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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 284 - (show annotations) (download)
Fri Dec 7 20:01:16 2007 UTC (6 years, 9 months ago) by ph10
File size: 145779 byte(s)
Oops, forgot the ChangeLog.

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 9. Fix two occurrences of "backslash" in comments that should have been
43 "backspace".
44
45 10. Remove two redundant lines of code that can never be obeyed (their function
46 was moved elsewhere).
47
48 11. The program that makes PCRE's Unicode character property table had a bug
49 which caused it to generate incorrect table entries for sequences of
50 characters that have the same character type, but are in different scripts.
51 It amalgamated them into a single range, with the script of the first of
52 them. In other words, some characters were in the wrong script. There were
53 thirteen such cases, affecting characters in the following ranges:
54
55 U+002b0 - U+002c1
56 U+0060c - U+0060d
57 U+0061e - U+00612
58 U+0064b - U+0065e
59 U+0074d - U+0076d
60 U+01800 - U+01805
61 U+01d00 - U+01d77
62 U+01d9b - U+01dbf
63 U+0200b - U+0200f
64 U+030fc - U+030fe
65 U+03260 - U+0327f
66 U+0fb46 - U+0fbb1
67 U+10450 - U+1049d
68
69 12. The -o option (show only the matching part of a line) for pcregrep was not
70 compatible with GNU grep in that, if there was more than one match in a
71 line, it showed only the first of them. It now behaves in the same way as
72 GNU grep.
73
74 13. If the -o and -v options were combined for pcregrep, it printed a blank
75 line for every non-matching line. GNU grep prints nothing, and pcregrep now
76 does the same. The return code can be used to tell if there were any
77 non-matching lines.
78
79 14. The pattern (?=something)(?R) was not being diagnosed as a potentially
80 infinitely looping recursion. The bug was that positive lookaheads were not
81 being skipped when checking for a possible empty match (negative lookaheads
82 and both kinds of lookbehind were skipped).
83
84 15. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
85 inclusion of <windows.h> to before rather than after the definition of
86 INVALID_FILE_ATTRIBUTES (patch from David Byron).
87
88
89 Version 7.4 21-Sep-07
90 ---------------------
91
92 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
93 means that a class such as [\s] counted as "explicit reference to CR or
94 LF". That isn't really right - the whole point of the change was to try to
95 help when there was an actual mention of one of the two characters. So now
96 the change happens only if \r or \n (or a literal CR or LF) character is
97 encountered.
98
99 2. The 32-bit options word was also used for 6 internal flags, but the numbers
100 of both had grown to the point where there were only 3 bits left.
101 Fortunately, there was spare space in the data structure, and so I have
102 moved the internal flags into a new 16-bit field to free up more option
103 bits.
104
105 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
106 but did not set the internal JCHANGED flag - either of these is enough to
107 control the way the "get" function works - but the PCRE_INFO_JCHANGED
108 facility is supposed to tell if (?J) was ever used, so now (?J) at the
109 start sets both bits.
110
111 4. Added options (at build time, compile time, exec time) to change \R from
112 matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
113
114 5. doc/pcresyntax.html was missing from the distribution.
115
116 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
117 compatibility, even though it is no longer used.
118
119 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
120 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
121 windows.h file is present (where different names are used). [This was
122 reversed later after testing - see 16 below.]
123
124 8. Changed all #include <config.h> to #include "config.h". There were also
125 some further <pcre.h> cases that I changed to "pcre.h".
126
127 9. When pcregrep was used with the --colour option, it missed the line ending
128 sequence off the lines that it output.
129
130 10. It was pointed out to me that arrays of string pointers cause lots of
131 relocations when a shared library is dynamically loaded. A technique of
132 using a single long string with a table of offsets can drastically reduce
133 these. I have refactored PCRE in four places to do this. The result is
134 dramatic:
135
136 Originally: 290
137 After changing UCP table: 187
138 After changing error message table: 43
139 After changing table of "verbs" 36
140 After changing table of Posix names 22
141
142 Thanks to the folks working on Gregex for glib for this insight.
143
144 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
145 unicode-properties was also set.
146
147 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
148
149 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
150 checked only for CRLF.
151
152 14. Added casts to pcretest.c to avoid compiler warnings.
153
154 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
155
156 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
157 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
158 entirely. This removes changes made in 7 above.
159
160 17. The CMake files have been updated, and there is now more information about
161 building with CMake in the NON-UNIX-USE document.
162
163
164 Version 7.3 28-Aug-07
165 ---------------------
166
167 1. In the rejigging of the build system that eventually resulted in 7.1, the
168 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
169 brackets there is not right, since it causes compilers to look for an
170 installed pcre.h, not the version that is in the source that is being
171 compiled (which of course may be different). I have changed it back to:
172
173 #include "pcre.h"
174
175 I have a vague recollection that the change was concerned with compiling in
176 different directories, but in the new build system, that is taken care of
177 by the VPATH setting the Makefile.
178
179 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
180 when the subject happened to end in the byte 0x85 (e.g. if the last
181 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
182 characters but of course it shouldn't be taken as a newline when it is part
183 of another character. The bug was that, for an unlimited repeat of . in
184 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
185 characters when looking for a newline.
186
187 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
188
189 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
190 in debug output.
191
192 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
193 long printing in the pcrecpp unittest when running under MinGW.
194
195 6. ESC_K was left out of the EBCDIC table.
196
197 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
198 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
199 limit also applies to "virtual nesting" when a pattern is recursive, and in
200 this case 1000 isn't so big. I have been able to remove this limit at the
201 expense of backing off one optimization in certain circumstances. Normally,
202 when pcre_exec() would call its internal match() function recursively and
203 immediately return the result unconditionally, it uses a "tail recursion"
204 feature to save stack. However, when a subpattern that can match an empty
205 string has an unlimited repetition quantifier, it no longer makes this
206 optimization. That gives it a stack frame in which to save the data for
207 checking that an empty string has been matched. Previously this was taken
208 from the 1000-entry workspace that had been reserved. So now there is no
209 explicit limit, but more stack is used.
210
211 8. Applied Daniel's patches to solve problems with the import/export magic
212 syntax that is required for Windows, and which was going wrong for the
213 pcreposix and pcrecpp parts of the library. These were overlooked when this
214 problem was solved for the main library.
215
216 9. There were some crude static tests to avoid integer overflow when computing
217 the size of patterns that contain repeated groups with explicit upper
218 limits. As the maximum quantifier is 65535, the maximum group length was
219 set at 30,000 so that the product of these two numbers did not overflow a
220 32-bit integer. However, it turns out that people want to use groups that
221 are longer than 30,000 bytes (though not repeat them that many times).
222 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
223 made it possible to implement the integer overflow checks in a much more
224 dynamic way, which I have now done. The artificial limitation on group
225 length has been removed - we now have only the limit on the total length of
226 the compiled pattern, which depends on the LINK_SIZE setting.
227
228 10. Fixed a bug in the documentation for get/copy named substring when
229 duplicate names are permitted. If none of the named substrings are set, the
230 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
231 empty string.
232
233 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
234 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
235 because the ] is interpreted as the first data character and the
236 terminating ] is not found. PCRE has been made compatible with Perl in this
237 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
238 cause memory overwriting.
239
240 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
241 string has been matched (to stop an infinite loop). It was not recognizing
242 a conditional subpattern that could match an empty string if that
243 subpattern was within another subpattern. For example, it looped when
244 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
245 condition was not nested. This bug has been fixed.
246
247 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
248 past the start of the subject in the presence of bytes with the top bit
249 set, for example "\x8aBCD".
250
251 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
252 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
253
254 14. Optimized (?!) to (*FAIL).
255
256 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
257 This restricts code points to be within the range 0 to 0x10FFFF, excluding
258 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
259 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
260 does: it's just the validity check that is more restrictive.
261
262 16. Inserted checks for integer overflows during escape sequence (backslash)
263 processing, and also fixed erroneous offset values for syntax errors during
264 backslash processing.
265
266 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
267 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
268
269 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
270 caused an overrun.
271
272 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
273 something other than just ASCII characters) inside a group that had an
274 unlimited repeat caused a loop at compile time (while checking to see
275 whether the group could match an empty string).
276
277 20. Debugging a pattern containing \p or \P could cause a crash. For example,
278 [\P{Any}] did so. (Error in the code for printing property names.)
279
280 21. An orphan \E inside a character class could cause a crash.
281
282 22. A repeated capturing bracket such as (A)? could cause a wild memory
283 reference during compilation.
284
285 23. There are several functions in pcre_compile() that scan along a compiled
286 expression for various reasons (e.g. to see if it's fixed length for look
287 behind). There were bugs in these functions when a repeated \p or \P was
288 present in the pattern. These operators have additional parameters compared
289 with \d, etc, and these were not being taken into account when moving along
290 the compiled data. Specifically:
291
292 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
293 length.
294
295 (b) An item such as \pL+ within a repeated group could cause crashes or
296 loops.
297
298 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
299 "reference to non-existent subpattern" error.
300
301 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
302
303 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
304 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
305
306 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
307
308 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
309 character were causing crashes (broken optimization).
310
311 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
312 \p or \P) caused a compile-time loop.
313
314 28. More problems have arisen in unanchored patterns when CRLF is a valid line
315 break. For example, the unstudied pattern [\r\n]A does not match the string
316 "\r\nA" because change 7.0/46 below moves the current point on by two
317 characters after failing to match at the start. However, the pattern \nA
318 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
319 the same is true. There doesn't seem any very clean way out of this, but
320 what I have chosen to do makes the common cases work: PCRE now takes note
321 of whether there can be an explicit match for \r or \n anywhere in the
322 pattern, and if so, 7.0/46 no longer applies. As part of this change,
323 there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
324 pattern has explicit CR or LF references.
325
326 29. Added (*CR) etc for changing newline setting at start of pattern.
327
328
329 Version 7.2 19-Jun-07
330 ---------------------
331
332 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
333 which is apparently normally available under Windows.
334
335 2. Re-jig the pcregrep tests with different newline settings in an attempt
336 to make them independent of the local environment's newline setting.
337
338 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
339
340 4. Some of the "internals" tests were previously cut out when the link size
341 was not 2, because the output contained actual offsets. The recent new
342 "Z" feature of pcretest means that these can be cut out, making the tests
343 usable with all link sizes.
344
345 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
346 stack recursion. This gives a massive performance boost under BSD, but just
347 a small improvement under Linux. However, it saves one field in the frame
348 in all cases.
349
350 6. Added more features from the forthcoming Perl 5.10:
351
352 (a) (?-n) (where n is a string of digits) is a relative subroutine or
353 recursion call. It refers to the nth most recently opened parentheses.
354
355 (b) (?+n) is also a relative subroutine call; it refers to the nth next
356 to be opened parentheses.
357
358 (c) Conditions that refer to capturing parentheses can be specified
359 relatively, for example, (?(-2)... or (?(+3)...
360
361 (d) \K resets the start of the current match so that everything before
362 is not part of it.
363
364 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
365
366 (f) \g{name} is another synonym - part of Perl 5.10's unification of
367 reference syntax.
368
369 (g) (?| introduces a group in which the numbering of parentheses in each
370 alternative starts with the same number.
371
372 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
373
374 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
375 PCRE_INFO_JCHANGED.
376
377 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
378 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
379 for detecting groups that can match an empty string.
380
381 9. A pattern with a very large number of alternatives (more than several
382 hundred) was running out of internal workspace during the pre-compile
383 phase, where pcre_compile() figures out how much memory will be needed. A
384 bit of new cunning has reduced the workspace needed for groups with
385 alternatives. The 1000-alternative test pattern now uses 12 bytes of
386 workspace instead of running out of the 4096 that are available.
387
388 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
389
390 11. Applied patch from Google to remove an optimization that didn't quite work.
391 The report of the bug said:
392
393 pcrecpp::RE("a*").FullMatch("aaa") matches, while
394 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
395 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
396
397 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
398 it matched the wrong number of bytes.
399
400
401 Version 7.1 24-Apr-07
402 ---------------------
403
404 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
405 that is more "standard", making use of automake and other Autotools. There
406 is some re-arrangement of the files and adjustment of comments consequent
407 on this.
408
409 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
410 for recursive directory scanning broke on some systems because the files
411 are not scanned in any specific order and on different systems the order
412 was different. A call to "sort" has been inserted into RunGrepTest for the
413 approprate test as a short-term fix. In the longer term there may be an
414 alternative.
415
416 3. I had an email from Eric Raymond about problems translating some of PCRE's
417 man pages to HTML (despite the fact that I distribute HTML pages, some
418 people do their own conversions for various reasons). The problems
419 concerned the use of low-level troff macros .br and .in. I have therefore
420 removed all such uses from the man pages (some were redundant, some could
421 be replaced by .nf/.fi pairs). The 132html script that I use to generate
422 HTML has been updated to handle .nf/.fi and to complain if it encounters
423 .br or .in.
424
425 4. Updated comments in configure.ac that get placed in config.h.in and also
426 arranged for config.h to be included in the distribution, with the name
427 config.h.generic, for the benefit of those who have to compile without
428 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
429
430 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
431 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
432 makevp.bat for the current PCRE, using the additional files
433 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
434
435 6. A Windows user reported a minor discrepancy with test 2, which turned out
436 to be caused by a trailing space on an input line that had got lost in his
437 copy. The trailing space was an accident, so I've just removed it.
438
439 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
440 that is needed.
441
442 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
443 as "const" (a) because they are and (b) because it helps the PHP
444 maintainers who have recently made a script to detect big data structures
445 in the php code that should be moved to the .rodata section. I remembered
446 to update Builducptable as well, so it won't revert if ucptable.h is ever
447 re-created.
448
449 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
450 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
451 order to be able to cut out the UTF-8 tables in the latter when UTF-8
452 support is not required. This saves 1.5-2K of code, which is important in
453 some applications.
454
455 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
456 so as not to refer to the tables, even though these functions will never be
457 called when UTF-8 support is disabled. Otherwise there are problems with a
458 shared library.
459
460 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
461
462 (a) It was defining its arguments as char * instead of void *.
463
464 (b) It was assuming that all moves were upwards in memory; this was true
465 a long time ago when I wrote it, but is no longer the case.
466
467 The emulated memove() is provided for those environments that have neither
468 memmove() nor bcopy(). I didn't think anyone used it these days, but that
469 is clearly not the case, as these two bugs were recently reported.
470
471 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
472 and Detrail to create the HTML documentation, the .txt form of the man
473 pages, and it removes trailing spaces from listed files. It also creates
474 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
475 case, it wraps all the #defines with #ifndefs. This script should be run
476 before "make dist".
477
478 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
479 with Unicode property support.
480
481 (a) For a maximizing quantifier, if the two different cases of the
482 character were of different lengths in their UTF-8 codings (there are
483 some cases like this - I found 11), and the matching function had to
484 back up over a mixture of the two cases, it incorrectly assumed they
485 were both the same length.
486
487 (b) When PCRE was configured to use the heap rather than the stack for
488 recursion during matching, it was not correctly preserving the data for
489 the other case of a UTF-8 character when checking ahead for a match
490 while processing a minimizing repeat. If the check also involved
491 matching a wide character, but failed, corruption could cause an
492 erroneous result when trying to check for a repeat of the original
493 character.
494
495 13. Some tidying changes to the testing mechanism:
496
497 (a) The RunTest script now detects the internal link size and whether there
498 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
499 values substituted by "configure". (The RunGrepTest script already did
500 this for UTF-8.) The configure.ac script no longer substitutes the
501 relevant variables.
502
503 (b) The debugging options /B and /D in pcretest show the compiled bytecode
504 with length and offset values. This means that the output is different
505 for different internal link sizes. Test 2 is skipped for link sizes
506 other than 2 because of this, bypassing the problem. Unfortunately,
507 there was also a test in test 3 (the locale tests) that used /B and
508 failed for link sizes other than 2. Rather than cut the whole test out,
509 I have added a new /Z option to pcretest that replaces the length and
510 offset values with spaces. This is now used to make test 3 independent
511 of link size. (Test 2 will be tidied up later.)
512
513 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
514 segmentation fault instead of returning the appropriate error message.
515
516 15. In multiline mode when the newline sequence was set to "any", the pattern
517 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
518 This doesn't seem right; it now treats the CRLF combination as the line
519 ending, and so does not match in that case. It's only a pattern such as ^$
520 that would hit this one: something like ^ABC$ would have failed after \r
521 and then tried again after \r\n.
522
523 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
524 in an attempt to make files that differ only in their line terminators
525 compare equal. This works on Linux.
526
527 17. Under certain error circumstances pcregrep might try to free random memory
528 as it exited. This is now fixed, thanks to valgrind.
529
530 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
531 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
532 was because its rules for how to advance for /g after matching an empty
533 string at the end of a line did not allow for this case. They now check for
534 it specially.
535
536 20. pcretest is supposed to handle patterns and data of any length, by
537 extending its buffers when necessary. It was getting this wrong when the
538 buffer for a data line had to be extended.
539
540 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
541 CRLF as a newline sequence.
542
543 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
544 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
545 I have nevertheless tidied it up.
546
547 23. Added some casts to kill warnings from HP-UX ia64 compiler.
548
549 24. Added a man page for pcre-config.
550
551
552 Version 7.0 19-Dec-06
553 ---------------------
554
555 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
556 moving to gcc 4.1.1.
557
558 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
559 sys/time.h, which is documented as needed for this function. It doesn't
560 seem to matter on Linux, but it showed up on some releases of OS X.
561
562 3. It seems that there are systems where bytes whose values are greater than
563 127 match isprint() in the "C" locale. The "C" locale should be the
564 default when a C program starts up. In most systems, only ASCII printing
565 characters match isprint(). This difference caused the output from pcretest
566 to vary, making some of the tests fail. I have changed pcretest so that:
567
568 (a) When it is outputting text in the compiled version of a pattern, bytes
569 other than 32-126 are always shown as hex escapes.
570
571 (b) When it is outputting text that is a matched part of a subject string,
572 it does the same, unless a different locale has been set for the match
573 (using the /L modifier). In this case, it uses isprint() to decide.
574
575 4. Fixed a major bug that caused incorrect computation of the amount of memory
576 required for a compiled pattern when options that changed within the
577 pattern affected the logic of the preliminary scan that determines the
578 length. The relevant options are -x, and -i in UTF-8 mode. The result was
579 that the computed length was too small. The symptoms of this bug were
580 either the PCRE error "internal error: code overflow" from pcre_compile(),
581 or a glibc crash with a message such as "pcretest: free(): invalid next
582 size (fast)". Examples of patterns that provoked this bug (shown in
583 pcretest format) are:
584
585 /(?-x: )/x
586 /(?x)(?-x: \s*#\s*)/
587 /((?i)[\x{c0}])/8
588 /(?i:[\x{c0}])/8
589
590 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
591 is now done differently.
592
593 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
594 wrapper classes; (b) implement a new function in the C++ scanner that is
595 more efficient than the old way of doing things because it avoids levels of
596 recursion in the regex matching; (c) add a paragraph to the documentation
597 for the FullMatch() function.
598
599 6. The escape sequence \n was being treated as whatever was defined as
600 "newline". Not only was this contrary to the documentation, which states
601 that \n is character 10 (hex 0A), but it also went horribly wrong when
602 "newline" was defined as CRLF. This has been fixed.
603
604 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
605 was being set to -1 for the "end of line" case (supposedly a value that no
606 character can have). Though this value is never used (the check for end of
607 line is "zero bytes in current character"), it caused compiler complaints.
608 I've changed it to 0xffffffff.
609
610 8. In pcre_version.c, the version string was being built by a sequence of
611 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
612 string (as it is for production releases) called a macro with an empty
613 argument. The C standard says the result of this is undefined. The gcc
614 compiler treats it as an empty string (which was what was wanted) but it is
615 reported that Visual C gives an error. The source has been hacked around to
616 avoid this problem.
617
618 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
619 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
620 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
621 of them did).
622
623 10. Originally, pcretest opened its input and output without "b"; then I was
624 told that "b" was needed in some environments, so it was added for release
625 5.0 to both the input and output. (It makes no difference on Unix-like
626 systems.) Later I was told that it is wrong for the input on Windows. I've
627 now abstracted the modes into two macros, to make it easier to fiddle with
628 them, and removed "b" from the input mode under Windows.
629
630 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
631
632 12. Added -help and --help to pcretest as an official way of being reminded
633 of the options.
634
635 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
636 and pcrecpp.cc because they annoy compilers at high warning levels.
637
638 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
639
640 15. Fixed an occurrence of == in configure.ac that should have been = (shell
641 scripts are not C programs :-) and which was not noticed because it works
642 on Linux.
643
644 16. pcretest is supposed to handle any length of pattern and data line (as one
645 line or as a continued sequence of lines) by extending its input buffer if
646 necessary. This feature was broken for very long pattern lines, leading to
647 a string of junk being passed to pcre_compile() if the pattern was longer
648 than about 50K.
649
650 17. I have done a major re-factoring of the way pcre_compile() computes the
651 amount of memory needed for a compiled pattern. Previously, there was code
652 that made a preliminary scan of the pattern in order to do this. That was
653 OK when PCRE was new, but as the facilities have expanded, it has become
654 harder and harder to keep it in step with the real compile phase, and there
655 have been a number of bugs (see for example, 4 above). I have now found a
656 cunning way of running the real compile function in a "fake" mode that
657 enables it to compute how much memory it would need, while actually only
658 ever using a few hundred bytes of working memory and without too many
659 tests of the mode. This should make future maintenance and development
660 easier. A side effect of this work is that the limit of 200 on the nesting
661 depth of parentheses has been removed (though this was never a serious
662 limitation, I suspect). However, there is a downside: pcre_compile() now
663 runs more slowly than before (30% or more, depending on the pattern). I
664 hope this isn't a big issue. There is no effect on runtime performance.
665
666 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
667 newline (only possible for the last line of a file) and it was a
668 pattern that set a locale (followed by /Lsomething), pcretest crashed.
669
670 19. Added additional timing features to pcretest. (1) The -tm option now times
671 matching only, not compiling. (2) Both -t and -tm can be followed, as a
672 separate command line item, by a number that specifies the number of
673 repeats to use when timing. The default is 50000; this gives better
674 precision, but takes uncomfortably long for very large patterns.
675
676 20. Extended pcre_study() to be more clever in cases where a branch of a
677 subpattern has no definite first character. For example, (a*|b*)[cd] would
678 previously give no result from pcre_study(). Now it recognizes that the
679 first character must be a, b, c, or d.
680
681 21. There was an incorrect error "recursive call could loop indefinitely" if
682 a subpattern (or the entire pattern) that was being tested for matching an
683 empty string contained only one non-empty item after a nested subpattern.
684 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
685 incorrectly, because the \d was being skipped in the check.
686
687 22. The pcretest program now has a new pattern option /B and a command line
688 option -b, which is equivalent to adding /B to every pattern. This causes
689 it to show the compiled bytecode, without the additional information that
690 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
691 is the same as /B/I).
692
693 23. A new optimization is now able automatically to treat some sequences such
694 as a*b as a*+b. More specifically, if something simple (such as a character
695 or a simple class like \d) has an unlimited quantifier, and is followed by
696 something that cannot possibly match the quantified thing, the quantifier
697 is automatically "possessified".
698
699 24. A recursive reference to a subpattern whose number was greater than 39
700 went wrong under certain circumstances in UTF-8 mode. This bug could also
701 have affected the operation of pcre_study().
702
703 25. Realized that a little bit of performance could be had by replacing
704 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
705
706 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
707
708 27. Possessive quantifiers such as a++ were previously implemented by turning
709 them into atomic groups such as ($>a+). Now they have their own opcodes,
710 which improves performance. This includes the automatically created ones
711 from 23 above.
712
713 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
714 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
715 the first matched character to be a colon. This applied both to named and
716 numbered groups.
717
718 29. The ucpinternal.h header file was missing its idempotency #ifdef.
719
720 30. I was sent a "project" file called libpcre.a.dev which I understand makes
721 building PCRE on Windows easier, so I have included it in the distribution.
722
723 31. There is now a check in pcretest against a ridiculously large number being
724 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
725 loop, the loop is abandoned.
726
727 32. Forward references to subpatterns in conditions such as (?(2)...) where
728 subpattern 2 is defined later cause pcre_compile() to search forwards in
729 the pattern for the relevant set of parentheses. This search went wrong
730 when there were unescaped parentheses in a character class, parentheses
731 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
732
733 33. "Subroutine" calls and backreferences were previously restricted to
734 referencing subpatterns earlier in the regex. This restriction has now
735 been removed.
736
737 34. Added a number of extra features that are going to be in Perl 5.10. On the
738 whole, these are just syntactic alternatives for features that PCRE had
739 previously implemented using the Python syntax or my own invention. The
740 other formats are all retained for compatibility.
741
742 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
743 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
744 also .NET compatible.
745
746 (b) A recursion or subroutine call to a named group can now be defined as
747 (?&name) as well as (?P>name).
748
749 (c) A backreference to a named group can now be defined as \k<name> or
750 \k'name' as well as (?P=name). The new forms, as well as being in Perl
751 5.10, are also .NET compatible.
752
753 (d) A conditional reference to a named group can now use the syntax
754 (?(<name>) or (?('name') as well as (?(name).
755
756 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
757 groups (named and numbered) that are never evaluated inline, but can be
758 called as "subroutines" from elsewhere. In effect, the DEFINE condition
759 is always false. There may be only one alternative in such a group.
760
761 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
762 as the simple (?(R). The condition is true only if the most recent
763 recursion is that of the given number or name. It does not search out
764 through the entire recursion stack.
765
766 (g) The escape \gN or \g{N} has been added, where N is a positive or
767 negative number, specifying an absolute or relative reference.
768
769 35. Tidied to get rid of some further signed/unsigned compiler warnings and
770 some "unreachable code" warnings.
771
772 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
773 things, this adds five new scripts.
774
775 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
776 There were also incompatibilities regarding the handling of \Q..\E inside
777 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
778 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
779
780 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
781 matches an empty string, and forcibly breaks the loop. There were bugs in
782 this code in non-simple cases. For a pattern such as ^(a()*)* matched
783 against aaaa the result was just "a" rather than "aaaa", for example. Two
784 separate and independent bugs (that affected different cases) have been
785 fixed.
786
787 39. Refactored the code to abolish the use of different opcodes for small
788 capturing bracket numbers. This is a tidy that I avoided doing when I
789 removed the limit on the number of capturing brackets for 3.5 back in 2001.
790 The new approach is not only tidier, it makes it possible to reduce the
791 memory needed to fix the previous bug (38).
792
793 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
794 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
795 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
796 mode.
797
798 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
799 report.
800
801 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
802 copy construction and assignment in the C++ wrapper.
803
804 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
805 couple of bugs that could have given wrong results in the "--newline=crlf"
806 case.
807
808 44. Added a number of casts and did some reorganization of signed/unsigned int
809 variables following suggestions from Dair Grant. Also renamed the variable
810 "this" as "item" because it is a C++ keyword.
811
812 45. Arranged for dftables to add
813
814 #include "pcre_internal.h"
815
816 to pcre_chartables.c because without it, gcc 4.x may remove the array
817 definition from the final binary if PCRE is built into a static library and
818 dead code stripping is activated.
819
820 46. For an unanchored pattern, if a match attempt fails at the start of a
821 newline sequence, and the newline setting is CRLF or ANY, and the next two
822 characters are CRLF, advance by two characters instead of one.
823
824
825 Version 6.7 04-Jul-06
826 ---------------------
827
828 1. In order to handle tests when input lines are enormously long, pcretest has
829 been re-factored so that it automatically extends its buffers when
830 necessary. The code is crude, but this _is_ just a test program. The
831 default size has been increased from 32K to 50K.
832
833 2. The code in pcre_study() was using the value of the re argument before
834 testing it for NULL. (Of course, in any sensible call of the function, it
835 won't be NULL.)
836
837 3. The memmove() emulation function in pcre_internal.h, which is used on
838 systems that lack both memmove() and bcopy() - that is, hardly ever -
839 was missing a "static" storage class specifier.
840
841 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
842 containing an extended class (one that cannot be represented by a bitmap
843 because it contains high-valued characters or Unicode property items, e.g.
844 [\pZ]). Almost always one would set UTF-8 mode when processing such a
845 pattern, but PCRE should not loop if you do not (it no longer does).
846 [Detail: two cases were found: (a) a repeated subpattern containing an
847 extended class; (b) a recursive reference to a subpattern that followed a
848 previous extended class. It wasn't skipping over the extended class
849 correctly when UTF-8 mode was not set.]
850
851 5. A negated single-character class was not being recognized as fixed-length
852 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
853 compile error "lookbehind assertion is not fixed length".
854
855 6. The RunPerlTest auxiliary script was showing an unexpected difference
856 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
857 write a Perl script that can interpret lines of an input file either as
858 byte characters or as UTF-8, which is what "perltest" was being required to
859 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
860 can't do is switch easily at run time between having the "use utf8;" pragma
861 or not. In the end, I fudged it by using the RunPerlTest script to insert
862 "use utf8;" explicitly for the UTF-8 tests.
863
864 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
865 the end of the subject string, contrary to the documentation and to what
866 Perl does. This was true of both matching functions. Now it matches only at
867 the start of the subject and immediately after *internal* newlines.
868
869 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
870 a pointer to an int instead of a pointer to an unsigned long int. This
871 caused problems on 64-bit systems.
872
873 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
874 instance of the 'standard' template library not being so standard".
875
876 10. There was no check on the number of named subpatterns nor the maximum
877 length of a subpattern name. The product of these values is used to compute
878 the size of the memory block for a compiled pattern. By supplying a very
879 long subpattern name and a large number of named subpatterns, the size
880 computation could be caused to overflow. This is now prevented by limiting
881 the length of names to 32 characters, and the number of named subpatterns
882 to 10,000.
883
884 11. Subpatterns that are repeated with specific counts have to be replicated in
885 the compiled pattern. The size of memory for this was computed from the
886 length of the subpattern and the repeat count. The latter is limited to
887 65535, but there was no limit on the former, meaning that integer overflow
888 could in principle occur. The compiled length of a repeated subpattern is
889 now limited to 30,000 bytes in order to prevent this.
890
891 12. Added the optional facility to have named substrings with the same name.
892
893 13. Added the ability to use a named substring as a condition, using the
894 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
895 are numbers (not recommended). Forward references are permitted.
896
897 14. Added forward references in named backreferences (if you see what I mean).
898
899 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
900 pattern could run off the end of the subject. For example, the pattern
901 "(?s)(.{1,5})"8 did this with the subject "ab".
902
903 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
904 PCRE_CASELESS was set when matching characters that were quantified with ?
905 or *.
906
907 17. A character class other than a single negated character that had a minimum
908 but no maximum quantifier - for example [ab]{6,} - was not handled
909 correctly by pce_dfa_exec(). It would match only one character.
910
911 18. A valid (though odd) pattern that looked like a POSIX character
912 class but used an invalid character after [ (for example [[,abc,]]) caused
913 pcre_compile() to give the error "Failed: internal error: code overflow" or
914 in some cases to crash with a glibc free() error. This could even happen if
915 the pattern terminated after [[ but there just happened to be a sequence of
916 letters, a binary zero, and a closing ] in the memory that followed.
917
918 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
919 over the years. Originally (before any Unicode support), just the bottom 8
920 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
921 output from "man perlunicode" includes this:
922
923 The regular expression compiler produces polymorphic opcodes. That
924 is, the pattern adapts to the data and automatically switches to
925 the Unicode character scheme when presented with Unicode data--or
926 instead uses a traditional byte scheme when presented with byte
927 data.
928
929 Sadly, a wide octal escape does not cause a switch, and in a string with
930 no other multibyte characters, these octal escapes are treated as before.
931 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
932 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
933 Unicode string.
934
935 I have not perpetrated such confusion in PCRE. Up till now, it took just
936 the bottom 8 bits, as in old Perl. I have now made octal escapes with
937 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
938 translate to the appropriate multibyte character.
939
940 29. Applied some refactoring to reduce the number of warnings from Microsoft
941 and Borland compilers. This has included removing the fudge introduced
942 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
943 a warning about an unused variable.
944
945 21. PCRE has not included VT (character 0x0b) in the set of whitespace
946 characters since release 4.0, because Perl (from release 5.004) does not.
947 [Or at least, is documented not to: some releases seem to be in conflict
948 with the documentation.] However, when a pattern was studied with
949 pcre_study() and all its branches started with \s, PCRE still included VT
950 as a possible starting character. Of course, this did no harm; it just
951 caused an unnecessary match attempt.
952
953 22. Removed a now-redundant internal flag bit that recorded the fact that case
954 dependency changed within the pattern. This was once needed for "required
955 byte" processing, but is no longer used. This recovers a now-scarce options
956 bit. Also moved the least significant internal flag bit to the most-
957 significant bit of the word, which was not previously used (hangover from
958 the days when it was an int rather than a uint) to free up another bit for
959 the future.
960
961 23. Added support for CRLF line endings as well as CR and LF. As well as the
962 default being selectable at build time, it can now be changed at runtime
963 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
964 specify that it is scanning data with non-default line endings.
965
966 24. Changed the definition of CXXLINK to make it agree with the definition of
967 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
968
969 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
970 recursions. This makes a big different to stack usage for some patterns.
971
972 26. If a subpattern containing a named recursion or subroutine reference such
973 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
974 the space required for the compiled pattern went wrong and gave too small a
975 value. Depending on the environment, this could lead to "Failed: internal
976 error: code overflow at offset 49" or "glibc detected double free or
977 corruption" errors.
978
979 27. Applied patches from Google (a) to support the new newline modes and (b) to
980 advance over multibyte UTF-8 characters in GlobalReplace.
981
982 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
983 difference for some implementation of PCRE in some Windows version.
984
985 29. Added some extra testing facilities to pcretest:
986
987 \q<number> in a data line sets the "match limit" value
988 \Q<number> in a data line sets the "match recursion limt" value
989 -S <number> sets the stack size, where <number> is in megabytes
990
991 The -S option isn't available for Windows.
992
993
994 Version 6.6 06-Feb-06
995 ---------------------
996
997 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
998 in pcreposix.h. I have copied the definition from pcre.h.
999
1000 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1001 because pcre.h is no longer a built file.
1002
1003 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1004 not normally included in the compiled code.
1005
1006
1007 Version 6.5 01-Feb-06
1008 ---------------------
1009
1010 1. When using the partial match feature with pcre_dfa_exec(), it was not
1011 anchoring the second and subsequent partial matches at the new starting
1012 point. This could lead to incorrect results. For example, with the pattern
1013 /1234/, partially matching against "123" and then "a4" gave a match.
1014
1015 2. Changes to pcregrep:
1016
1017 (a) All non-match returns from pcre_exec() were being treated as failures
1018 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1019 error message is output. Some extra information is given for the
1020 PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1021 probably the only errors that are likely to be caused by users (by
1022 specifying a regex that has nested indefinite repeats, for instance).
1023 If there are more than 20 of these errors, pcregrep is abandoned.
1024
1025 (b) A binary zero was treated as data while matching, but terminated the
1026 output line if it was written out. This has been fixed: binary zeroes
1027 are now no different to any other data bytes.
1028
1029 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1030 used to set a locale for matching. The --locale=xxxx long option has
1031 been added (no short equivalent) to specify a locale explicitly on the
1032 pcregrep command, overriding the environment variables.
1033
1034 (d) When -B was used with -n, some line numbers in the output were one less
1035 than they should have been.
1036
1037 (e) Added the -o (--only-matching) option.
1038
1039 (f) If -A or -C was used with -c (count only), some lines of context were
1040 accidentally printed for the final match.
1041
1042 (g) Added the -H (--with-filename) option.
1043
1044 (h) The combination of options -rh failed to suppress file names for files
1045 that were found from directory arguments.
1046
1047 (i) Added the -D (--devices) and -d (--directories) options.
1048
1049 (j) Added the -F (--fixed-strings) option.
1050
1051 (k) Allow "-" to be used as a file name for -f as well as for a data file.
1052
1053 (l) Added the --colo(u)r option.
1054
1055 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1056 is not present by default.
1057
1058 3. A nasty bug was discovered in the handling of recursive patterns, that is,
1059 items such as (?R) or (?1), when the recursion could match a number of
1060 alternatives. If it matched one of the alternatives, but subsequently,
1061 outside the recursion, there was a failure, the code tried to back up into
1062 the recursion. However, because of the way PCRE is implemented, this is not
1063 possible, and the result was an incorrect result from the match.
1064
1065 In order to prevent this happening, the specification of recursion has
1066 been changed so that all such subpatterns are automatically treated as
1067 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1068
1069 4. I had overlooked the fact that, in some locales, there are characters for
1070 which isalpha() is true but neither isupper() nor islower() are true. In
1071 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1072 and ordfeminine) are like this. This affected the treatment of \w and \W
1073 when they appeared in character classes, but not when they appeared outside
1074 a character class. The bit map for "word" characters is now created
1075 separately from the results of isalnum() instead of just taking it from the
1076 upper, lower, and digit maps. (Plus the underscore character, of course.)
1077
1078 5. The above bug also affected the handling of POSIX character classes such as
1079 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1080 permanent tables. Instead, the bit maps for such a class were previously
1081 created as the appropriate unions of the upper, lower, and digit bitmaps.
1082 Now they are created by subtraction from the [[:word:]] class, which has
1083 its own bitmap.
1084
1085 6. The [[:blank:]] character class matches horizontal, but not vertical space.
1086 It is created by subtracting the vertical space characters (\x09, \x0a,
1087 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1088 subtraction was done in the overall bitmap for a character class, meaning
1089 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1090 be recognized. This bug has been fixed.
1091
1092 7. Patches from the folks at Google:
1093
1094 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1095 real life, but is still worth protecting against".
1096
1097 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1098 regular expressions".
1099
1100 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1101 have it.
1102
1103 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1104 "configure" and the latter not, in order to fix a problem somebody had
1105 with compiling the Arg class on HP-UX.
1106
1107 (e) Improve the error-handling of the C++ wrapper a little bit.
1108
1109 (f) New tests for checking recursion limiting.
1110
1111 8. The pcre_memmove() function, which is used only if the environment does not
1112 have a standard memmove() function (and is therefore rarely compiled),
1113 contained two bugs: (a) use of int instead of size_t, and (b) it was not
1114 returning a result (though PCRE never actually uses the result).
1115
1116 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1117 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1118 returned instead of calling malloc() with an overflowing number that would
1119 most likely cause subsequent chaos.
1120
1121 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1122
1123 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1124 with this option is matched, the nmatch and pmatch options of regexec() are
1125 ignored.
1126
1127 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1128 provided in case anyone wants to the the POSIX interface with UTF-8
1129 strings.
1130
1131 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1132 C++ linking (needed for some HP-UX environments).
1133
1134 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1135 (unused parameter) and in the pcre_printint() function (omitted "default"
1136 switch label when the default is to do nothing).
1137
1138 15. Added some code to make it possible, when PCRE is compiled as a C++
1139 library, to replace subject pointers for pcre_exec() with a smart pointer
1140 class, thus making it possible to process discontinuous strings.
1141
1142 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1143 much the same function. They were added by different people who were trying
1144 to make PCRE easy to compile on non-Unix systems. It has been suggested
1145 that PCRE_EXPORT be abolished now that there is more automatic apparatus
1146 for compiling on Windows systems. I have therefore replaced it with
1147 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1148 defaults to "extern" for C or "extern C" for C++, which works fine on
1149 Unix-like systems. It is now possible to override the value of PCRE_DATA_
1150 SCOPE with something explicit in config.h. In addition:
1151
1152 (a) pcreposix.h still had just "extern" instead of either of these macros;
1153 I have replaced it with PCRE_DATA_SCOPE.
1154
1155 (b) Functions such as _pcre_xclass(), which are internal to the library,
1156 but external in the C sense, all had PCRE_EXPORT in their definitions.
1157 This is apparently wrong for the Windows case, so I have removed it.
1158 (It makes no difference on Unix-like systems.)
1159
1160 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1161 of recursive calls to match(). This is different to MATCH_LIMIT because
1162 that limits the total number of calls to match(), not all of which increase
1163 the depth of recursion. Limiting the recursion depth limits the amount of
1164 stack (or heap if NO_RECURSE is set) that is used. The default can be set
1165 when PCRE is compiled, and changed at run time. A patch from Google adds
1166 this functionality to the C++ interface.
1167
1168 18. Changes to the handling of Unicode character properties:
1169
1170 (a) Updated the table to Unicode 4.1.0.
1171
1172 (b) Recognize characters that are not in the table as "Cn" (undefined).
1173
1174 (c) I revised the way the table is implemented to a much improved format
1175 which includes recognition of ranges. It now supports the ranges that
1176 are defined in UnicodeData.txt, and it also amalgamates other
1177 characters into ranges. This has reduced the number of entries in the
1178 table from around 16,000 to around 3,000, thus reducing its size
1179 considerably. I realized I did not need to use a tree structure after
1180 all - a binary chop search is just as efficient. Having reduced the
1181 number of entries, I extended their size from 6 bytes to 8 bytes to
1182 allow for more data.
1183
1184 (d) Added support for Unicode script names via properties such as \p{Han}.
1185
1186 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1187 matching that character.
1188
1189 20. When matching a repeated Unicode property with a minimum greater than zero,
1190 (for example \pL{2,}), PCRE could look past the end of the subject if it
1191 reached it while seeking the minimum number of characters. This could
1192 happen only if some of the characters were more than one byte long, because
1193 there is a check for at least the minimum number of bytes.
1194
1195 21. Refactored the implementation of \p and \P so as to be more general, to
1196 allow for more different types of property in future. This has changed the
1197 compiled form incompatibly. Anybody with saved compiled patterns that use
1198 \p or \P will have to recompile them.
1199
1200 22. Added "Any" and "L&" to the supported property types.
1201
1202 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1203 but give a compile time error if the value is greater than 0xff.
1204
1205 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1206 accidentally not being installed or uninstalled.
1207
1208 25. The pcre.h file was built from pcre.h.in, but the only changes that were
1209 made were to insert the current release number. This seemed silly, because
1210 it made things harder for people building PCRE on systems that don't run
1211 "configure". I have turned pcre.h into a distributed file, no longer built
1212 by "configure", with the version identification directly included. There is
1213 no longer a pcre.h.in file.
1214
1215 However, this change necessitated a change to the pcre-config script as
1216 well. It is built from pcre-config.in, and one of the substitutions was the
1217 release number. I have updated configure.ac so that ./configure now finds
1218 the release number by grepping pcre.h.
1219
1220 26. Added the ability to run the tests under valgrind.
1221
1222
1223 Version 6.4 05-Sep-05
1224 ---------------------
1225
1226 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1227 "--" to be printed when multiple files were scanned, even when none of the
1228 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1229 consider it to be a bug, and have restored the previous behaviour.
1230
1231 2. A couple of code tidies to get rid of compiler warnings.
1232
1233 3. The pcretest program used to cheat by referring to symbols in the library
1234 whose names begin with _pcre_. These are internal symbols that are not
1235 really supposed to be visible externally, and in some environments it is
1236 possible to suppress them. The cheating is now confined to including
1237 certain files from the library's source, which is a bit cleaner.
1238
1239 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1240 file's purpose clearer.
1241
1242 5. Reorganized pcre_ucp_findchar().
1243
1244
1245 Version 6.3 15-Aug-05
1246 ---------------------
1247
1248 1. The file libpcre.pc.in did not have general read permission in the tarball.
1249
1250 2. There were some problems when building without C++ support:
1251
1252 (a) If C++ support was not built, "make install" and "make test" still
1253 tried to test it.
1254
1255 (b) There were problems when the value of CXX was explicitly set. Some
1256 changes have been made to try to fix these, and ...
1257
1258 (c) --disable-cpp can now be used to explicitly disable C++ support.
1259
1260 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1261 backslash in a target when C++ was disabled. This confuses some
1262 versions of "make", apparently. Using an intermediate variable solves
1263 this. (Same for CPP_LOBJ.)
1264
1265 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1266 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1267 necessary on certain architectures.
1268
1269 4. Added a setting of -export-symbols-regex to the link command to remove
1270 those symbols that are exported in the C sense, but actually are local
1271 within the library, and not documented. Their names all begin with
1272 "_pcre_". This is not a perfect job, because (a) we have to except some
1273 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1274 available (and never for static libraries). I have made a note to try to
1275 find a way round (a) in the future.
1276
1277
1278 Version 6.2 01-Aug-05
1279 ---------------------
1280
1281 1. There was no test for integer overflow of quantifier values. A construction
1282 such as {1111111111111111} would give undefined results. What is worse, if
1283 a minimum quantifier for a parenthesized subpattern overflowed and became
1284 negative, the calculation of the memory size went wrong. This could have
1285 led to memory overwriting.
1286
1287 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1288
1289 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1290 operating environments where this matters.
1291
1292 4. Applied Giuseppe Maxia's patch to add additional features for controlling
1293 PCRE options from within the C++ wrapper.
1294
1295 5. Named capturing subpatterns were not being correctly counted when a pattern
1296 was compiled. This caused two problems: (a) If there were more than 100
1297 such subpatterns, the calculation of the memory needed for the whole
1298 compiled pattern went wrong, leading to an overflow error. (b) Numerical
1299 back references of the form \12, where the number was greater than 9, were
1300 not recognized as back references, even though there were sufficient
1301 previous subpatterns.
1302
1303 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1304 versions of gcc, e.g. 2.95.4.
1305
1306
1307 Version 6.1 21-Jun-05
1308 ---------------------
1309
1310 1. There was one reference to the variable "posix" in pcretest.c that was not
1311 surrounded by "#if !defined NOPOSIX".
1312
1313 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1314 the cross-check on the old pcre_info() function, for the benefit of the
1315 cut-down version of PCRE that is currently imported into Exim.
1316
1317 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1318 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1319 patterns that start (?i) but I don't think that matters. The use of (?i) is
1320 just an example; this all applies to the other options as well.
1321
1322 4. Since libtool seems to echo the compile commands it is issuing, the output
1323 from "make" can be reduced a bit by putting "@" in front of each libtool
1324 compile command.
1325
1326 5. Patch from the folks at Google for configure.in to be a bit more thorough
1327 in checking for a suitable C++ installation before trying to compile the
1328 C++ stuff. This should fix a reported problem when a compiler was present,
1329 but no suitable headers.
1330
1331 6. The man pages all had just "PCRE" as their title. I have changed them to
1332 be the relevant file name. I have also arranged that these names are
1333 retained in the file doc/pcre.txt, which is a concatenation in text format
1334 of all the man pages except the little individual ones for each function.
1335
1336 7. The NON-UNIX-USE file had not been updated for the different set of source
1337 files that come with release 6. I also added a few comments about the C++
1338 wrapper.
1339
1340
1341 Version 6.0 07-Jun-05
1342 ---------------------
1343
1344 1. Some minor internal re-organization to help with my DFA experiments.
1345
1346 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1347 didn't matter for the library itself when fully configured, but did matter
1348 when compiling without UCP support, or within Exim, where the ucp files are
1349 not imported.
1350
1351 3. Refactoring of the library code to split up the various functions into
1352 different source modules. The addition of the new DFA matching code (see
1353 below) to a single monolithic source would have made it really too
1354 unwieldy, quite apart from causing all the code to be include in a
1355 statically linked application, when only some functions are used. This is
1356 relevant even without the DFA addition now that patterns can be compiled in
1357 one application and matched in another.
1358
1359 The downside of splitting up is that there have to be some external
1360 functions and data tables that are used internally in different modules of
1361 the library but which are not part of the API. These have all had their
1362 names changed to start with "_pcre_" so that they are unlikely to clash
1363 with other external names.
1364
1365 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1366 a different (DFA) algorithm. Although it is slower than the original
1367 function, it does have some advantages for certain types of matching
1368 problem.
1369
1370 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1371 including restarting after a partial match.
1372
1373 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1374 defined when compiling for Windows was sent to me. I have put it into the
1375 code, though I have no means of testing or verifying it.
1376
1377 7. Added the pcre_refcount() auxiliary function.
1378
1379 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1380 match before or at the first newline in the subject string. In pcretest,
1381 the /f option on a pattern can be used to set this.
1382
1383 9. A repeated \w when used in UTF-8 mode with characters greater than 256
1384 would behave wrongly. This has been present in PCRE since release 4.0.
1385
1386 10. A number of changes to the pcregrep command:
1387
1388 (a) Refactored how -x works; insert ^(...)$ instead of setting
1389 PCRE_ANCHORED and checking the length, in preparation for adding
1390 something similar for -w.
1391
1392 (b) Added the -w (match as a word) option.
1393
1394 (c) Refactored the way lines are read and buffered so as to have more
1395 than one at a time available.
1396
1397 (d) Implemented a pcregrep test script.
1398
1399 (e) Added the -M (multiline match) option. This allows patterns to match
1400 over several lines of the subject. The buffering ensures that at least
1401 8K, or the rest of the document (whichever is the shorter) is available
1402 for matching (and similarly the previous 8K for lookbehind assertions).
1403
1404 (f) Changed the --help output so that it now says
1405
1406 -w, --word-regex(p)
1407
1408 instead of two lines, one with "regex" and the other with "regexp"
1409 because that confused at least one person since the short forms are the
1410 same. (This required a bit of code, as the output is generated
1411 automatically from a table. It wasn't just a text change.)
1412
1413 (g) -- can be used to terminate pcregrep options if the next thing isn't an
1414 option but starts with a hyphen. Could be a pattern or a path name
1415 starting with a hyphen, for instance.
1416
1417 (h) "-" can be given as a file name to represent stdin.
1418
1419 (i) When file names are being printed, "(standard input)" is used for
1420 the standard input, for compatibility with GNU grep. Previously
1421 "<stdin>" was used.
1422
1423 (j) The option --label=xxx can be used to supply a name to be used for
1424 stdin when file names are being printed. There is no short form.
1425
1426 (k) Re-factored the options decoding logic because we are going to add
1427 two more options that take data. Such options can now be given in four
1428 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1429
1430 (l) Added the -A, -B, and -C options for requesting that lines of context
1431 around matches be printed.
1432
1433 (m) Added the -L option to print the names of files that do not contain
1434 any matching lines, that is, the complement of -l.
1435
1436 (n) The return code is 2 if any file cannot be opened, but pcregrep does
1437 continue to scan other files.
1438
1439 (o) The -s option was incorrectly implemented. For compatibility with other
1440 greps, it now suppresses the error message for a non-existent or non-
1441 accessible file (but not the return code). There is a new option called
1442 -q that suppresses the output of matching lines, which was what -s was
1443 previously doing.
1444
1445 (p) Added --include and --exclude options to specify files for inclusion
1446 and exclusion when recursing.
1447
1448 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1449 Hopefully, it now does.
1450
1451 12. Missing cast in pcre_study().
1452
1453 13. Added an "uninstall" target to the makefile.
1454
1455 14. Replaced "extern" in the function prototypes in Makefile.in with
1456 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1457 world, but is set differently for Windows.
1458
1459 15. Added a second compiling function called pcre_compile2(). The only
1460 difference is that it has an extra argument, which is a pointer to an
1461 integer error code. When there is a compile-time failure, this is set
1462 non-zero, in addition to the error test pointer being set to point to an
1463 error message. The new argument may be NULL if no error number is required
1464 (but then you may as well call pcre_compile(), which is now just a
1465 wrapper). This facility is provided because some applications need a
1466 numeric error indication, but it has also enabled me to tidy up the way
1467 compile-time errors are handled in the POSIX wrapper.
1468
1469 16. Added VPATH=.libs to the makefile; this should help when building with one
1470 prefix path and installing with another. (Or so I'm told by someone who
1471 knows more about this stuff than I do.)
1472
1473 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1474 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1475 match everything, including newlines. This is not POSIX-compatible, but
1476 somebody wanted the feature. From pcretest it can be activated by using
1477 both the P and the s flags.
1478
1479 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1480
1481 19. libpcre.pc was being incorrectly installed as executable.
1482
1483 20. A couple of places in pcretest check for end-of-line by looking for '\n';
1484 it now also looks for '\r' so that it will work unmodified on Windows.
1485
1486 21. Added Google's contributed C++ wrapper to the distribution.
1487
1488 22. Added some untidy missing memory free() calls in pcretest, to keep
1489 Electric Fence happy when testing.
1490
1491
1492
1493 Version 5.0 13-Sep-04
1494 ---------------------
1495
1496 1. Internal change: literal characters are no longer packed up into items
1497 containing multiple characters in a single byte-string. Each character
1498 is now matched using a separate opcode. However, there may be more than one
1499 byte in the character in UTF-8 mode.
1500
1501 2. The pcre_callout_block structure has two new fields: pattern_position and
1502 next_item_length. These contain the offset in the pattern to the next match
1503 item, and its length, respectively.
1504
1505 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1506 insertion of callouts before each pattern item. Added the /C option to
1507 pcretest to make use of this.
1508
1509 4. On the advice of a Windows user, the lines
1510
1511 #if defined(_WIN32) || defined(WIN32)
1512 _setmode( _fileno( stdout ), 0x8000 );
1513 #endif /* defined(_WIN32) || defined(WIN32) */
1514
1515 have been added to the source of pcretest. This apparently does useful
1516 magic in relation to line terminators.
1517
1518 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1519 for the benefit of those environments where the "b" makes a difference.
1520
1521 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1522 to know about it. I have put a hack into configure.in that adds in code
1523 to set GCC=yes if CC=icc. This seems to end up at a point in the
1524 generated configure script that is early enough to affect the setting of
1525 compiler options, which is what is needed, but I have no means of testing
1526 whether it really works. (The user who reported this had patched the
1527 generated configure script, which of course I cannot do.)
1528
1529 LATER: After change 22 below (new libtool files), the configure script
1530 seems to know about icc (and also ecc). Therefore, I have commented out
1531 this hack in configure.in.
1532
1533 7. Added support for pkg-config (2 patches were sent in).
1534
1535 8. Negated POSIX character classes that used a combination of internal tables
1536 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1537 [[:^ascii]]. Typically, they would match almost any characters. The other
1538 POSIX classes were not broken in this way.
1539
1540 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1541 to find the match, as PCRE was deluded into thinking that the match had to
1542 start at the start point or following a newline. The same bug applied to
1543 patterns with negative forward assertions or any backward assertions
1544 preceding ".*" at the start, unless the pattern required a fixed first
1545 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1546
1547 10. In UTF-8 mode, when moving forwards in the subject after a failed match
1548 starting at the last subject character, bytes beyond the end of the subject
1549 string were read.
1550
1551 11. Renamed the variable "class" as "classbits" to make life easier for C++
1552 users. (Previously there was a macro definition, but it apparently wasn't
1553 enough.)
1554
1555 12. Added the new field "tables" to the extra data so that tables can be passed
1556 in at exec time, or the internal tables can be re-selected. This allows
1557 a compiled regex to be saved and re-used at a later time by a different
1558 program that might have everything at different addresses.
1559
1560 13. Modified the pcre-config script so that, when run on Solaris, it shows a
1561 -R library as well as a -L library.
1562
1563 14. The debugging options of pcretest (-d on the command line or D on a
1564 pattern) showed incorrect output for anything following an extended class
1565 that contained multibyte characters and which was followed by a quantifier.
1566
1567 15. Added optional support for general category Unicode character properties
1568 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1569 support. It adds about 90K to the size of the library. The meanings of the
1570 inbuilt class escapes such as \d and \s have NOT been changed.
1571
1572 16. Updated pcredemo.c to include calls to free() to release the memory for the
1573 compiled pattern.
1574
1575 17. The generated file chartables.c was being created in the source directory
1576 instead of in the building directory. This caused the build to fail if the
1577 source directory was different from the building directory, and was
1578 read-only.
1579
1580 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1581 file. No doubt somebody will tell me if they don't make sense... Also added
1582 Dan Mooney's comments about building on OpenVMS.
1583
1584 19. Added support for partial matching via the PCRE_PARTIAL option for
1585 pcre_exec() and the \P data escape in pcretest.
1586
1587 20. Extended pcretest with 3 new pattern features:
1588
1589 (i) A pattern option of the form ">rest-of-line" causes pcretest to
1590 write the compiled pattern to the file whose name is "rest-of-line".
1591 This is a straight binary dump of the data, with the saved pointer to
1592 the character tables forced to be NULL. The study data, if any, is
1593 written too. After writing, pcretest reads a new pattern.
1594
1595 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1596 compiled pattern from the given file. There must not be any
1597 occurrences of "<" in the file name (pretty unlikely); if there are,
1598 pcretest will instead treat the initial "<" as a pattern delimiter.
1599 After reading in the pattern, pcretest goes on to read data lines as
1600 usual.
1601
1602 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1603 and 16-bit fields in a compiled pattern, to simulate a pattern that
1604 was compiled on a host of opposite endianness.
1605
1606 21. The pcre-exec() function can now cope with patterns that were compiled on
1607 hosts of opposite endianness, with this restriction:
1608
1609 As for any compiled expression that is saved and used later, the tables
1610 pointer field cannot be preserved; the extra_data field in the arguments
1611 to pcre_exec() should be used to pass in a tables address if a value
1612 other than the default internal tables were used at compile time.
1613
1614 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1615 now diagnosed as an error. Previously, most of the time, a negative number
1616 would have been treated as zero, but if in addition "ovector" was passed as
1617 NULL, a crash could occur.
1618
1619 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1620 new versions from the libtool 1.5 distribution (the last one is a copy of
1621 a file called libtool.m4). This seems to have fixed the need to patch
1622 "configure" to support Darwin 1.3 (which I used to do). However, I still
1623 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1624 workstation).
1625
1626 24. Changed the PCRE licence to be the more standard "BSD" licence.
1627
1628
1629 Version 4.5 01-Dec-03
1630 ---------------------
1631
1632 1. There has been some re-arrangement of the code for the match() function so
1633 that it can be compiled in a version that does not call itself recursively.
1634 Instead, it keeps those local variables that need separate instances for
1635 each "recursion" in a frame on the heap, and gets/frees frames whenever it
1636 needs to "recurse". Keeping track of where control must go is done by means
1637 of setjmp/longjmp. The whole thing is implemented by a set of macros that
1638 hide most of the details from the main code, and operates only if
1639 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1640 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1641 operating.
1642
1643 To make it easier for callers to provide specially tailored get/free
1644 functions for this usage, two new functions, pcre_stack_malloc, and
1645 pcre_stack_free, are used. They are always called in strict stacking order,
1646 and the size of block requested is always the same.
1647
1648 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1649 PCRE has been compiled to use the stack or the heap for recursion. The
1650 -C option of pcretest uses this to show which version is compiled.
1651
1652 A new data escape \S, is added to pcretest; it causes the amounts of store
1653 obtained and freed by both kinds of malloc/free at match time to be added
1654 to the output.
1655
1656 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1657 what's available on my current Linux desktop machine.
1658
1659 3. When matching a UTF-8 string, the test for a valid string at the start has
1660 been extended. If start_offset is not zero, PCRE now checks that it points
1661 to a byte that is the start of a UTF-8 character. If not, it returns
1662 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1663 this is necessary because there may be backward assertions in the pattern.
1664 When matching the same subject several times, it may save resources to use
1665 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1666
1667 4. The code for checking the validity of UTF-8 strings has been tightened so
1668 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1669 containing "overlong sequences".
1670
1671 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1672 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1673 should have been "&", but it just so happened that all the cases this let
1674 through by mistake were picked up later in the function.
1675
1676 6. I had used a variable called "isblank" - this is a C99 function, causing
1677 some compilers to warn. To avoid this, I renamed it (as "blankclass").
1678
1679 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1680 prompting; (b) run "./pcretest /dev/null" at the start of the test script
1681 so the version is shown; (c) stop "make test" echoing "./RunTest".
1682
1683 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1684
1685 9. The prototype for memmove() for systems that don't have it was using
1686 size_t, but the inclusion of the header that defines size_t was later. I've
1687 moved the #includes for the C headers earlier to avoid this.
1688
1689 10. Added some adjustments to the code to make it easier to compiler on certain
1690 special systems:
1691
1692 (a) Some "const" qualifiers were missing.
1693 (b) Added the macro EXPORT before all exported functions; by default this
1694 is defined to be empty.
1695 (c) Changed the dftables auxiliary program (that builds chartables.c) so
1696 that it reads its output file name as an argument instead of writing
1697 to the standard output and assuming this can be redirected.
1698
1699 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1700 class containing characters with values greater than 255, PCRE compilation
1701 went into a loop.
1702
1703 12. A recursive reference to a subpattern that was within another subpattern
1704 that had a minimum quantifier of zero caused PCRE to crash. For example,
1705 (x(y(?2))z)? provoked this bug with a subject that got as far as the
1706 recursion. If the recursively-called subpattern itself had a zero repeat,
1707 that was OK.
1708
1709 13. In pcretest, the buffer for reading a data line was set at 30K, but the
1710 buffer into which it was copied (for escape processing) was still set at
1711 1024, so long lines caused crashes.
1712
1713 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1714 "internal error: code overflow...". This applied to any character class
1715 that was followed by a possessive quantifier.
1716
1717 15. Modified the Makefile to add libpcre.la as a prerequisite for
1718 libpcreposix.la because I was told this is needed for a parallel build to
1719 work.
1720
1721 16. If a pattern that contained .* following optional items at the start was
1722 studied, the wrong optimizing data was generated, leading to matching
1723 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1724 matching string must start with a or b or c. The correct conclusion for
1725 this pattern is that a match can start with any character.
1726
1727
1728 Version 4.4 13-Aug-03
1729 ---------------------
1730
1731 1. In UTF-8 mode, a character class containing characters with values between
1732 127 and 255 was not handled correctly if the compiled pattern was studied.
1733 In fixing this, I have also improved the studying algorithm for such
1734 classes (slightly).
1735
1736 2. Three internal functions had redundant arguments passed to them. Removal
1737 might give a very teeny performance improvement.
1738
1739 3. Documentation bug: the value of the capture_top field in a callout is *one
1740 more than* the number of the hightest numbered captured substring.
1741
1742 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1743 in incorrectly linking with a previously installed version. They now link
1744 explicitly with libpcre.la.
1745
1746 5. configure.in no longer needs to recognize Cygwin specially.
1747
1748 6. A problem in pcre.in for Windows platforms is fixed.
1749
1750 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1751 pcretest, it used to include the size of the study block as part of its
1752 output. Unfortunately, the structure contains a field that has a different
1753 size on different hardware architectures. This meant that the tests that
1754 showed this size failed. As the block is currently always of a fixed size,
1755 this information isn't actually particularly useful in pcretest output, so
1756 I have just removed it.
1757
1758 8. Three pre-processor statements accidentally did not start in column 1.
1759 Sadly, there are *still* compilers around that complain, even though
1760 standard C has not required this for well over a decade. Sigh.
1761
1762 9. In pcretest, the code for checking callouts passed small integers in the
1763 callout_data field, which is a void * field. However, some picky compilers
1764 complained about the casts involved for this on 64-bit systems. Now
1765 pcretest passes the address of the small integer instead, which should get
1766 rid of the warnings.
1767
1768 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1769 both compile and run time, and gives an error if an invalid UTF-8 sequence
1770 is found. There is a option for disabling this check in cases where the
1771 string is known to be correct and/or the maximum performance is wanted.
1772
1773 11. In response to a bug report, I changed one line in Makefile.in from
1774
1775 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1776 to
1777 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1778
1779 to look similar to other lines, but I have no way of telling whether this
1780 is the right thing to do, as I do not use Windows. No doubt I'll get told
1781 if it's wrong...
1782
1783
1784 Version 4.3 21-May-03
1785 ---------------------
1786
1787 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1788 Makefile.
1789
1790 2. Some refactoring to improve the quality of the code:
1791
1792 (i) The utf8_table... variables are now declared "const".
1793
1794 (ii) The code for \cx, which used the "case flipping" table to upper case
1795 lower case letters, now just substracts 32. This is ASCII-specific,
1796 but the whole concept of \cx is ASCII-specific, so it seems
1797 reasonable.
1798
1799 (iii) PCRE was using its character types table to recognize decimal and
1800 hexadecimal digits in the pattern. This is silly, because it handles
1801 only 0-9, a-f, and A-F, but the character types table is locale-
1802 specific, which means strange things might happen. A private
1803 table is now used for this - though it costs 256 bytes, a table is
1804 much faster than multiple explicit tests. Of course, the standard
1805 character types table is still used for matching digits in subject
1806 strings against \d.
1807
1808 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1809 ending in _t are). So I've renamed it as ESC_tee.
1810
1811 3. The first argument for regexec() in the POSIX wrapper should have been
1812 defined as "const".
1813
1814 4. Changed pcretest to use malloc() for its buffers so that they can be
1815 Electric Fenced for debugging.
1816
1817 5. There were several places in the code where, in UTF-8 mode, PCRE would try
1818 to read one or more bytes before the start of the subject string. Often this
1819 had no effect on PCRE's behaviour, but in some circumstances it could
1820 provoke a segmentation fault.
1821
1822 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1823 to try to read one or more bytes before the start of the subject string.
1824
1825 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1826 UTF-8 support could misbehave in various ways if the subject string
1827 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1828 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1829 back over UTF-8 characters.)
1830
1831
1832 Version 4.2 14-Apr-03
1833 ---------------------
1834
1835 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1836
1837 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1838 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1839 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1840 [WIN_PREFIX]: new variable, "cyg" for Cygwin
1841 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1842 and BUILD_EXEEXT
1843 Note: automatic setting of the BUILD variables is not yet working
1844 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1845 compile-time but not at link-time
1846 [LINK]: use for linking executables only
1847 make different versions for Windows and non-Windows
1848 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1849 libraries
1850 [LINK_FOR_BUILD]: new variable
1851 [OBJEXT]: use throughout
1852 [EXEEXT]: use throughout
1853 <winshared>: new target
1854 <wininstall>: new target
1855 <dftables.o>: use native compiler
1856 <dftables>: use native linker
1857 <install>: handle Windows platform correctly
1858 <clean>: ditto
1859 <check>: ditto
1860 copy DLL to top builddir before testing
1861
1862 As part of these changes, -no-undefined was removed again. This was reported
1863 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1864 in any case.
1865
1866 3. Some tidies to get rid of compiler warnings:
1867
1868 . In the match_data structure, match_limit was an unsigned long int, whereas
1869 match_call_count was an int. I've made them both unsigned long ints.
1870
1871 . In pcretest the fact that a const uschar * doesn't automatically cast to
1872 a void * provoked a warning.
1873
1874 . Turning on some more compiler warnings threw up some "shadow" variables
1875 and a few more missing casts.
1876
1877 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1878 option, a class that contained a single character with a value between 128
1879 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1880
1881 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1882 option, a class that contained several characters, but with at least one
1883 whose value was between 128 and 255 caused PCRE to crash.
1884
1885
1886 Version 4.1 12-Mar-03
1887 ---------------------
1888
1889 1. Compiling with gcc -pedantic found a couple of places where casts were
1890 needed, and a string in dftables.c that was longer than standard compilers are
1891 required to support.
1892
1893 2. Compiling with Sun's compiler found a few more places where the code could
1894 be tidied up in order to avoid warnings.
1895
1896 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1897 first of these names is deprecated in the latest Autoconf in favour of the name
1898 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1899 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1900 analogy I have changed it to CFLAGS_FOR_BUILD.
1901
1902 4. Added -no-undefined to the linking command in the Makefile, because this is
1903 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1904 linking step for the pcreposix library.
1905
1906 5. PCRE was failing to diagnose the case of two named groups with the same
1907 name.
1908
1909 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1910 literal character that is needed in the subject for a match, and scans along to
1911 ensure that it is present before embarking on the full matching process. This
1912 saves time in cases of nested unlimited repeats that are never going to match.
1913 Problem: the scan can take a lot of time if the subject is very long (e.g.
1914 megabytes), thus penalizing straightforward matches. It is now done only if the
1915 amount of subject to be scanned is less than 1000 bytes.
1916
1917 7. A lesser problem with the same optimization is that it was recording the
1918 first character of an anchored pattern as "needed", thus provoking a search
1919 right along the subject, even when the first match of the pattern was going to
1920 fail. The "needed" character is now not set for anchored patterns, unless it
1921 follows something in the pattern that is of non-fixed length. Thus, it still
1922 fulfils its original purpose of finding quick non-matches in cases of nested
1923 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1924
1925
1926 Version 4.0 17-Feb-03
1927 ---------------------
1928
1929 1. If a comment in an extended regex that started immediately after a meta-item
1930 extended to the end of string, PCRE compiled incorrect data. This could lead to
1931 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1932
1933 2. Moved to autoconf 2.53 and libtool 1.4.2.
1934
1935 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1936 the special perltest8 script is no longer needed - all the tests can be run
1937 from a single perltest script.
1938
1939 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1940 by \s. It has now been removed in PCRE. This means it isn't recognized as
1941 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1942 class [:space:] *does* include VT, thereby creating a mess.
1943
1944 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1945 space and tab.
1946
1947 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1948 its new features into the main test script, reducing the number of scripts.
1949
1950 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1951 were backward compatible, and made the (?i) apply to the whole pattern, as if
1952 /i were given. Now it behaves more logically, and applies the option setting
1953 only to what follows. PCRE has been changed to follow suit. However, if it
1954 finds options settings right at the start of the pattern, it extracts them into
1955 the global options, as before. Thus, they show up in the info data.
1956
1957 8. Added support for the \Q...\E escape sequence. Characters in between are
1958 treated as literals. This is slightly different from Perl in that $ and @ are
1959 also handled as literals inside the quotes. In Perl, they will cause variable
1960 interpolation. Note the following examples:
1961
1962 Pattern PCRE matches Perl matches
1963
1964 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
1965 \Qabc\$xyz\E abc\$xyz abc\$xyz
1966 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
1967
1968 For compatibility with Perl, \Q...\E sequences are recognized inside character
1969 classes as well as outside them.
1970
1971 9. Re-organized 3 code statements in pcretest to avoid "overflow in
1972 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1973 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1974 signed/unsigned warnings.
1975
1976 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1977 option for pcretest, so I've replaced it by a simple function that does just
1978 that job.
1979
1980 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1981 "pcregrep -".
1982
1983 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1984 Java package. This provides some syntactic sugar for simple cases of what my
1985 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1986 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1987 item, you can use this simplified notation. Note that only makes sense with
1988 greedy quantifiers. Consequently, the use of the possessive quantifier forces
1989 greediness, whatever the setting of the PCRE_UNGREEDY option.
1990
1991 13. A change of greediness default within a pattern was not taking effect at
1992 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1993 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1994 was abstracted outside.
1995
1996 14. PCRE now supports the \G assertion. It is true when the current matching
1997 position is at the start point of the match. This differs from \A when the
1998 starting offset is non-zero. Used with the /g option of pcretest (or similar
1999 code), it works in the same way as it does for Perl's /g option. If all
2000 alternatives of a regex begin with \G, the expression is anchored to the start
2001 match position, and the "anchored" flag is set in the compiled expression.
2002
2003 15. Some bugs concerning the handling of certain option changes within patterns
2004 have been fixed. These applied to options other than (?ims). For example,
2005 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2006 been the other way round. Some of this was related to change 7 above.
2007
2008 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2009 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2010 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2011 POSIX classes only within a class (e.g. /[[:alpha:]]/).
2012
2013 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2014 mode. Unlike ".", it always matches newline, whatever the setting of
2015 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2016 assertions. Perl allows it, but it doesn't (in general) work because it can't
2017 calculate the length of the lookbehind. At least, that's the case for Perl
2018 5.8.0 - I've been told they are going to document that it doesn't work in
2019 future.
2020
2021 18. Added an error diagnosis for escapes that PCRE does not support: these are
2022 \L, \l, \N, \P, \p, \U, \u, and \X.
2023
2024 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2025 reading past the end of the pattern in cases such as /[abcd/.
2026
2027 20. PCRE was getting more memory than necessary for patterns with classes that
2028 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2029
2030 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2031 compiling PCRE for use with Virtual Pascal.
2032
2033 22. Small fix to the Makefile to make it work properly if the build is done
2034 outside the source tree.
2035
2036 23. Added a new extension: a condition to go with recursion. If a conditional
2037 subpattern starts with (?(R) the "true" branch is used if recursion has
2038 happened, whereas the "false" branch is used only at the top level.
2039
2040 24. When there was a very long string of literal characters (over 255 bytes
2041 without UTF support, over 250 bytes with UTF support), the computation of how
2042 much memory was required could be incorrect, leading to segfaults or other
2043 strange effects.
2044
2045 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2046 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2047 there was a subsequent back reference to those brackets. This meant that, for
2048 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2049 possible to check for precisely this case. All we can do is abandon the
2050 optimization if .* occurs inside capturing brackets when there are any back
2051 references whatsoever. (See below for a better fix that came later.)
2052
2053 26. The handling of the optimization for finding the first character of a
2054 non-anchored pattern, and for finding a character that is required later in the
2055 match were failing in some cases. This didn't break the matching; it just
2056 failed to optimize when it could. The way this is done has been re-implemented.
2057
2058 27. Fixed typo in error message for invalid (?R item (it said "(?p").
2059
2060 28. Added a new feature that provides some of the functionality that Perl
2061 provides with (?{...}). The facility is termed a "callout". The way it is done
2062 in PCRE is for the caller to provide an optional function, by setting
2063 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2064 global variable. By default it is unset, which disables all calling out. To get
2065 the function called, the regex must include (?C) at appropriate points. This
2066 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2067 This provides a means of identifying different callout points. When PCRE
2068 reaches such a point in the regex, if pcre_callout has been set, the external
2069 function is called. It is provided with data in a structure called
2070 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2071 matching continues; if it returns a non-zero value, the match at the current
2072 point fails. However, backtracking will occur if possible. [This was changed
2073 later and other features added - see item 49 below.]
2074
2075 29. pcretest is upgraded to test the callout functionality. It provides a
2076 callout function that displays information. By default, it shows the start of
2077 the match and the current position in the text. There are some new data escapes
2078 to vary what happens:
2079
2080 \C+ in addition, show current contents of captured substrings
2081 \C- do not supply a callout function
2082 \C!n return 1 when callout number n is reached
2083 \C!n!m return 1 when callout number n is reached for the mth time
2084
2085 30. If pcregrep was called with the -l option and just a single file name, it
2086 output "<stdin>" if a match was found, instead of the file name.
2087
2088 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2089 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2090 pcre_exec(). This saves a malloc/free per call. The default value of
2091 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2092 when configuring.
2093
2094 32. The default maximum size of a compiled pattern is 64K. There have been a
2095 few cases of people hitting this limit. The code now uses macros to handle the
2096 storing of links as offsets within the compiled pattern. It defaults to 2-byte
2097 links, but this can be changed to 3 or 4 bytes by --with-link-size when
2098 configuring. Tests 2 and 5 work only with 2-byte links because they output
2099 debugging information about compiled patterns.
2100
2101 33. Internal code re-arrangements:
2102
2103 (a) Moved the debugging function for printing out a compiled regex into
2104 its own source file (printint.c) and used #include to pull it into
2105 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2106 separate copies.
2107
2108 (b) Defined the list of op-code names for debugging as a macro in
2109 internal.h so that it is next to the definition of the opcodes.
2110
2111 (c) Defined a table of op-code lengths for simpler skipping along compiled
2112 code. This is again a macro in internal.h so that it is next to the
2113 definition of the opcodes.
2114
2115 34. Added support for recursive calls to individual subpatterns, along the
2116 lines of Robin Houston's patch (but implemented somewhat differently).
2117
2118 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2119 allow it to read and process whole directories in Win32. This code was
2120 contributed by Lionel Fourquaux; it has not been tested by me.
2121
2122 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2123 used to name a group. Names consist of alphanumerics and underscores, and must
2124 be unique. Back references use the syntax (?P=name) and recursive calls use
2125 (?P>name) which is a PCRE extension to the Python extension. Groups still have
2126 numbers. The function pcre_fullinfo() can be used after compilation to extract
2127 a name/number map. There are three relevant calls:
2128
2129 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
2130 PCRE_INFO_NAMECOUNT yields the number of entries
2131 PCRE_INFO_NAMETABLE yields a pointer to the map.
2132
2133 The map is a vector of fixed-size entries. The size of each entry depends on
2134 the length of the longest name used. The first two bytes of each entry are the
2135 group number, most significant byte first. There follows the corresponding
2136 name, zero terminated. The names are in alphabetical order.
2137
2138 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2139 case instead of 255. Making it the same both with and without UTF-8 support
2140 means that the same test output works with both.
2141
2142 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2143 calling malloc() with a zero argument.
2144
2145 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2146 optimization. I've improved things by keeping a bitmap of backreferences with
2147 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2148 fact referenced, the optimization can be applied. It is unlikely that a
2149 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2150 the match to follow \n) will appear inside brackets with a number greater than
2151 31, but if it does, any back reference > 31 suppresses the optimization.
2152
2153 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2154 of disabling numbered capturing parentheses. Any opening parenthesis that is
2155 not followed by ? behaves as if it were followed by ?: but named parentheses
2156 can still be used for capturing (and they will acquire numbers in the usual
2157 way).
2158
2159 41. Redesigned the return codes from the match() function into yes/no/error so
2160 that errors can be passed back from deep inside the nested calls. A malloc
2161 failure while inside a recursive subpattern call now causes the
2162 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2163
2164 42. It is now possible to set a limit on the number of times the match()
2165 function is called in a call to pcre_exec(). This facility makes it possible to
2166 limit the amount of recursion and backtracking, though not in a directly
2167 obvious way, because the match() function is used in a number of different
2168 circumstances. The count starts from zero for each position in the subject
2169 string (for non-anchored patterns). The default limit is, for compatibility, a
2170 large number, namely 10 000 000. You can change this in two ways:
2171
2172 (a) When configuring PCRE before making, you can use --with-match-limit=n
2173 to set a default value for the compiled library.
2174
2175 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2176 a different value is set. See 45 below.
2177
2178 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2179
2180 43. Added a new function pcre_config(int, void *) to enable run-time extraction
2181 of things that can be changed at compile time. The first argument specifies
2182 what is wanted and the second points to where the information is to be placed.
2183 The current list of available information is:
2184
2185 PCRE_CONFIG_UTF8
2186
2187 The output is an integer that is set to one if UTF-8 support is available;
2188 otherwise it is set to zero.
2189
2190 PCRE_CONFIG_NEWLINE
2191
2192 The output is an integer that it set to the value of the code that is used for
2193 newline. It is either LF (10) or CR (13).
2194
2195 PCRE_CONFIG_LINK_SIZE
2196
2197 The output is an integer that contains the number of bytes used for internal
2198 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2199
2200 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2201
2202 The output is an integer that contains the threshold above which the POSIX
2203 interface uses malloc() for output vectors. See item 31 above.
2204
2205 PCRE_CONFIG_MATCH_LIMIT
2206
2207 The output is an unsigned integer that contains the default limit of the number
2208 of match() calls in a pcre_exec() execution. See 42 above.
2209
2210 44. pcretest has been upgraded by the addition of the -C option. This causes it
2211 to extract all the available output from the new pcre_config() function, and to
2212 output it. The program then exits immediately.
2213
2214 45. A need has arisen to pass over additional data with calls to pcre_exec() in
2215 order to support additional features. One way would have been to define
2216 pcre_exec2() (for example) with extra arguments, but this would not have been
2217 extensible, and would also have required all calls to the original function to
2218 be mapped to the new one. Instead, I have chosen to extend the mechanism that
2219 is used for passing in "extra" data from pcre_study().
2220
2221 The pcre_extra structure is now exposed and defined in pcre.h. It currently
2222 contains the following fields:
2223
2224 flags a bitmap indicating which of the following fields are set
2225 study_data opaque data from pcre_study()
2226 match_limit a way of specifying a limit on match() calls for a specific
2227 call to pcre_exec()
2228 callout_data data for callouts (see 49 below)
2229
2230 The flag bits are also defined in pcre.h, and are
2231
2232 PCRE_EXTRA_STUDY_DATA
2233 PCRE_EXTRA_MATCH_LIMIT
2234 PCRE_EXTRA_CALLOUT_DATA
2235
2236 The pcre_study() function now returns one of these new pcre_extra blocks, with
2237 the actual study data pointed to by the study_data field, and the
2238 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2239 before. That is, this change is entirely upwards-compatible and requires no
2240 change to existing code.
2241
2242 If you want to pass in additional data to pcre_exec(), you can either place it
2243 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2244 block.
2245
2246 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2247 data string contains the escape sequence \M, pcretest calls pcre_exec() several
2248 times with different match limits, until it finds the minimum value needed for
2249 pcre_exec() to complete. The value is then output. This can be instructive; for
2250 most simple matches the number is quite small, but for pathological cases it
2251 gets very large very quickly.
2252
2253 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2254 returns the size of the data block pointed to by the study_data field in a
2255 pcre_extra block, that is, the value that was passed as the argument to
2256 pcre_malloc() when PCRE was getting memory in which to place the information
2257 created by pcre_study(). The fourth argument should point to a size_t variable.
2258 pcretest has been extended so that this information is shown after a successful
2259 pcre_study() call when information about the compiled regex is being displayed.
2260
2261 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2262 because what follows is always an absolute path. (Later: it turns out that this
2263 is more than cosmetic for MinGW, because it doesn't like empty path
2264 components.)
2265
2266 49. Some changes have been made to the callout feature (see 28 above):
2267
2268 (i) A callout function now has three choices for what it returns:
2269
2270 0 => success, carry on matching
2271 > 0 => failure at this point, but backtrack if possible
2272 < 0 => serious error, return this value from pcre_exec()
2273
2274 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2275 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2276 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2277 use by callout functions. It will never be used by PCRE itself.
2278
2279 (ii) The pcre_extra structure (see 45 above) has a void * field called
2280 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2281 pcre_callout_block structure has a field of the same name. The contents of
2282 the field passed in the pcre_extra structure are passed to the callout
2283 function in the corresponding field in the callout block. This makes it
2284 easier to use the same callout-containing regex from multiple threads. For
2285 testing, the pcretest program has a new data escape
2286
2287 \C*n pass the number n (may be negative) as callout_data
2288
2289 If the callout function in pcretest receives a non-zero value as
2290 callout_data, it returns that value.
2291
2292 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2293 there were some redundant $(CFLAGS) in commands that are now specified as
2294 $(LINK), which already includes $(CFLAGS).
2295
2296 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2297 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2298 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2299 one-byte characters throughout. Note that case-insensitive matching applies
2300 only to characters whose values are less than 256. PCRE doesn't support the
2301 notion of cases for higher-valued characters.
2302
2303 (i) A character class whose characters are all within 0-255 is handled as
2304 a bit map, and the map is inverted for negative classes. Previously, a
2305 character > 255 always failed to match such a class; however it should
2306 match if the class was a negative one (e.g. [^ab]). This has been fixed.
2307
2308 (ii) A negated character class with a single character < 255 is coded as
2309 "not this character" (OP_NOT). This wasn't working properly when the test
2310 character was multibyte, either singly or repeated.
2311
2312 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2313 mode, for example: \x{100}{2,3}.
2314
2315 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2316 singly or repeated) now correctly test multibyte characters. However,
2317 PCRE doesn't recognize any characters with values greater than 255 as
2318 digits, spaces, or word characters. Such characters always match \D, \S,
2319 and \W, and never match \d, \s, or \w.
2320
2321 (v) Classes may now contain characters and character ranges with values
2322 greater than 255. For example: [ab\x{100}-\x{400}].
2323
2324 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
2325 PCRE in UTF-8 mode.
2326
2327 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2328 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2329 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2330 value.)
2331
2332 53. The single man page has become too large. I have therefore split it up into
2333 a number of separate man pages. These also give rise to individual HTML pages;
2334 these are now put in a separate directory, and there is an index.html page that
2335 lists them all. Some hyperlinking between the pages has been installed.
2336
2337 54. Added convenience functions for handling named capturing parentheses.
2338
2339 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2340 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2341 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2342 are faulted.
2343
2344 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2345 calling configure. These values are used when compiling the dftables.c program
2346 which is run to generate the source of the default character tables. They
2347 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2348 you will need to set these values.
2349
2350 57. Updated the building process for Windows DLL, as provided by Fred Cox.
2351
2352
2353 Version 3.9 02-Jan-02
2354 ---------------------
2355
2356 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2357
2358 2. If --disable-static was given, the building process failed when trying to
2359 build pcretest and pcregrep. (For some reason it was using libtool to compile
2360 them, which is not right, as they aren't part of the library.)
2361
2362
2363 Version 3.8 18-Dec-01
2364 ---------------------
2365
2366 1. The experimental UTF-8 code was completely screwed up. It was packing the
2367 bytes in the wrong order. How dumb can you get?
2368
2369
2370 Version 3.7 29-Oct-01
2371 ---------------------
2372
2373 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2374 This caused pcretest, when used on the test data, to segfault. Unfortunately,
2375 this didn't happen under Solaris 8, where I normally test things.
2376
2377 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2378 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2379 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2380 this fix an hour or so after the initial 3.7 release.)
2381
2382
2383 Version 3.6 23-Oct-01
2384 ---------------------
2385
2386 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2387 offsets passed as NULL with zero offset count.
2388
2389 2. The config.guess and config.sub files had not been updated when I moved to
2390 the latest autoconf.
2391
2392
2393 Version 3.5 15-Aug-01
2394 ---------------------
2395
2396 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2397 had been forgotten.
2398
2399 2. By using declared but undefined structures, we can avoid using "void"
2400 definitions in pcre.h while keeping the internal definitions of the structures
2401 private.
2402
2403 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2404 user point of view, this means that both static and shared libraries are built
2405 by default, but this can be individually controlled. More of the work of
2406 handling this static/shared cases is now inside libtool instead of PCRE's make
2407 file.
2408
2409 4. The pcretest utility is now installed along with pcregrep because it is
2410 useful for users (to test regexs) and by doing this, it automatically gets
2411 relinked by libtool. The documentation has been turned into a man page, so
2412 there are now .1, .txt, and .html versions in /doc.
2413
2414 5. Upgrades to pcregrep:
2415 (i) Added long-form option names like gnu grep.
2416 (ii) Added --help to list all options with an explanatory phrase.
2417 (iii) Added -r, --recursive to recurse into sub-directories.
2418 (iv) Added -f, --file to read patterns from a file.
2419
2420 6. pcre_exec() was referring to its "code" argument before testing that
2421 argument for NULL (and giving an error if it was NULL).
2422
2423 7. Upgraded Makefile.in to allow for compiling in a different directory from
2424 the source directory.
2425
2426 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2427 options bits, the pointer it was passed was to an int instead of to an unsigned
2428 long int. This mattered only on 64-bit systems.
2429
2430 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2431 generated) instead of pcre.in, which it its source. Also made the same change
2432 in several of the .c files.
2433
2434 10. A new release of gcc defines printf() as a macro, which broke pcretest
2435 because it had an ifdef in the middle of a string argument for printf(). Fixed
2436 by using separate calls to printf().
2437
2438 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2439 script, to force use of CR or LF instead of \n in the source. On non-Unix
2440 systems, the value can be set in config.h.
2441
2442 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2443 absolute limit. Changed the text of the error message to make this clear, and
2444 likewise updated the man page.
2445
2446 13. The limit of 99 on the number of capturing subpatterns has been removed.
2447 The new limit is 65535, which I hope will not be a "real" limit.
2448
2449
2450 Version 3.4 22-Aug-00
2451 ---------------------
2452
2453 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2454
2455 2. Diagnose condition (?(0) as an error instead of crashing on matching.
2456
2457
2458 Version 3.3 01-Aug-00
2459 ---------------------
2460
2461 1. If an octal character was given, but the value was greater than \377, it
2462 was not getting masked to the least significant bits, as documented. This could
2463 lead to crashes in some systems.
2464
2465 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2466 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2467
2468 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2469 These just pass their arguments on to (pcre_free)(), but they are provided
2470 because some uses of PCRE bind it to non-C systems that can call its functions,
2471 but cannot call free() or pcre_free() directly.
2472
2473 4. Add "make test" as a synonym for "make check". Corrected some comments in
2474 the Makefile.
2475
2476 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2477 Makefile.
2478
2479 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2480 command called pgrep for grepping around the active processes.
2481
2482 7. Added the beginnings of support for UTF-8 character strings.
2483
2484 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2485 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2486 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2487 out for the ar command.)
2488
2489
2490 Version 3.2 12-May-00
2491 ---------------------
2492
2493 This is purely a bug fixing release.
2494
2495 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2496 of ZA. This was just one example of several cases that could provoke this bug,
2497 which was introduced by change 9 of version 2.00. The code for breaking
2498 infinite loops after an iteration that matches an empty string was't working
2499 correctly.
2500
2501 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2502 when matched against abbab (for example). After matching an empty string, it
2503 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2504 caused it to match further down the string than it should.
2505
2506 3. The code contained an inclusion of sys/types.h. It isn't clear why this
2507 was there because it doesn't seem to be needed, and it causes trouble on some
2508 systems, as it is not a Standard C header. It has been removed.
2509
2510 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2511 were reported on the Macintosh. The changes were from
2512
2513 while ((c = *(++ptr)) != 0 && c != '\n');
2514 to
2515 while ((c = *(++ptr)) != 0 && c != '\n') ;
2516
2517 Totally extraordinary, but if that's what it takes...
2518
2519 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2520 available. Added HAVE_BCOPY and an autoconf test for it; if neither
2521 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2522 assumes the way PCRE uses memmove() (always moving upwards).
2523
2524 6. PCRE is being used in one environment where strchr() is not available. There
2525 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2526 faster code anyway.
2527
2528
2529 Version 3.1 09-Feb-00
2530 ---------------------
2531
2532 The only change in this release is the fixing of some bugs in Makefile.in for
2533 the "install" target:
2534
2535 (1) It was failing to install pcreposix.h.
2536
2537 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2538
2539
2540 Version 3.0 01-Feb-00
2541 ---------------------
2542
2543 1. Add support for the /+ modifier to perltest (to output $` like it does in
2544 pcretest).
2545
2546 2. Add support for the /g modifier to perltest.
2547
2548 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2549 matches null strings.
2550
2551 4. Fix perltest so that it doesn't do unwanted things when fed an empty
2552 pattern. Perl treats empty patterns specially - it reuses the most recent
2553 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2554 effect.
2555
2556 5. The POSIX interface was broken in that it was just handing over the POSIX
2557 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2558 required a bigger vector, with some working space on the end. This means that
2559 the POSIX wrapper now has to get and free some memory, and copy the results.
2560
2561 6. Added some simple autoconf support, placing the test data and the
2562 documentation in separate directories, re-organizing some of the
2563 information files, and making it build pcre-config (a GNU standard). Also added
2564 libtool support for building PCRE as a shared library, which is now the
2565 default.
2566
2567 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2568 09 are not valid octal constants. Single digits will be used for minor values
2569 less than 10.
2570
2571 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2572 existing programs that set these in the POSIX interface can use PCRE without
2573 modification.
2574
2575 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2576 return all that pcre_info() returns, plus additional data. The pcre_info()
2577 function is retained for compatibility, but is considered to be obsolete.
2578
2579 10. Added experimental recursion feature (?R) to handle one common case that
2580 Perl 5.6 will be able to do with (?p{...}).
2581
2582 11. Added support for POSIX character classes like [:alpha:], which Perl is
2583 adopting.
2584
2585
2586 Version 2.08 31-Aug-99
2587 ----------------------
2588
2589 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2590 trying to match at the startoffset position, but instead was moving forward to
2591 the next newline as if a previous match had failed.
2592
2593 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2594 and could get into a loop if a null string was matched other than at the start
2595 of the subject.
2596
2597 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2598 be distinguished at compile time, and for completeness also added PCRE_DATE.
2599
2600 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2601 in GnuWin32 environments.
2602
2603
2604 Version 2.07 29-Jul-99
2605 ----------------------
2606
2607 1. The documentation is now supplied in plain text form and HTML as well as in
2608 the form of man page sources.
2609
2610 2. C++ compilers don't like assigning (void *) values to other pointer types.
2611 In particular this affects malloc(). Although there is no problem in Standard
2612 C, I've put in casts to keep C++ compilers happy.
2613
2614 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2615 should be (const char *).
2616
2617 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2618 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2619 However, I haven't made this a standard facility. The documentation doesn't
2620 mention it, and the Makefile doesn't support it.
2621
2622 5. The Makefile now contains an "install" target, with editable destinations at
2623 the top of the file. The pcretest program is not installed.
2624
2625 6. pgrep -V now gives the PCRE version number and date.
2626
2627 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2628 causing the entire string to be ignored, instead of just the last character.
2629
2630 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2631 non-matching string, it can take a very, very long time, even for strings of
2632 quite modest length, because of the nested recursion. PCRE now does better in
2633 some of these cases. It does this by remembering the last required literal
2634 character in the pattern, and pre-searching the subject to ensure it is present
2635 before running the real match. In other words, it applies a heuristic to detect
2636 some types of certain failure quickly, and in the above example, if presented
2637 with a string that has no trailing " it gives "no match" very quickly.
2638
2639 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2640 other alternatives are tried instead.
2641
2642
2643 Version 2.06 09-Jun-99
2644 ----------------------
2645
2646 1. Change pcretest's output for amount of store used to show just the code
2647 space, because the remainder (the data block) varies in size between 32-bit and
2648 64-bit systems.
2649
2650 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2651 start matching at. This allows lookbehinds to work when searching for multiple
2652 occurrences in a string.
2653
2654 3. Added additional options to pcretest for testing multiple occurrences:
2655
2656 /+ outputs the rest of the string that follows a match
2657 /g loops for multiple occurrences, using the new startoffset argument
2658 /G loops for multiple occurrences by passing an incremented pointer
2659
2660 4. PCRE wasn't doing the "first character" optimization for patterns starting
2661 with \b or \B, though it was doing it for other lookbehind assertions. That is,
2662 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2663 the letter 'x'. On long subject strings, this gives a significant speed-up.
2664
2665
2666 Version 2.05 21-Apr-99
2667 ----------------------
2668
2669 1. Changed the type of magic_number from int to long int so that it works
2670 properly on 16-bit systems.
2671
2672 2. Fixed a bug which caused patterns starting with .* not to work correctly
2673 when the subject string contained newline characters. PCRE was assuming
2674 anchoring for such patterns in all cases, which is not correct because .* will
2675 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2676 DOTALL is set at top level; otherwise it knows that patterns starting with .*
2677 must be retried after every newline in the subject.
2678
2679
2680 Version 2.04 18-Feb-99
2681 ----------------------
2682
2683 1. For parenthesized subpatterns with repeats whose minimum was zero, the
2684 computation of the store needed to hold the pattern was incorrect (too large).
2685 If such patterns were nested a few deep, this could multiply and become a real
2686 problem.
2687
2688 2. Added /M option to pcretest to show the memory requirement of a specific
2689 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2690
2691 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2692 compiled in such a way that the backtracking after subsequent failure was
2693 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2694 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2695
2696
2697 Version 2.03 02-Feb-99
2698 ----------------------
2699
2700 1. Fixed typo and small mistake in man page.
2701
2702 2. Added 4th condition (GPL supersedes if conflict) and created separate
2703 LICENCE file containing the conditions.
2704
2705 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2706 Perl, that is the internal \ allows the delimiter to be included in the
2707 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2708 the final delimiter, add \ to the end of the pattern (to test the error).
2709
2710 4. Added the convenience functions for extracting substrings after a successful
2711 match. Updated pcretest to make it able to test these functions.
2712
2713
2714 Version 2.02 14-Jan-99
2715 ----------------------
2716
2717 1. Initialized the working variables associated with each extraction so that
2718 their saving and restoring doesn't refer to uninitialized store.
2719
2720 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2721 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2722 fix the problem.
2723
2724 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2725 calls, and wasn't printing the correct value for compiling calls. Increased the
2726 default value of LOOPREPEAT, and the number of significant figures in the
2727 times.
2728
2729 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2730
2731 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2732 a building problem on Windows NT with a FAT file system.
2733
2734
2735 Version 2.01 21-Oct-98
2736 ----------------------
2737
2738 1. Changed the API for pcre_compile() to allow for the provision of a pointer
2739 to character tables built by pcre_maketables() in the current locale. If NULL
2740 is passed, the default tables are used.
2741
2742
2743 Version 2.00 24-Sep-98
2744 ----------------------
2745
2746 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2747 it any more.
2748
2749 2. Allow quantification of (?>) groups, and make it work correctly.
2750
2751 3. The first character computation wasn't working for (?>) groups.
2752
2753 4. Correct the implementation of \Z (it is permitted to match on the \n at the
2754 end of the subject) and add 5.005's \z, which really does match only at the
2755 very end of the subject.
2756
2757 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2758
2759 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2760 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2761 localized options. All options to pcre_study() were also removed.
2762
2763 7. Add other new features from 5.005:
2764
2765 $(?<= positive lookbehind
2766 $(?<! negative lookbehind
2767 (?imsx-imsx) added the unsetting capability
2768 such a setting is global if at outer level; local otherwise
2769 (?imsx-imsx:) non-capturing groups with option setting
2770 (?(cond)re|re) conditional pattern matching
2771
2772 A backreference to itself in a repeated group matches the previous
2773 captured string.
2774
2775 8. General tidying up of studying (both automatic and via "study")
2776 consequential on the addition of new assertions.
2777
2778 9. As in 5.005, unlimited repeated groups that could match an empty substring
2779 are no longer faulted at compile time. Instead, the loop is forcibly broken at
2780 runtime if any iteration does actually match an empty substring.
2781
2782 10. Include the RunTest script in the distribution.
2783
2784 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2785 discrepancies, some of which were old and were also with respect to 5.004. They
2786 have now been fixed.
2787
2788
2789 Version 1.09 28-Apr-98
2790 ----------------------
2791
2792 1. A negated single character class followed by a quantifier with a minimum
2793 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
2794 program crashes, or just wrong answers. This did not apply to negated classes
2795 containing more than one character, or to minima other than one.
2796
2797
2798 Version 1.08 27-Mar-98
2799 ----------------------
2800
2801 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
2802
2803 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
2804 latter must appear before anything that relies on it in the pattern.
2805
2806
2807 Version 1.07 16-Feb-98
2808 ----------------------
2809
2810 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
2811 repeat of a potentially empty string).
2812
2813
2814 Version 1.06 23-Jan-98
2815 ----------------------
2816
2817 1. Added Markus Oberhumer's little patches for C++.
2818
2819 2. Literal strings longer than 255 characters were broken.
2820
2821
2822 Version 1.05 23-Dec-97
2823 ----------------------
2824
2825 1. Negated character classes containing more than one character were failing if
2826 PCRE_CASELESS was set at run time.
2827
2828
2829 Version 1.04 19-Dec-97
2830 ----------------------
2831
2832 1. Corrected the man page, where some "const" qualifiers had been omitted.
2833
2834 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
2835 input syntax.
2836
2837 3. Fixed memory leak which occurred when a regex with back references was
2838 matched with an offsets vector that wasn't big enough. The temporary memory
2839 that is used in this case wasn't being freed if the match failed.
2840
2841 4. Tidied pcretest to ensure it frees memory that it gets.
2842
2843 5. Temporary memory was being obtained in the case where the passed offsets
2844 vector was exactly big enough.
2845
2846 6. Corrected definition of offsetof() from change 5 below.
2847
2848 7. I had screwed up change 6 below and broken the rules for the use of
2849 setjmp(). Now fixed.
2850
2851
2852 Version 1.03 18-Dec-97
2853 ----------------------
2854
2855 1. A erroneous regex with a missing opening parenthesis was correctly
2856 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
2857 on some systems.
2858
2859 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
2860 it was reported that one broken compiler failed on the former because "code" is
2861 also an independent variable.
2862
2863 3. The erroneous regex a[]b caused an array overrun reference.
2864
2865 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
2866 fail on data ending with that character. (It was going on too far, and checking
2867 the next character, typically a binary zero.) This was specific to the
2868 optimized code for single-character negative classes.
2869
2870 5. Added a contributed patch from the TIN world which does the following:
2871
2872 + Add an undef for memmove, in case the the system defines a macro for it.
2873
2874 + Add a definition of offsetof(), in case there isn't one. (I don't know
2875 the reason behind this - offsetof() is part of the ANSI standard - but
2876 it does no harm).
2877
2878 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
2879 most of the places where whitespace preceded '#'. I have given up and
2880 allowed the remaining 2 cases to be at the margin.
2881
2882 + Rename some variables in pcre to eliminate shadowing. This seems very
2883 pedantic, but does no harm, of course.
2884
2885 6. Moved the call to setjmp() into its own function, to get rid of warnings
2886 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
2887
2888 7. Constructs such as \d{8,} were compiling into the equivalent of
2889 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
2890 outcome, but in this particular case used more store than had been allocated,
2891 which caused the bug to be discovered because it threw up an internal error.
2892
2893 8. The debugging code in both pcre and pcretest for outputting the compiled
2894 form of a regex was going wrong in the case of back references followed by
2895 curly-bracketed repeats.
2896
2897
2898 Version 1.02 12-Dec-97
2899 ----------------------
2900
2901 1. Typos in pcre.3 and comments in the source fixed.
2902
2903 2. Applied a contributed patch to get rid of places where it used to remove
2904 'const' from variables, and fixed some signed/unsigned and uninitialized
2905 variable warnings.
2906
2907 3. Added the "runtest" target to Makefile.
2908
2909 4. Set default compiler flag to -O2 rather than just -O.
2910
2911
2912 Version 1.01 19-Nov-97
2913 ----------------------
2914
2915 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
2916 like /([ab]*)*/, that is, for classes with more than one character in them.
2917
2918 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
2919 as /((?>a*))*/ (a PCRE_EXTRA facility).
2920
2921
2922 Version 1.00 18-Nov-97
2923 ----------------------
2924
2925 1. Added compile-time macros to support systems such as SunOS4 which don't have
2926 memmove() or strerror() but have other things that can be used instead.
2927
2928 2. Arranged that "make clean" removes the executables.
2929
2930
2931 Version 0.99 27-Oct-97
2932 ----------------------
2933
2934 1. Fixed bug in code for optimizing classes with only one character. It was
2935 initializing a 32-byte map regardless, which could cause it to run off the end
2936 of the memory it had got.
2937
2938 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
2939
2940
2941 Version 0.98 22-Oct-97
2942 ----------------------
2943
2944 1. Fixed bug in code for handling temporary memory usage when there are more
2945 back references than supplied space in the ovector. This could cause segfaults.
2946
2947
2948 Version 0.97 21-Oct-97
2949 ----------------------
2950
2951 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
2952
2953 2. Optimized negated single characters not to use a bit map.
2954
2955 3. Brought error texts together as macro definitions; clarified some of them;
2956 fixed one that was wrong - it said "range out of order" when it meant "invalid
2957 escape sequence".
2958
2959 4. Changed some char * arguments to const char *.
2960
2961 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
2962
2963 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
2964 pcretest.
2965
2966
2967 Version 0.96 16-Oct-97
2968 ----------------------
2969
2970 1. Added a simple "pgrep" utility to the distribution.
2971
2972 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
2973 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
2974 where "ddd" means "one or more decimal digits".
2975
2976 3. Fixed serious bug. If a pattern had a back reference, but the call to
2977 pcre_exec() didn't supply a large enough ovector to record the related
2978 identifying subpattern, the match always failed. PCRE now remembers the number
2979 of the largest back reference, and gets some temporary memory in which to save
2980 the offsets during matching if necessary, in order to ensure that
2981 backreferences always work.
2982
2983 4. Increased the compatibility with Perl in a number of ways:
2984
2985 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
2986 to request this handling. The option can be set at compile or exec time.
2987
2988 (b) $ matches before a terminating newline by default; an option
2989 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
2990 mode). The option can be set at compile or exec time.
2991
2992 (c) The handling of \ followed by a digit other than 0 is now supposed to be
2993 the same as Perl's. If the decimal number it represents is less than 10
2994 or there aren't that many previous left capturing parentheses, an octal
2995 escape is read. Inside a character class, it's always an octal escape,
2996 even if it is a single digit.
2997
2998 (d) An escaped but undefined alphabetic character is taken as a literal,
2999 unless PCRE_EXTRA is set. Currently this just reserves the remaining
3000 escapes.
3001
3002 (e) {0} is now permitted. (The previous item is removed from the compiled
3003 pattern).
3004
3005 5. Changed all the names of code files so that the basic parts are no longer
3006 than 10 characters, and abolished the teeny "globals.c" file.
3007
3008 6. Changed the handling of character classes; they are now done with a 32-byte
3009 bit map always.
3010
3011 7. Added the -d and /D options to pcretest to make it possible to look at the
3012 internals of compilation without having to recompile pcre.
3013
3014
3015 Version 0.95 23-Sep-97
3016 ----------------------
3017
3018 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
3019 \x20 at the start of a run of normal characters. These were being treated as
3020 real characters, instead of the source characters being re-checked.
3021
3022
3023 Version 0.94 18-Sep-97
3024 ----------------------
3025
3026 1. The functions are now thread-safe, with the caveat that the global variables
3027 containing pointers to malloc() and free() or alternative functions are the
3028 same for all threads.
3029
3030 2. Get pcre_study() to generate a bitmap of initial characters for non-
3031 anchored patterns when this is possible, and use it if passed to pcre_exec().
3032
3033
3034 Version 0.93 15-Sep-97
3035 ----------------------
3036
3037 1. /(b)|(:+)/ was computing an incorrect first character.
3038
3039 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
3040 but not actually doing anything yet.
3041
3042 3. Treat "-" characters in classes that cannot be part of ranges as literals,
3043 as Perl does (e.g. [-az] or [az-]).
3044
3045 4. Set the anchored flag if a branch starts with .* or .*? because that tests
3046 all possible positions.
3047
3048 5. Split up into different modules to avoid including unneeded functions in a
3049 compiled binary. However, compile and exec are still in one module. The "study"
3050 function is split off.
3051
3052 6. The character tables are now in a separate module whose source is generated
3053 by an auxiliary program - but can then be edited by hand if required. There are
3054 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
3055 toupper() in the code.
3056
3057 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
3058 make them global. Abolish the function for setting them, as the caller can now
3059 set them directly.
3060
3061
3062 Version 0.92 11-Sep-97
3063 ----------------------
3064
3065 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
3066 (e.g. /a{1,3}/) was broken (I mis-optimized it).
3067
3068 2. Caseless matching was not working in character classes if the characters in
3069 the pattern were in upper case.
3070
3071 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
3072
3073 4. Make PCRE_ANCHORED public and accept as a compile option.
3074
3075 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
3076 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
3077 pass them.
3078
3079 6. Give an error if bad option bits passed at compile or run time.
3080
3081 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
3082 pcretest to cause it to pass that flag.
3083
3084 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
3085 options, and the first character, if set.
3086
3087 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
3088
3089
3090 Version 0.91 10-Sep-97
3091 ----------------------
3092
3093 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
3094 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
3095
3096 2. PCRE was looping on encountering an indefinitely repeated back reference to
3097 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
3098 Perl does - treats the match as successful.
3099
3100 ****

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