312 pcre_compile().
314 21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
315 back references. Groups containing recursive back references were forced to
316 be atomic by that change, but in the case of named groups, the amount of
317 memory required was incorrectly computed, leading to "Failed: internal
318 error: code overflow". This has been fixed.
320 22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
321 pcretest.c, to avoid build problems in some Borland environments.
324 Version 8.10 25-Jun-2010
325 ------------------------
327 1. Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
328 THEN.
330 2. (*ACCEPT) was not working when inside an atomic group.
332 3. Inside a character class, \B is treated as a literal by default, but
333 faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
334 causes the error). The code is unchanged, but I tidied the documentation.
336 4. Inside a character class, PCRE always treated \R and \X as literals,
337 whereas Perl faults them if its -w option is set. I have changed PCRE so
338 that it faults them when PCRE_EXTRA is set.
340 5. Added support for \N, which always matches any character other than
341 newline. (It is the same as "." when PCRE_DOTALL is not set.)
343 6. When compiling pcregrep with newer versions of gcc which may have
344 FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
345 declared with attribute warn_unused_result" were given. Just casting the
346 result to (void) does not stop the warnings; a more elaborate fudge is
347 needed. I've used a macro to implement this.
349 7. Minor change to pcretest.c to avoid a compiler warning.
351 8. Added four artifical Unicode properties to help with an option to make
352 \s etc use properties (see next item). The new properties are: Xan
353 (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
355 9. Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
356 use Unicode properties. (*UCP) at the start of a pattern can be used to set
357 this option. Modified pcretest to add /W to test this facility. Added
358 REG_UCP to make it available via the POSIX interface.
360 10. Added --line-buffered to pcregrep.
362 11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
363 studied, and the match started with a letter with a code point greater than
364 127 whose first byte was different to the first byte of the other case of
365 the letter, the other case of this starting letter was not recognized
366 (#976).
368 12. If a pattern that was studied started with a repeated Unicode property
369 test, for example, \p{Nd}+, there was the theoretical possibility of
370 setting up an incorrect bitmap of starting bytes, but fortunately it could
371 not have actually happened in practice until change 8 above was made (it
372 added property types that matched character-matching opcodes).
374 13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
375 possible starting bytes for non-anchored patterns.
377 14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
378 \R, and also a number of cases that involve Unicode properties, both
379 explicit and implicit when PCRE_UCP is set.
381 15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
382 input, it could crash or give wrong results if characters with values
383 greater than 0xc0 were present in the subject string. (Detail: it assumed
384 UTF-8 input when processing these items.)
386 16. Added a lot of (int) casts to avoid compiler warnings in systems where
387 size_t is 64-bit (#991).
389 17. Added a check for running out of memory when PCRE is compiled with
390 --disable-stack-for-recursion (#990).
392 18. If the last data line in a file for pcretest does not have a newline on
393 the end, a newline was missing in the output.
395 19. The default pcre_chartables.c file recognizes only ASCII characters (values
396 less than 128) in its various bitmaps. However, there is a facility for
397 generating tables according to the current locale when PCRE is compiled. It
398 turns out that in some environments, 0x85 and 0xa0, which are Unicode space
399 characters, are recognized by isspace() and therefore were getting set in
400 these tables, and indeed these tables seem to approximate to ISO 8859. This
401 caused a problem in UTF-8 mode when pcre_study() was used to create a list
402 of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
403 which of course cannot start UTF-8 characters. I have changed the code so
404 that only real ASCII characters (less than 128) and the correct starting
405 bytes for UTF-8 encodings are set for characters greater than 127 when in
406 UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
407 altogether.)
409 20. Added the /T option to pcretest so as to be able to run tests with non-
410 standard character tables, thus making it possible to include the tests
411 used for 19 above in the standard set of tests.
413 21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
414 reference to a subpattern the other side of a comment that contains an
415 opening parenthesis caused either an internal compiling error, or a
416 reference to the wrong subpattern.
419 Version 8.02 19-Mar-2010
420 ------------------------
422 1. The Unicode data tables have been updated to Unicode 5.2.0.
424 2. Added the option --libs-cpp to pcre-config, but only when C++ support is
425 configured.
427 3. Updated the licensing terms in the pcregexp.pas file, as agreed with the
428 original author of that file, following a query about its status.
430 4. On systems that do not have stdint.h (e.g. Solaris), check for and include
431 inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
433 5. A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
434 quantifier applied to a forward-referencing subroutine call, could compile
435 incorrect code or give the error "internal error: previously-checked
436 referenced subpattern not found".
438 6. Both MS Visual Studio and Symbian OS have problems with initializing
439 variables to point to external functions. For these systems, therefore,
440 pcre_malloc etc. are now initialized to local functions that call the
441 relevant global functions.
443 7. There were two entries missing in the vectors called coptable and poptable
444 in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
445 I've fixed the data, and added a kludgy way of testing at compile time that
446 the lengths are correct (equal to the number of opcodes).
448 8. Following on from 7, I added a similar kludge to check the length of the
449 eint vector in pcreposix.c.
451 9. Error texts for pcre_compile() are held as one long string to avoid too
452 much relocation at load time. To find a text, the string is searched,
453 counting zeros. There was no check for running off the end of the string,
454 which could happen if a new error number was added without updating the
455 string.
457 10. \K gave a compile-time error if it appeared in a lookbehind assersion.
459 11. \K was not working if it appeared in an atomic group or in a group that
460 was called as a "subroutine", or in an assertion. Perl 5.11 documents that
461 \K is "not well defined" if used in an assertion. PCRE now accepts it if
462 the assertion is positive, but not if it is negative.
464 12. Change 11 fortuitously reduced the size of the stack frame used in the
465 "match()" function of pcre_exec.c by one pointer. Forthcoming
466 implementation of support for (*MARK) will need an extra pointer on the
467 stack; I have reserved it now, so that the stack frame size does not
468 decrease.
470 13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
471 item in branch that calls a recursion is a subroutine call - as in the
472 second branch in the above example - was incorrectly given the compile-
473 time error "recursive call could loop indefinitely" because pcre_compile()
474 was not correctly checking the subroutine for matching a non-empty string.
476 14. The checks for overrunning compiling workspace could trigger after an
477 overrun had occurred. This is a "should never occur" error, but it can be
478 triggered by pathological patterns such as hundreds of nested parentheses.
479 The checks now trigger 100 bytes before the end of the workspace.
481 15. Fix typo in configure.ac: "srtoq" should be "strtoq".
484 Version 8.01 19-Jan-2010
485 ------------------------
487 1. If a pattern contained a conditional subpattern with only one branch (in
488 particular, this includes all (*DEFINE) patterns), a call to pcre_study()
489 computed the wrong minimum data length (which is of course zero for such
490 subpatterns). This could cause incorrect "no match" results.
492 2. For patterns such as (?i)a(?-i)b|c where an option setting at the start of
493 the pattern is reset in the first branch, pcre_compile() failed with
494 "internal error: code overflow at offset...". This happened only when
495 the reset was to the original external option setting. (An optimization
496 abstracts leading options settings into an external setting, which was the
497 cause of this.)
499 3. A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
500 of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
501 assertion pattern did not match (meaning that the assertion was true), it
502 was incorrectly treated as false if the SKIP had been reached during the
503 matching. This also applied to assertions used as conditions.
505 4. If an item that is not supported by pcre_dfa_exec() was encountered in an
506 assertion subpattern, including such a pattern used as a condition,
507 unpredictable results occurred, instead of the error return
510 5. The C++ GlobalReplace function was not working like Perl for the special
511 situation when an empty string is matched. It now does the fancy magic
512 stuff that is necessary.
514 6. In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
515 removed. (These were left over from very, very early versions of PCRE.)
517 7. Some cosmetic changes to the code to make life easier when compiling it
518 as part of something else:
520 (a) Change DEBUG to PCRE_DEBUG.
522 (b) In pcre_compile(), rename the member of the "branch_chain" structure
523 called "current" as "current_branch", to prevent a collision with the
524 Linux macro when compiled as a kernel module.
526 (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
527 prevent a collision with the Linux macro when compiled as a kernel
528 module.
530 8. In pcre_compile() there are some checks for integer overflows that used to
531 cast potentially large values to (double). This has been changed to that
532 when building, a check for int64_t is made, and if it is found, it is used
533 instead, thus avoiding the use of floating point arithmetic. (There is no
534 other use of FP in PCRE.) If int64_t is not found, the fallback is to
535 double.
537 9. Added two casts to avoid signed/unsigned warnings from VS Studio Express
538 2005 (difference between two addresses compared to an unsigned value).
540 10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
541 custom one, because of the following reported problem in Windows:
543 - libbz2 uses the Pascal calling convention (WINAPI) for the functions
544 under Win32.
545 - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
546 therefore missing the function definition.
547 - The compiler thus generates a "C" signature for the test function.
548 - The linker fails to find the "C" function.
549 - PCRE fails to configure if asked to do so against libbz2.
551 11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
552 messages were output:
554 Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
555 rerunning libtoolize, to keep the correct libtool macros in-tree.
556 Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
558 I have done both of these things.
560 12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
561 most of the time, it *can* run out if it is given a pattern that contains a
562 runaway infinite recursion. I updated the discussion in the pcrestack man
563 page.
565 13. Now that we have gone to the x.xx style of version numbers, the minor
566 version may start with zero. Using 08 or 09 is a bad idea because users
567 might check the value of PCRE_MINOR in their code, and 08 or 09 may be
568 interpreted as invalid octal numbers. I've updated the previous comment in
569 configure.ac, and also added a check that gives an error if 08 or 09 are
570 used.
572 14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
573 causing partial matching to fail when the end of the subject matched \W
574 in a UTF-8 pattern where \W was quantified with a minimum of 3.
576 15. There were some discrepancies between the declarations in pcre_internal.h
577 of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
578 their definitions. The declarations used "const uschar *" and the
579 definitions used USPTR. Even though USPTR is normally defined as "const
580 unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
581 reported that: "This difference in casting confuses some C++ compilers, for
582 example, SunCC recognizes above declarations as different functions and
583 generates broken code for hbpcre." I have changed the declarations to use
584 USPTR.
586 16. GNU libtool is named differently on some systems. The autogen.sh script now
587 tries several variants such as glibtoolize (MacOSX) and libtoolize1x
588 (FreeBSD).
590 17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
591 (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
592 comment: "Figure out how to create a longlong from a string: strtoll and
593 equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
594 instance, but it only takes 2 args instead of 3!"
596 18. A subtle bug concerned with back references has been fixed by a change of
597 specification, with a corresponding code fix. A pattern such as
598 ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
599 refers, was giving matches when it shouldn't. For example, xa=xaaa would
600 match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
601 same bug. Such groups have to be quantified to be useful, or contained
602 inside another quantified group. (If there's no repetition, the reference
603 can never match.) The problem arises because, having left the group and
604 moved on to the rest of the pattern, a later failure that backtracks into
605 the group uses the captured value from the final iteration of the group
606 rather than the correct earlier one. I have fixed this in PCRE by forcing
607 any group that contains a reference to itself to be an atomic group; that
608 is, there cannot be any backtracking into it once it has completed. This is
609 similar to recursive and subroutine calls.
612 Version 8.00 19-Oct-09
613 ----------------------
615 1. The table for translating pcre_compile() error codes into POSIX error codes
616 was out-of-date, and there was no check on the pcre_compile() error code
617 being within the table. This could lead to an OK return being given in
618 error.
620 2. Changed the call to open a subject file in pcregrep from fopen(pathname,
621 "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
622 in a Windows environment.
624 3. The pcregrep --count option prints the count for each file even when it is
625 zero, as does GNU grep. However, pcregrep was also printing all files when
626 --files-with-matches was added. Now, when both options are given, it prints
627 counts only for those files that have at least one match. (GNU grep just
628 prints the file name in this circumstance, but including the count seems
629 more useful - otherwise, why use --count?) Also ensured that the
630 combination -clh just lists non-zero counts, with no names.
632 4. The long form of the pcregrep -F option was incorrectly implemented as
633 --fixed_strings instead of --fixed-strings. This is an incompatible change,
634 but it seems right to fix it, and I didn't think it was worth preserving
635 the old behaviour.
637 5. The command line items --regex=pattern and --regexp=pattern were not
638 recognized by pcregrep, which required --regex pattern or --regexp pattern
639 (with a space rather than an '='). The man page documented the '=' forms,
640 which are compatible with GNU grep; these now work.
642 6. No libpcreposix.pc file was created for pkg-config; there was just
643 libpcre.pc and libpcrecpp.pc. The omission has been rectified.
645 7. Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
646 when UCP support is not needed, by modifying the Python script that
647 generates it from Unicode data files. This should not matter if the module
648 is correctly used as a library, but I received one complaint about 50K of
649 unwanted data. My guess is that the person linked everything into his
650 program rather than using a library. Anyway, it does no harm.
652 8. A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
653 was a minimum greater than 1 for a wide character in a possessive
654 repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
655 which had an unlimited repeat of a nested, fixed maximum repeat of a wide
656 character. Chaos in the form of incorrect output or a compiling loop could
657 result.
659 9. The restrictions on what a pattern can contain when partial matching is
660 requested for pcre_exec() have been removed. All patterns can now be
661 partially matched by this function. In addition, if there are at least two
662 slots in the offset vector, the offset of the earliest inspected character
663 for the match and the offset of the end of the subject are set in them when
664 PCRE_ERROR_PARTIAL is returned.
666 10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
667 synonymous with PCRE_PARTIAL, for backwards compatibility, and
668 PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
669 and may be more useful for multi-segment matching.
671 11. Partial matching with pcre_exec() is now more intuitive. A partial match
672 used to be given if ever the end of the subject was reached; now it is
673 given only if matching could not proceed because another character was
674 needed. This makes a difference in some odd cases such as Z(*FAIL) with the
675 string "Z", which now yields "no match" instead of "partial match". In the
676 case of pcre_dfa_exec(), "no match" is given if every matching path for the
677 final character ended with (*FAIL).
679 12. Restarting a match using pcre_dfa_exec() after a partial match did not work
680 if the pattern had a "must contain" character that was already found in the
681 earlier partial match, unless partial matching was again requested. For
682 example, with the pattern /dog.(body)?/, the "must contain" character is
683 "g". If the first part-match was for the string "dog", restarting with
684 "sbody" failed. This bug has been fixed.
686 13. The string returned by pcre_dfa_exec() after a partial match has been
687 changed so that it starts at the first inspected character rather than the
688 first character of the match. This makes a difference only if the pattern
689 starts with a lookbehind assertion or \b or \B (\K is not supported by
690 pcre_dfa_exec()). It's an incompatible change, but it makes the two
691 matching functions compatible, and I think it's the right thing to do.
693 14. Added a pcredemo man page, created automatically from the pcredemo.c file,
694 so that the demonstration program is easily available in environments where
695 PCRE has not been installed from source.
697 15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
698 libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
699 library.
701 16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
702 It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
703 is not the first non-POSIX option to be added. Clearly some people find
704 these options useful.
706 17. If a caller to the POSIX matching function regexec() passes a non-zero
707 value for nmatch with a NULL value for pmatch, the value of
708 nmatch is forced to zero.
710 18. RunGrepTest did not have a test for the availability of the -u option of
711 the diff command, as RunTest does. It now checks in the same way as
712 RunTest, and also checks for the -b option.
714 19. If an odd number of negated classes containing just a single character
715 interposed, within parentheses, between a forward reference to a named
716 subpattern and the definition of the subpattern, compilation crashed with
717 an internal error, complaining that it could not find the referenced
718 subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
719 [The bug was that it was starting one character too far in when skipping
720 over the character class, thus treating the ] as data rather than
721 terminating the class. This meant it could skip too much.]
723 20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
724 /g option in pcretest when the pattern contains \K, which makes it possible
725 to have an empty string match not at the start, even when the pattern is
726 anchored. Updated pcretest and pcredemo to use this option.
728 21. If the maximum number of capturing subpatterns in a recursion was greater
729 than the maximum at the outer level, the higher number was returned, but
730 with unset values at the outer level. The correct (outer level) value is
731 now given.
733 22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
734 PCRE did not set those parentheses (unlike Perl). I have now found a way to
735 make it do so. The string so far is captured, making this feature
736 compatible with Perl.
738 23. The tests have been re-organized, adding tests 11 and 12, to make it
739 possible to check the Perl 5.10 features against Perl 5.10.
741 24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
742 pattern matches a fixed length string. PCRE did not allow this; now it
743 does. Neither allows recursion.
745 25. I finally figured out how to implement a request to provide the minimum
746 length of subject string that was needed in order to match a given pattern.
747 (It was back references and recursion that I had previously got hung up
748 on.) This code has now been added to pcre_study(); it finds a lower bound
749 to the length of subject needed. It is not necessarily the greatest lower
750 bound, but using it to avoid searching strings that are too short does give
751 some useful speed-ups. The value is available to calling programs via
752 pcre_fullinfo().
754 26. While implementing 25, I discovered to my embarrassment that pcretest had
755 not been passing the result of pcre_study() to pcre_dfa_exec(), so the
756 study optimizations had never been tested with that matching function.
757 Oops. What is worse, even when it was passed study data, there was a bug in
758 pcre_dfa_exec() that meant it never actually used it. Double oops. There
759 were also very few tests of studied patterns with pcre_dfa_exec().
761 27. If (?| is used to create subpatterns with duplicate numbers, they are now
762 allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
763 on the other side of the coin, they are no longer allowed to have different
764 names, because these cannot be distinguished in PCRE, and this has caused
765 confusion. (This is a difference from Perl.)
767 28. When duplicate subpattern names are present (necessarily with different
768 numbers, as required by 27 above), and a test is made by name in a
769 conditional pattern, either for a subpattern having been matched, or for
770 recursion in such a pattern, all the associated numbered subpatterns are
771 tested, and the overall condition is true if the condition is true for any
772 one of them. This is the way Perl works, and is also more like the way
773 testing by number works.
776 Version 7.9 11-Apr-09
777 ---------------------
779 1. When building with support for bzlib/zlib (pcregrep) and/or readline
780 (pcretest), all targets were linked against these libraries. This included
781 libpcre, libpcreposix, and libpcrecpp, even though they do not use these
782 libraries. This caused unwanted dependencies to be created. This problem
783 has been fixed, and now only pcregrep is linked with bzlib/zlib and only
784 pcretest is linked with readline.
786 2. The "typedef int BOOL" in pcre_internal.h that was included inside the
787 "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
788 moved outside it again, because FALSE and TRUE are already defined in AIX,
789 but BOOL is not.
791 3. The pcre_config() function was treating the PCRE_MATCH_LIMIT and
792 PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
794 4. The pcregrep documentation said spaces were inserted as well as colons (or
795 hyphens) following file names and line numbers when outputting matching
796 lines. This is not true; no spaces are inserted. I have also clarified the
797 wording for the --colour (or --color) option.
799 5. In pcregrep, when --colour was used with -o, the list of matching strings
800 was not coloured; this is different to GNU grep, so I have changed it to be
801 the same.
803 6. When --colo(u)r was used in pcregrep, only the first matching substring in
804 each matching line was coloured. Now it goes on to look for further matches
805 of any of the test patterns, which is the same behaviour as GNU grep.
807 7. A pattern that could match an empty string could cause pcregrep to loop; it
808 doesn't make sense to accept an empty string match in pcregrep, so I have
809 locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
810 seems to be how GNU grep behaves.
812 8. The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
813 start or after a newline", because the conditional assertion was not being
814 correctly handled. The rule now is that both the assertion and what follows
815 in the first alternative must satisfy the test.
817 9. If auto-callout was enabled in a pattern with a conditional group whose
818 condition was an assertion, PCRE could crash during matching, both with
819 pcre_exec() and pcre_dfa_exec().
821 10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
822 used for matching.
824 11. Unicode property support in character classes was not working for
825 characters (bytes) greater than 127 when not in UTF-8 mode.
827 12. Added the -M command line option to pcretest.
829 14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
831 15. Added the PCRE_NO_START_OPTIMIZE match-time option.
833 16. Added comments and documentation about mis-use of no_arg in the C++
834 wrapper.
836 17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
837 from Martin Jerabek that uses macro names for all relevant character and
838 string constants.
840 18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
841 SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
842 SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
843 these, but not everybody uses configure.
845 19. A conditional group that had only one branch was not being correctly
846 recognized as an item that could match an empty string. This meant that an
847 enclosing group might also not be so recognized, causing infinite looping
848 (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
849 with the subject "ab", where knowledge that the repeated group can match
850 nothing is needed in order to break the loop.
852 20. If a pattern that was compiled with callouts was matched using pcre_dfa_
853 exec(), but without supplying a callout function, matching went wrong.
855 21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
856 leak if the size of the offset vector was greater than 30. When the vector
857 is smaller, the saved offsets during recursion go onto a local stack
858 vector, but for larger vectors malloc() is used. It was failing to free
859 when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
860 error, in fact).
862 22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
863 heapframe that is used only when UTF-8 support is enabled. This caused no
864 problem, but was untidy.
866 23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
867 CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
868 included within another project.
870 24. Steven Van Ingelgem's patches to add more options to the CMake support,
871 slightly modified by me:
873 (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
874 not building pcregrep.
876 (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
877 if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
879 25. Forward references, both numeric and by name, in patterns that made use of
880 duplicate group numbers, could behave incorrectly or give incorrect errors,
881 because when scanning forward to find the reference group, PCRE was not
882 taking into account the duplicate group numbers. A pattern such as
883 ^X(?3)(a)(?|(b)|(q))(Y) is an example.
885 26. Changed a few more instances of "const unsigned char *" to USPTR, making
886 the feature of a custom pointer more persuasive (as requested by a user).
888 27. Wrapped the definitions of fileno and isatty for Windows, which appear in
889 pcretest.c, inside #ifndefs, because it seems they are sometimes already
890 pre-defined.
892 28. Added support for (*UTF8) at the start of a pattern.
894 29. Arrange for flags added by the "release type" setting in CMake to be shown
895 in the configuration summary.
898 Version 7.8 05-Sep-08
899 ---------------------
901 1. Replaced UCP searching code with optimized version as implemented for Ad
902 Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
903 stage table and inline lookup instead of a function, giving speed ups of 2
904 to 5 times on some simple patterns that I tested. Permission was given to
905 distribute the MultiStage2.py script that generates the tables (it's not in
906 the tarball, but is in the Subversion repository).
908 2. Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
909 scripts.
911 3. Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
912 a group with a zero qualifier. The result of the study could be incorrect,
913 or the function might crash, depending on the pattern.
915 4. Caseless matching was not working for non-ASCII characters in back
916 references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
917 It now works when Unicode Property Support is available.
919 5. In pcretest, an escape such as \x{de} in the data was always generating
920 a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
921 non-UTF-8 mode. If the value is greater than 255, it gives a warning about
922 truncation.
924 6. Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
926 7. Added two (int) casts to pcregrep when printing the difference of two
927 pointers, in case they are 64-bit values.
929 8. Added comments about Mac OS X stack usage to the pcrestack man page and to
930 test 2 if it fails.
932 9. Added PCRE_CALL_CONVENTION just before the names of all exported functions,
933 and a #define of that name to empty if it is not externally set. This is to
934 allow users of MSVC to set it if necessary.
936 10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
937 the convenience functions in the pcre_get.c source file.
939 11. An option change at the start of a pattern that had top-level alternatives
940 could cause overwriting and/or a crash. This command provoked a crash in
941 some environments:
943 printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
945 This potential security problem was recorded as CVE-2008-2371.
947 12. For a pattern where the match had to start at the beginning or immediately
948 after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
949 pcre_dfa_exec() could read past the end of the passed subject if there was
950 no match. To help with detecting such bugs (e.g. with valgrind), I modified
951 pcretest so that it places the subject at the end of its malloc-ed buffer.
953 13. The change to pcretest in 12 above threw up a couple more cases when pcre_
954 exec() might read past the end of the data buffer in UTF-8 mode.
956 14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
957 the data contained the byte 0x85 as part of a UTF-8 character within its
958 first line. This applied both to normal and DFA matching.
960 15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
961 /^[^d]*?$/8 failed to match "abc".
963 16. Added a missing copyright notice to pcrecpp_internal.h.
965 17. Make it more clear in the documentation that values returned from
966 pcre_exec() in ovector are byte offsets, not character counts.
968 18. Tidied a few places to stop certain compilers from issuing warnings.
970 19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
971 supplied by Stefan Weber. I made a further small update for 7.8 because
972 there is a change of source arrangements: the pcre_searchfuncs.c module is
973 replaced by pcre_ucd.c.
976 Version 7.7 07-May-08
977 ---------------------
979 1. Applied Craig's patch to sort out a long long problem: "If we can't convert
980 a string to a long long, pretend we don't even have a long long." This is
981 done by checking for the strtoq, strtoll, and _strtoi64 functions.
983 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
984 pre-7.6 versions, which defined a global no_arg variable instead of putting
985 it in the RE class. (See also #8 below.)
987 3. Remove a line of dead code, identified by coverity and reported by Nuno
988 Lopes.
990 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
992 (1) The include/exclude patterns were being applied to the whole pathnames
993 of files, instead of just to the final components.
995 (2) If there was more than one level of directory, the subdirectories were
996 skipped unless they satisfied the include/exclude conditions. This is
997 inconsistent with GNU grep (and could even be seen as contrary to the
998 pcregrep specification - which I improved to make it absolutely clear).
999 The action now is always to scan all levels of directory, and just
1000 apply the include/exclude patterns to regular files.
1002 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1003 --exclude_dir in the tests to avoid scanning .svn directories.
1005 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
1006 NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1007 doesn't support NULs in patterns.
1009 7. Added some missing "const"s to declarations of static tables in
1010 pcre_compile.c and pcre_dfa_exec.c.
1012 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1013 caused by fix #2 above. (Subsequently also a second patch to fix the
1014 first patch. And a third patch - this was a messy problem.)
1016 9. Applied Craig's patch to remove the use of push_back().
1018 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1019 matching function regexec().
1021 11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1022 which, however, unlike Perl's \g{...}, are subroutine calls, not back
1023 references. PCRE supports relative numbers with this syntax (I don't think
1024 Oniguruma does).
1026 12. Previously, a group with a zero repeat such as (...){0} was completely
1027 omitted from the compiled regex. However, this means that if the group
1028 was called as a subroutine from elsewhere in the pattern, things went wrong
1029 (an internal error was given). Such groups are now left in the compiled
1030 pattern, with a new opcode that causes them to be skipped at execution
1031 time.
1033 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1034 to the way PCRE behaves:
1036 (a) A lone ] character is dis-allowed (Perl treats it as data).
1038 (b) A back reference to an unmatched subpattern matches an empty string
1039 (Perl fails the current match path).
1041 (c) A data ] in a character class must be notated as \] because if the
1042 first data character in a class is ], it defines an empty class. (In
1043 Perl it is not possible to have an empty class.) The empty class []
1044 never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1045 The negative empty class [^] matches any one character, independently
1046 of the DOTALL setting.
1048 14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1049 non-existent subpattern following a character class starting with ']' and
1050 containing () gave an internal compiling error instead of "reference to
1051 non-existent subpattern". Fortunately, when the pattern did exist, the
1052 compiled code was correct. (When scanning forwards to check for the
1053 existencd of the subpattern, it was treating the data ']' as terminating
1054 the class, so got the count wrong. When actually compiling, the reference
1055 was subsequently set up correctly.)
1057 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1058 it was being rejected as not supported by pcre_dfa_exec(), even though
1059 other assertions are supported. I have made pcre_dfa_exec() support
1060 (*FAIL).
1062 16. The implementation of 13c above involved the invention of a new opcode,
1063 OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1064 cannot be changed at match time, I realized I could make a small
1065 improvement to matching performance by compiling OP_ALLANY instead of
1066 OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1067 on the OP_ANY path.
1069 17. Compiling pcretest on Windows with readline support failed without the
1070 following two fixes: (1) Make the unistd.h include conditional on
1071 HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1073 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1074 ncurses library to be included for pcretest when ReadLine support is
1075 requested, but also to allow for it to be overridden. This patch came from
1076 Daniel Bergström.
1078 19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1079 as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1080 any errors with the current Unicode tables. Thanks to Peter Kankowski for
1081 spotting this.
1084 Version 7.6 28-Jan-08
1085 ---------------------
1087 1. A character class containing a very large number of characters with
1088 codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1089 overflow.
1091 2. Patch to cut out the "long long" test in pcrecpp_unittest when
1092 HAVE_LONG_LONG is not defined.
1094 3. Applied Christian Ehrlicher's patch to update the CMake build files to
1095 bring them up to date and include new features. This patch includes:
1097 - Fixed PH's badly added libz and libbz2 support.
1098 - Fixed a problem with static linking.
1099 - Added pcredemo. [But later removed - see 7 below.]
1100 - Fixed dftables problem and added an option.
1101 - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1103 - Added readline support for pcretest.
1104 - Added an listing of the option settings after cmake has run.
1106 4. A user submitted a patch to Makefile that makes it easy to create
1107 "pcre.dll" under mingw when using Configure/Make. I added stuff to
1108 Makefile.am that cause it to include this special target, without
1109 affecting anything else. Note that the same mingw target plus all
1110 the other distribution libraries and programs are now supported
1111 when configuring with CMake (see 6 below) instead of with
1112 Configure/Make.
1114 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1115 This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1116 exported in the Windows port". It has not yet been confirmed that the patch
1117 solves the problem, but it does no harm.
1119 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1120 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1121 with CMake, and also correct the comment about stack recursion.
1123 7. Remove the automatic building of pcredemo from the ./configure system and
1124 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1125 of a program that users should build themselves after PCRE is installed, so
1126 building it automatically is not really right. What is more, it gave
1127 trouble in some build environments.
1129 8. Further tidies to CMakeLists.txt from Sheri and Christian.
1132 Version 7.5 10-Jan-08
1133 ---------------------
1135 1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
1136 values in parens when parsing an RE using the C++ wrapper."
1138 2. Negative specials like \S did not work in character classes in UTF-8 mode.
1139 Characters greater than 255 were excluded from the class instead of being
1140 included.
1142 3. The same bug as (2) above applied to negated POSIX classes such as
1143 [:^space:].
1145 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1146 defined or documented. It seems to have been a typo for PCRE_STATIC, so
1147 I have changed it.
1149 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
1150 first named subpattern) and a construct such as (?&a) would reference the
1151 first named subpattern whose name started with "a" (in other words, the
1152 length check was missing). Both these problems are fixed. "Subpattern name
1153 expected" is now given for (?&) (a zero-length name), and this patch also
1154 makes it give the same error for \k'' (previously it complained that that
1155 was a reference to a non-existent subpattern).
1157 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
1158 this is right because (?- can be followed by option settings as well as by
1159 digits. I have, however, made the messages clearer.
1161 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1162 than the number used in the conditional) now cause a compile-time error.
1163 This is actually not compatible with Perl, which accepts such patterns, but
1164 treats the conditional as always being FALSE (as PCRE used to), but it
1165 seems to me that giving a diagnostic is better.
1167 8. Change "alphameric" to the more common word "alphanumeric" in comments
1168 and messages.
1170 9. Fix two occurrences of "backslash" in comments that should have been
1171 "backspace".
1173 10. Remove two redundant lines of code that can never be obeyed (their function
1174 was moved elsewhere).
1176 11. The program that makes PCRE's Unicode character property table had a bug
1177 which caused it to generate incorrect table entries for sequences of
1178 characters that have the same character type, but are in different scripts.
1179 It amalgamated them into a single range, with the script of the first of
1180 them. In other words, some characters were in the wrong script. There were
1181 thirteen such cases, affecting characters in the following ranges:
1183 U+002b0 - U+002c1
1184 U+0060c - U+0060d
1185 U+0061e - U+00612
1186 U+0064b - U+0065e
1187 U+0074d - U+0076d
1188 U+01800 - U+01805
1189 U+01d00 - U+01d77
1190 U+01d9b - U+01dbf
1191 U+0200b - U+0200f
1192 U+030fc - U+030fe
1193 U+03260 - U+0327f
1194 U+0fb46 - U+0fbb1
1195 U+10450 - U+1049d
1197 12. The -o option (show only the matching part of a line) for pcregrep was not
1198 compatible with GNU grep in that, if there was more than one match in a
1199 line, it showed only the first of them. It now behaves in the same way as
1200 GNU grep.
1202 13. If the -o and -v options were combined for pcregrep, it printed a blank
1203 line for every non-matching line. GNU grep prints nothing, and pcregrep now
1204 does the same. The return code can be used to tell if there were any
1205 non-matching lines.
1207 14. Added --file-offsets and --line-offsets to pcregrep.
1209 15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1210 infinitely looping recursion. The bug was that positive lookaheads were not
1211 being skipped when checking for a possible empty match (negative lookaheads
1212 and both kinds of lookbehind were skipped).
1214 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1215 inclusion of <windows.h> to before rather than after the definition of
1216 INVALID_FILE_ATTRIBUTES (patch from David Byron).
1218 17. Specifying a possessive quantifier with a specific limit for a Unicode
1219 character property caused pcre_compile() to compile bad code, which led at
1220 runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1221 are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1222 caused the error; without that there was no problem.
1224 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1226 19. Added --enable-pcretest-libreadline.
1228 20. In pcrecpp.cc, the variable 'count' was incremented twice in
1229 RE::GlobalReplace(). As a result, the number of replacements returned was
1230 double what it should be. I removed one of the increments, but Craig sent a
1231 later patch that removed the other one (the right fix) and added unit tests
1232 that check the return values (which was not done before).
1234 21. Several CMake things:
1236 (1) Arranged that, when cmake is used on Unix, the libraries end up with
1237 the names libpcre and libpcreposix, not just pcre and pcreposix.
1239 (2) The above change means that pcretest and pcregrep are now correctly
1240 linked with the newly-built libraries, not previously installed ones.
1244 22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1245 crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1246 UTF-8 newline character). The key issue is that the pattern starts .*;
1247 this means that the match must be either at the beginning, or after a
1248 newline. The bug was in the code for advancing after a failed match and
1249 checking that the new position followed a newline. It was not taking
1250 account of UTF-8 characters correctly.
1252 23. PCRE was behaving differently from Perl in the way it recognized POSIX
1253 character classes. PCRE was not treating the sequence [:...:] as a
1254 character class unless the ... were all letters. Perl, however, seems to
1255 allow any characters between [: and :], though of course it rejects as
1256 unknown any "names" that contain non-letters, because all the known class
1257 names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1258 for example, whereas PCRE did not - it did not recognize a POSIX character
1259 class. This seemed a bit dangerous, so the code has been changed to be
1260 closer to Perl. The behaviour is not identical to Perl, because PCRE will
1261 diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1262 treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1263 Perl does, and where it didn't before.
1265 24. Rewrite so as to remove the single use of %n from pcregrep because in some
1266 Windows environments %n is disabled by default.
1269 Version 7.4 21-Sep-07
1270 ---------------------
1272 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
1273 means that a class such as [\s] counted as "explicit reference to CR or
1274 LF". That isn't really right - the whole point of the change was to try to
1275 help when there was an actual mention of one of the two characters. So now
1276 the change happens only if \r or \n (or a literal CR or LF) character is
1277 encountered.
1279 2. The 32-bit options word was also used for 6 internal flags, but the numbers
1280 of both had grown to the point where there were only 3 bits left.
1281 Fortunately, there was spare space in the data structure, and so I have
1282 moved the internal flags into a new 16-bit field to free up more option
1283 bits.
1285 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1286 but did not set the internal JCHANGED flag - either of these is enough to
1287 control the way the "get" function works - but the PCRE_INFO_JCHANGED
1288 facility is supposed to tell if (?J) was ever used, so now (?J) at the
1289 start sets both bits.
1291 4. Added options (at build time, compile time, exec time) to change \R from
1292 matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1294 5. doc/pcresyntax.html was missing from the distribution.
1296 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1297 compatibility, even though it is no longer used.
1299 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1300 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1301 windows.h file is present (where different names are used). [This was
1302 reversed later after testing - see 16 below.]
1304 8. Changed all #include <config.h> to #include "config.h". There were also
1305 some further <pcre.h> cases that I changed to "pcre.h".
1307 9. When pcregrep was used with the --colour option, it missed the line ending
1308 sequence off the lines that it output.
1310 10. It was pointed out to me that arrays of string pointers cause lots of
1311 relocations when a shared library is dynamically loaded. A technique of
1312 using a single long string with a table of offsets can drastically reduce
1313 these. I have refactored PCRE in four places to do this. The result is
1314 dramatic:
1316 Originally: 290
1317 After changing UCP table: 187
1318 After changing error message table: 43
1319 After changing table of "verbs" 36
1320 After changing table of Posix names 22
1322 Thanks to the folks working on Gregex for glib for this insight.
1324 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1325 unicode-properties was also set.
1327 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1329 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1330 checked only for CRLF.
1332 14. Added casts to pcretest.c to avoid compiler warnings.
1334 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1336 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1337 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1338 entirely. This removes changes made in 7 above.
1340 17. The CMake files have been updated, and there is now more information about
1341 building with CMake in the NON-UNIX-USE document.
1344 Version 7.3 28-Aug-07
1345 ---------------------
1347 1. In the rejigging of the build system that eventually resulted in 7.1, the
1348 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1349 brackets there is not right, since it causes compilers to look for an
1350 installed pcre.h, not the version that is in the source that is being
1351 compiled (which of course may be different). I have changed it back to:
1353 #include "pcre.h"
1355 I have a vague recollection that the change was concerned with compiling in
1356 different directories, but in the new build system, that is taken care of
1357 by the VPATH setting the Makefile.
1359 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1360 when the subject happened to end in the byte 0x85 (e.g. if the last
1361 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1362 characters but of course it shouldn't be taken as a newline when it is part
1363 of another character. The bug was that, for an unlimited repeat of . in
1364 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1365 characters when looking for a newline.
1367 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1369 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1370 in debug output.
1372 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1373 long printing in the pcrecpp unittest when running under MinGW.
1375 6. ESC_K was left out of the EBCDIC table.
1377 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1378 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1379 limit also applies to "virtual nesting" when a pattern is recursive, and in
1380 this case 1000 isn't so big. I have been able to remove this limit at the
1381 expense of backing off one optimization in certain circumstances. Normally,
1382 when pcre_exec() would call its internal match() function recursively and
1383 immediately return the result unconditionally, it uses a "tail recursion"
1384 feature to save stack. However, when a subpattern that can match an empty
1385 string has an unlimited repetition quantifier, it no longer makes this
1386 optimization. That gives it a stack frame in which to save the data for
1387 checking that an empty string has been matched. Previously this was taken
1388 from the 1000-entry workspace that had been reserved. So now there is no
1389 explicit limit, but more stack is used.
1391 8. Applied Daniel's patches to solve problems with the import/export magic
1392 syntax that is required for Windows, and which was going wrong for the
1393 pcreposix and pcrecpp parts of the library. These were overlooked when this
1394 problem was solved for the main library.
1396 9. There were some crude static tests to avoid integer overflow when computing
1397 the size of patterns that contain repeated groups with explicit upper
1398 limits. As the maximum quantifier is 65535, the maximum group length was
1399 set at 30,000 so that the product of these two numbers did not overflow a
1400 32-bit integer. However, it turns out that people want to use groups that
1401 are longer than 30,000 bytes (though not repeat them that many times).
1402 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1403 made it possible to implement the integer overflow checks in a much more
1404 dynamic way, which I have now done. The artificial limitation on group
1405 length has been removed - we now have only the limit on the total length of
1406 the compiled pattern, which depends on the LINK_SIZE setting.
1408 10. Fixed a bug in the documentation for get/copy named substring when
1409 duplicate names are permitted. If none of the named substrings are set, the
1410 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1411 empty string.
1413 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1414 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1415 because the ] is interpreted as the first data character and the
1416 terminating ] is not found. PCRE has been made compatible with Perl in this
1417 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1418 cause memory overwriting.
1420 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1421 string has been matched (to stop an infinite loop). It was not recognizing
1422 a conditional subpattern that could match an empty string if that
1423 subpattern was within another subpattern. For example, it looped when
1424 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
1425 condition was not nested. This bug has been fixed.
1427 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1428 past the start of the subject in the presence of bytes with the top bit
1429 set, for example "\x8aBCD".
1431 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1432 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1434 14. Optimized (?!) to (*FAIL).
1436 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1437 This restricts code points to be within the range 0 to 0x10FFFF, excluding
1438 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1439 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1440 does: it's just the validity check that is more restrictive.
1442 16. Inserted checks for integer overflows during escape sequence (backslash)
1443 processing, and also fixed erroneous offset values for syntax errors during
1444 backslash processing.
1446 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1447 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1449 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1450 caused an overrun.
1452 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1453 something other than just ASCII characters) inside a group that had an
1454 unlimited repeat caused a loop at compile time (while checking to see
1455 whether the group could match an empty string).
1457 20. Debugging a pattern containing \p or \P could cause a crash. For example,
1458 [\P{Any}] did so. (Error in the code for printing property names.)
1460 21. An orphan \E inside a character class could cause a crash.
1462 22. A repeated capturing bracket such as (A)? could cause a wild memory
1463 reference during compilation.
1465 23. There are several functions in pcre_compile() that scan along a compiled
1466 expression for various reasons (e.g. to see if it's fixed length for look
1467 behind). There were bugs in these functions when a repeated \p or \P was
1468 present in the pattern. These operators have additional parameters compared
1469 with \d, etc, and these were not being taken into account when moving along
1470 the compiled data. Specifically:
1472 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1473 length.
1475 (b) An item such as \pL+ within a repeated group could cause crashes or
1476 loops.
1478 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1479 "reference to non-existent subpattern" error.
1481 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1483 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1484 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1486 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1488 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1489 character were causing crashes (broken optimization).
1491 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1492 \p or \P) caused a compile-time loop.
1494 28. More problems have arisen in unanchored patterns when CRLF is a valid line
1495 break. For example, the unstudied pattern [\r\n]A does not match the string
1496 "\r\nA" because change 7.0/46 below moves the current point on by two
1497 characters after failing to match at the start. However, the pattern \nA
1498 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1499 the same is true. There doesn't seem any very clean way out of this, but
1500 what I have chosen to do makes the common cases work: PCRE now takes note
1501 of whether there can be an explicit match for \r or \n anywhere in the
1502 pattern, and if so, 7.0/46 no longer applies. As part of this change,
1503 there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1504 pattern has explicit CR or LF references.
1506 29. Added (*CR) etc for changing newline setting at start of pattern.
1509 Version 7.2 19-Jun-07
1510 ---------------------
1512 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1513 which is apparently normally available under Windows.
1515 2. Re-jig the pcregrep tests with different newline settings in an attempt
1516 to make them independent of the local environment's newline setting.
1518 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1520 4. Some of the "internals" tests were previously cut out when the link size
1521 was not 2, because the output contained actual offsets. The recent new
1522 "Z" feature of pcretest means that these can be cut out, making the tests
1523 usable with all link sizes.
1525 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1526 stack recursion. This gives a massive performance boost under BSD, but just
1527 a small improvement under Linux. However, it saves one field in the frame
1528 in all cases.
1530 6. Added more features from the forthcoming Perl 5.10:
1532 (a) (?-n) (where n is a string of digits) is a relative subroutine or
1533 recursion call. It refers to the nth most recently opened parentheses.
1535 (b) (?+n) is also a relative subroutine call; it refers to the nth next
1536 to be opened parentheses.
1538 (c) Conditions that refer to capturing parentheses can be specified
1539 relatively, for example, (?(-2)... or (?(+3)...
1541 (d) \K resets the start of the current match so that everything before
1542 is not part of it.
1544 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1546 (f) \g{name} is another synonym - part of Perl 5.10's unification of
1547 reference syntax.
1549 (g) (?| introduces a group in which the numbering of parentheses in each
1550 alternative starts with the same number.
1552 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1554 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1557 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
1558 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1559 for detecting groups that can match an empty string.
1561 9. A pattern with a very large number of alternatives (more than several
1562 hundred) was running out of internal workspace during the pre-compile
1563 phase, where pcre_compile() figures out how much memory will be needed. A
1564 bit of new cunning has reduced the workspace needed for groups with
1565 alternatives. The 1000-alternative test pattern now uses 12 bytes of
1566 workspace instead of running out of the 4096 that are available.
1568 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1570 11. Applied patch from Google to remove an optimization that didn't quite work.
1571 The report of the bug said:
1573 pcrecpp::RE("a*").FullMatch("aaa") matches, while
1574 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1575 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1577 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1578 it matched the wrong number of bytes.
1581 Version 7.1 24-Apr-07
1582 ---------------------
1584 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1585 that is more "standard", making use of automake and other Autotools. There
1586 is some re-arrangement of the files and adjustment of comments consequent
1587 on this.
1589 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1590 for recursive directory scanning broke on some systems because the files
1591 are not scanned in any specific order and on different systems the order
1592 was different. A call to "sort" has been inserted into RunGrepTest for the
1593 approprate test as a short-term fix. In the longer term there may be an
1594 alternative.
1596 3. I had an email from Eric Raymond about problems translating some of PCRE's
1597 man pages to HTML (despite the fact that I distribute HTML pages, some
1598 people do their own conversions for various reasons). The problems
1599 concerned the use of low-level troff macros .br and .in. I have therefore
1600 removed all such uses from the man pages (some were redundant, some could
1601 be replaced by .nf/.fi pairs). The 132html script that I use to generate
1602 HTML has been updated to handle .nf/.fi and to complain if it encounters
1603 .br or .in.
1605 4. Updated comments in configure.ac that get placed in config.h.in and also
1606 arranged for config.h to be included in the distribution, with the name
1607 config.h.generic, for the benefit of those who have to compile without
1608 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1610 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1611 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1612 makevp.bat for the current PCRE, using the additional files
1613 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1615 6. A Windows user reported a minor discrepancy with test 2, which turned out
1616 to be caused by a trailing space on an input line that had got lost in his
1617 copy. The trailing space was an accident, so I've just removed it.
1619 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1620 that is needed.
1622 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1623 as "const" (a) because they are and (b) because it helps the PHP
1624 maintainers who have recently made a script to detect big data structures
1625 in the php code that should be moved to the .rodata section. I remembered
1626 to update Builducptable as well, so it won't revert if ucptable.h is ever
1627 re-created.
1629 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1630 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1631 order to be able to cut out the UTF-8 tables in the latter when UTF-8
1632 support is not required. This saves 1.5-2K of code, which is important in
1633 some applications.
1635 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1636 so as not to refer to the tables, even though these functions will never be
1637 called when UTF-8 support is disabled. Otherwise there are problems with a
1638 shared library.
1640 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1642 (a) It was defining its arguments as char * instead of void *.
1644 (b) It was assuming that all moves were upwards in memory; this was true
1645 a long time ago when I wrote it, but is no longer the case.
1647 The emulated memove() is provided for those environments that have neither
1648 memmove() nor bcopy(). I didn't think anyone used it these days, but that
1649 is clearly not the case, as these two bugs were recently reported.
1651 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1652 and Detrail to create the HTML documentation, the .txt form of the man
1653 pages, and it removes trailing spaces from listed files. It also creates
1654 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1655 case, it wraps all the #defines with #ifndefs. This script should be run
1656 before "make dist".
1658 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1659 with Unicode property support.
1661 (a) For a maximizing quantifier, if the two different cases of the
1662 character were of different lengths in their UTF-8 codings (there are
1663 some cases like this - I found 11), and the matching function had to
1664 back up over a mixture of the two cases, it incorrectly assumed they
1665 were both the same length.
1667 (b) When PCRE was configured to use the heap rather than the stack for
1668 recursion during matching, it was not correctly preserving the data for
1669 the other case of a UTF-8 character when checking ahead for a match
1670 while processing a minimizing repeat. If the check also involved
1671 matching a wide character, but failed, corruption could cause an
1672 erroneous result when trying to check for a repeat of the original
1673 character.
1675 13. Some tidying changes to the testing mechanism:
1677 (a) The RunTest script now detects the internal link size and whether there
1678 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1679 values substituted by "configure". (The RunGrepTest script already did
1680 this for UTF-8.) The configure.ac script no longer substitutes the
1681 relevant variables.
1683 (b) The debugging options /B and /D in pcretest show the compiled bytecode
1684 with length and offset values. This means that the output is different
1685 for different internal link sizes. Test 2 is skipped for link sizes
1686 other than 2 because of this, bypassing the problem. Unfortunately,
1687 there was also a test in test 3 (the locale tests) that used /B and
1688 failed for link sizes other than 2. Rather than cut the whole test out,
1689 I have added a new /Z option to pcretest that replaces the length and
1690 offset values with spaces. This is now used to make test 3 independent
1691 of link size. (Test 2 will be tidied up later.)
1693 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1694 segmentation fault instead of returning the appropriate error message.
1696 15. In multiline mode when the newline sequence was set to "any", the pattern
1697 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1698 This doesn't seem right; it now treats the CRLF combination as the line
1699 ending, and so does not match in that case. It's only a pattern such as ^$
1700 that would hit this one: something like ^ABC$ would have failed after \r
1701 and then tried again after \r\n.
1703 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1704 in an attempt to make files that differ only in their line terminators
1705 compare equal. This works on Linux.
1707 17. Under certain error circumstances pcregrep might try to free random memory
1708 as it exited. This is now fixed, thanks to valgrind.
1710 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1711 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1712 was because its rules for how to advance for /g after matching an empty
1713 string at the end of a line did not allow for this case. They now check for
1714 it specially.
1716 20. pcretest is supposed to handle patterns and data of any length, by
1717 extending its buffers when necessary. It was getting this wrong when the
1718 buffer for a data line had to be extended.
1720 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1721 CRLF as a newline sequence.
1723 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1724 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1725 I have nevertheless tidied it up.
1727 23. Added some casts to kill warnings from HP-UX ia64 compiler.
1729 24. Added a man page for pcre-config.
1732 Version 7.0 19-Dec-06
1733 ---------------------
1735 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1736 moving to gcc 4.1.1.
1738 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1739 sys/time.h, which is documented as needed for this function. It doesn't
1740 seem to matter on Linux, but it showed up on some releases of OS X.
1742 3. It seems that there are systems where bytes whose values are greater than
1743 127 match isprint() in the "C" locale. The "C" locale should be the
1744 default when a C program starts up. In most systems, only ASCII printing
1745 characters match isprint(). This difference caused the output from pcretest
1746 to vary, making some of the tests fail. I have changed pcretest so that:
1748 (a) When it is outputting text in the compiled version of a pattern, bytes
1749 other than 32-126 are always shown as hex escapes.
1751 (b) When it is outputting text that is a matched part of a subject string,
1752 it does the same, unless a different locale has been set for the match
1753 (using the /L modifier). In this case, it uses isprint() to decide.
1755 4. Fixed a major bug that caused incorrect computation of the amount of memory
1756 required for a compiled pattern when options that changed within the
1757 pattern affected the logic of the preliminary scan that determines the
1758 length. The relevant options are -x, and -i in UTF-8 mode. The result was
1759 that the computed length was too small. The symptoms of this bug were
1760 either the PCRE error "internal error: code overflow" from pcre_compile(),
1761 or a glibc crash with a message such as "pcretest: free(): invalid next
1762 size (fast)". Examples of patterns that provoked this bug (shown in
1763 pcretest format) are:
1765 /(?-x: )/x
1766 /(?x)(?-x: \s*#\s*)/
1767 /((?i)[\x{c0}])/8
1768 /(?i:[\x{c0}])/8
1770 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1771 is now done differently.
1773 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1774 wrapper classes; (b) implement a new function in the C++ scanner that is
1775 more efficient than the old way of doing things because it avoids levels of
1776 recursion in the regex matching; (c) add a paragraph to the documentation
1777 for the FullMatch() function.
1779 6. The escape sequence \n was being treated as whatever was defined as
1780 "newline". Not only was this contrary to the documentation, which states
1781 that \n is character 10 (hex 0A), but it also went horribly wrong when
1782 "newline" was defined as CRLF. This has been fixed.
1784 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1785 was being set to -1 for the "end of line" case (supposedly a value that no
1786 character can have). Though this value is never used (the check for end of
1787 line is "zero bytes in current character"), it caused compiler complaints.
1788 I've changed it to 0xffffffff.
1790 8. In pcre_version.c, the version string was being built by a sequence of
1791 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1792 string (as it is for production releases) called a macro with an empty
1793 argument. The C standard says the result of this is undefined. The gcc
1794 compiler treats it as an empty string (which was what was wanted) but it is
1795 reported that Visual C gives an error. The source has been hacked around to
1796 avoid this problem.
1798 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1799 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1800 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1801 of them did).
1803 10. Originally, pcretest opened its input and output without "b"; then I was
1804 told that "b" was needed in some environments, so it was added for release
1805 5.0 to both the input and output. (It makes no difference on Unix-like
1806 systems.) Later I was told that it is wrong for the input on Windows. I've
1807 now abstracted the modes into two macros, to make it easier to fiddle with
1808 them, and removed "b" from the input mode under Windows.
1810 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1812 12. Added -help and --help to pcretest as an official way of being reminded
1813 of the options.
1815 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1816 and pcrecpp.cc because they annoy compilers at high warning levels.
1818 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1820 15. Fixed an occurrence of == in configure.ac that should have been = (shell
1821 scripts are not C programs :-) and which was not noticed because it works
1822 on Linux.
1824 16. pcretest is supposed to handle any length of pattern and data line (as one
1825 line or as a continued sequence of lines) by extending its input buffer if
1826 necessary. This feature was broken for very long pattern lines, leading to
1827 a string of junk being passed to pcre_compile() if the pattern was longer
1828 than about 50K.
1830 17. I have done a major re-factoring of the way pcre_compile() computes the
1831 amount of memory needed for a compiled pattern. Previously, there was code
1832 that made a preliminary scan of the pattern in order to do this. That was
1833 OK when PCRE was new, but as the facilities have expanded, it has become
1834 harder and harder to keep it in step with the real compile phase, and there
1835 have been a number of bugs (see for example, 4 above). I have now found a
1836 cunning way of running the real compile function in a "fake" mode that
1837 enables it to compute how much memory it would need, while actually only
1838 ever using a few hundred bytes of working memory and without too many
1839 tests of the mode. This should make future maintenance and development
1840 easier. A side effect of this work is that the limit of 200 on the nesting
1841 depth of parentheses has been removed (though this was never a serious
1842 limitation, I suspect). However, there is a downside: pcre_compile() now
1843 runs more slowly than before (30% or more, depending on the pattern). I
1844 hope this isn't a big issue. There is no effect on runtime performance.
1846 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1847 newline (only possible for the last line of a file) and it was a
1848 pattern that set a locale (followed by /Lsomething), pcretest crashed.
1850 19. Added additional timing features to pcretest. (1) The -tm option now times
1851 matching only, not compiling. (2) Both -t and -tm can be followed, as a
1852 separate command line item, by a number that specifies the number of
1853 repeats to use when timing. The default is 50000; this gives better
1854 precision, but takes uncomfortably long for very large patterns.
1856 20. Extended pcre_study() to be more clever in cases where a branch of a
1857 subpattern has no definite first character. For example, (a*|b*)[cd] would
1858 previously give no result from pcre_study(). Now it recognizes that the
1859 first character must be a, b, c, or d.
1861 21. There was an incorrect error "recursive call could loop indefinitely" if
1862 a subpattern (or the entire pattern) that was being tested for matching an
1863 empty string contained only one non-empty item after a nested subpattern.
1864 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1865 incorrectly, because the \d was being skipped in the check.
1867 22. The pcretest program now has a new pattern option /B and a command line
1868 option -b, which is equivalent to adding /B to every pattern. This causes
1869 it to show the compiled bytecode, without the additional information that
1870 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1871 is the same as /B/I).
1873 23. A new optimization is now able automatically to treat some sequences such
1874 as a*b as a*+b. More specifically, if something simple (such as a character
1875 or a simple class like \d) has an unlimited quantifier, and is followed by
1876 something that cannot possibly match the quantified thing, the quantifier
1877 is automatically "possessified".
1879 24. A recursive reference to a subpattern whose number was greater than 39
1880 went wrong under certain circumstances in UTF-8 mode. This bug could also
1881 have affected the operation of pcre_study().
1883 25. Realized that a little bit of performance could be had by replacing
1884 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1886 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1888 27. Possessive quantifiers such as a++ were previously implemented by turning
1889 them into atomic groups such as ($>a+). Now they have their own opcodes,
1890 which improves performance. This includes the automatically created ones
1891 from 23 above.
1893 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1894 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1895 the first matched character to be a colon. This applied both to named and
1896 numbered groups.
1898 29. The ucpinternal.h header file was missing its idempotency #ifdef.
1900 30. I was sent a "project" file called libpcre.a.dev which I understand makes
1901 building PCRE on Windows easier, so I have included it in the distribution.
1903 31. There is now a check in pcretest against a ridiculously large number being
1904 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1905 loop, the loop is abandoned.
1907 32. Forward references to subpatterns in conditions such as (?(2)...) where
1908 subpattern 2 is defined later cause pcre_compile() to search forwards in
1909 the pattern for the relevant set of parentheses. This search went wrong
1910 when there were unescaped parentheses in a character class, parentheses
1911 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1913 33. "Subroutine" calls and backreferences were previously restricted to
1914 referencing subpatterns earlier in the regex. This restriction has now
1915 been removed.
1917 34. Added a number of extra features that are going to be in Perl 5.10. On the
1918 whole, these are just syntactic alternatives for features that PCRE had
1919 previously implemented using the Python syntax or my own invention. The
1920 other formats are all retained for compatibility.
1922 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1923 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1924 also .NET compatible.
1926 (b) A recursion or subroutine call to a named group can now be defined as
1927 (?&name) as well as (?P>name).
1929 (c) A backreference to a named group can now be defined as \k<name> or
1930 \k'name' as well as (?P=name). The new forms, as well as being in Perl
1931 5.10, are also .NET compatible.
1933 (d) A conditional reference to a named group can now use the syntax
1934 (?(<name>) or (?('name') as well as (?(name).
1936 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1937 groups (named and numbered) that are never evaluated inline, but can be
1938 called as "subroutines" from elsewhere. In effect, the DEFINE condition
1939 is always false. There may be only one alternative in such a group.
1941 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1942 as the simple (?(R). The condition is true only if the most recent
1943 recursion is that of the given number or name. It does not search out
1944 through the entire recursion stack.
1946 (g) The escape \gN or \g{N} has been added, where N is a positive or
1947 negative number, specifying an absolute or relative reference.
1949 35. Tidied to get rid of some further signed/unsigned compiler warnings and
1950 some "unreachable code" warnings.
1952 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1953 things, this adds five new scripts.
1955 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1956 There were also incompatibilities regarding the handling of \Q..\E inside
1957 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1958 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1960 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1961 matches an empty string, and forcibly breaks the loop. There were bugs in
1962 this code in non-simple cases. For a pattern such as ^(a()*)* matched
1963 against aaaa the result was just "a" rather than "aaaa", for example. Two
1964 separate and independent bugs (that affected different cases) have been
1965 fixed.
1967 39. Refactored the code to abolish the use of different opcodes for small
1968 capturing bracket numbers. This is a tidy that I avoided doing when I
1969 removed the limit on the number of capturing brackets for 3.5 back in 2001.
1970 The new approach is not only tidier, it makes it possible to reduce the
1971 memory needed to fix the previous bug (38).
1973 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1974 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1975 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1976 mode.
1978 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1979 report.
1981 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1982 copy construction and assignment in the C++ wrapper.
1984 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1985 couple of bugs that could have given wrong results in the "--newline=crlf"
1986 case.
1988 44. Added a number of casts and did some reorganization of signed/unsigned int
1989 variables following suggestions from Dair Grant. Also renamed the variable
1990 "this" as "item" because it is a C++ keyword.
1992 45. Arranged for dftables to add
1994 #include "pcre_internal.h"
1996 to pcre_chartables.c because without it, gcc 4.x may remove the array
1997 definition from the final binary if PCRE is built into a static library and
1998 dead code stripping is activated.
2000 46. For an unanchored pattern, if a match attempt fails at the start of a
2001 newline sequence, and the newline setting is CRLF or ANY, and the next two
2002 characters are CRLF, advance by two characters instead of one.
2005 Version 6.7 04-Jul-06
2006 ---------------------
2008 1. In order to handle tests when input lines are enormously long, pcretest has
2009 been re-factored so that it automatically extends its buffers when
2010 necessary. The code is crude, but this _is_ just a test program. The
2011 default size has been increased from 32K to 50K.
2013 2. The code in pcre_study() was using the value of the re argument before
2014 testing it for NULL. (Of course, in any sensible call of the function, it
2015 won't be NULL.)
2017 3. The memmove() emulation function in pcre_internal.h, which is used on
2018 systems that lack both memmove() and bcopy() - that is, hardly ever -
2019 was missing a "static" storage class specifier.
2021 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
2022 containing an extended class (one that cannot be represented by a bitmap
2023 because it contains high-valued characters or Unicode property items, e.g.
2024 [\pZ]). Almost always one would set UTF-8 mode when processing such a
2025 pattern, but PCRE should not loop if you do not (it no longer does).
2026 [Detail: two cases were found: (a) a repeated subpattern containing an
2027 extended class; (b) a recursive reference to a subpattern that followed a
2028 previous extended class. It wasn't skipping over the extended class
2029 correctly when UTF-8 mode was not set.]
2031 5. A negated single-character class was not being recognized as fixed-length
2032 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
2033 compile error "lookbehind assertion is not fixed length".
2035 6. The RunPerlTest auxiliary script was showing an unexpected difference
2036 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
2037 write a Perl script that can interpret lines of an input file either as
2038 byte characters or as UTF-8, which is what "perltest" was being required to
2039 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
2040 can't do is switch easily at run time between having the "use utf8;" pragma
2041 or not. In the end, I fudged it by using the RunPerlTest script to insert
2042 "use utf8;" explicitly for the UTF-8 tests.
2044 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
2045 the end of the subject string, contrary to the documentation and to what
2046 Perl does. This was true of both matching functions. Now it matches only at
2047 the start of the subject and immediately after *internal* newlines.
2049 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
2050 a pointer to an int instead of a pointer to an unsigned long int. This
2051 caused problems on 64-bit systems.
2053 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
2054 instance of the 'standard' template library not being so standard".
2056 10. There was no check on the number of named subpatterns nor the maximum
2057 length of a subpattern name. The product of these values is used to compute
2058 the size of the memory block for a compiled pattern. By supplying a very
2059 long subpattern name and a large number of named subpatterns, the size
2060 computation could be caused to overflow. This is now prevented by limiting
2061 the length of names to 32 characters, and the number of named subpatterns
2062 to 10,000.
2064 11. Subpatterns that are repeated with specific counts have to be replicated in
2065 the compiled pattern. The size of memory for this was computed from the
2066 length of the subpattern and the repeat count. The latter is limited to
2067 65535, but there was no limit on the former, meaning that integer overflow
2068 could in principle occur. The compiled length of a repeated subpattern is
2069 now limited to 30,000 bytes in order to prevent this.
2071 12. Added the optional facility to have named substrings with the same name.
2073 13. Added the ability to use a named substring as a condition, using the
2074 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2075 are numbers (not recommended). Forward references are permitted.
2077 14. Added forward references in named backreferences (if you see what I mean).
2079 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2080 pattern could run off the end of the subject. For example, the pattern
2081 "(?s)(.{1,5})"8 did this with the subject "ab".
2083 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2084 PCRE_CASELESS was set when matching characters that were quantified with ?
2085 or *.
2087 17. A character class other than a single negated character that had a minimum
2088 but no maximum quantifier - for example [ab]{6,} - was not handled
2089 correctly by pce_dfa_exec(). It would match only one character.
2091 18. A valid (though odd) pattern that looked like a POSIX character
2092 class but used an invalid character after [ (for example [[,abc,]]) caused
2093 pcre_compile() to give the error "Failed: internal error: code overflow" or
2094 in some cases to crash with a glibc free() error. This could even happen if
2095 the pattern terminated after [[ but there just happened to be a sequence of
2096 letters, a binary zero, and a closing ] in the memory that followed.
2098 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2099 over the years. Originally (before any Unicode support), just the bottom 8
2100 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2101 output from "man perlunicode" includes this:
2103 The regular expression compiler produces polymorphic opcodes. That
2104 is, the pattern adapts to the data and automatically switches to
2105 the Unicode character scheme when presented with Unicode data--or
2106 instead uses a traditional byte scheme when presented with byte
2107 data.
2109 Sadly, a wide octal escape does not cause a switch, and in a string with
2110 no other multibyte characters, these octal escapes are treated as before.
2111 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
2112 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2113 Unicode string.
2115 I have not perpetrated such confusion in PCRE. Up till now, it took just
2116 the bottom 8 bits, as in old Perl. I have now made octal escapes with
2117 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2118 translate to the appropriate multibyte character.
2120 29. Applied some refactoring to reduce the number of warnings from Microsoft
2121 and Borland compilers. This has included removing the fudge introduced
2122 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2123 a warning about an unused variable.
2125 21. PCRE has not included VT (character 0x0b) in the set of whitespace
2126 characters since release 4.0, because Perl (from release 5.004) does not.
2127 [Or at least, is documented not to: some releases seem to be in conflict
2128 with the documentation.] However, when a pattern was studied with
2129 pcre_study() and all its branches started with \s, PCRE still included VT
2130 as a possible starting character. Of course, this did no harm; it just
2131 caused an unnecessary match attempt.
2133 22. Removed a now-redundant internal flag bit that recorded the fact that case
2134 dependency changed within the pattern. This was once needed for "required
2135 byte" processing, but is no longer used. This recovers a now-scarce options
2136 bit. Also moved the least significant internal flag bit to the most-
2137 significant bit of the word, which was not previously used (hangover from
2138 the days when it was an int rather than a uint) to free up another bit for
2139 the future.
2141 23. Added support for CRLF line endings as well as CR and LF. As well as the
2142 default being selectable at build time, it can now be changed at runtime
2143 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2144 specify that it is scanning data with non-default line endings.
2146 24. Changed the definition of CXXLINK to make it agree with the definition of
2147 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2149 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2150 recursions. This makes a big different to stack usage for some patterns.
2152 26. If a subpattern containing a named recursion or subroutine reference such
2153 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2154 the space required for the compiled pattern went wrong and gave too small a
2155 value. Depending on the environment, this could lead to "Failed: internal
2156 error: code overflow at offset 49" or "glibc detected double free or
2157 corruption" errors.
2159 27. Applied patches from Google (a) to support the new newline modes and (b) to
2160 advance over multibyte UTF-8 characters in GlobalReplace.
2162 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2163 difference for some implementation of PCRE in some Windows version.
2165 29. Added some extra testing facilities to pcretest:
2167 \q<number> in a data line sets the "match limit" value
2168 \Q<number> in a data line sets the "match recursion limt" value
2169 -S <number> sets the stack size, where <number> is in megabytes
2171 The -S option isn't available for Windows.
2174 Version 6.6 06-Feb-06
2175 ---------------------
2177 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2178 in pcreposix.h. I have copied the definition from pcre.h.
2180 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2181 because pcre.h is no longer a built file.
2183 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2184 not normally included in the compiled code.
2187 Version 6.5 01-Feb-06
2188 ---------------------
2190 1. When using the partial match feature with pcre_dfa_exec(), it was not
2191 anchoring the second and subsequent partial matches at the new starting
2192 point. This could lead to incorrect results. For example, with the pattern
2193 /1234/, partially matching against "123" and then "a4" gave a match.
2195 2. Changes to pcregrep:
2197 (a) All non-match returns from pcre_exec() were being treated as failures
2198 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2199 error message is output. Some extra information is given for the
2201 probably the only errors that are likely to be caused by users (by
2202 specifying a regex that has nested indefinite repeats, for instance).
2203 If there are more than 20 of these errors, pcregrep is abandoned.
2205 (b) A binary zero was treated as data while matching, but terminated the
2206 output line if it was written out. This has been fixed: binary zeroes
2207 are now no different to any other data bytes.
2209 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2210 used to set a locale for matching. The --locale=xxxx long option has
2211 been added (no short equivalent) to specify a locale explicitly on the
2212 pcregrep command, overriding the environment variables.
2214 (d) When -B was used with -n, some line numbers in the output were one less
2215 than they should have been.
2217 (e) Added the -o (--only-matching) option.
2219 (f) If -A or -C was used with -c (count only), some lines of context were
2220 accidentally printed for the final match.
2222 (g) Added the -H (--with-filename) option.
2224 (h) The combination of options -rh failed to suppress file names for files
2225 that were found from directory arguments.
2227 (i) Added the -D (--devices) and -d (--directories) options.
2229 (j) Added the -F (--fixed-strings) option.
2231 (k) Allow "-" to be used as a file name for -f as well as for a data file.
2233 (l) Added the --colo(u)r option.
2235 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2236 is not present by default.
2238 3. A nasty bug was discovered in the handling of recursive patterns, that is,
2239 items such as (?R) or (?1), when the recursion could match a number of
2240 alternatives. If it matched one of the alternatives, but subsequently,
2241 outside the recursion, there was a failure, the code tried to back up into
2242 the recursion. However, because of the way PCRE is implemented, this is not
2243 possible, and the result was an incorrect result from the match.
2245 In order to prevent this happening, the specification of recursion has
2246 been changed so that all such subpatterns are automatically treated as
2247 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2249 4. I had overlooked the fact that, in some locales, there are characters for
2250 which isalpha() is true but neither isupper() nor islower() are true. In
2251 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2252 and ordfeminine) are like this. This affected the treatment of \w and \W
2253 when they appeared in character classes, but not when they appeared outside
2254 a character class. The bit map for "word" characters is now created
2255 separately from the results of isalnum() instead of just taking it from the
2256 upper, lower, and digit maps. (Plus the underscore character, of course.)
2258 5. The above bug also affected the handling of POSIX character classes such as
2259 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2260 permanent tables. Instead, the bit maps for such a class were previously
2261 created as the appropriate unions of the upper, lower, and digit bitmaps.
2262 Now they are created by subtraction from the [[:word:]] class, which has
2263 its own bitmap.
2265 6. The [[:blank:]] character class matches horizontal, but not vertical space.
2266 It is created by subtracting the vertical space characters (\x09, \x0a,
2267 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2268 subtraction was done in the overall bitmap for a character class, meaning
2269 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2270 be recognized. This bug has been fixed.
2272 7. Patches from the folks at Google:
2274 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2275 real life, but is still worth protecting against".
2277 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2278 regular expressions".
2280 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2281 have it.
2283 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2284 "configure" and the latter not, in order to fix a problem somebody had
2285 with compiling the Arg class on HP-UX.
2287 (e) Improve the error-handling of the C++ wrapper a little bit.
2289 (f) New tests for checking recursion limiting.
2291 8. The pcre_memmove() function, which is used only if the environment does not
2292 have a standard memmove() function (and is therefore rarely compiled),
2293 contained two bugs: (a) use of int instead of size_t, and (b) it was not
2294 returning a result (though PCRE never actually uses the result).
2296 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2297 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2298 returned instead of calling malloc() with an overflowing number that would
2299 most likely cause subsequent chaos.
2301 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2303 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2304 with this option is matched, the nmatch and pmatch options of regexec() are
2305 ignored.
2307 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2308 provided in case anyone wants to the the POSIX interface with UTF-8
2309 strings.
2311 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2312 C++ linking (needed for some HP-UX environments).
2314 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2315 (unused parameter) and in the pcre_printint() function (omitted "default"
2316 switch label when the default is to do nothing).
2318 15. Added some code to make it possible, when PCRE is compiled as a C++
2319 library, to replace subject pointers for pcre_exec() with a smart pointer
2320 class, thus making it possible to process discontinuous strings.
2322 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2323 much the same function. They were added by different people who were trying
2324 to make PCRE easy to compile on non-Unix systems. It has been suggested
2325 that PCRE_EXPORT be abolished now that there is more automatic apparatus
2326 for compiling on Windows systems. I have therefore replaced it with
2327 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2328 defaults to "extern" for C or "extern C" for C++, which works fine on
2329 Unix-like systems. It is now possible to override the value of PCRE_DATA_
2330 SCOPE with something explicit in config.h. In addition:
2332 (a) pcreposix.h still had just "extern" instead of either of these macros;
2333 I have replaced it with PCRE_DATA_SCOPE.
2335 (b) Functions such as _pcre_xclass(), which are internal to the library,
2336 but external in the C sense, all had PCRE_EXPORT in their definitions.
2337 This is apparently wrong for the Windows case, so I have removed it.
2338 (It makes no difference on Unix-like systems.)
2340 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2341 of recursive calls to match(). This is different to MATCH_LIMIT because
2342 that limits the total number of calls to match(), not all of which increase
2343 the depth of recursion. Limiting the recursion depth limits the amount of
2344 stack (or heap if NO_RECURSE is set) that is used. The default can be set
2345 when PCRE is compiled, and changed at run time. A patch from Google adds
2346 this functionality to the C++ interface.
2348 18. Changes to the handling of Unicode character properties:
2350 (a) Updated the table to Unicode 4.1.0.
2352 (b) Recognize characters that are not in the table as "Cn" (undefined).
2354 (c) I revised the way the table is implemented to a much improved format
2355 which includes recognition of ranges. It now supports the ranges that
2356 are defined in UnicodeData.txt, and it also amalgamates other
2357 characters into ranges. This has reduced the number of entries in the
2358 table from around 16,000 to around 3,000, thus reducing its size
2359 considerably. I realized I did not need to use a tree structure after
2360 all - a binary chop search is just as efficient. Having reduced the
2361 number of entries, I extended their size from 6 bytes to 8 bytes to
2362 allow for more data.
2364 (d) Added support for Unicode script names via properties such as \p{Han}.
2366 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2367 matching that character.
2369 20. When matching a repeated Unicode property with a minimum greater than zero,
2370 (for example \pL{2,}), PCRE could look past the end of the subject if it
2371 reached it while seeking the minimum number of characters. This could
2372 happen only if some of the characters were more than one byte long, because
2373 there is a check for at least the minimum number of bytes.
2375 21. Refactored the implementation of \p and \P so as to be more general, to
2376 allow for more different types of property in future. This has changed the
2377 compiled form incompatibly. Anybody with saved compiled patterns that use
2378 \p or \P will have to recompile them.
2380 22. Added "Any" and "L&" to the supported property types.
2382 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2383 but give a compile time error if the value is greater than 0xff.
2385 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2386 accidentally not being installed or uninstalled.
2388 25. The pcre.h file was built from pcre.h.in, but the only changes that were
2389 made were to insert the current release number. This seemed silly, because
2390 it made things harder for people building PCRE on systems that don't run
2391 "configure". I have turned pcre.h into a distributed file, no longer built
2392 by "configure", with the version identification directly included. There is
2393 no longer a pcre.h.in file.
2395 However, this change necessitated a change to the pcre-config script as
2396 well. It is built from pcre-config.in, and one of the substitutions was the
2397 release number. I have updated configure.ac so that ./configure now finds
2398 the release number by grepping pcre.h.
2400 26. Added the ability to run the tests under valgrind.
2403 Version 6.4 05-Sep-05
2404 ---------------------
2406 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2407 "--" to be printed when multiple files were scanned, even when none of the
2408 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2409 consider it to be a bug, and have restored the previous behaviour.
2411 2. A couple of code tidies to get rid of compiler warnings.
2413 3. The pcretest program used to cheat by referring to symbols in the library
2414 whose names begin with _pcre_. These are internal symbols that are not
2415 really supposed to be visible externally, and in some environments it is
2416 possible to suppress them. The cheating is now confined to including
2417 certain files from the library's source, which is a bit cleaner.
2419 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2420 file's purpose clearer.
2422 5. Reorganized pcre_ucp_findchar().
2425 Version 6.3 15-Aug-05
2426 ---------------------
2428 1. The file libpcre.pc.in did not have general read permission in the tarball.
2430 2. There were some problems when building without C++ support:
2432 (a) If C++ support was not built, "make install" and "make test" still
2433 tried to test it.
2435 (b) There were problems when the value of CXX was explicitly set. Some
2436 changes have been made to try to fix these, and ...
2438 (c) --disable-cpp can now be used to explicitly disable C++ support.
2440 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2441 backslash in a target when C++ was disabled. This confuses some
2442 versions of "make", apparently. Using an intermediate variable solves
2443 this. (Same for CPP_LOBJ.)
2445 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2446 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2447 necessary on certain architectures.
2449 4. Added a setting of -export-symbols-regex to the link command to remove
2450 those symbols that are exported in the C sense, but actually are local
2451 within the library, and not documented. Their names all begin with
2452 "_pcre_". This is not a perfect job, because (a) we have to except some
2453 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2454 available (and never for static libraries). I have made a note to try to
2455 find a way round (a) in the future.
2458 Version 6.2 01-Aug-05
2459 ---------------------
2461 1. There was no test for integer overflow of quantifier values. A construction
2462 such as {1111111111111111} would give undefined results. What is worse, if
2463 a minimum quantifier for a parenthesized subpattern overflowed and became
2464 negative, the calculation of the memory size went wrong. This could have
2465 led to memory overwriting.
2467 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2469 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2470 operating environments where this matters.
2472 4. Applied Giuseppe Maxia's patch to add additional features for controlling
2473 PCRE options from within the C++ wrapper.
2475 5. Named capturing subpatterns were not being correctly counted when a pattern
2476 was compiled. This caused two problems: (a) If there were more than 100
2477 such subpatterns, the calculation of the memory needed for the whole
2478 compiled pattern went wrong, leading to an overflow error. (b) Numerical
2479 back references of the form \12, where the number was greater than 9, were
2480 not recognized as back references, even though there were sufficient
2481 previous subpatterns.
2483 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2484 versions of gcc, e.g. 2.95.4.
2487 Version 6.1 21-Jun-05
2488 ---------------------
2490 1. There was one reference to the variable "posix" in pcretest.c that was not
2491 surrounded by "#if !defined NOPOSIX".
2493 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2494 the cross-check on the old pcre_info() function, for the benefit of the
2495 cut-down version of PCRE that is currently imported into Exim.
2497 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2498 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2499 patterns that start (?i) but I don't think that matters. The use of (?i) is
2500 just an example; this all applies to the other options as well.
2502 4. Since libtool seems to echo the compile commands it is issuing, the output
2503 from "make" can be reduced a bit by putting "@" in front of each libtool
2504 compile command.
2506 5. Patch from the folks at Google for configure.in to be a bit more thorough
2507 in checking for a suitable C++ installation before trying to compile the
2508 C++ stuff. This should fix a reported problem when a compiler was present,
2509 but no suitable headers.
2511 6. The man pages all had just "PCRE" as their title. I have changed them to
2512 be the relevant file name. I have also arranged that these names are
2513 retained in the file doc/pcre.txt, which is a concatenation in text format
2514 of all the man pages except the little individual ones for each function.
2516 7. The NON-UNIX-USE file had not been updated for the different set of source
2517 files that come with release 6. I also added a few comments about the C++
2518 wrapper.
2521 Version 6.0 07-Jun-05
2522 ---------------------
2524 1. Some minor internal re-organization to help with my DFA experiments.
2526 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2527 didn't matter for the library itself when fully configured, but did matter
2528 when compiling without UCP support, or within Exim, where the ucp files are
2529 not imported.
2531 3. Refactoring of the library code to split up the various functions into
2532 different source modules. The addition of the new DFA matching code (see
2533 below) to a single monolithic source would have made it really too
2534 unwieldy, quite apart from causing all the code to be include in a
2535 statically linked application, when only some functions are used. This is
2536 relevant even without the DFA addition now that patterns can be compiled in
2537 one application and matched in another.
2539 The downside of splitting up is that there have to be some external
2540 functions and data tables that are used internally in different modules of
2541 the library but which are not part of the API. These have all had their
2542 names changed to start with "_pcre_" so that they are unlikely to clash
2543 with other external names.
2545 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2546 a different (DFA) algorithm. Although it is slower than the original
2547 function, it does have some advantages for certain types of matching
2548 problem.
2550 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2551 including restarting after a partial match.
2553 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2554 defined when compiling for Windows was sent to me. I have put it into the
2555 code, though I have no means of testing or verifying it.
2557 7. Added the pcre_refcount() auxiliary function.
2559 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2560 match before or at the first newline in the subject string. In pcretest,
2561 the /f option on a pattern can be used to set this.
2563 9. A repeated \w when used in UTF-8 mode with characters greater than 256
2564 would behave wrongly. This has been present in PCRE since release 4.0.
2566 10. A number of changes to the pcregrep command:
2568 (a) Refactored how -x works; insert ^(...)$ instead of setting
2569 PCRE_ANCHORED and checking the length, in preparation for adding
2570 something similar for -w.
2572 (b) Added the -w (match as a word) option.
2574 (c) Refactored the way lines are read and buffered so as to have more
2575 than one at a time available.
2577 (d) Implemented a pcregrep test script.
2579 (e) Added the -M (multiline match) option. This allows patterns to match
2580 over several lines of the subject. The buffering ensures that at least
2581 8K, or the rest of the document (whichever is the shorter) is available
2582 for matching (and similarly the previous 8K for lookbehind assertions).
2584 (f) Changed the --help output so that it now says
2586 -w, --word-regex(p)
2588 instead of two lines, one with "regex" and the other with "regexp"
2589 because that confused at least one person since the short forms are the
2590 same. (This required a bit of code, as the output is generated
2591 automatically from a table. It wasn't just a text change.)
2593 (g) -- can be used to terminate pcregrep options if the next thing isn't an
2594 option but starts with a hyphen. Could be a pattern or a path name
2595 starting with a hyphen, for instance.
2597 (h) "-" can be given as a file name to represent stdin.
2599 (i) When file names are being printed, "(standard input)" is used for
2600 the standard input, for compatibility with GNU grep. Previously
2601 "<stdin>" was used.
2603 (j) The option --label=xxx can be used to supply a name to be used for
2604 stdin when file names are being printed. There is no short form.
2606 (k) Re-factored the options decoding logic because we are going to add
2607 two more options that take data. Such options can now be given in four
2608 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2610 (l) Added the -A, -B, and -C options for requesting that lines of context
2611 around matches be printed.
2613 (m) Added the -L option to print the names of files that do not contain
2614 any matching lines, that is, the complement of -l.
2616 (n) The return code is 2 if any file cannot be opened, but pcregrep does
2617 continue to scan other files.
2619 (o) The -s option was incorrectly implemented. For compatibility with other
2620 greps, it now suppresses the error message for a non-existent or non-
2621 accessible file (but not the return code). There is a new option called
2622 -q that suppresses the output of matching lines, which was what -s was
2623 previously doing.
2625 (p) Added --include and --exclude options to specify files for inclusion
2626 and exclusion when recursing.
2628 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2629 Hopefully, it now does.
2631 12. Missing cast in pcre_study().
2633 13. Added an "uninstall" target to the makefile.
2635 14. Replaced "extern" in the function prototypes in Makefile.in with
2636 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2637 world, but is set differently for Windows.
2639 15. Added a second compiling function called pcre_compile2(). The only
2640 difference is that it has an extra argument, which is a pointer to an
2641 integer error code. When there is a compile-time failure, this is set
2642 non-zero, in addition to the error test pointer being set to point to an
2643 error message. The new argument may be NULL if no error number is required
2644 (but then you may as well call pcre_compile(), which is now just a
2645 wrapper). This facility is provided because some applications need a
2646 numeric error indication, but it has also enabled me to tidy up the way
2647 compile-time errors are handled in the POSIX wrapper.
2649 16. Added VPATH=.libs to the makefile; this should help when building with one
2650 prefix path and installing with another. (Or so I'm told by someone who
2651 knows more about this stuff than I do.)
2653 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2654 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2655 match everything, including newlines. This is not POSIX-compatible, but
2656 somebody wanted the feature. From pcretest it can be activated by using
2657 both the P and the s flags.
2659 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2661 19. libpcre.pc was being incorrectly installed as executable.
2663 20. A couple of places in pcretest check for end-of-line by looking for '\n';
2664 it now also looks for '\r' so that it will work unmodified on Windows.
2666 21. Added Google's contributed C++ wrapper to the distribution.
2668 22. Added some untidy missing memory free() calls in pcretest, to keep
2669 Electric Fence happy when testing.
2673 Version 5.0 13-Sep-04
2674 ---------------------
2676 1. Internal change: literal characters are no longer packed up into items
2677 containing multiple characters in a single byte-string. Each character
2678 is now matched using a separate opcode. However, there may be more than one
2679 byte in the character in UTF-8 mode.
2681 2. The pcre_callout_block structure has two new fields: pattern_position and
2682 next_item_length. These contain the offset in the pattern to the next match
2683 item, and its length, respectively.
2685 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2686 insertion of callouts before each pattern item. Added the /C option to
2687 pcretest to make use of this.
2689 4. On the advice of a Windows user, the lines
2691 #if defined(_WIN32) || defined(WIN32)
2692 _setmode( _fileno( stdout ), 0x8000 );
2693 #endif /* defined(_WIN32) || defined(WIN32) */
2695 have been added to the source of pcretest. This apparently does useful
2696 magic in relation to line terminators.
2698 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2699 for the benefit of those environments where the "b" makes a difference.
2701 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2702 to know about it. I have put a hack into configure.in that adds in code
2703 to set GCC=yes if CC=icc. This seems to end up at a point in the
2704 generated configure script that is early enough to affect the setting of
2705 compiler options, which is what is needed, but I have no means of testing
2706 whether it really works. (The user who reported this had patched the
2707 generated configure script, which of course I cannot do.)
2709 LATER: After change 22 below (new libtool files), the configure script
2710 seems to know about icc (and also ecc). Therefore, I have commented out
2711 this hack in configure.in.
2713 7. Added support for pkg-config (2 patches were sent in).
2715 8. Negated POSIX character classes that used a combination of internal tables
2716 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2717 [[:^ascii]]. Typically, they would match almost any characters. The other
2718 POSIX classes were not broken in this way.
2720 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2721 to find the match, as PCRE was deluded into thinking that the match had to
2722 start at the start point or following a newline. The same bug applied to
2723 patterns with negative forward assertions or any backward assertions
2724 preceding ".*" at the start, unless the pattern required a fixed first
2725 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2727 10. In UTF-8 mode, when moving forwards in the subject after a failed match
2728 starting at the last subject character, bytes beyond the end of the subject
2729 string were read.
2731 11. Renamed the variable "class" as "classbits" to make life easier for C++
2732 users. (Previously there was a macro definition, but it apparently wasn't
2733 enough.)
2735 12. Added the new field "tables" to the extra data so that tables can be passed
2736 in at exec time, or the internal tables can be re-selected. This allows
2737 a compiled regex to be saved and re-used at a later time by a different
2738 program that might have everything at different addresses.
2740 13. Modified the pcre-config script so that, when run on Solaris, it shows a
2741 -R library as well as a -L library.
2743 14. The debugging options of pcretest (-d on the command line or D on a
2744 pattern) showed incorrect output for anything following an extended class
2745 that contained multibyte characters and which was followed by a quantifier.
2747 15. Added optional support for general category Unicode character properties
2748 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2749 support. It adds about 90K to the size of the library. The meanings of the
2750 inbuilt class escapes such as \d and \s have NOT been changed.
2752 16. Updated pcredemo.c to include calls to free() to release the memory for the
2753 compiled pattern.
2755 17. The generated file chartables.c was being created in the source directory
2756 instead of in the building directory. This caused the build to fail if the
2757 source directory was different from the building directory, and was
2758 read-only.
2760 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2761 file. No doubt somebody will tell me if they don't make sense... Also added
2762 Dan Mooney's comments about building on OpenVMS.
2764 19. Added support for partial matching via the PCRE_PARTIAL option for
2765 pcre_exec() and the \P data escape in pcretest.
2767 20. Extended pcretest with 3 new pattern features:
2769 (i) A pattern option of the form ">rest-of-line" causes pcretest to
2770 write the compiled pattern to the file whose name is "rest-of-line".
2771 This is a straight binary dump of the data, with the saved pointer to
2772 the character tables forced to be NULL. The study data, if any, is
2773 written too. After writing, pcretest reads a new pattern.
2775 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2776 compiled pattern from the given file. There must not be any
2777 occurrences of "<" in the file name (pretty unlikely); if there are,
2778 pcretest will instead treat the initial "<" as a pattern delimiter.
2779 After reading in the pattern, pcretest goes on to read data lines as
2780 usual.
2782 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2783 and 16-bit fields in a compiled pattern, to simulate a pattern that
2784 was compiled on a host of opposite endianness.
2786 21. The pcre-exec() function can now cope with patterns that were compiled on
2787 hosts of opposite endianness, with this restriction:
2789 As for any compiled expression that is saved and used later, the tables
2790 pointer field cannot be preserved; the extra_data field in the arguments
2791 to pcre_exec() should be used to pass in a tables address if a value
2792 other than the default internal tables were used at compile time.
2794 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2795 now diagnosed as an error. Previously, most of the time, a negative number
2796 would have been treated as zero, but if in addition "ovector" was passed as
2797 NULL, a crash could occur.
2799 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2800 new versions from the libtool 1.5 distribution (the last one is a copy of
2801 a file called libtool.m4). This seems to have fixed the need to patch
2802 "configure" to support Darwin 1.3 (which I used to do). However, I still
2803 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2804 workstation).
2806 24. Changed the PCRE licence to be the more standard "BSD" licence.
2809 Version 4.5 01-Dec-03
2810 ---------------------
2812 1. There has been some re-arrangement of the code for the match() function so
2813 that it can be compiled in a version that does not call itself recursively.
2814 Instead, it keeps those local variables that need separate instances for
2815 each "recursion" in a frame on the heap, and gets/frees frames whenever it
2816 needs to "recurse". Keeping track of where control must go is done by means
2817 of setjmp/longjmp. The whole thing is implemented by a set of macros that
2818 hide most of the details from the main code, and operates only if
2819 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2820 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2821 operating.
2823 To make it easier for callers to provide specially tailored get/free
2824 functions for this usage, two new functions, pcre_stack_malloc, and
2825 pcre_stack_free, are used. They are always called in strict stacking order,
2826 and the size of block requested is always the same.
2828 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2829 PCRE has been compiled to use the stack or the heap for recursion. The
2830 -C option of pcretest uses this to show which version is compiled.
2832 A new data escape \S, is added to pcretest; it causes the amounts of store
2833 obtained and freed by both kinds of malloc/free at match time to be added
2834 to the output.
2836 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2837 what's available on my current Linux desktop machine.
2839 3. When matching a UTF-8 string, the test for a valid string at the start has
2840 been extended. If start_offset is not zero, PCRE now checks that it points
2841 to a byte that is the start of a UTF-8 character. If not, it returns
2842 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2843 this is necessary because there may be backward assertions in the pattern.
2844 When matching the same subject several times, it may save resources to use
2845 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2847 4. The code for checking the validity of UTF-8 strings has been tightened so
2848 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2849 containing "overlong sequences".
2851 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2852 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2853 should have been "&", but it just so happened that all the cases this let
2854 through by mistake were picked up later in the function.
2856 6. I had used a variable called "isblank" - this is a C99 function, causing
2857 some compilers to warn. To avoid this, I renamed it (as "blankclass").
2859 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2860 prompting; (b) run "./pcretest /dev/null" at the start of the test script
2861 so the version is shown; (c) stop "make test" echoing "./RunTest".
2863 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2865 9. The prototype for memmove() for systems that don't have it was using
2866 size_t, but the inclusion of the header that defines size_t was later. I've
2867 moved the #includes for the C headers earlier to avoid this.
2869 10. Added some adjustments to the code to make it easier to compiler on certain
2870 special systems:
2872 (a) Some "const" qualifiers were missing.
2873 (b) Added the macro EXPORT before all exported functions; by default this
2874 is defined to be empty.
2875 (c) Changed the dftables auxiliary program (that builds chartables.c) so
2876 that it reads its output file name as an argument instead of writing
2877 to the standard output and assuming this can be redirected.
2879 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2880 class containing characters with values greater than 255, PCRE compilation
2881 went into a loop.
2883 12. A recursive reference to a subpattern that was within another subpattern
2884 that had a minimum quantifier of zero caused PCRE to crash. For example,
2885 (x(y(?2))z)? provoked this bug with a subject that got as far as the
2886 recursion. If the recursively-called subpattern itself had a zero repeat,
2887 that was OK.
2889 13. In pcretest, the buffer for reading a data line was set at 30K, but the
2890 buffer into which it was copied (for escape processing) was still set at
2891 1024, so long lines caused crashes.
2893 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2894 "internal error: code overflow...". This applied to any character class
2895 that was followed by a possessive quantifier.
2897 15. Modified the Makefile to add libpcre.la as a prerequisite for
2898 libpcreposix.la because I was told this is needed for a parallel build to
2899 work.
2901 16. If a pattern that contained .* following optional items at the start was
2902 studied, the wrong optimizing data was generated, leading to matching
2903 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2904 matching string must start with a or b or c. The correct conclusion for
2905 this pattern is that a match can start with any character.
2908 Version 4.4 13-Aug-03
2909 ---------------------
2911 1. In UTF-8 mode, a character class containing characters with values between
2912 127 and 255 was not handled correctly if the compiled pattern was studied.
2913 In fixing this, I have also improved the studying algorithm for such
2914 classes (slightly).
2916 2. Three internal functions had redundant arguments passed to them. Removal
2917 might give a very teeny performance improvement.
2919 3. Documentation bug: the value of the capture_top field in a callout is *one
2920 more than* the number of the hightest numbered captured substring.
2922 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2923 in incorrectly linking with a previously installed version. They now link
2924 explicitly with libpcre.la.
2926 5. configure.in no longer needs to recognize Cygwin specially.
2928 6. A problem in pcre.in for Windows platforms is fixed.
2930 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2931 pcretest, it used to include the size of the study block as part of its
2932 output. Unfortunately, the structure contains a field that has a different
2933 size on different hardware architectures. This meant that the tests that
2934 showed this size failed. As the block is currently always of a fixed size,
2935 this information isn't actually particularly useful in pcretest output, so
2936 I have just removed it.
2938 8. Three pre-processor statements accidentally did not start in column 1.
2939 Sadly, there are *still* compilers around that complain, even though
2940 standard C has not required this for well over a decade. Sigh.
2942 9. In pcretest, the code for checking callouts passed small integers in the
2943 callout_data field, which is a void * field. However, some picky compilers
2944 complained about the casts involved for this on 64-bit systems. Now
2945 pcretest passes the address of the small integer instead, which should get
2946 rid of the warnings.
2948 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2949 both compile and run time, and gives an error if an invalid UTF-8 sequence
2950 is found. There is a option for disabling this check in cases where the
2951 string is known to be correct and/or the maximum performance is wanted.
2953 11. In response to a bug report, I changed one line in Makefile.in from
2955 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2956 to
2957 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2959 to look similar to other lines, but I have no way of telling whether this
2960 is the right thing to do, as I do not use Windows. No doubt I'll get told
2961 if it's wrong...
2964 Version 4.3 21-May-03
2965 ---------------------
2967 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2968 Makefile.
2970 2. Some refactoring to improve the quality of the code:
2972 (i) The utf8_table... variables are now declared "const".
2974 (ii) The code for \cx, which used the "case flipping" table to upper case
2975 lower case letters, now just substracts 32. This is ASCII-specific,
2976 but the whole concept of \cx is ASCII-specific, so it seems
2977 reasonable.
2979 (iii) PCRE was using its character types table to recognize decimal and
2980 hexadecimal digits in the pattern. This is silly, because it handles
2981 only 0-9, a-f, and A-F, but the character types table is locale-
2982 specific, which means strange things might happen. A private
2983 table is now used for this - though it costs 256 bytes, a table is
2984 much faster than multiple explicit tests. Of course, the standard
2985 character types table is still used for matching digits in subject
2986 strings against \d.
2988 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2989 ending in _t are). So I've renamed it as ESC_tee.
2991 3. The first argument for regexec() in the POSIX wrapper should have been
2992 defined as "const".
2994 4. Changed pcretest to use malloc() for its buffers so that they can be
2995 Electric Fenced for debugging.
2997 5. There were several places in the code where, in UTF-8 mode, PCRE would try
2998 to read one or more bytes before the start of the subject string. Often this
2999 had no effect on PCRE's behaviour, but in some circumstances it could
3000 provoke a segmentation fault.
3002 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
3003 to try to read one or more bytes before the start of the subject string.
3005 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
3006 UTF-8 support could misbehave in various ways if the subject string
3007 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
3008 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
3009 back over UTF-8 characters.)
3012 Version 4.2 14-Apr-03
3013 ---------------------
3015 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
3017 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
3018 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
3019 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
3020 [WIN_PREFIX]: new variable, "cyg" for Cygwin
3021 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
3023 Note: automatic setting of the BUILD variables is not yet working
3024 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
3025 compile-time but not at link-time
3026 [LINK]: use for linking executables only
3027 make different versions for Windows and non-Windows
3028 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
3029 libraries
3030 [LINK_FOR_BUILD]: new variable
3031 [OBJEXT]: use throughout
3032 [EXEEXT]: use throughout
3033 <winshared>: new target
3034 <wininstall>: new target
3035 <dftables.o>: use native compiler
3036 <dftables>: use native linker
3037 <install>: handle Windows platform correctly
3038 <clean>: ditto
3039 <check>: ditto
3040 copy DLL to top builddir before testing
3042 As part of these changes, -no-undefined was removed again. This was reported
3043 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
3044 in any case.
3046 3. Some tidies to get rid of compiler warnings:
3048 . In the match_data structure, match_limit was an unsigned long int, whereas
3049 match_call_count was an int. I've made them both unsigned long ints.
3051 . In pcretest the fact that a const uschar * doesn't automatically cast to
3052 a void * provoked a warning.
3054 . Turning on some more compiler warnings threw up some "shadow" variables
3055 and a few more missing casts.
3057 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
3058 option, a class that contained a single character with a value between 128
3059 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
3061 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
3062 option, a class that contained several characters, but with at least one
3063 whose value was between 128 and 255 caused PCRE to crash.
3066 Version 4.1 12-Mar-03
3067 ---------------------
3069 1. Compiling with gcc -pedantic found a couple of places where casts were
3070 needed, and a string in dftables.c that was longer than standard compilers are
3071 required to support.
3073 2. Compiling with Sun's compiler found a few more places where the code could
3074 be tidied up in order to avoid warnings.
3076 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
3077 first of these names is deprecated in the latest Autoconf in favour of the name
3078 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
3079 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
3080 analogy I have changed it to CFLAGS_FOR_BUILD.
3082 4. Added -no-undefined to the linking command in the Makefile, because this is
3083 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
3084 linking step for the pcreposix library.
3086 5. PCRE was failing to diagnose the case of two named groups with the same
3087 name.
3089 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
3090 literal character that is needed in the subject for a match, and scans along to
3091 ensure that it is present before embarking on the full matching process. This
3092 saves time in cases of nested unlimited repeats that are never going to match.
3093 Problem: the scan can take a lot of time if the subject is very long (e.g.
3094 megabytes), thus penalizing straightforward matches. It is now done only if the
3095 amount of subject to be scanned is less than 1000 bytes.
3097 7. A lesser problem with the same optimization is that it was recording the
3098 first character of an anchored pattern as "needed", thus provoking a search
3099 right along the subject, even when the first match of the pattern was going to
3100 fail. The "needed" character is now not set for anchored patterns, unless it
3101 follows something in the pattern that is of non-fixed length. Thus, it still
3102 fulfils its original purpose of finding quick non-matches in cases of nested
3103 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
3106 Version 4.0 17-Feb-03
3107 ---------------------
3109 1. If a comment in an extended regex that started immediately after a meta-item
3110 extended to the end of string, PCRE compiled incorrect data. This could lead to
3111 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
3113 2. Moved to autoconf 2.53 and libtool 1.4.2.
3115 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
3116 the special perltest8 script is no longer needed - all the tests can be run
3117 from a single perltest script.
3119 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
3120 by \s. It has now been removed in PCRE. This means it isn't recognized as
3121 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
3122 class [:space:] *does* include VT, thereby creating a mess.
3124 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
3125 space and tab.
3127 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
3128 its new features into the main test script, reducing the number of scripts.
3130 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
3131 were backward compatible, and made the (?i) apply to the whole pattern, as if
3132 /i were given. Now it behaves more logically, and applies the option setting
3133 only to what follows. PCRE has been changed to follow suit. However, if it
3134 finds options settings right at the start of the pattern, it extracts them into
3135 the global options, as before. Thus, they show up in the info data.
3137 8. Added support for the \Q...\E escape sequence. Characters in between are
3138 treated as literals. This is slightly different from Perl in that $ and @ are
3139 also handled as literals inside the quotes. In Perl, they will cause variable
3140 interpolation. Note the following examples:
3142 Pattern PCRE matches Perl matches
3144 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
3145 \Qabc\$xyz\E abc\$xyz abc\$xyz
3146 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
3148 For compatibility with Perl, \Q...\E sequences are recognized inside character
3149 classes as well as outside them.
3151 9. Re-organized 3 code statements in pcretest to avoid "overflow in
3152 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
3153 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
3154 signed/unsigned warnings.
3156 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
3157 option for pcretest, so I've replaced it by a simple function that does just
3158 that job.
3160 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
3161 "pcregrep -".
3163 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
3164 Java package. This provides some syntactic sugar for simple cases of what my
3165 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
3166 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
3167 item, you can use this simplified notation. Note that only makes sense with
3168 greedy quantifiers. Consequently, the use of the possessive quantifier forces
3169 greediness, whatever the setting of the PCRE_UNGREEDY option.
3171 13. A change of greediness default within a pattern was not taking effect at
3172 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
3173 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
3174 was abstracted outside.
3176 14. PCRE now supports the \G assertion. It is true when the current matching
3177 position is at the start point of the match. This differs from \A when the
3178 starting offset is non-zero. Used with the /g option of pcretest (or similar
3179 code), it works in the same way as it does for Perl's /g option. If all
3180 alternatives of a regex begin with \G, the expression is anchored to the start
3181 match position, and the "anchored" flag is set in the compiled expression.
3183 15. Some bugs concerning the handling of certain option changes within patterns
3184 have been fixed. These applied to options other than (?ims). For example,
3185 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
3186 been the other way round. Some of this was related to change 7 above.
3188 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
3189 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
3190 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
3191 POSIX classes only within a class (e.g. /[[:alpha:]]/).
3193 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
3194 mode. Unlike ".", it always matches newline, whatever the setting of
3195 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
3196 assertions. Perl allows it, but it doesn't (in general) work because it can't
3197 calculate the length of the lookbehind. At least, that's the case for Perl
3198 5.8.0 - I've been told they are going to document that it doesn't work in
3199 future.
3201 18. Added an error diagnosis for escapes that PCRE does not support: these are
3202 \L, \l, \N, \P, \p, \U, \u, and \X.
3204 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
3205 reading past the end of the pattern in cases such as /[abcd/.
3207 20. PCRE was getting more memory than necessary for patterns with classes that
3208 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
3210 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
3211 compiling PCRE for use with Virtual Pascal.
3213 22. Small fix to the Makefile to make it work properly if the build is done
3214 outside the source tree.
3216 23. Added a new extension: a condition to go with recursion. If a conditional
3217 subpattern starts with (?(R) the "true" branch is used if recursion has
3218 happened, whereas the "false" branch is used only at the top level.
3220 24. When there was a very long string of literal characters (over 255 bytes
3221 without UTF support, over 250 bytes with UTF support), the computation of how
3222 much memory was required could be incorrect, leading to segfaults or other
3223 strange effects.
3225 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
3226 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
3227 there was a subsequent back reference to those brackets. This meant that, for
3228 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
3229 possible to check for precisely this case. All we can do is abandon the
3230 optimization if .* occurs inside capturing brackets when there are any back
3231 references whatsoever. (See below for a better fix that came later.)
3233 26. The handling of the optimization for finding the first character of a
3234 non-anchored pattern, and for finding a character that is required later in the
3235 match were failing in some cases. This didn't break the matching; it just
3236 failed to optimize when it could. The way this is done has been re-implemented.
3238 27. Fixed typo in error message for invalid (?R item (it said "(?p").
3240 28. Added a new feature that provides some of the functionality that Perl
3241 provides with (?{...}). The facility is termed a "callout". The way it is done
3242 in PCRE is for the caller to provide an optional function, by setting
3243 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
3244 global variable. By default it is unset, which disables all calling out. To get
3245 the function called, the regex must include (?C) at appropriate points. This
3246 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
3247 This provides a means of identifying different callout points. When PCRE
3248 reaches such a point in the regex, if pcre_callout has been set, the external
3249 function is called. It is provided with data in a structure called
3250 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
3251 matching continues; if it returns a non-zero value, the match at the current
3252 point fails. However, backtracking will occur if possible. [This was changed
3253 later and other features added - see item 49 below.]
3255 29. pcretest is upgraded to test the callout functionality. It provides a
3256 callout function that displays information. By default, it shows the start of
3257 the match and the current position in the text. There are some new data escapes
3258 to vary what happens:
3260 \C+ in addition, show current contents of captured substrings
3261 \C- do not supply a callout function
3262 \C!n return 1 when callout number n is reached
3263 \C!n!m return 1 when callout number n is reached for the mth time
3265 30. If pcregrep was called with the -l option and just a single file name, it
3266 output "<stdin>" if a match was found, instead of the file name.
3268 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
3269 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
3270 pcre_exec(). This saves a malloc/free per call. The default value of
3271 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
3272 when configuring.
3274 32. The default maximum size of a compiled pattern is 64K. There have been a
3275 few cases of people hitting this limit. The code now uses macros to handle the
3276 storing of links as offsets within the compiled pattern. It defaults to 2-byte
3277 links, but this can be changed to 3 or 4 bytes by --with-link-size when
3278 configuring. Tests 2 and 5 work only with 2-byte links because they output
3279 debugging information about compiled patterns.
3281 33. Internal code re-arrangements:
3283 (a) Moved the debugging function for printing out a compiled regex into
3284 its own source file (printint.c) and used #include to pull it into
3285 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
3286 separate copies.
3288 (b) Defined the list of op-code names for debugging as a macro in
3289 internal.h so that it is next to the definition of the opcodes.
3291 (c) Defined a table of op-code lengths for simpler skipping along compiled
3292 code. This is again a macro in internal.h so that it is next to the
3293 definition of the opcodes.
3295 34. Added support for recursive calls to individual subpatterns, along the
3296 lines of Robin Houston's patch (but implemented somewhat differently).
3298 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
3299 allow it to read and process whole directories in Win32. This code was
3300 contributed by Lionel Fourquaux; it has not been tested by me.
3302 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
3303 used to name a group. Names consist of alphanumerics and underscores, and must
3304 be unique. Back references use the syntax (?P=name) and recursive calls use
3305 (?P>name) which is a PCRE extension to the Python extension. Groups still have
3306 numbers. The function pcre_fullinfo() can be used after compilation to extract
3307 a name/number map. There are three relevant calls:
3309 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
3310 PCRE_INFO_NAMECOUNT yields the number of entries
3311 PCRE_INFO_NAMETABLE yields a pointer to the map.
3313 The map is a vector of fixed-size entries. The size of each entry depends on
3314 the length of the longest name used. The first two bytes of each entry are the
3315 group number, most significant byte first. There follows the corresponding
3316 name, zero terminated. The names are in alphabetical order.
3318 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
3319 case instead of 255. Making it the same both with and without UTF-8 support
3320 means that the same test output works with both.
3322 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
3323 calling malloc() with a zero argument.
3325 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
3326 optimization. I've improved things by keeping a bitmap of backreferences with
3327 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
3328 fact referenced, the optimization can be applied. It is unlikely that a
3329 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
3330 the match to follow \n) will appear inside brackets with a number greater than
3331 31, but if it does, any back reference > 31 suppresses the optimization.
3333 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
3334 of disabling numbered capturing parentheses. Any opening parenthesis that is
3335 not followed by ? behaves as if it were followed by ?: but named parentheses
3336 can still be used for capturing (and they will acquire numbers in the usual
3337 way).
3339 41. Redesigned the return codes from the match() function into yes/no/error so
3340 that errors can be passed back from deep inside the nested calls. A malloc
3341 failure while inside a recursive subpattern call now causes the
3342 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
3344 42. It is now possible to set a limit on the number of times the match()
3345 function is called in a call to pcre_exec(). This facility makes it possible to
3346 limit the amount of recursion and backtracking, though not in a directly
3347 obvious way, because the match() function is used in a number of different
3348 circumstances. The count starts from zero for each position in the subject
3349 string (for non-anchored patterns). The default limit is, for compatibility, a
3350 large number, namely 10 000 000. You can change this in two ways:
3352 (a) When configuring PCRE before making, you can use --with-match-limit=n
3353 to set a default value for the compiled library.
3355 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
3356 a different value is set. See 45 below.
3358 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3360 43. Added a new function pcre_config(int, void *) to enable run-time extraction
3361 of things that can be changed at compile time. The first argument specifies
3362 what is wanted and the second points to where the information is to be placed.
3363 The current list of available information is:
3367 The output is an integer that is set to one if UTF-8 support is available;
3368 otherwise it is set to zero.
3372 The output is an integer that it set to the value of the code that is used for
3373 newline. It is either LF (10) or CR (13).
3377 The output is an integer that contains the number of bytes used for internal
3378 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3382 The output is an integer that contains the threshold above which the POSIX
3383 interface uses malloc() for output vectors. See item 31 above.
3387 The output is an unsigned integer that contains the default limit of the number
3388 of match() calls in a pcre_exec() execution. See 42 above.
3390 44. pcretest has been upgraded by the addition of the -C option. This causes it
3391 to extract all the available output from the new pcre_config() function, and to
3392 output it. The program then exits immediately.
3394 45. A need has arisen to pass over additional data with calls to pcre_exec() in
3395 order to support additional features. One way would have been to define
3396 pcre_exec2() (for example) with extra arguments, but this would not have been
3397 extensible, and would also have required all calls to the original function to
3398 be mapped to the new one. Instead, I have chosen to extend the mechanism that
3399 is used for passing in "extra" data from pcre_study().
3401 The pcre_extra structure is now exposed and defined in pcre.h. It currently
3402 contains the following fields:
3404 flags a bitmap indicating which of the following fields are set
3405 study_data opaque data from pcre_study()
3406 match_limit a way of specifying a limit on match() calls for a specific
3407 call to pcre_exec()
3408 callout_data data for callouts (see 49 below)
3410 The flag bits are also defined in pcre.h, and are
3416 The pcre_study() function now returns one of these new pcre_extra blocks, with
3417 the actual study data pointed to by the study_data field, and the
3418 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3419 before. That is, this change is entirely upwards-compatible and requires no
3420 change to existing code.
3422 If you want to pass in additional data to pcre_exec(), you can either place it
3423 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3424 block.
3426 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3427 data string contains the escape sequence \M, pcretest calls pcre_exec() several
3428 times with different match limits, until it finds the minimum value needed for
3429 pcre_exec() to complete. The value is then output. This can be instructive; for
3430 most simple matches the number is quite small, but for pathological cases it
3431 gets very large very quickly.
3433 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3434 returns the size of the data block pointed to by the study_data field in a
3435 pcre_extra block, that is, the value that was passed as the argument to
3436 pcre_malloc() when PCRE was getting memory in which to place the information
3437 created by pcre_study(). The fourth argument should point to a size_t variable.
3438 pcretest has been extended so that this information is shown after a successful
3439 pcre_study() call when information about the compiled regex is being displayed.
3441 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3442 because what follows is always an absolute path. (Later: it turns out that this
3443 is more than cosmetic for MinGW, because it doesn't like empty path
3444 components.)
3446 49. Some changes have been made to the callout feature (see 28 above):
3448 (i) A callout function now has three choices for what it returns:
3450 0 => success, carry on matching
3451 > 0 => failure at this point, but backtrack if possible
3452 < 0 => serious error, return this value from pcre_exec()
3454 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3455 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3456 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3457 use by callout functions. It will never be used by PCRE itself.
3459 (ii) The pcre_extra structure (see 45 above) has a void * field called
3460 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3461 pcre_callout_block structure has a field of the same name. The contents of
3462 the field passed in the pcre_extra structure are passed to the callout
3463 function in the corresponding field in the callout block. This makes it
3464 easier to use the same callout-containing regex from multiple threads. For
3465 testing, the pcretest program has a new data escape
3467 \C*n pass the number n (may be negative) as callout_data
3469 If the callout function in pcretest receives a non-zero value as
3470 callout_data, it returns that value.
3472 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3473 there were some redundant $(CFLAGS) in commands that are now specified as
3474 $(LINK), which already includes $(CFLAGS).
3476 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3477 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3478 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3479 one-byte characters throughout. Note that case-insensitive matching applies
3480 only to characters whose values are less than 256. PCRE doesn't support the
3481 notion of cases for higher-valued characters.
3483 (i) A character class whose characters are all within 0-255 is handled as
3484 a bit map, and the map is inverted for negative classes. Previously, a
3485 character > 255 always failed to match such a class; however it should
3486 match if the class was a negative one (e.g. [^ab]). This has been fixed.
3488 (ii) A negated character class with a single character < 255 is coded as
3489 "not this character" (OP_NOT). This wasn't working properly when the test
3490 character was multibyte, either singly or repeated.
3492 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3493 mode, for example: \x{100}{2,3}.
3495 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3496 singly or repeated) now correctly test multibyte characters. However,
3497 PCRE doesn't recognize any characters with values greater than 255 as
3498 digits, spaces, or word characters. Such characters always match \D, \S,
3499 and \W, and never match \d, \s, or \w.
3501 (v) Classes may now contain characters and character ranges with values
3502 greater than 255. For example: [ab\x{100}-\x{400}].
3504 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
3505 PCRE in UTF-8 mode.
3507 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3508 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3509 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3510 value.)
3512 53. The single man page has become too large. I have therefore split it up into
3513 a number of separate man pages. These also give rise to individual HTML pages;
3514 these are now put in a separate directory, and there is an index.html page that
3515 lists them all. Some hyperlinking between the pages has been installed.
3517 54. Added convenience functions for handling named capturing parentheses.
3519 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3520 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3521 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3522 are faulted.
3524 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3525 calling configure. These values are used when compiling the dftables.c program
3526 which is run to generate the source of the default character tables. They
3527 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3528 you will need to set these values.
3530 57. Updated the building process for Windows DLL, as provided by Fred Cox.
3533 Version 3.9 02-Jan-02
3534 ---------------------
3536 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3538 2. If --disable-static was given, the building process failed when trying to
3539 build pcretest and pcregrep. (For some reason it was using libtool to compile
3540 them, which is not right, as they aren't part of the library.)
3543 Version 3.8 18-Dec-01
3544 ---------------------
3546 1. The experimental UTF-8 code was completely screwed up. It was packing the
3547 bytes in the wrong order. How dumb can you get?
3550 Version 3.7 29-Oct-01
3551 ---------------------
3553 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3554 This caused pcretest, when used on the test data, to segfault. Unfortunately,
3555 this didn't happen under Solaris 8, where I normally test things.
3557 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3558 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3559 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3560 this fix an hour or so after the initial 3.7 release.)
3563 Version 3.6 23-Oct-01
3564 ---------------------
3566 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3567 offsets passed as NULL with zero offset count.
3569 2. The config.guess and config.sub files had not been updated when I moved to
3570 the latest autoconf.
3573 Version 3.5 15-Aug-01
3574 ---------------------
3576 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
3577 had been forgotten.
3579 2. By using declared but undefined structures, we can avoid using "void"
3580 definitions in pcre.h while keeping the internal definitions of the structures
3581 private.
3583 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
3584 user point of view, this means that both static and shared libraries are built
3585 by default, but this can be individually controlled. More of the work of
3586 handling this static/shared cases is now inside libtool instead of PCRE's make
3587 file.
3589 4. The pcretest utility is now installed along with pcregrep because it is
3590 useful for users (to test regexs) and by doing this, it automatically gets
3591 relinked by libtool. The documentation has been turned into a man page, so
3592 there are now .1, .txt, and .html versions in /doc.
3594 5. Upgrades to pcregrep:
3595 (i) Added long-form option names like gnu grep.
3596 (ii) Added --help to list all options with an explanatory phrase.
3597 (iii) Added -r, --recursive to recurse into sub-directories.
3598 (iv) Added -f, --file to read patterns from a file.
3600 6. pcre_exec() was referring to its "code" argument before testing that
3601 argument for NULL (and giving an error if it was NULL).
3603 7. Upgraded Makefile.in to allow for compiling in a different directory from
3604 the source directory.
3606 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
3607 options bits, the pointer it was passed was to an int instead of to an unsigned
3608 long int. This mattered only on 64-bit systems.
3610 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3611 generated) instead of pcre.in, which it its source. Also made the same change
3612 in several of the .c files.
3614 10. A new release of gcc defines printf() as a macro, which broke pcretest
3615 because it had an ifdef in the middle of a string argument for printf(). Fixed
3616 by using separate calls to printf().
3618 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
3619 script, to force use of CR or LF instead of \n in the source. On non-Unix
3620 systems, the value can be set in config.h.
3622 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
3623 absolute limit. Changed the text of the error message to make this clear, and
3624 likewise updated the man page.
3626 13. The limit of 99 on the number of capturing subpatterns has been removed.
3627 The new limit is 65535, which I hope will not be a "real" limit.
3630 Version 3.4 22-Aug-00
3631 ---------------------
3633 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
3635 2. Diagnose condition (?(0) as an error instead of crashing on matching.
3638 Version 3.3 01-Aug-00
3639 ---------------------
3641 1. If an octal character was given, but the value was greater than \377, it
3642 was not getting masked to the least significant bits, as documented. This could
3643 lead to crashes in some systems.
3645 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
3646 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
3648 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
3649 These just pass their arguments on to (pcre_free)(), but they are provided
3650 because some uses of PCRE bind it to non-C systems that can call its functions,
3651 but cannot call free() or pcre_free() directly.
3653 4. Add "make test" as a synonym for "make check". Corrected some comments in
3654 the Makefile.
3656 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
3657 Makefile.
3659 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
3660 command called pgrep for grepping around the active processes.
3662 7. Added the beginnings of support for UTF-8 character strings.
3664 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
3665 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
3666 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
3667 out for the ar command.)
3670 Version 3.2 12-May-00
3671 ---------------------
3673 This is purely a bug fixing release.
3675 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
3676 of ZA. This was just one example of several cases that could provoke this bug,
3677 which was introduced by change 9 of version 2.00. The code for breaking
3678 infinite loops after an iteration that matches an empty string was't working
3679 correctly.
3681 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3682 when matched against abbab (for example). After matching an empty string, it
3683 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3684 caused it to match further down the string than it should.
3686 3. The code contained an inclusion of sys/types.h. It isn't clear why this
3687 was there because it doesn't seem to be needed, and it causes trouble on some
3688 systems, as it is not a Standard C header. It has been removed.
3690 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3691 were reported on the Macintosh. The changes were from
3693 while ((c = *(++ptr)) != 0 && c != '\n');
3694 to
3695 while ((c = *(++ptr)) != 0 && c != '\n') ;
3697 Totally extraordinary, but if that's what it takes...
3699 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3700 available. Added HAVE_BCOPY and an autoconf test for it; if neither
3701 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3702 assumes the way PCRE uses memmove() (always moving upwards).
3704 6. PCRE is being used in one environment where strchr() is not available. There
3705 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3706 faster code anyway.
3709 Version 3.1 09-Feb-00
3710 ---------------------
3712 The only change in this release is the fixing of some bugs in Makefile.in for
3713 the "install" target:
3715 (1) It was failing to install pcreposix.h.
3717 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3720 Version 3.0 01-Feb-00
3721 ---------------------
3723 1. Add support for the /+ modifier to perltest (to output $` like it does in
3724 pcretest).
3726 2. Add support for the /g modifier to perltest.
3728 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3729 matches null strings.
3731 4. Fix perltest so that it doesn't do unwanted things when fed an empty
3732 pattern. Perl treats empty patterns specially - it reuses the most recent
3733 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3734 effect.
3736 5. The POSIX interface was broken in that it was just handing over the POSIX
3737 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3738 required a bigger vector, with some working space on the end. This means that
3739 the POSIX wrapper now has to get and free some memory, and copy the results.
3741 6. Added some simple autoconf support, placing the test data and the
3742 documentation in separate directories, re-organizing some of the
3743 information files, and making it build pcre-config (a GNU standard). Also added
3744 libtool support for building PCRE as a shared library, which is now the
3745 default.
3747 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3748 09 are not valid octal constants. Single digits will be used for minor values
3749 less than 10.
3751 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3752 existing programs that set these in the POSIX interface can use PCRE without
3753 modification.
3755 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3756 return all that pcre_info() returns, plus additional data. The pcre_info()
3757 function is retained for compatibility, but is considered to be obsolete.
3759 10. Added experimental recursion feature (?R) to handle one common case that
3760 Perl 5.6 will be able to do with (?p{...}).
3762 11. Added support for POSIX character classes like [:alpha:], which Perl is
3763 adopting.
3766 Version 2.08 31-Aug-99
3767 ----------------------
3769 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3770 trying to match at the startoffset position, but instead was moving forward to
3771 the next newline as if a previous match had failed.
3773 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3774 and could get into a loop if a null string was matched other than at the start
3775 of the subject.
3777 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3778 be distinguished at compile time, and for completeness also added PCRE_DATE.
3780 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3781 in GnuWin32 environments.
3784 Version 2.07 29-Jul-99
3785 ----------------------
3787 1. The documentation is now supplied in plain text form and HTML as well as in
3788 the form of man page sources.
3790 2. C++ compilers don't like assigning (void *) values to other pointer types.
3791 In particular this affects malloc(). Although there is no problem in Standard
3792 C, I've put in casts to keep C++ compilers happy.
3794 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3795 should be (const char *).
3797 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3798 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3799 However, I haven't made this a standard facility. The documentation doesn't
3800 mention it, and the Makefile doesn't support it.
3802 5. The Makefile now contains an "install" target, with editable destinations at
3803 the top of the file. The pcretest program is not installed.
3805 6. pgrep -V now gives the PCRE version number and date.
3807 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3808 causing the entire string to be ignored, instead of just the last character.
3810 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3811 non-matching string, it can take a very, very long time, even for strings of
3812 quite modest length, because of the nested recursion. PCRE now does better in
3813 some of these cases. It does this by remembering the last required literal
3814 character in the pattern, and pre-searching the subject to ensure it is present
3815 before running the real match. In other words, it applies a heuristic to detect
3816 some types of certain failure quickly, and in the above example, if presented
3817 with a string that has no trailing " it gives "no match" very quickly.
3819 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3820 other alternatives are tried instead.
3823 Version 2.06 09-Jun-99
3824 ----------------------
3826 1. Change pcretest's output for amount of store used to show just the code
3827 space, because the remainder (the data block) varies in size between 32-bit and
3828 64-bit systems.
3830 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3831 start matching at. This allows lookbehinds to work when searching for multiple
3832 occurrences in a string.
3834 3. Added additional options to pcretest for testing multiple occurrences:
3836 /+ outputs the rest of the string that follows a match
3837 /g loops for multiple occurrences, using the new startoffset argument
3838 /G loops for multiple occurrences by passing an incremented pointer
3840 4. PCRE wasn't doing the "first character" optimization for patterns starting
3841 with \b or \B, though it was doing it for other lookbehind assertions. That is,
3842 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3843 the letter 'x'. On long subject strings, this gives a significant speed-up.
3846 Version 2.05 21-Apr-99
3847 ----------------------
3849 1. Changed the type of magic_number from int to long int so that it works
3850 properly on 16-bit systems.
3852 2. Fixed a bug which caused patterns starting with .* not to work correctly
3853 when the subject string contained newline characters. PCRE was assuming
3854 anchoring for such patterns in all cases, which is not correct because .* will
3855 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3856 DOTALL is set at top level; otherwise it knows that patterns starting with .*
3857 must be retried after every newline in the subject.
3860 Version 2.04 18-Feb-99
3861 ----------------------
3863 1. For parenthesized subpatterns with repeats whose minimum was zero, the
3864 computation of the store needed to hold the pattern was incorrect (too large).
3865 If such patterns were nested a few deep, this could multiply and become a real
3866 problem.
3868 2. Added /M option to pcretest to show the memory requirement of a specific
3869 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3871 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3872 compiled in such a way that the backtracking after subsequent failure was
3873 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3874 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3877 Version 2.03 02-Feb-99
3878 ----------------------
3880 1. Fixed typo and small mistake in man page.
3882 2. Added 4th condition (GPL supersedes if conflict) and created separate
3883 LICENCE file containing the conditions.
3885 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3886 Perl, that is the internal \ allows the delimiter to be included in the
3887 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3888 the final delimiter, add \ to the end of the pattern (to test the error).
3890 4. Added the convenience functions for extracting substrings after a successful
3891 match. Updated pcretest to make it able to test these functions.
3894 Version 2.02 14-Jan-99
3895 ----------------------
3897 1. Initialized the working variables associated with each extraction so that
3898 their saving and restoring doesn't refer to uninitialized store.
3900 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3901 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3902 fix the problem.
3904 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3905 calls, and wasn't printing the correct value for compiling calls. Increased the
3906 default value of LOOPREPEAT, and the number of significant figures in the
3907 times.
3909 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3911 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3912 a building problem on Windows NT with a FAT file system.
3915 Version 2.01 21-Oct-98
3916 ----------------------
3918 1. Changed the API for pcre_compile() to allow for the provision of a pointer
3919 to character tables built by pcre_maketables() in the current locale. If NULL
3920 is passed, the default tables are used.
3923 Version 2.00 24-Sep-98
3924 ----------------------
3926 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3927 it any more.
3929 2. Allow quantification of (?>) groups, and make it work correctly.
3931 3. The first character computation wasn't working for (?>) groups.
3933 4. Correct the implementation of \Z (it is permitted to match on the \n at the
3934 end of the subject) and add 5.005's \z, which really does match only at the
3935 very end of the subject.
3937 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3939 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3940 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3941 localized options. All options to pcre_study() were also removed.
3943 7. Add other new features from 5.005:
3945 $(?<= positive lookbehind
3946 $(?<! negative lookbehind
3947 (?imsx-imsx) added the unsetting capability
3948 such a setting is global if at outer level; local otherwise
3949 (?imsx-imsx:) non-capturing groups with option setting
3950 (?(cond)re|re) conditional pattern matching
3952 A backreference to itself in a repeated group matches the previous
3953 captured string.
3955 8. General tidying up of studying (both automatic and via "study")
3956 consequential on the addition of new assertions.
3958 9. As in 5.005, unlimited repeated groups that could match an empty substring
3959 are no longer faulted at compile time. Instead, the loop is forcibly broken at
3960 runtime if any iteration does actually match an empty substring.
3962 10. Include the RunTest script in the distribution.
3964 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3965 discrepancies, some of which were old and were also with respect to 5.004. They
3966 have now been fixed.
3969 Version 1.09 28-Apr-98
3970 ----------------------
3972 1. A negated single character class followed by a quantifier with a minimum
3973 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
3974 program crashes, or just wrong answers. This did not apply to negated classes
3975 containing more than one character, or to minima other than one.
3978 Version 1.08 27-Mar-98
3979 ----------------------
3981 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
3983 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
3984 latter must appear before anything that relies on it in the pattern.
3987 Version 1.07 16-Feb-98
3988 ----------------------
3990 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
3991 repeat of a potentially empty string).
3994 Version 1.06 23-Jan-98
3995 ----------------------
3997 1. Added Markus Oberhumer's little patches for C++.
3999 2. Literal strings longer than 255 characters were broken.
4002 Version 1.05 23-Dec-97
4003 ----------------------
4005 1. Negated character classes containing more than one character were failing if
4006 PCRE_CASELESS was set at run time.
4009 Version 1.04 19-Dec-97
4010 ----------------------
4012 1. Corrected the man page, where some "const" qualifiers had been omitted.
4014 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
4015 input syntax.
4017 3. Fixed memory leak which occurred when a regex with back references was
4018 matched with an offsets vector that wasn't big enough. The temporary memory
4019 that is used in this case wasn't being freed if the match failed.
4021 4. Tidied pcretest to ensure it frees memory that it gets.
4023 5. Temporary memory was being obtained in the case where the passed offsets
4024 vector was exactly big enough.
4026 6. Corrected definition of offsetof() from change 5 below.
4028 7. I had screwed up change 6 below and broken the rules for the use of
4029 setjmp(). Now fixed.
4032 Version 1.03 18-Dec-97
4033 ----------------------
4035 1. A erroneous regex with a missing opening parenthesis was correctly
4036 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
4037 on some systems.
4039 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
4040 it was reported that one broken compiler failed on the former because "code" is
4041 also an independent variable.
4043 3. The erroneous regex a[]b caused an array overrun reference.
4045 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
4046 fail on data ending with that character. (It was going on too far, and checking
4047 the next character, typically a binary zero.) This was specific to the
4048 optimized code for single-character negative classes.
4050 5. Added a contributed patch from the TIN world which does the following:
4052 + Add an undef for memmove, in case the the system defines a macro for it.
4054 + Add a definition of offsetof(), in case there isn't one. (I don't know
4055 the reason behind this - offsetof() is part of the ANSI standard - but
4056 it does no harm).
4058 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
4059 most of the places where whitespace preceded '#'. I have given up and
4060 allowed the remaining 2 cases to be at the margin.
4062 + Rename some variables in pcre to eliminate shadowing. This seems very
4063 pedantic, but does no harm, of course.
4065 6. Moved the call to setjmp() into its own function, to get rid of warnings
4066 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
4068 7. Constructs such as \d{8,} were compiling into the equivalent of
4069 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
4070 outcome, but in this particular case used more store than had been allocated,
4071 which caused the bug to be discovered because it threw up an internal error.
4073 8. The debugging code in both pcre and pcretest for outputting the compiled
4074 form of a regex was going wrong in the case of back references followed by
4075 curly-bracketed repeats.
4078 Version 1.02 12-Dec-97
4079 ----------------------
4081 1. Typos in pcre.3 and comments in the source fixed.
4083 2. Applied a contributed patch to get rid of places where it used to remove
4084 'const' from variables, and fixed some signed/unsigned and uninitialized
4085 variable warnings.
4087 3. Added the "runtest" target to Makefile.
4089 4. Set default compiler flag to -O2 rather than just -O.
4092 Version 1.01 19-Nov-97
4093 ----------------------
4095 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
4096 like /([ab]*)*/, that is, for classes with more than one character in them.
4098 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
4099 as /((?>a*))*/ (a PCRE_EXTRA facility).
4102 Version 1.00 18-Nov-97
4103 ----------------------
4105 1. Added compile-time macros to support systems such as SunOS4 which don't have
4106 memmove() or strerror() but have other things that can be used instead.
4108 2. Arranged that "make clean" removes the executables.
4111 Version 0.99 27-Oct-97
4112 ----------------------
4114 1. Fixed bug in code for optimizing classes with only one character. It was
4115 initializing a 32-byte map regardless, which could cause it to run off the end
4116 of the memory it had got.
4118 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
4121 Version 0.98 22-Oct-97
4122 ----------------------
4124 1. Fixed bug in code for handling temporary memory usage when there are more
4125 back references than supplied space in the ovector. This could cause segfaults.
4128 Version 0.97 21-Oct-97
4129 ----------------------
4131 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
4133 2. Optimized negated single characters not to use a bit map.
4135 3. Brought error texts together as macro definitions; clarified some of them;
4136 fixed one that was wrong - it said "range out of order" when it meant "invalid
4137 escape sequence".
4139 4. Changed some char * arguments to const char *.
4141 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
4143 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
4144 pcretest.
4147 Version 0.96 16-Oct-97
4148 ----------------------
4150 1. Added a simple "pgrep" utility to the distribution.
4152 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
4153 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
4154 where "ddd" means "one or more decimal digits".
4156 3. Fixed serious bug. If a pattern had a back reference, but the call to
4157 pcre_exec() didn't supply a large enough ovector to record the related
4158 identifying subpattern, the match always failed. PCRE now remembers the number
4159 of the largest back reference, and gets some temporary memory in which to save
4160 the offsets during matching if necessary, in order to ensure that
4161 backreferences always work.
4163 4. Increased the compatibility with Perl in a number of ways:
4165 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
4166 to request this handling. The option can be set at compile or exec time.
4168 (b) $ matches before a terminating newline by default; an option
4169 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
4170 mode). The option can be set at compile or exec time.
4172 (c) The handling of \ followed by a digit other than 0 is now supposed to be
4173 the same as Perl's. If the decimal number it represents is less than 10
4174 or there aren't that many previous left capturing parentheses, an octal
4175 escape is read. Inside a character class, it's always an octal escape,
4176 even if it is a single digit.
4178 (d) An escaped but undefined alphabetic character is taken as a literal,
4179 unless PCRE_EXTRA is set. Currently this just reserves the remaining
4180 escapes.
4182 (e) {0} is now permitted. (The previous item is removed from the compiled
4183 pattern).
4185 5. Changed all the names of code files so that the basic parts are no longer
4186 than 10 characters, and abolished the teeny "globals.c" file.
4188 6. Changed the handling of character classes; they are now done with a 32-byte
4189 bit map always.
4191 7. Added the -d and /D options to pcretest to make it possible to look at the
4192 internals of compilation without having to recompile pcre.
4195 Version 0.95 23-Sep-97
4196 ----------------------
4198 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
4199 \x20 at the start of a run of normal characters. These were being treated as
4200 real characters, instead of the source characters being re-checked.
4203 Version 0.94 18-Sep-97
4204 ----------------------
4206 1. The functions are now thread-safe, with the caveat that the global variables
4207 containing pointers to malloc() and free() or alternative functions are the
4208 same for all threads.
4210 2. Get pcre_study() to generate a bitmap of initial characters for non-
4211 anchored patterns when this is possible, and use it if passed to pcre_exec().
4214 Version 0.93 15-Sep-97
4215 ----------------------
4217 1. /(b)|(:+)/ was computing an incorrect first character.
4219 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
4220 but not actually doing anything yet.
4222 3. Treat "-" characters in classes that cannot be part of ranges as literals,
4223 as Perl does (e.g. [-az] or [az-]).
4225 4. Set the anchored flag if a branch starts with .* or .*? because that tests
4226 all possible positions.
4228 5. Split up into different modules to avoid including unneeded functions in a
4229 compiled binary. However, compile and exec are still in one module. The "study"
4230 function is split off.
4232 6. The character tables are now in a separate module whose source is generated
4233 by an auxiliary program - but can then be edited by hand if required. There are
4234 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
4235 toupper() in the code.
4237 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
4238 make them global. Abolish the function for setting them, as the caller can now
4239 set them directly.
4242 Version 0.92 11-Sep-97
4243 ----------------------
4245 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
4246 (e.g. /a{1,3}/) was broken (I mis-optimized it).
4248 2. Caseless matching was not working in character classes if the characters in
4249 the pattern were in upper case.
4251 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
4253 4. Make PCRE_ANCHORED public and accept as a compile option.
4255 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
4256 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
4257 pass them.
4259 6. Give an error if bad option bits passed at compile or run time.
4261 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
4262 pcretest to cause it to pass that flag.
4264 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
4265 options, and the first character, if set.
4267 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
4270 Version 0.91 10-Sep-97
4271 ----------------------
4273 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
4274 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
4276 2. PCRE was looping on encountering an indefinitely repeated back reference to
4277 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
4278 Perl does - treats the match as successful.
4280 ****


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

  ViewVC Help
Powered by ViewVC 1.1.5