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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 274 - (show annotations) (download)
Tue Nov 20 10:05:23 2007 UTC (6 years, 8 months ago) by ph10
File size: 143771 byte(s)
Change "alphameric" to "alphanumeric".

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

Properties

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

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12