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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 294 - (show annotations) (download)
Sat Dec 29 16:55:59 2007 UTC (6 years, 10 months ago) by ph10
File size: 147358 byte(s)
Fix UTF-8/newline=ANY with .* crash.

1 ChangeLog for PCRE
2 ------------------
3
4 Version 7.5 29-Dec-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. Added --file-offsets and --line-offsets to pcregrep.
80
81 15. The pattern (?=something)(?R) was not being diagnosed as a potentially
82 infinitely looping recursion. The bug was that positive lookaheads were not
83 being skipped when checking for a possible empty match (negative lookaheads
84 and both kinds of lookbehind were skipped).
85
86 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
87 inclusion of <windows.h> to before rather than after the definition of
88 INVALID_FILE_ATTRIBUTES (patch from David Byron).
89
90 17. Specifying a possessive quantifier with a specific limit for a Unicode
91 character property caused pcre_compile() to compile bad code, which led at
92 runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
93 are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
94 caused the error; without that there was no problem.
95
96 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
97
98 19. Added --enable-pcretest-libreadline.
99
100 20. In pcrecpp.cc, the variable 'count' was incremented twice in
101 RE::GlobalReplace(). As a result, the number of replacements returned was
102 double what it should be. I have removed one of the increments.
103
104 21. Several CMake things:
105
106 (1) Arranged that, when cmake is used on Unix, the libraries end up with
107 the names libpcre and libpcreposix, not just pcre and pcreposix.
108
109 (2) The above change means that pcretest and pcregrep are now correctly
110 linked with the newly-built libraries, not previously installed ones.
111
112 (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
113
114 22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
115 crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
116 UTF-8 newline character). The key issue is that the pattern starts .*;
117 this means that the match must be either at the beginning, or after a
118 newline. The bug was in the code for advancing after a failed match and
119 checking that the new position followed a newline. It was not taking
120 account of UTF-8 characters correctly.
121
122
123 Version 7.4 21-Sep-07
124 ---------------------
125
126 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
127 means that a class such as [\s] counted as "explicit reference to CR or
128 LF". That isn't really right - the whole point of the change was to try to
129 help when there was an actual mention of one of the two characters. So now
130 the change happens only if \r or \n (or a literal CR or LF) character is
131 encountered.
132
133 2. The 32-bit options word was also used for 6 internal flags, but the numbers
134 of both had grown to the point where there were only 3 bits left.
135 Fortunately, there was spare space in the data structure, and so I have
136 moved the internal flags into a new 16-bit field to free up more option
137 bits.
138
139 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
140 but did not set the internal JCHANGED flag - either of these is enough to
141 control the way the "get" function works - but the PCRE_INFO_JCHANGED
142 facility is supposed to tell if (?J) was ever used, so now (?J) at the
143 start sets both bits.
144
145 4. Added options (at build time, compile time, exec time) to change \R from
146 matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
147
148 5. doc/pcresyntax.html was missing from the distribution.
149
150 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
151 compatibility, even though it is no longer used.
152
153 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
154 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
155 windows.h file is present (where different names are used). [This was
156 reversed later after testing - see 16 below.]
157
158 8. Changed all #include <config.h> to #include "config.h". There were also
159 some further <pcre.h> cases that I changed to "pcre.h".
160
161 9. When pcregrep was used with the --colour option, it missed the line ending
162 sequence off the lines that it output.
163
164 10. It was pointed out to me that arrays of string pointers cause lots of
165 relocations when a shared library is dynamically loaded. A technique of
166 using a single long string with a table of offsets can drastically reduce
167 these. I have refactored PCRE in four places to do this. The result is
168 dramatic:
169
170 Originally: 290
171 After changing UCP table: 187
172 After changing error message table: 43
173 After changing table of "verbs" 36
174 After changing table of Posix names 22
175
176 Thanks to the folks working on Gregex for glib for this insight.
177
178 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
179 unicode-properties was also set.
180
181 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
182
183 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
184 checked only for CRLF.
185
186 14. Added casts to pcretest.c to avoid compiler warnings.
187
188 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
189
190 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
191 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
192 entirely. This removes changes made in 7 above.
193
194 17. The CMake files have been updated, and there is now more information about
195 building with CMake in the NON-UNIX-USE document.
196
197
198 Version 7.3 28-Aug-07
199 ---------------------
200
201 1. In the rejigging of the build system that eventually resulted in 7.1, the
202 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
203 brackets there is not right, since it causes compilers to look for an
204 installed pcre.h, not the version that is in the source that is being
205 compiled (which of course may be different). I have changed it back to:
206
207 #include "pcre.h"
208
209 I have a vague recollection that the change was concerned with compiling in
210 different directories, but in the new build system, that is taken care of
211 by the VPATH setting the Makefile.
212
213 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
214 when the subject happened to end in the byte 0x85 (e.g. if the last
215 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
216 characters but of course it shouldn't be taken as a newline when it is part
217 of another character. The bug was that, for an unlimited repeat of . in
218 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
219 characters when looking for a newline.
220
221 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
222
223 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
224 in debug output.
225
226 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
227 long printing in the pcrecpp unittest when running under MinGW.
228
229 6. ESC_K was left out of the EBCDIC table.
230
231 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
232 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
233 limit also applies to "virtual nesting" when a pattern is recursive, and in
234 this case 1000 isn't so big. I have been able to remove this limit at the
235 expense of backing off one optimization in certain circumstances. Normally,
236 when pcre_exec() would call its internal match() function recursively and
237 immediately return the result unconditionally, it uses a "tail recursion"
238 feature to save stack. However, when a subpattern that can match an empty
239 string has an unlimited repetition quantifier, it no longer makes this
240 optimization. That gives it a stack frame in which to save the data for
241 checking that an empty string has been matched. Previously this was taken
242 from the 1000-entry workspace that had been reserved. So now there is no
243 explicit limit, but more stack is used.
244
245 8. Applied Daniel's patches to solve problems with the import/export magic
246 syntax that is required for Windows, and which was going wrong for the
247 pcreposix and pcrecpp parts of the library. These were overlooked when this
248 problem was solved for the main library.
249
250 9. There were some crude static tests to avoid integer overflow when computing
251 the size of patterns that contain repeated groups with explicit upper
252 limits. As the maximum quantifier is 65535, the maximum group length was
253 set at 30,000 so that the product of these two numbers did not overflow a
254 32-bit integer. However, it turns out that people want to use groups that
255 are longer than 30,000 bytes (though not repeat them that many times).
256 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
257 made it possible to implement the integer overflow checks in a much more
258 dynamic way, which I have now done. The artificial limitation on group
259 length has been removed - we now have only the limit on the total length of
260 the compiled pattern, which depends on the LINK_SIZE setting.
261
262 10. Fixed a bug in the documentation for get/copy named substring when
263 duplicate names are permitted. If none of the named substrings are set, the
264 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
265 empty string.
266
267 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
268 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
269 because the ] is interpreted as the first data character and the
270 terminating ] is not found. PCRE has been made compatible with Perl in this
271 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
272 cause memory overwriting.
273
274 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
275 string has been matched (to stop an infinite loop). It was not recognizing
276 a conditional subpattern that could match an empty string if that
277 subpattern was within another subpattern. For example, it looped when
278 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
279 condition was not nested. This bug has been fixed.
280
281 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
282 past the start of the subject in the presence of bytes with the top bit
283 set, for example "\x8aBCD".
284
285 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
286 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
287
288 14. Optimized (?!) to (*FAIL).
289
290 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
291 This restricts code points to be within the range 0 to 0x10FFFF, excluding
292 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
293 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
294 does: it's just the validity check that is more restrictive.
295
296 16. Inserted checks for integer overflows during escape sequence (backslash)
297 processing, and also fixed erroneous offset values for syntax errors during
298 backslash processing.
299
300 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
301 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
302
303 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
304 caused an overrun.
305
306 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
307 something other than just ASCII characters) inside a group that had an
308 unlimited repeat caused a loop at compile time (while checking to see
309 whether the group could match an empty string).
310
311 20. Debugging a pattern containing \p or \P could cause a crash. For example,
312 [\P{Any}] did so. (Error in the code for printing property names.)
313
314 21. An orphan \E inside a character class could cause a crash.
315
316 22. A repeated capturing bracket such as (A)? could cause a wild memory
317 reference during compilation.
318
319 23. There are several functions in pcre_compile() that scan along a compiled
320 expression for various reasons (e.g. to see if it's fixed length for look
321 behind). There were bugs in these functions when a repeated \p or \P was
322 present in the pattern. These operators have additional parameters compared
323 with \d, etc, and these were not being taken into account when moving along
324 the compiled data. Specifically:
325
326 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
327 length.
328
329 (b) An item such as \pL+ within a repeated group could cause crashes or
330 loops.
331
332 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
333 "reference to non-existent subpattern" error.
334
335 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
336
337 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
338 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
339
340 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
341
342 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
343 character were causing crashes (broken optimization).
344
345 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
346 \p or \P) caused a compile-time loop.
347
348 28. More problems have arisen in unanchored patterns when CRLF is a valid line
349 break. For example, the unstudied pattern [\r\n]A does not match the string
350 "\r\nA" because change 7.0/46 below moves the current point on by two
351 characters after failing to match at the start. However, the pattern \nA
352 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
353 the same is true. There doesn't seem any very clean way out of this, but
354 what I have chosen to do makes the common cases work: PCRE now takes note
355 of whether there can be an explicit match for \r or \n anywhere in the
356 pattern, and if so, 7.0/46 no longer applies. As part of this change,
357 there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
358 pattern has explicit CR or LF references.
359
360 29. Added (*CR) etc for changing newline setting at start of pattern.
361
362
363 Version 7.2 19-Jun-07
364 ---------------------
365
366 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
367 which is apparently normally available under Windows.
368
369 2. Re-jig the pcregrep tests with different newline settings in an attempt
370 to make them independent of the local environment's newline setting.
371
372 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
373
374 4. Some of the "internals" tests were previously cut out when the link size
375 was not 2, because the output contained actual offsets. The recent new
376 "Z" feature of pcretest means that these can be cut out, making the tests
377 usable with all link sizes.
378
379 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
380 stack recursion. This gives a massive performance boost under BSD, but just
381 a small improvement under Linux. However, it saves one field in the frame
382 in all cases.
383
384 6. Added more features from the forthcoming Perl 5.10:
385
386 (a) (?-n) (where n is a string of digits) is a relative subroutine or
387 recursion call. It refers to the nth most recently opened parentheses.
388
389 (b) (?+n) is also a relative subroutine call; it refers to the nth next
390 to be opened parentheses.
391
392 (c) Conditions that refer to capturing parentheses can be specified
393 relatively, for example, (?(-2)... or (?(+3)...
394
395 (d) \K resets the start of the current match so that everything before
396 is not part of it.
397
398 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
399
400 (f) \g{name} is another synonym - part of Perl 5.10's unification of
401 reference syntax.
402
403 (g) (?| introduces a group in which the numbering of parentheses in each
404 alternative starts with the same number.
405
406 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
407
408 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
409 PCRE_INFO_JCHANGED.
410
411 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
412 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
413 for detecting groups that can match an empty string.
414
415 9. A pattern with a very large number of alternatives (more than several
416 hundred) was running out of internal workspace during the pre-compile
417 phase, where pcre_compile() figures out how much memory will be needed. A
418 bit of new cunning has reduced the workspace needed for groups with
419 alternatives. The 1000-alternative test pattern now uses 12 bytes of
420 workspace instead of running out of the 4096 that are available.
421
422 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
423
424 11. Applied patch from Google to remove an optimization that didn't quite work.
425 The report of the bug said:
426
427 pcrecpp::RE("a*").FullMatch("aaa") matches, while
428 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
429 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
430
431 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
432 it matched the wrong number of bytes.
433
434
435 Version 7.1 24-Apr-07
436 ---------------------
437
438 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
439 that is more "standard", making use of automake and other Autotools. There
440 is some re-arrangement of the files and adjustment of comments consequent
441 on this.
442
443 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
444 for recursive directory scanning broke on some systems because the files
445 are not scanned in any specific order and on different systems the order
446 was different. A call to "sort" has been inserted into RunGrepTest for the
447 approprate test as a short-term fix. In the longer term there may be an
448 alternative.
449
450 3. I had an email from Eric Raymond about problems translating some of PCRE's
451 man pages to HTML (despite the fact that I distribute HTML pages, some
452 people do their own conversions for various reasons). The problems
453 concerned the use of low-level troff macros .br and .in. I have therefore
454 removed all such uses from the man pages (some were redundant, some could
455 be replaced by .nf/.fi pairs). The 132html script that I use to generate
456 HTML has been updated to handle .nf/.fi and to complain if it encounters
457 .br or .in.
458
459 4. Updated comments in configure.ac that get placed in config.h.in and also
460 arranged for config.h to be included in the distribution, with the name
461 config.h.generic, for the benefit of those who have to compile without
462 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
463
464 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
465 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
466 makevp.bat for the current PCRE, using the additional files
467 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
468
469 6. A Windows user reported a minor discrepancy with test 2, which turned out
470 to be caused by a trailing space on an input line that had got lost in his
471 copy. The trailing space was an accident, so I've just removed it.
472
473 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
474 that is needed.
475
476 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
477 as "const" (a) because they are and (b) because it helps the PHP
478 maintainers who have recently made a script to detect big data structures
479 in the php code that should be moved to the .rodata section. I remembered
480 to update Builducptable as well, so it won't revert if ucptable.h is ever
481 re-created.
482
483 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
484 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
485 order to be able to cut out the UTF-8 tables in the latter when UTF-8
486 support is not required. This saves 1.5-2K of code, which is important in
487 some applications.
488
489 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
490 so as not to refer to the tables, even though these functions will never be
491 called when UTF-8 support is disabled. Otherwise there are problems with a
492 shared library.
493
494 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
495
496 (a) It was defining its arguments as char * instead of void *.
497
498 (b) It was assuming that all moves were upwards in memory; this was true
499 a long time ago when I wrote it, but is no longer the case.
500
501 The emulated memove() is provided for those environments that have neither
502 memmove() nor bcopy(). I didn't think anyone used it these days, but that
503 is clearly not the case, as these two bugs were recently reported.
504
505 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
506 and Detrail to create the HTML documentation, the .txt form of the man
507 pages, and it removes trailing spaces from listed files. It also creates
508 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
509 case, it wraps all the #defines with #ifndefs. This script should be run
510 before "make dist".
511
512 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
513 with Unicode property support.
514
515 (a) For a maximizing quantifier, if the two different cases of the
516 character were of different lengths in their UTF-8 codings (there are
517 some cases like this - I found 11), and the matching function had to
518 back up over a mixture of the two cases, it incorrectly assumed they
519 were both the same length.
520
521 (b) When PCRE was configured to use the heap rather than the stack for
522 recursion during matching, it was not correctly preserving the data for
523 the other case of a UTF-8 character when checking ahead for a match
524 while processing a minimizing repeat. If the check also involved
525 matching a wide character, but failed, corruption could cause an
526 erroneous result when trying to check for a repeat of the original
527 character.
528
529 13. Some tidying changes to the testing mechanism:
530
531 (a) The RunTest script now detects the internal link size and whether there
532 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
533 values substituted by "configure". (The RunGrepTest script already did
534 this for UTF-8.) The configure.ac script no longer substitutes the
535 relevant variables.
536
537 (b) The debugging options /B and /D in pcretest show the compiled bytecode
538 with length and offset values. This means that the output is different
539 for different internal link sizes. Test 2 is skipped for link sizes
540 other than 2 because of this, bypassing the problem. Unfortunately,
541 there was also a test in test 3 (the locale tests) that used /B and
542 failed for link sizes other than 2. Rather than cut the whole test out,
543 I have added a new /Z option to pcretest that replaces the length and
544 offset values with spaces. This is now used to make test 3 independent
545 of link size. (Test 2 will be tidied up later.)
546
547 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
548 segmentation fault instead of returning the appropriate error message.
549
550 15. In multiline mode when the newline sequence was set to "any", the pattern
551 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
552 This doesn't seem right; it now treats the CRLF combination as the line
553 ending, and so does not match in that case. It's only a pattern such as ^$
554 that would hit this one: something like ^ABC$ would have failed after \r
555 and then tried again after \r\n.
556
557 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
558 in an attempt to make files that differ only in their line terminators
559 compare equal. This works on Linux.
560
561 17. Under certain error circumstances pcregrep might try to free random memory
562 as it exited. This is now fixed, thanks to valgrind.
563
564 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
565 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
566 was because its rules for how to advance for /g after matching an empty
567 string at the end of a line did not allow for this case. They now check for
568 it specially.
569
570 20. pcretest is supposed to handle patterns and data of any length, by
571 extending its buffers when necessary. It was getting this wrong when the
572 buffer for a data line had to be extended.
573
574 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
575 CRLF as a newline sequence.
576
577 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
578 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
579 I have nevertheless tidied it up.
580
581 23. Added some casts to kill warnings from HP-UX ia64 compiler.
582
583 24. Added a man page for pcre-config.
584
585
586 Version 7.0 19-Dec-06
587 ---------------------
588
589 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
590 moving to gcc 4.1.1.
591
592 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
593 sys/time.h, which is documented as needed for this function. It doesn't
594 seem to matter on Linux, but it showed up on some releases of OS X.
595
596 3. It seems that there are systems where bytes whose values are greater than
597 127 match isprint() in the "C" locale. The "C" locale should be the
598 default when a C program starts up. In most systems, only ASCII printing
599 characters match isprint(). This difference caused the output from pcretest
600 to vary, making some of the tests fail. I have changed pcretest so that:
601
602 (a) When it is outputting text in the compiled version of a pattern, bytes
603 other than 32-126 are always shown as hex escapes.
604
605 (b) When it is outputting text that is a matched part of a subject string,
606 it does the same, unless a different locale has been set for the match
607 (using the /L modifier). In this case, it uses isprint() to decide.
608
609 4. Fixed a major bug that caused incorrect computation of the amount of memory
610 required for a compiled pattern when options that changed within the
611 pattern affected the logic of the preliminary scan that determines the
612 length. The relevant options are -x, and -i in UTF-8 mode. The result was
613 that the computed length was too small. The symptoms of this bug were
614 either the PCRE error "internal error: code overflow" from pcre_compile(),
615 or a glibc crash with a message such as "pcretest: free(): invalid next
616 size (fast)". Examples of patterns that provoked this bug (shown in
617 pcretest format) are:
618
619 /(?-x: )/x
620 /(?x)(?-x: \s*#\s*)/
621 /((?i)[\x{c0}])/8
622 /(?i:[\x{c0}])/8
623
624 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
625 is now done differently.
626
627 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
628 wrapper classes; (b) implement a new function in the C++ scanner that is
629 more efficient than the old way of doing things because it avoids levels of
630 recursion in the regex matching; (c) add a paragraph to the documentation
631 for the FullMatch() function.
632
633 6. The escape sequence \n was being treated as whatever was defined as
634 "newline". Not only was this contrary to the documentation, which states
635 that \n is character 10 (hex 0A), but it also went horribly wrong when
636 "newline" was defined as CRLF. This has been fixed.
637
638 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
639 was being set to -1 for the "end of line" case (supposedly a value that no
640 character can have). Though this value is never used (the check for end of
641 line is "zero bytes in current character"), it caused compiler complaints.
642 I've changed it to 0xffffffff.
643
644 8. In pcre_version.c, the version string was being built by a sequence of
645 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
646 string (as it is for production releases) called a macro with an empty
647 argument. The C standard says the result of this is undefined. The gcc
648 compiler treats it as an empty string (which was what was wanted) but it is
649 reported that Visual C gives an error. The source has been hacked around to
650 avoid this problem.
651
652 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
653 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
654 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
655 of them did).
656
657 10. Originally, pcretest opened its input and output without "b"; then I was
658 told that "b" was needed in some environments, so it was added for release
659 5.0 to both the input and output. (It makes no difference on Unix-like
660 systems.) Later I was told that it is wrong for the input on Windows. I've
661 now abstracted the modes into two macros, to make it easier to fiddle with
662 them, and removed "b" from the input mode under Windows.
663
664 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
665
666 12. Added -help and --help to pcretest as an official way of being reminded
667 of the options.
668
669 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
670 and pcrecpp.cc because they annoy compilers at high warning levels.
671
672 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
673
674 15. Fixed an occurrence of == in configure.ac that should have been = (shell
675 scripts are not C programs :-) and which was not noticed because it works
676 on Linux.
677
678 16. pcretest is supposed to handle any length of pattern and data line (as one
679 line or as a continued sequence of lines) by extending its input buffer if
680 necessary. This feature was broken for very long pattern lines, leading to
681 a string of junk being passed to pcre_compile() if the pattern was longer
682 than about 50K.
683
684 17. I have done a major re-factoring of the way pcre_compile() computes the
685 amount of memory needed for a compiled pattern. Previously, there was code
686 that made a preliminary scan of the pattern in order to do this. That was
687 OK when PCRE was new, but as the facilities have expanded, it has become
688 harder and harder to keep it in step with the real compile phase, and there
689 have been a number of bugs (see for example, 4 above). I have now found a
690 cunning way of running the real compile function in a "fake" mode that
691 enables it to compute how much memory it would need, while actually only
692 ever using a few hundred bytes of working memory and without too many
693 tests of the mode. This should make future maintenance and development
694 easier. A side effect of this work is that the limit of 200 on the nesting
695 depth of parentheses has been removed (though this was never a serious
696 limitation, I suspect). However, there is a downside: pcre_compile() now
697 runs more slowly than before (30% or more, depending on the pattern). I
698 hope this isn't a big issue. There is no effect on runtime performance.
699
700 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
701 newline (only possible for the last line of a file) and it was a
702 pattern that set a locale (followed by /Lsomething), pcretest crashed.
703
704 19. Added additional timing features to pcretest. (1) The -tm option now times
705 matching only, not compiling. (2) Both -t and -tm can be followed, as a
706 separate command line item, by a number that specifies the number of
707 repeats to use when timing. The default is 50000; this gives better
708 precision, but takes uncomfortably long for very large patterns.
709
710 20. Extended pcre_study() to be more clever in cases where a branch of a
711 subpattern has no definite first character. For example, (a*|b*)[cd] would
712 previously give no result from pcre_study(). Now it recognizes that the
713 first character must be a, b, c, or d.
714
715 21. There was an incorrect error "recursive call could loop indefinitely" if
716 a subpattern (or the entire pattern) that was being tested for matching an
717 empty string contained only one non-empty item after a nested subpattern.
718 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
719 incorrectly, because the \d was being skipped in the check.
720
721 22. The pcretest program now has a new pattern option /B and a command line
722 option -b, which is equivalent to adding /B to every pattern. This causes
723 it to show the compiled bytecode, without the additional information that
724 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
725 is the same as /B/I).
726
727 23. A new optimization is now able automatically to treat some sequences such
728 as a*b as a*+b. More specifically, if something simple (such as a character
729 or a simple class like \d) has an unlimited quantifier, and is followed by
730 something that cannot possibly match the quantified thing, the quantifier
731 is automatically "possessified".
732
733 24. A recursive reference to a subpattern whose number was greater than 39
734 went wrong under certain circumstances in UTF-8 mode. This bug could also
735 have affected the operation of pcre_study().
736
737 25. Realized that a little bit of performance could be had by replacing
738 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
739
740 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
741
742 27. Possessive quantifiers such as a++ were previously implemented by turning
743 them into atomic groups such as ($>a+). Now they have their own opcodes,
744 which improves performance. This includes the automatically created ones
745 from 23 above.
746
747 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
748 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
749 the first matched character to be a colon. This applied both to named and
750 numbered groups.
751
752 29. The ucpinternal.h header file was missing its idempotency #ifdef.
753
754 30. I was sent a "project" file called libpcre.a.dev which I understand makes
755 building PCRE on Windows easier, so I have included it in the distribution.
756
757 31. There is now a check in pcretest against a ridiculously large number being
758 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
759 loop, the loop is abandoned.
760
761 32. Forward references to subpatterns in conditions such as (?(2)...) where
762 subpattern 2 is defined later cause pcre_compile() to search forwards in
763 the pattern for the relevant set of parentheses. This search went wrong
764 when there were unescaped parentheses in a character class, parentheses
765 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
766
767 33. "Subroutine" calls and backreferences were previously restricted to
768 referencing subpatterns earlier in the regex. This restriction has now
769 been removed.
770
771 34. Added a number of extra features that are going to be in Perl 5.10. On the
772 whole, these are just syntactic alternatives for features that PCRE had
773 previously implemented using the Python syntax or my own invention. The
774 other formats are all retained for compatibility.
775
776 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
777 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
778 also .NET compatible.
779
780 (b) A recursion or subroutine call to a named group can now be defined as
781 (?&name) as well as (?P>name).
782
783 (c) A backreference to a named group can now be defined as \k<name> or
784 \k'name' as well as (?P=name). The new forms, as well as being in Perl
785 5.10, are also .NET compatible.
786
787 (d) A conditional reference to a named group can now use the syntax
788 (?(<name>) or (?('name') as well as (?(name).
789
790 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
791 groups (named and numbered) that are never evaluated inline, but can be
792 called as "subroutines" from elsewhere. In effect, the DEFINE condition
793 is always false. There may be only one alternative in such a group.
794
795 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
796 as the simple (?(R). The condition is true only if the most recent
797 recursion is that of the given number or name. It does not search out
798 through the entire recursion stack.
799
800 (g) The escape \gN or \g{N} has been added, where N is a positive or
801 negative number, specifying an absolute or relative reference.
802
803 35. Tidied to get rid of some further signed/unsigned compiler warnings and
804 some "unreachable code" warnings.
805
806 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
807 things, this adds five new scripts.
808
809 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
810 There were also incompatibilities regarding the handling of \Q..\E inside
811 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
812 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
813
814 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
815 matches an empty string, and forcibly breaks the loop. There were bugs in
816 this code in non-simple cases. For a pattern such as ^(a()*)* matched
817 against aaaa the result was just "a" rather than "aaaa", for example. Two
818 separate and independent bugs (that affected different cases) have been
819 fixed.
820
821 39. Refactored the code to abolish the use of different opcodes for small
822 capturing bracket numbers. This is a tidy that I avoided doing when I
823 removed the limit on the number of capturing brackets for 3.5 back in 2001.
824 The new approach is not only tidier, it makes it possible to reduce the
825 memory needed to fix the previous bug (38).
826
827 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
828 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
829 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
830 mode.
831
832 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
833 report.
834
835 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
836 copy construction and assignment in the C++ wrapper.
837
838 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
839 couple of bugs that could have given wrong results in the "--newline=crlf"
840 case.
841
842 44. Added a number of casts and did some reorganization of signed/unsigned int
843 variables following suggestions from Dair Grant. Also renamed the variable
844 "this" as "item" because it is a C++ keyword.
845
846 45. Arranged for dftables to add
847
848 #include "pcre_internal.h"
849
850 to pcre_chartables.c because without it, gcc 4.x may remove the array
851 definition from the final binary if PCRE is built into a static library and
852 dead code stripping is activated.
853
854 46. For an unanchored pattern, if a match attempt fails at the start of a
855 newline sequence, and the newline setting is CRLF or ANY, and the next two
856 characters are CRLF, advance by two characters instead of one.
857
858
859 Version 6.7 04-Jul-06
860 ---------------------
861
862 1. In order to handle tests when input lines are enormously long, pcretest has
863 been re-factored so that it automatically extends its buffers when
864 necessary. The code is crude, but this _is_ just a test program. The
865 default size has been increased from 32K to 50K.
866
867 2. The code in pcre_study() was using the value of the re argument before
868 testing it for NULL. (Of course, in any sensible call of the function, it
869 won't be NULL.)
870
871 3. The memmove() emulation function in pcre_internal.h, which is used on
872 systems that lack both memmove() and bcopy() - that is, hardly ever -
873 was missing a "static" storage class specifier.
874
875 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
876 containing an extended class (one that cannot be represented by a bitmap
877 because it contains high-valued characters or Unicode property items, e.g.
878 [\pZ]). Almost always one would set UTF-8 mode when processing such a
879 pattern, but PCRE should not loop if you do not (it no longer does).
880 [Detail: two cases were found: (a) a repeated subpattern containing an
881 extended class; (b) a recursive reference to a subpattern that followed a
882 previous extended class. It wasn't skipping over the extended class
883 correctly when UTF-8 mode was not set.]
884
885 5. A negated single-character class was not being recognized as fixed-length
886 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
887 compile error "lookbehind assertion is not fixed length".
888
889 6. The RunPerlTest auxiliary script was showing an unexpected difference
890 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
891 write a Perl script that can interpret lines of an input file either as
892 byte characters or as UTF-8, which is what "perltest" was being required to
893 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
894 can't do is switch easily at run time between having the "use utf8;" pragma
895 or not. In the end, I fudged it by using the RunPerlTest script to insert
896 "use utf8;" explicitly for the UTF-8 tests.
897
898 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
899 the end of the subject string, contrary to the documentation and to what
900 Perl does. This was true of both matching functions. Now it matches only at
901 the start of the subject and immediately after *internal* newlines.
902
903 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
904 a pointer to an int instead of a pointer to an unsigned long int. This
905 caused problems on 64-bit systems.
906
907 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
908 instance of the 'standard' template library not being so standard".
909
910 10. There was no check on the number of named subpatterns nor the maximum
911 length of a subpattern name. The product of these values is used to compute
912 the size of the memory block for a compiled pattern. By supplying a very
913 long subpattern name and a large number of named subpatterns, the size
914 computation could be caused to overflow. This is now prevented by limiting
915 the length of names to 32 characters, and the number of named subpatterns
916 to 10,000.
917
918 11. Subpatterns that are repeated with specific counts have to be replicated in
919 the compiled pattern. The size of memory for this was computed from the
920 length of the subpattern and the repeat count. The latter is limited to
921 65535, but there was no limit on the former, meaning that integer overflow
922 could in principle occur. The compiled length of a repeated subpattern is
923 now limited to 30,000 bytes in order to prevent this.
924
925 12. Added the optional facility to have named substrings with the same name.
926
927 13. Added the ability to use a named substring as a condition, using the
928 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
929 are numbers (not recommended). Forward references are permitted.
930
931 14. Added forward references in named backreferences (if you see what I mean).
932
933 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
934 pattern could run off the end of the subject. For example, the pattern
935 "(?s)(.{1,5})"8 did this with the subject "ab".
936
937 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
938 PCRE_CASELESS was set when matching characters that were quantified with ?
939 or *.
940
941 17. A character class other than a single negated character that had a minimum
942 but no maximum quantifier - for example [ab]{6,} - was not handled
943 correctly by pce_dfa_exec(). It would match only one character.
944
945 18. A valid (though odd) pattern that looked like a POSIX character
946 class but used an invalid character after [ (for example [[,abc,]]) caused
947 pcre_compile() to give the error "Failed: internal error: code overflow" or
948 in some cases to crash with a glibc free() error. This could even happen if
949 the pattern terminated after [[ but there just happened to be a sequence of
950 letters, a binary zero, and a closing ] in the memory that followed.
951
952 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
953 over the years. Originally (before any Unicode support), just the bottom 8
954 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
955 output from "man perlunicode" includes this:
956
957 The regular expression compiler produces polymorphic opcodes. That
958 is, the pattern adapts to the data and automatically switches to
959 the Unicode character scheme when presented with Unicode data--or
960 instead uses a traditional byte scheme when presented with byte
961 data.
962
963 Sadly, a wide octal escape does not cause a switch, and in a string with
964 no other multibyte characters, these octal escapes are treated as before.
965 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
966 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
967 Unicode string.
968
969 I have not perpetrated such confusion in PCRE. Up till now, it took just
970 the bottom 8 bits, as in old Perl. I have now made octal escapes with
971 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
972 translate to the appropriate multibyte character.
973
974 29. Applied some refactoring to reduce the number of warnings from Microsoft
975 and Borland compilers. This has included removing the fudge introduced
976 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
977 a warning about an unused variable.
978
979 21. PCRE has not included VT (character 0x0b) in the set of whitespace
980 characters since release 4.0, because Perl (from release 5.004) does not.
981 [Or at least, is documented not to: some releases seem to be in conflict
982 with the documentation.] However, when a pattern was studied with
983 pcre_study() and all its branches started with \s, PCRE still included VT
984 as a possible starting character. Of course, this did no harm; it just
985 caused an unnecessary match attempt.
986
987 22. Removed a now-redundant internal flag bit that recorded the fact that case
988 dependency changed within the pattern. This was once needed for "required
989 byte" processing, but is no longer used. This recovers a now-scarce options
990 bit. Also moved the least significant internal flag bit to the most-
991 significant bit of the word, which was not previously used (hangover from
992 the days when it was an int rather than a uint) to free up another bit for
993 the future.
994
995 23. Added support for CRLF line endings as well as CR and LF. As well as the
996 default being selectable at build time, it can now be changed at runtime
997 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
998 specify that it is scanning data with non-default line endings.
999
1000 24. Changed the definition of CXXLINK to make it agree with the definition of
1001 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1002
1003 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1004 recursions. This makes a big different to stack usage for some patterns.
1005
1006 26. If a subpattern containing a named recursion or subroutine reference such
1007 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1008 the space required for the compiled pattern went wrong and gave too small a
1009 value. Depending on the environment, this could lead to "Failed: internal
1010 error: code overflow at offset 49" or "glibc detected double free or
1011 corruption" errors.
1012
1013 27. Applied patches from Google (a) to support the new newline modes and (b) to
1014 advance over multibyte UTF-8 characters in GlobalReplace.
1015
1016 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1017 difference for some implementation of PCRE in some Windows version.
1018
1019 29. Added some extra testing facilities to pcretest:
1020
1021 \q<number> in a data line sets the "match limit" value
1022 \Q<number> in a data line sets the "match recursion limt" value
1023 -S <number> sets the stack size, where <number> is in megabytes
1024
1025 The -S option isn't available for Windows.
1026
1027
1028 Version 6.6 06-Feb-06
1029 ---------------------
1030
1031 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1032 in pcreposix.h. I have copied the definition from pcre.h.
1033
1034 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1035 because pcre.h is no longer a built file.
1036
1037 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1038 not normally included in the compiled code.
1039
1040
1041 Version 6.5 01-Feb-06
1042 ---------------------
1043
1044 1. When using the partial match feature with pcre_dfa_exec(), it was not
1045 anchoring the second and subsequent partial matches at the new starting
1046 point. This could lead to incorrect results. For example, with the pattern
1047 /1234/, partially matching against "123" and then "a4" gave a match.
1048
1049 2. Changes to pcregrep:
1050
1051 (a) All non-match returns from pcre_exec() were being treated as failures
1052 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1053 error message is output. Some extra information is given for the
1054 PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1055 probably the only errors that are likely to be caused by users (by
1056 specifying a regex that has nested indefinite repeats, for instance).
1057 If there are more than 20 of these errors, pcregrep is abandoned.
1058
1059 (b) A binary zero was treated as data while matching, but terminated the
1060 output line if it was written out. This has been fixed: binary zeroes
1061 are now no different to any other data bytes.
1062
1063 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1064 used to set a locale for matching. The --locale=xxxx long option has
1065 been added (no short equivalent) to specify a locale explicitly on the
1066 pcregrep command, overriding the environment variables.
1067
1068 (d) When -B was used with -n, some line numbers in the output were one less
1069 than they should have been.
1070
1071 (e) Added the -o (--only-matching) option.
1072
1073 (f) If -A or -C was used with -c (count only), some lines of context were
1074 accidentally printed for the final match.
1075
1076 (g) Added the -H (--with-filename) option.
1077
1078 (h) The combination of options -rh failed to suppress file names for files
1079 that were found from directory arguments.
1080
1081 (i) Added the -D (--devices) and -d (--directories) options.
1082
1083 (j) Added the -F (--fixed-strings) option.
1084
1085 (k) Allow "-" to be used as a file name for -f as well as for a data file.
1086
1087 (l) Added the --colo(u)r option.
1088
1089 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1090 is not present by default.
1091
1092 3. A nasty bug was discovered in the handling of recursive patterns, that is,
1093 items such as (?R) or (?1), when the recursion could match a number of
1094 alternatives. If it matched one of the alternatives, but subsequently,
1095 outside the recursion, there was a failure, the code tried to back up into
1096 the recursion. However, because of the way PCRE is implemented, this is not
1097 possible, and the result was an incorrect result from the match.
1098
1099 In order to prevent this happening, the specification of recursion has
1100 been changed so that all such subpatterns are automatically treated as
1101 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1102
1103 4. I had overlooked the fact that, in some locales, there are characters for
1104 which isalpha() is true but neither isupper() nor islower() are true. In
1105 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1106 and ordfeminine) are like this. This affected the treatment of \w and \W
1107 when they appeared in character classes, but not when they appeared outside
1108 a character class. The bit map for "word" characters is now created
1109 separately from the results of isalnum() instead of just taking it from the
1110 upper, lower, and digit maps. (Plus the underscore character, of course.)
1111
1112 5. The above bug also affected the handling of POSIX character classes such as
1113 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1114 permanent tables. Instead, the bit maps for such a class were previously
1115 created as the appropriate unions of the upper, lower, and digit bitmaps.
1116 Now they are created by subtraction from the [[:word:]] class, which has
1117 its own bitmap.
1118
1119 6. The [[:blank:]] character class matches horizontal, but not vertical space.
1120 It is created by subtracting the vertical space characters (\x09, \x0a,
1121 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1122 subtraction was done in the overall bitmap for a character class, meaning
1123 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1124 be recognized. This bug has been fixed.
1125
1126 7. Patches from the folks at Google:
1127
1128 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1129 real life, but is still worth protecting against".
1130
1131 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1132 regular expressions".
1133
1134 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1135 have it.
1136
1137 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1138 "configure" and the latter not, in order to fix a problem somebody had
1139 with compiling the Arg class on HP-UX.
1140
1141 (e) Improve the error-handling of the C++ wrapper a little bit.
1142
1143 (f) New tests for checking recursion limiting.
1144
1145 8. The pcre_memmove() function, which is used only if the environment does not
1146 have a standard memmove() function (and is therefore rarely compiled),
1147 contained two bugs: (a) use of int instead of size_t, and (b) it was not
1148 returning a result (though PCRE never actually uses the result).
1149
1150 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1151 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1152 returned instead of calling malloc() with an overflowing number that would
1153 most likely cause subsequent chaos.
1154
1155 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1156
1157 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1158 with this option is matched, the nmatch and pmatch options of regexec() are
1159 ignored.
1160
1161 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1162 provided in case anyone wants to the the POSIX interface with UTF-8
1163 strings.
1164
1165 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1166 C++ linking (needed for some HP-UX environments).
1167
1168 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1169 (unused parameter) and in the pcre_printint() function (omitted "default"
1170 switch label when the default is to do nothing).
1171
1172 15. Added some code to make it possible, when PCRE is compiled as a C++
1173 library, to replace subject pointers for pcre_exec() with a smart pointer
1174 class, thus making it possible to process discontinuous strings.
1175
1176 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1177 much the same function. They were added by different people who were trying
1178 to make PCRE easy to compile on non-Unix systems. It has been suggested
1179 that PCRE_EXPORT be abolished now that there is more automatic apparatus
1180 for compiling on Windows systems. I have therefore replaced it with
1181 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1182 defaults to "extern" for C or "extern C" for C++, which works fine on
1183 Unix-like systems. It is now possible to override the value of PCRE_DATA_
1184 SCOPE with something explicit in config.h. In addition:
1185
1186 (a) pcreposix.h still had just "extern" instead of either of these macros;
1187 I have replaced it with PCRE_DATA_SCOPE.
1188
1189 (b) Functions such as _pcre_xclass(), which are internal to the library,
1190 but external in the C sense, all had PCRE_EXPORT in their definitions.
1191 This is apparently wrong for the Windows case, so I have removed it.
1192 (It makes no difference on Unix-like systems.)
1193
1194 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1195 of recursive calls to match(). This is different to MATCH_LIMIT because
1196 that limits the total number of calls to match(), not all of which increase
1197 the depth of recursion. Limiting the recursion depth limits the amount of
1198 stack (or heap if NO_RECURSE is set) that is used. The default can be set
1199 when PCRE is compiled, and changed at run time. A patch from Google adds
1200 this functionality to the C++ interface.
1201
1202 18. Changes to the handling of Unicode character properties:
1203
1204 (a) Updated the table to Unicode 4.1.0.
1205
1206 (b) Recognize characters that are not in the table as "Cn" (undefined).
1207
1208 (c) I revised the way the table is implemented to a much improved format
1209 which includes recognition of ranges. It now supports the ranges that
1210 are defined in UnicodeData.txt, and it also amalgamates other
1211 characters into ranges. This has reduced the number of entries in the
1212 table from around 16,000 to around 3,000, thus reducing its size
1213 considerably. I realized I did not need to use a tree structure after
1214 all - a binary chop search is just as efficient. Having reduced the
1215 number of entries, I extended their size from 6 bytes to 8 bytes to
1216 allow for more data.
1217
1218 (d) Added support for Unicode script names via properties such as \p{Han}.
1219
1220 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1221 matching that character.
1222
1223 20. When matching a repeated Unicode property with a minimum greater than zero,
1224 (for example \pL{2,}), PCRE could look past the end of the subject if it
1225 reached it while seeking the minimum number of characters. This could
1226 happen only if some of the characters were more than one byte long, because
1227 there is a check for at least the minimum number of bytes.
1228
1229 21. Refactored the implementation of \p and \P so as to be more general, to
1230 allow for more different types of property in future. This has changed the
1231 compiled form incompatibly. Anybody with saved compiled patterns that use
1232 \p or \P will have to recompile them.
1233
1234 22. Added "Any" and "L&" to the supported property types.
1235
1236 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1237 but give a compile time error if the value is greater than 0xff.
1238
1239 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1240 accidentally not being installed or uninstalled.
1241
1242 25. The pcre.h file was built from pcre.h.in, but the only changes that were
1243 made were to insert the current release number. This seemed silly, because
1244 it made things harder for people building PCRE on systems that don't run
1245 "configure". I have turned pcre.h into a distributed file, no longer built
1246 by "configure", with the version identification directly included. There is
1247 no longer a pcre.h.in file.
1248
1249 However, this change necessitated a change to the pcre-config script as
1250 well. It is built from pcre-config.in, and one of the substitutions was the
1251 release number. I have updated configure.ac so that ./configure now finds
1252 the release number by grepping pcre.h.
1253
1254 26. Added the ability to run the tests under valgrind.
1255
1256
1257 Version 6.4 05-Sep-05
1258 ---------------------
1259
1260 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1261 "--" to be printed when multiple files were scanned, even when none of the
1262 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1263 consider it to be a bug, and have restored the previous behaviour.
1264
1265 2. A couple of code tidies to get rid of compiler warnings.
1266
1267 3. The pcretest program used to cheat by referring to symbols in the library
1268 whose names begin with _pcre_. These are internal symbols that are not
1269 really supposed to be visible externally, and in some environments it is
1270 possible to suppress them. The cheating is now confined to including
1271 certain files from the library's source, which is a bit cleaner.
1272
1273 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1274 file's purpose clearer.
1275
1276 5. Reorganized pcre_ucp_findchar().
1277
1278
1279 Version 6.3 15-Aug-05
1280 ---------------------
1281
1282 1. The file libpcre.pc.in did not have general read permission in the tarball.
1283
1284 2. There were some problems when building without C++ support:
1285
1286 (a) If C++ support was not built, "make install" and "make test" still
1287 tried to test it.
1288
1289 (b) There were problems when the value of CXX was explicitly set. Some
1290 changes have been made to try to fix these, and ...
1291
1292 (c) --disable-cpp can now be used to explicitly disable C++ support.
1293
1294 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1295 backslash in a target when C++ was disabled. This confuses some
1296 versions of "make", apparently. Using an intermediate variable solves
1297 this. (Same for CPP_LOBJ.)
1298
1299 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1300 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1301 necessary on certain architectures.
1302
1303 4. Added a setting of -export-symbols-regex to the link command to remove
1304 those symbols that are exported in the C sense, but actually are local
1305 within the library, and not documented. Their names all begin with
1306 "_pcre_". This is not a perfect job, because (a) we have to except some
1307 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1308 available (and never for static libraries). I have made a note to try to
1309 find a way round (a) in the future.
1310
1311
1312 Version 6.2 01-Aug-05
1313 ---------------------
1314
1315 1. There was no test for integer overflow of quantifier values. A construction
1316 such as {1111111111111111} would give undefined results. What is worse, if
1317 a minimum quantifier for a parenthesized subpattern overflowed and became
1318 negative, the calculation of the memory size went wrong. This could have
1319 led to memory overwriting.
1320
1321 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1322
1323 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1324 operating environments where this matters.
1325
1326 4. Applied Giuseppe Maxia's patch to add additional features for controlling
1327 PCRE options from within the C++ wrapper.
1328
1329 5. Named capturing subpatterns were not being correctly counted when a pattern
1330 was compiled. This caused two problems: (a) If there were more than 100
1331 such subpatterns, the calculation of the memory needed for the whole
1332 compiled pattern went wrong, leading to an overflow error. (b) Numerical
1333 back references of the form \12, where the number was greater than 9, were
1334 not recognized as back references, even though there were sufficient
1335 previous subpatterns.
1336
1337 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1338 versions of gcc, e.g. 2.95.4.
1339
1340
1341 Version 6.1 21-Jun-05
1342 ---------------------
1343
1344 1. There was one reference to the variable "posix" in pcretest.c that was not
1345 surrounded by "#if !defined NOPOSIX".
1346
1347 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1348 the cross-check on the old pcre_info() function, for the benefit of the
1349 cut-down version of PCRE that is currently imported into Exim.
1350
1351 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1352 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1353 patterns that start (?i) but I don't think that matters. The use of (?i) is
1354 just an example; this all applies to the other options as well.
1355
1356 4. Since libtool seems to echo the compile commands it is issuing, the output
1357 from "make" can be reduced a bit by putting "@" in front of each libtool
1358 compile command.
1359
1360 5. Patch from the folks at Google for configure.in to be a bit more thorough
1361 in checking for a suitable C++ installation before trying to compile the
1362 C++ stuff. This should fix a reported problem when a compiler was present,
1363 but no suitable headers.
1364
1365 6. The man pages all had just "PCRE" as their title. I have changed them to
1366 be the relevant file name. I have also arranged that these names are
1367 retained in the file doc/pcre.txt, which is a concatenation in text format
1368 of all the man pages except the little individual ones for each function.
1369
1370 7. The NON-UNIX-USE file had not been updated for the different set of source
1371 files that come with release 6. I also added a few comments about the C++
1372 wrapper.
1373
1374
1375 Version 6.0 07-Jun-05
1376 ---------------------
1377
1378 1. Some minor internal re-organization to help with my DFA experiments.
1379
1380 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1381 didn't matter for the library itself when fully configured, but did matter
1382 when compiling without UCP support, or within Exim, where the ucp files are
1383 not imported.
1384
1385 3. Refactoring of the library code to split up the various functions into
1386 different source modules. The addition of the new DFA matching code (see
1387 below) to a single monolithic source would have made it really too
1388 unwieldy, quite apart from causing all the code to be include in a
1389 statically linked application, when only some functions are used. This is
1390 relevant even without the DFA addition now that patterns can be compiled in
1391 one application and matched in another.
1392
1393 The downside of splitting up is that there have to be some external
1394 functions and data tables that are used internally in different modules of
1395 the library but which are not part of the API. These have all had their
1396 names changed to start with "_pcre_" so that they are unlikely to clash
1397 with other external names.
1398
1399 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1400 a different (DFA) algorithm. Although it is slower than the original
1401 function, it does have some advantages for certain types of matching
1402 problem.
1403
1404 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1405 including restarting after a partial match.
1406
1407 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1408 defined when compiling for Windows was sent to me. I have put it into the
1409 code, though I have no means of testing or verifying it.
1410
1411 7. Added the pcre_refcount() auxiliary function.
1412
1413 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1414 match before or at the first newline in the subject string. In pcretest,
1415 the /f option on a pattern can be used to set this.
1416
1417 9. A repeated \w when used in UTF-8 mode with characters greater than 256
1418 would behave wrongly. This has been present in PCRE since release 4.0.
1419
1420 10. A number of changes to the pcregrep command:
1421
1422 (a) Refactored how -x works; insert ^(...)$ instead of setting
1423 PCRE_ANCHORED and checking the length, in preparation for adding
1424 something similar for -w.
1425
1426 (b) Added the -w (match as a word) option.
1427
1428 (c) Refactored the way lines are read and buffered so as to have more
1429 than one at a time available.
1430
1431 (d) Implemented a pcregrep test script.
1432
1433 (e) Added the -M (multiline match) option. This allows patterns to match
1434 over several lines of the subject. The buffering ensures that at least
1435 8K, or the rest of the document (whichever is the shorter) is available
1436 for matching (and similarly the previous 8K for lookbehind assertions).
1437
1438 (f) Changed the --help output so that it now says
1439
1440 -w, --word-regex(p)
1441
1442 instead of two lines, one with "regex" and the other with "regexp"
1443 because that confused at least one person since the short forms are the
1444 same. (This required a bit of code, as the output is generated
1445 automatically from a table. It wasn't just a text change.)
1446
1447 (g) -- can be used to terminate pcregrep options if the next thing isn't an
1448 option but starts with a hyphen. Could be a pattern or a path name
1449 starting with a hyphen, for instance.
1450
1451 (h) "-" can be given as a file name to represent stdin.
1452
1453 (i) When file names are being printed, "(standard input)" is used for
1454 the standard input, for compatibility with GNU grep. Previously
1455 "<stdin>" was used.
1456
1457 (j) The option --label=xxx can be used to supply a name to be used for
1458 stdin when file names are being printed. There is no short form.
1459
1460 (k) Re-factored the options decoding logic because we are going to add
1461 two more options that take data. Such options can now be given in four
1462 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1463
1464 (l) Added the -A, -B, and -C options for requesting that lines of context
1465 around matches be printed.
1466
1467 (m) Added the -L option to print the names of files that do not contain
1468 any matching lines, that is, the complement of -l.
1469
1470 (n) The return code is 2 if any file cannot be opened, but pcregrep does
1471 continue to scan other files.
1472
1473 (o) The -s option was incorrectly implemented. For compatibility with other
1474 greps, it now suppresses the error message for a non-existent or non-
1475 accessible file (but not the return code). There is a new option called
1476 -q that suppresses the output of matching lines, which was what -s was
1477 previously doing.
1478
1479 (p) Added --include and --exclude options to specify files for inclusion
1480 and exclusion when recursing.
1481
1482 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1483 Hopefully, it now does.
1484
1485 12. Missing cast in pcre_study().
1486
1487 13. Added an "uninstall" target to the makefile.
1488
1489 14. Replaced "extern" in the function prototypes in Makefile.in with
1490 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1491 world, but is set differently for Windows.
1492
1493 15. Added a second compiling function called pcre_compile2(). The only
1494 difference is that it has an extra argument, which is a pointer to an
1495 integer error code. When there is a compile-time failure, this is set
1496 non-zero, in addition to the error test pointer being set to point to an
1497 error message. The new argument may be NULL if no error number is required
1498 (but then you may as well call pcre_compile(), which is now just a
1499 wrapper). This facility is provided because some applications need a
1500 numeric error indication, but it has also enabled me to tidy up the way
1501 compile-time errors are handled in the POSIX wrapper.
1502
1503 16. Added VPATH=.libs to the makefile; this should help when building with one
1504 prefix path and installing with another. (Or so I'm told by someone who
1505 knows more about this stuff than I do.)
1506
1507 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1508 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1509 match everything, including newlines. This is not POSIX-compatible, but
1510 somebody wanted the feature. From pcretest it can be activated by using
1511 both the P and the s flags.
1512
1513 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1514
1515 19. libpcre.pc was being incorrectly installed as executable.
1516
1517 20. A couple of places in pcretest check for end-of-line by looking for '\n';
1518 it now also looks for '\r' so that it will work unmodified on Windows.
1519
1520 21. Added Google's contributed C++ wrapper to the distribution.
1521
1522 22. Added some untidy missing memory free() calls in pcretest, to keep
1523 Electric Fence happy when testing.
1524
1525
1526
1527 Version 5.0 13-Sep-04
1528 ---------------------
1529
1530 1. Internal change: literal characters are no longer packed up into items
1531 containing multiple characters in a single byte-string. Each character
1532 is now matched using a separate opcode. However, there may be more than one
1533 byte in the character in UTF-8 mode.
1534
1535 2. The pcre_callout_block structure has two new fields: pattern_position and
1536 next_item_length. These contain the offset in the pattern to the next match
1537 item, and its length, respectively.
1538
1539 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1540 insertion of callouts before each pattern item. Added the /C option to
1541 pcretest to make use of this.
1542
1543 4. On the advice of a Windows user, the lines
1544
1545 #if defined(_WIN32) || defined(WIN32)
1546 _setmode( _fileno( stdout ), 0x8000 );
1547 #endif /* defined(_WIN32) || defined(WIN32) */
1548
1549 have been added to the source of pcretest. This apparently does useful
1550 magic in relation to line terminators.
1551
1552 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1553 for the benefit of those environments where the "b" makes a difference.
1554
1555 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1556 to know about it. I have put a hack into configure.in that adds in code
1557 to set GCC=yes if CC=icc. This seems to end up at a point in the
1558 generated configure script that is early enough to affect the setting of
1559 compiler options, which is what is needed, but I have no means of testing
1560 whether it really works. (The user who reported this had patched the
1561 generated configure script, which of course I cannot do.)
1562
1563 LATER: After change 22 below (new libtool files), the configure script
1564 seems to know about icc (and also ecc). Therefore, I have commented out
1565 this hack in configure.in.
1566
1567 7. Added support for pkg-config (2 patches were sent in).
1568
1569 8. Negated POSIX character classes that used a combination of internal tables
1570 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1571 [[:^ascii]]. Typically, they would match almost any characters. The other
1572 POSIX classes were not broken in this way.
1573
1574 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1575 to find the match, as PCRE was deluded into thinking that the match had to
1576 start at the start point or following a newline. The same bug applied to
1577 patterns with negative forward assertions or any backward assertions
1578 preceding ".*" at the start, unless the pattern required a fixed first
1579 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1580
1581 10. In UTF-8 mode, when moving forwards in the subject after a failed match
1582 starting at the last subject character, bytes beyond the end of the subject
1583 string were read.
1584
1585 11. Renamed the variable "class" as "classbits" to make life easier for C++
1586 users. (Previously there was a macro definition, but it apparently wasn't
1587 enough.)
1588
1589 12. Added the new field "tables" to the extra data so that tables can be passed
1590 in at exec time, or the internal tables can be re-selected. This allows
1591 a compiled regex to be saved and re-used at a later time by a different
1592 program that might have everything at different addresses.
1593
1594 13. Modified the pcre-config script so that, when run on Solaris, it shows a
1595 -R library as well as a -L library.
1596
1597 14. The debugging options of pcretest (-d on the command line or D on a
1598 pattern) showed incorrect output for anything following an extended class
1599 that contained multibyte characters and which was followed by a quantifier.
1600
1601 15. Added optional support for general category Unicode character properties
1602 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1603 support. It adds about 90K to the size of the library. The meanings of the
1604 inbuilt class escapes such as \d and \s have NOT been changed.
1605
1606 16. Updated pcredemo.c to include calls to free() to release the memory for the
1607 compiled pattern.
1608
1609 17. The generated file chartables.c was being created in the source directory
1610 instead of in the building directory. This caused the build to fail if the
1611 source directory was different from the building directory, and was
1612 read-only.
1613
1614 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1615 file. No doubt somebody will tell me if they don't make sense... Also added
1616 Dan Mooney's comments about building on OpenVMS.
1617
1618 19. Added support for partial matching via the PCRE_PARTIAL option for
1619 pcre_exec() and the \P data escape in pcretest.
1620
1621 20. Extended pcretest with 3 new pattern features:
1622
1623 (i) A pattern option of the form ">rest-of-line" causes pcretest to
1624 write the compiled pattern to the file whose name is "rest-of-line".
1625 This is a straight binary dump of the data, with the saved pointer to
1626 the character tables forced to be NULL. The study data, if any, is
1627 written too. After writing, pcretest reads a new pattern.
1628
1629 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1630 compiled pattern from the given file. There must not be any
1631 occurrences of "<" in the file name (pretty unlikely); if there are,
1632 pcretest will instead treat the initial "<" as a pattern delimiter.
1633 After reading in the pattern, pcretest goes on to read data lines as
1634 usual.
1635
1636 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1637 and 16-bit fields in a compiled pattern, to simulate a pattern that
1638 was compiled on a host of opposite endianness.
1639
1640 21. The pcre-exec() function can now cope with patterns that were compiled on
1641 hosts of opposite endianness, with this restriction:
1642
1643 As for any compiled expression that is saved and used later, the tables
1644 pointer field cannot be preserved; the extra_data field in the arguments
1645 to pcre_exec() should be used to pass in a tables address if a value
1646 other than the default internal tables were used at compile time.
1647
1648 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1649 now diagnosed as an error. Previously, most of the time, a negative number
1650 would have been treated as zero, but if in addition "ovector" was passed as
1651 NULL, a crash could occur.
1652
1653 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1654 new versions from the libtool 1.5 distribution (the last one is a copy of
1655 a file called libtool.m4). This seems to have fixed the need to patch
1656 "configure" to support Darwin 1.3 (which I used to do). However, I still
1657 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1658 workstation).
1659
1660 24. Changed the PCRE licence to be the more standard "BSD" licence.
1661
1662
1663 Version 4.5 01-Dec-03
1664 ---------------------
1665
1666 1. There has been some re-arrangement of the code for the match() function so
1667 that it can be compiled in a version that does not call itself recursively.
1668 Instead, it keeps those local variables that need separate instances for
1669 each "recursion" in a frame on the heap, and gets/frees frames whenever it
1670 needs to "recurse". Keeping track of where control must go is done by means
1671 of setjmp/longjmp. The whole thing is implemented by a set of macros that
1672 hide most of the details from the main code, and operates only if
1673 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1674 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1675 operating.
1676
1677 To make it easier for callers to provide specially tailored get/free
1678 functions for this usage, two new functions, pcre_stack_malloc, and
1679 pcre_stack_free, are used. They are always called in strict stacking order,
1680 and the size of block requested is always the same.
1681
1682 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1683 PCRE has been compiled to use the stack or the heap for recursion. The
1684 -C option of pcretest uses this to show which version is compiled.
1685
1686 A new data escape \S, is added to pcretest; it causes the amounts of store
1687 obtained and freed by both kinds of malloc/free at match time to be added
1688 to the output.
1689
1690 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1691 what's available on my current Linux desktop machine.
1692
1693 3. When matching a UTF-8 string, the test for a valid string at the start has
1694 been extended. If start_offset is not zero, PCRE now checks that it points
1695 to a byte that is the start of a UTF-8 character. If not, it returns
1696 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1697 this is necessary because there may be backward assertions in the pattern.
1698 When matching the same subject several times, it may save resources to use
1699 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1700
1701 4. The code for checking the validity of UTF-8 strings has been tightened so
1702 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1703 containing "overlong sequences".
1704
1705 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1706 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1707 should have been "&", but it just so happened that all the cases this let
1708 through by mistake were picked up later in the function.
1709
1710 6. I had used a variable called "isblank" - this is a C99 function, causing
1711 some compilers to warn. To avoid this, I renamed it (as "blankclass").
1712
1713 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1714 prompting; (b) run "./pcretest /dev/null" at the start of the test script
1715 so the version is shown; (c) stop "make test" echoing "./RunTest".
1716
1717 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1718
1719 9. The prototype for memmove() for systems that don't have it was using
1720 size_t, but the inclusion of the header that defines size_t was later. I've
1721 moved the #includes for the C headers earlier to avoid this.
1722
1723 10. Added some adjustments to the code to make it easier to compiler on certain
1724 special systems:
1725
1726 (a) Some "const" qualifiers were missing.
1727 (b) Added the macro EXPORT before all exported functions; by default this
1728 is defined to be empty.
1729 (c) Changed the dftables auxiliary program (that builds chartables.c) so
1730 that it reads its output file name as an argument instead of writing
1731 to the standard output and assuming this can be redirected.
1732
1733 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1734 class containing characters with values greater than 255, PCRE compilation
1735 went into a loop.
1736
1737 12. A recursive reference to a subpattern that was within another subpattern
1738 that had a minimum quantifier of zero caused PCRE to crash. For example,
1739 (x(y(?2))z)? provoked this bug with a subject that got as far as the
1740 recursion. If the recursively-called subpattern itself had a zero repeat,
1741 that was OK.
1742
1743 13. In pcretest, the buffer for reading a data line was set at 30K, but the
1744 buffer into which it was copied (for escape processing) was still set at
1745 1024, so long lines caused crashes.
1746
1747 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1748 "internal error: code overflow...". This applied to any character class
1749 that was followed by a possessive quantifier.
1750
1751 15. Modified the Makefile to add libpcre.la as a prerequisite for
1752 libpcreposix.la because I was told this is needed for a parallel build to
1753 work.
1754
1755 16. If a pattern that contained .* following optional items at the start was
1756 studied, the wrong optimizing data was generated, leading to matching
1757 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1758 matching string must start with a or b or c. The correct conclusion for
1759 this pattern is that a match can start with any character.
1760
1761
1762 Version 4.4 13-Aug-03
1763 ---------------------
1764
1765 1. In UTF-8 mode, a character class containing characters with values between
1766 127 and 255 was not handled correctly if the compiled pattern was studied.
1767 In fixing this, I have also improved the studying algorithm for such
1768 classes (slightly).
1769
1770 2. Three internal functions had redundant arguments passed to them. Removal
1771 might give a very teeny performance improvement.
1772
1773 3. Documentation bug: the value of the capture_top field in a callout is *one
1774 more than* the number of the hightest numbered captured substring.
1775
1776 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1777 in incorrectly linking with a previously installed version. They now link
1778 explicitly with libpcre.la.
1779
1780 5. configure.in no longer needs to recognize Cygwin specially.
1781
1782 6. A problem in pcre.in for Windows platforms is fixed.
1783
1784 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1785 pcretest, it used to include the size of the study block as part of its
1786 output. Unfortunately, the structure contains a field that has a different
1787 size on different hardware architectures. This meant that the tests that
1788 showed this size failed. As the block is currently always of a fixed size,
1789 this information isn't actually particularly useful in pcretest output, so
1790 I have just removed it.
1791
1792 8. Three pre-processor statements accidentally did not start in column 1.
1793 Sadly, there are *still* compilers around that complain, even though
1794 standard C has not required this for well over a decade. Sigh.
1795
1796 9. In pcretest, the code for checking callouts passed small integers in the
1797 callout_data field, which is a void * field. However, some picky compilers
1798 complained about the casts involved for this on 64-bit systems. Now
1799 pcretest passes the address of the small integer instead, which should get
1800 rid of the warnings.
1801
1802 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1803 both compile and run time, and gives an error if an invalid UTF-8 sequence
1804 is found. There is a option for disabling this check in cases where the
1805 string is known to be correct and/or the maximum performance is wanted.
1806
1807 11. In response to a bug report, I changed one line in Makefile.in from
1808
1809 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1810 to
1811 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1812
1813 to look similar to other lines, but I have no way of telling whether this
1814 is the right thing to do, as I do not use Windows. No doubt I'll get told
1815 if it's wrong...
1816
1817
1818 Version 4.3 21-May-03
1819 ---------------------
1820
1821 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1822 Makefile.
1823
1824 2. Some refactoring to improve the quality of the code:
1825
1826 (i) The utf8_table... variables are now declared "const".
1827
1828 (ii) The code for \cx, which used the "case flipping" table to upper case
1829 lower case letters, now just substracts 32. This is ASCII-specific,
1830 but the whole concept of \cx is ASCII-specific, so it seems
1831 reasonable.
1832
1833 (iii) PCRE was using its character types table to recognize decimal and
1834 hexadecimal digits in the pattern. This is silly, because it handles
1835 only 0-9, a-f, and A-F, but the character types table is locale-
1836 specific, which means strange things might happen. A private
1837 table is now used for this - though it costs 256 bytes, a table is
1838 much faster than multiple explicit tests. Of course, the standard
1839 character types table is still used for matching digits in subject
1840 strings against \d.
1841
1842 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1843 ending in _t are). So I've renamed it as ESC_tee.
1844
1845 3. The first argument for regexec() in the POSIX wrapper should have been
1846 defined as "const".
1847
1848 4. Changed pcretest to use malloc() for its buffers so that they can be
1849 Electric Fenced for debugging.
1850
1851 5. There were several places in the code where, in UTF-8 mode, PCRE would try
1852 to read one or more bytes before the start of the subject string. Often this
1853 had no effect on PCRE's behaviour, but in some circumstances it could
1854 provoke a segmentation fault.
1855
1856 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1857 to try to read one or more bytes before the start of the subject string.
1858
1859 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1860 UTF-8 support could misbehave in various ways if the subject string
1861 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1862 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1863 back over UTF-8 characters.)
1864
1865
1866 Version 4.2 14-Apr-03
1867 ---------------------
1868
1869 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1870
1871 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1872 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1873 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1874 [WIN_PREFIX]: new variable, "cyg" for Cygwin
1875 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1876 and BUILD_EXEEXT
1877 Note: automatic setting of the BUILD variables is not yet working
1878 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1879 compile-time but not at link-time
1880 [LINK]: use for linking executables only
1881 make different versions for Windows and non-Windows
1882 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1883 libraries
1884 [LINK_FOR_BUILD]: new variable
1885 [OBJEXT]: use throughout
1886 [EXEEXT]: use throughout
1887 <winshared>: new target
1888 <wininstall>: new target
1889 <dftables.o>: use native compiler
1890 <dftables>: use native linker
1891 <install>: handle Windows platform correctly
1892 <clean>: ditto
1893 <check>: ditto
1894 copy DLL to top builddir before testing
1895
1896 As part of these changes, -no-undefined was removed again. This was reported
1897 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1898 in any case.
1899
1900 3. Some tidies to get rid of compiler warnings:
1901
1902 . In the match_data structure, match_limit was an unsigned long int, whereas
1903 match_call_count was an int. I've made them both unsigned long ints.
1904
1905 . In pcretest the fact that a const uschar * doesn't automatically cast to
1906 a void * provoked a warning.
1907
1908 . Turning on some more compiler warnings threw up some "shadow" variables
1909 and a few more missing casts.
1910
1911 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1912 option, a class that contained a single character with a value between 128
1913 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1914
1915 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1916 option, a class that contained several characters, but with at least one
1917 whose value was between 128 and 255 caused PCRE to crash.
1918
1919
1920 Version 4.1 12-Mar-03
1921 ---------------------
1922
1923 1. Compiling with gcc -pedantic found a couple of places where casts were
1924 needed, and a string in dftables.c that was longer than standard compilers are
1925 required to support.
1926
1927 2. Compiling with Sun's compiler found a few more places where the code could
1928 be tidied up in order to avoid warnings.
1929
1930 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1931 first of these names is deprecated in the latest Autoconf in favour of the name
1932 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1933 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1934 analogy I have changed it to CFLAGS_FOR_BUILD.
1935
1936 4. Added -no-undefined to the linking command in the Makefile, because this is
1937 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1938 linking step for the pcreposix library.
1939
1940 5. PCRE was failing to diagnose the case of two named groups with the same
1941 name.
1942
1943 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1944 literal character that is needed in the subject for a match, and scans along to
1945 ensure that it is present before embarking on the full matching process. This
1946 saves time in cases of nested unlimited repeats that are never going to match.
1947 Problem: the scan can take a lot of time if the subject is very long (e.g.
1948 megabytes), thus penalizing straightforward matches. It is now done only if the
1949 amount of subject to be scanned is less than 1000 bytes.
1950
1951 7. A lesser problem with the same optimization is that it was recording the
1952 first character of an anchored pattern as "needed", thus provoking a search
1953 right along the subject, even when the first match of the pattern was going to
1954 fail. The "needed" character is now not set for anchored patterns, unless it
1955 follows something in the pattern that is of non-fixed length. Thus, it still
1956 fulfils its original purpose of finding quick non-matches in cases of nested
1957 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1958
1959
1960 Version 4.0 17-Feb-03
1961 ---------------------
1962
1963 1. If a comment in an extended regex that started immediately after a meta-item
1964 extended to the end of string, PCRE compiled incorrect data. This could lead to
1965 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1966
1967 2. Moved to autoconf 2.53 and libtool 1.4.2.
1968
1969 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1970 the special perltest8 script is no longer needed - all the tests can be run
1971 from a single perltest script.
1972
1973 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1974 by \s. It has now been removed in PCRE. This means it isn't recognized as
1975 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1976 class [:space:] *does* include VT, thereby creating a mess.
1977
1978 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1979 space and tab.
1980
1981 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1982 its new features into the main test script, reducing the number of scripts.
1983
1984 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1985 were backward compatible, and made the (?i) apply to the whole pattern, as if
1986 /i were given. Now it behaves more logically, and applies the option setting
1987 only to what follows. PCRE has been changed to follow suit. However, if it
1988 finds options settings right at the start of the pattern, it extracts them into
1989 the global options, as before. Thus, they show up in the info data.
1990
1991 8. Added support for the \Q...\E escape sequence. Characters in between are
1992 treated as literals. This is slightly different from Perl in that $ and @ are
1993 also handled as literals inside the quotes. In Perl, they will cause variable
1994 interpolation. Note the following examples:
1995
1996 Pattern PCRE matches Perl matches
1997
1998 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
1999 \Qabc\$xyz\E abc\$xyz abc\$xyz
2000 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
2001
2002 For compatibility with Perl, \Q...\E sequences are recognized inside character
2003 classes as well as outside them.
2004
2005 9. Re-organized 3 code statements in pcretest to avoid "overflow in
2006 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2007 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2008 signed/unsigned warnings.
2009
2010 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2011 option for pcretest, so I've replaced it by a simple function that does just
2012 that job.
2013
2014 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2015 "pcregrep -".
2016
2017 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2018 Java package. This provides some syntactic sugar for simple cases of what my
2019 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2020 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2021 item, you can use this simplified notation. Note that only makes sense with
2022 greedy quantifiers. Consequently, the use of the possessive quantifier forces
2023 greediness, whatever the setting of the PCRE_UNGREEDY option.
2024
2025 13. A change of greediness default within a pattern was not taking effect at
2026 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2027 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2028 was abstracted outside.
2029
2030 14. PCRE now supports the \G assertion. It is true when the current matching
2031 position is at the start point of the match. This differs from \A when the
2032 starting offset is non-zero. Used with the /g option of pcretest (or similar
2033 code), it works in the same way as it does for Perl's /g option. If all
2034 alternatives of a regex begin with \G, the expression is anchored to the start
2035 match position, and the "anchored" flag is set in the compiled expression.
2036
2037 15. Some bugs concerning the handling of certain option changes within patterns
2038 have been fixed. These applied to options other than (?ims). For example,
2039 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2040 been the other way round. Some of this was related to change 7 above.
2041
2042 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2043 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2044 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2045 POSIX classes only within a class (e.g. /[[:alpha:]]/).
2046
2047 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2048 mode. Unlike ".", it always matches newline, whatever the setting of
2049 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2050 assertions. Perl allows it, but it doesn't (in general) work because it can't
2051 calculate the length of the lookbehind. At least, that's the case for Perl
2052 5.8.0 - I've been told they are going to document that it doesn't work in
2053 future.
2054
2055 18. Added an error diagnosis for escapes that PCRE does not support: these are
2056 \L, \l, \N, \P, \p, \U, \u, and \X.
2057
2058 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2059 reading past the end of the pattern in cases such as /[abcd/.
2060
2061 20. PCRE was getting more memory than necessary for patterns with classes that
2062 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2063
2064 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2065 compiling PCRE for use with Virtual Pascal.
2066
2067 22. Small fix to the Makefile to make it work properly if the build is done
2068 outside the source tree.
2069
2070 23. Added a new extension: a condition to go with recursion. If a conditional
2071 subpattern starts with (?(R) the "true" branch is used if recursion has
2072 happened, whereas the "false" branch is used only at the top level.
2073
2074 24. When there was a very long string of literal characters (over 255 bytes
2075 without UTF support, over 250 bytes with UTF support), the computation of how
2076 much memory was required could be incorrect, leading to segfaults or other
2077 strange effects.
2078
2079 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2080 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2081 there was a subsequent back reference to those brackets. This meant that, for
2082 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2083 possible to check for precisely this case. All we can do is abandon the
2084 optimization if .* occurs inside capturing brackets when there are any back
2085 references whatsoever. (See below for a better fix that came later.)
2086
2087 26. The handling of the optimization for finding the first character of a
2088 non-anchored pattern, and for finding a character that is required later in the
2089 match were failing in some cases. This didn't break the matching; it just
2090 failed to optimize when it could. The way this is done has been re-implemented.
2091
2092 27. Fixed typo in error message for invalid (?R item (it said "(?p").
2093
2094 28. Added a new feature that provides some of the functionality that Perl
2095 provides with (?{...}). The facility is termed a "callout". The way it is done
2096 in PCRE is for the caller to provide an optional function, by setting
2097 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2098 global variable. By default it is unset, which disables all calling out. To get
2099 the function called, the regex must include (?C) at appropriate points. This
2100 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2101 This provides a means of identifying different callout points. When PCRE
2102 reaches such a point in the regex, if pcre_callout has been set, the external
2103 function is called. It is provided with data in a structure called
2104 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2105 matching continues; if it returns a non-zero value, the match at the current
2106 point fails. However, backtracking will occur if possible. [This was changed
2107 later and other features added - see item 49 below.]
2108
2109 29. pcretest is upgraded to test the callout functionality. It provides a
2110 callout function that displays information. By default, it shows the start of
2111 the match and the current position in the text. There are some new data escapes
2112 to vary what happens:
2113
2114 \C+ in addition, show current contents of captured substrings
2115 \C- do not supply a callout function
2116 \C!n return 1 when callout number n is reached
2117 \C!n!m return 1 when callout number n is reached for the mth time
2118
2119 30. If pcregrep was called with the -l option and just a single file name, it
2120 output "<stdin>" if a match was found, instead of the file name.
2121
2122 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2123 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2124 pcre_exec(). This saves a malloc/free per call. The default value of
2125 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2126 when configuring.
2127
2128 32. The default maximum size of a compiled pattern is 64K. There have been a
2129 few cases of people hitting this limit. The code now uses macros to handle the
2130 storing of links as offsets within the compiled pattern. It defaults to 2-byte
2131 links, but this can be changed to 3 or 4 bytes by --with-link-size when
2132 configuring. Tests 2 and 5 work only with 2-byte links because they output
2133 debugging information about compiled patterns.
2134
2135 33. Internal code re-arrangements:
2136
2137 (a) Moved the debugging function for printing out a compiled regex into
2138 its own source file (printint.c) and used #include to pull it into
2139 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2140 separate copies.
2141
2142 (b) Defined the list of op-code names for debugging as a macro in
2143 internal.h so that it is next to the definition of the opcodes.
2144
2145 (c) Defined a table of op-code lengths for simpler skipping along compiled
2146 code. This is again a macro in internal.h so that it is next to the
2147 definition of the opcodes.
2148
2149 34. Added support for recursive calls to individual subpatterns, along the
2150 lines of Robin Houston's patch (but implemented somewhat differently).
2151
2152 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2153 allow it to read and process whole directories in Win32. This code was
2154 contributed by Lionel Fourquaux; it has not been tested by me.
2155
2156 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2157 used to name a group. Names consist of alphanumerics and underscores, and must
2158 be unique. Back references use the syntax (?P=name) and recursive calls use
2159 (?P>name) which is a PCRE extension to the Python extension. Groups still have
2160 numbers. The function pcre_fullinfo() can be used after compilation to extract
2161 a name/number map. There are three relevant calls:
2162
2163 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
2164 PCRE_INFO_NAMECOUNT yields the number of entries
2165 PCRE_INFO_NAMETABLE yields a pointer to the map.
2166
2167 The map is a vector of fixed-size entries. The size of each entry depends on
2168 the length of the longest name used. The first two bytes of each entry are the
2169 group number, most significant byte first. There follows the corresponding
2170 name, zero terminated. The names are in alphabetical order.
2171
2172 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2173 case instead of 255. Making it the same both with and without UTF-8 support
2174 means that the same test output works with both.
2175
2176 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2177 calling malloc() with a zero argument.
2178
2179 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2180 optimization. I've improved things by keeping a bitmap of backreferences with
2181 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2182 fact referenced, the optimization can be applied. It is unlikely that a
2183 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2184 the match to follow \n) will appear inside brackets with a number greater than
2185 31, but if it does, any back reference > 31 suppresses the optimization.
2186
2187 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2188 of disabling numbered capturing parentheses. Any opening parenthesis that is
2189 not followed by ? behaves as if it were followed by ?: but named parentheses
2190 can still be used for capturing (and they will acquire numbers in the usual
2191 way).
2192
2193 41. Redesigned the return codes from the match() function into yes/no/error so
2194 that errors can be passed back from deep inside the nested calls. A malloc
2195 failure while inside a recursive subpattern call now causes the
2196 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2197
2198 42. It is now possible to set a limit on the number of times the match()
2199 function is called in a call to pcre_exec(). This facility makes it possible to
2200 limit the amount of recursion and backtracking, though not in a directly
2201 obvious way, because the match() function is used in a number of different
2202 circumstances. The count starts from zero for each position in the subject
2203 string (for non-anchored patterns). The default limit is, for compatibility, a
2204 large number, namely 10 000 000. You can change this in two ways:
2205
2206 (a) When configuring PCRE before making, you can use --with-match-limit=n
2207 to set a default value for the compiled library.
2208
2209 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2210 a different value is set. See 45 below.
2211
2212 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2213
2214 43. Added a new function pcre_config(int, void *) to enable run-time extraction
2215 of things that can be changed at compile time. The first argument specifies
2216 what is wanted and the second points to where the information is to be placed.
2217 The current list of available information is:
2218
2219 PCRE_CONFIG_UTF8
2220
2221 The output is an integer that is set to one if UTF-8 support is available;
2222 otherwise it is set to zero.
2223
2224 PCRE_CONFIG_NEWLINE
2225
2226 The output is an integer that it set to the value of the code that is used for
2227 newline. It is either LF (10) or CR (13).
2228
2229 PCRE_CONFIG_LINK_SIZE
2230
2231 The output is an integer that contains the number of bytes used for internal
2232 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2233
2234 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2235
2236 The output is an integer that contains the threshold above which the POSIX
2237 interface uses malloc() for output vectors. See item 31 above.
2238
2239 PCRE_CONFIG_MATCH_LIMIT
2240
2241 The output is an unsigned integer that contains the default limit of the number
2242 of match() calls in a pcre_exec() execution. See 42 above.
2243
2244 44. pcretest has been upgraded by the addition of the -C option. This causes it
2245 to extract all the available output from the new pcre_config() function, and to
2246 output it. The program then exits immediately.
2247
2248 45. A need has arisen to pass over additional data with calls to pcre_exec() in
2249 order to support additional features. One way would have been to define
2250 pcre_exec2() (for example) with extra arguments, but this would not have been
2251 extensible, and would also have required all calls to the original function to
2252 be mapped to the new one. Instead, I have chosen to extend the mechanism that
2253 is used for passing in "extra" data from pcre_study().
2254
2255 The pcre_extra structure is now exposed and defined in pcre.h. It currently
2256 contains the following fields:
2257
2258 flags a bitmap indicating which of the following fields are set
2259 study_data opaque data from pcre_study()
2260 match_limit a way of specifying a limit on match() calls for a specific
2261 call to pcre_exec()
2262 callout_data data for callouts (see 49 below)
2263
2264 The flag bits are also defined in pcre.h, and are
2265
2266 PCRE_EXTRA_STUDY_DATA
2267 PCRE_EXTRA_MATCH_LIMIT
2268 PCRE_EXTRA_CALLOUT_DATA
2269
2270 The pcre_study() function now returns one of these new pcre_extra blocks, with
2271 the actual study data pointed to by the study_data field, and the
2272 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2273 before. That is, this change is entirely upwards-compatible and requires no
2274 change to existing code.
2275
2276 If you want to pass in additional data to pcre_exec(), you can either place it
2277 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2278 block.
2279
2280 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2281 data string contains the escape sequence \M, pcretest calls pcre_exec() several
2282 times with different match limits, until it finds the minimum value needed for
2283 pcre_exec() to complete. The value is then output. This can be instructive; for
2284 most simple matches the number is quite small, but for pathological cases it
2285 gets very large very quickly.
2286
2287 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2288 returns the size of the data block pointed to by the study_data field in a
2289 pcre_extra block, that is, the value that was passed as the argument to
2290 pcre_malloc() when PCRE was getting memory in which to place the information
2291 created by pcre_study(). The fourth argument should point to a size_t variable.
2292 pcretest has been extended so that this information is shown after a successful
2293 pcre_study() call when information about the compiled regex is being displayed.
2294
2295 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2296 because what follows is always an absolute path. (Later: it turns out that this
2297 is more than cosmetic for MinGW, because it doesn't like empty path
2298 components.)
2299
2300 49. Some changes have been made to the callout feature (see 28 above):
2301
2302 (i) A callout function now has three choices for what it returns:
2303
2304 0 => success, carry on matching
2305 > 0 => failure at this point, but backtrack if possible
2306 < 0 => serious error, return this value from pcre_exec()
2307
2308 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2309 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2310 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2311 use by callout functions. It will never be used by PCRE itself.
2312
2313 (ii) The pcre_extra structure (see 45 above) has a void * field called
2314 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2315 pcre_callout_block structure has a field of the same name. The contents of
2316 the field passed in the pcre_extra structure are passed to the callout
2317 function in the corresponding field in the callout block. This makes it
2318 easier to use the same callout-containing regex from multiple threads. For
2319 testing, the pcretest program has a new data escape
2320
2321 \C*n pass the number n (may be negative) as callout_data
2322
2323 If the callout function in pcretest receives a non-zero value as
2324 callout_data, it returns that value.
2325
2326 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2327 there were some redundant $(CFLAGS) in commands that are now specified as
2328 $(LINK), which already includes $(CFLAGS).
2329
2330 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2331 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2332 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2333 one-byte characters throughout. Note that case-insensitive matching applies
2334 only to characters whose values are less than 256. PCRE doesn't support the
2335 notion of cases for higher-valued characters.
2336
2337 (i) A character class whose characters are all within 0-255 is handled as
2338 a bit map, and the map is inverted for negative classes. Previously, a
2339 character > 255 always failed to match such a class; however it should
2340 match if the class was a negative one (e.g. [^ab]). This has been fixed.
2341
2342 (ii) A negated character class with a single character < 255 is coded as
2343 "not this character" (OP_NOT). This wasn't working properly when the test
2344 character was multibyte, either singly or repeated.
2345
2346 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2347 mode, for example: \x{100}{2,3}.
2348
2349 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2350 singly or repeated) now correctly test multibyte characters. However,
2351 PCRE doesn't recognize any characters with values greater than 255 as
2352 digits, spaces, or word characters. Such characters always match \D, \S,
2353 and \W, and never match \d, \s, or \w.
2354
2355 (v) Classes may now contain characters and character ranges with values
2356 greater than 255. For example: [ab\x{100}-\x{400}].
2357
2358 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
2359 PCRE in UTF-8 mode.
2360
2361 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2362 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2363 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2364 value.)
2365
2366 53. The single man page has become too large. I have therefore split it up into
2367 a number of separate man pages. These also give rise to individual HTML pages;
2368 these are now put in a separate directory, and there is an index.html page that
2369 lists them all. Some hyperlinking between the pages has been installed.
2370
2371 54. Added convenience functions for handling named capturing parentheses.
2372
2373 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2374 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2375 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2376 are faulted.
2377
2378 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2379 calling configure. These values are used when compiling the dftables.c program
2380 which is run to generate the source of the default character tables. They
2381 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2382 you will need to set these values.
2383
2384 57. Updated the building process for Windows DLL, as provided by Fred Cox.
2385
2386
2387 Version 3.9 02-Jan-02
2388 ---------------------
2389
2390 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2391
2392 2. If --disable-static was given, the building process failed when trying to
2393 build pcretest and pcregrep. (For some reason it was using libtool to compile
2394 them, which is not right, as they aren't part of the library.)
2395
2396
2397 Version 3.8 18-Dec-01
2398 ---------------------
2399
2400 1. The experimental UTF-8 code was completely screwed up. It was packing the
2401 bytes in the wrong order. How dumb can you get?
2402
2403
2404 Version 3.7 29-Oct-01
2405 ---------------------
2406
2407 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2408 This caused pcretest, when used on the test data, to segfault. Unfortunately,
2409 this didn't happen under Solaris 8, where I normally test things.
2410
2411 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2412 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2413 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2414 this fix an hour or so after the initial 3.7 release.)
2415
2416
2417 Version 3.6 23-Oct-01
2418 ---------------------
2419
2420 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2421 offsets passed as NULL with zero offset count.
2422
2423 2. The config.guess and config.sub files had not been updated when I moved to
2424 the latest autoconf.
2425
2426
2427 Version 3.5 15-Aug-01
2428 ---------------------
2429
2430 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2431 had been forgotten.
2432
2433 2. By using declared but undefined structures, we can avoid using "void"
2434 definitions in pcre.h while keeping the internal definitions of the structures
2435 private.
2436
2437 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2438 user point of view, this means that both static and shared libraries are built
2439 by default, but this can be individually controlled. More of the work of
2440 handling this static/shared cases is now inside libtool instead of PCRE's make
2441 file.
2442
2443 4. The pcretest utility is now installed along with pcregrep because it is
2444 useful for users (to test regexs) and by doing this, it automatically gets
2445 relinked by libtool. The documentation has been turned into a man page, so
2446 there are now .1, .txt, and .html versions in /doc.
2447
2448 5. Upgrades to pcregrep:
2449 (i) Added long-form option names like gnu grep.
2450 (ii) Added --help to list all options with an explanatory phrase.
2451 (iii) Added -r, --recursive to recurse into sub-directories.
2452 (iv) Added -f, --file to read patterns from a file.
2453
2454 6. pcre_exec() was referring to its "code" argument before testing that
2455 argument for NULL (and giving an error if it was NULL).
2456
2457 7. Upgraded Makefile.in to allow for compiling in a different directory from
2458 the source directory.
2459
2460 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2461 options bits, the pointer it was passed was to an int instead of to an unsigned
2462 long int. This mattered only on 64-bit systems.
2463
2464 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2465 generated) instead of pcre.in, which it its source. Also made the same change
2466 in several of the .c files.
2467
2468 10. A new release of gcc defines printf() as a macro, which broke pcretest
2469 because it had an ifdef in the middle of a string argument for printf(). Fixed
2470 by using separate calls to printf().
2471
2472 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2473 script, to force use of CR or LF instead of \n in the source. On non-Unix
2474 systems, the value can be set in config.h.
2475
2476 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2477 absolute limit. Changed the text of the error message to make this clear, and
2478 likewise updated the man page.
2479
2480 13. The limit of 99 on the number of capturing subpatterns has been removed.
2481 The new limit is 65535, which I hope will not be a "real" limit.
2482
2483
2484 Version 3.4 22-Aug-00
2485 ---------------------
2486
2487 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2488
2489 2. Diagnose condition (?(0) as an error instead of crashing on matching.
2490
2491
2492 Version 3.3 01-Aug-00
2493 ---------------------
2494
2495 1. If an octal character was given, but the value was greater than \377, it
2496 was not getting masked to the least significant bits, as documented. This could
2497 lead to crashes in some systems.
2498
2499 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2500 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2501
2502 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2503 These just pass their arguments on to (pcre_free)(), but they are provided
2504 because some uses of PCRE bind it to non-C systems that can call its functions,
2505 but cannot call free() or pcre_free() directly.
2506
2507 4. Add "make test" as a synonym for "make check". Corrected some comments in
2508 the Makefile.
2509
2510 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2511 Makefile.
2512
2513 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2514 command called pgrep for grepping around the active processes.
2515
2516 7. Added the beginnings of support for UTF-8 character strings.
2517
2518 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2519 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2520 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2521 out for the ar command.)
2522
2523
2524 Version 3.2 12-May-00
2525 ---------------------
2526
2527 This is purely a bug fixing release.
2528
2529 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2530 of ZA. This was just one example of several cases that could provoke this bug,
2531 which was introduced by change 9 of version 2.00. The code for breaking
2532 infinite loops after an iteration that matches an empty string was't working
2533 correctly.
2534
2535 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2536 when matched against abbab (for example). After matching an empty string, it
2537 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2538 caused it to match further down the string than it should.
2539
2540 3. The code contained an inclusion of sys/types.h. It isn't clear why this
2541 was there because it doesn't seem to be needed, and it causes trouble on some
2542 systems, as it is not a Standard C header. It has been removed.
2543
2544 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2545 were reported on the Macintosh. The changes were from
2546
2547 while ((c = *(++ptr)) != 0 && c != '\n');
2548 to
2549 while ((c = *(++ptr)) != 0 && c != '\n') ;
2550
2551 Totally extraordinary, but if that's what it takes...
2552
2553 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2554 available. Added HAVE_BCOPY and an autoconf test for it; if neither
2555 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2556 assumes the way PCRE uses memmove() (always moving upwards).
2557
2558 6. PCRE is being used in one environment where strchr() is not available. There
2559 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2560 faster code anyway.
2561
2562
2563 Version 3.1 09-Feb-00
2564 ---------------------
2565
2566 The only change in this release is the fixing of some bugs in Makefile.in for
2567 the "install" target:
2568
2569 (1) It was failing to install pcreposix.h.
2570
2571 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2572
2573
2574 Version 3.0 01-Feb-00
2575 ---------------------
2576
2577 1. Add support for the /+ modifier to perltest (to output $` like it does in
2578 pcretest).
2579
2580 2. Add support for the /g modifier to perltest.
2581
2582 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2583 matches null strings.
2584
2585 4. Fix perltest so that it doesn't do unwanted things when fed an empty
2586 pattern. Perl treats empty patterns specially - it reuses the most recent
2587 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2588 effect.
2589
2590 5. The POSIX interface was broken in that it was just handing over the POSIX
2591 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2592 required a bigger vector, with some working space on the end. This means that
2593 the POSIX wrapper now has to get and free some memory, and copy the results.
2594
2595 6. Added some simple autoconf support, placing the test data and the
2596 documentation in separate directories, re-organizing some of the
2597 information files, and making it build pcre-config (a GNU standard). Also added
2598 libtool support for building PCRE as a shared library, which is now the
2599 default.
2600
2601 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2602 09 are not valid octal constants. Single digits will be used for minor values
2603 less than 10.
2604
2605 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2606 existing programs that set these in the POSIX interface can use PCRE without
2607 modification.
2608
2609 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2610 return all that pcre_info() returns, plus additional data. The pcre_info()
2611 function is retained for compatibility, but is considered to be obsolete.
2612
2613 10. Added experimental recursion feature (?R) to handle one common case that
2614 Perl 5.6 will be able to do with (?p{...}).
2615
2616 11. Added support for POSIX character classes like [:alpha:], which Perl is
2617 adopting.
2618
2619
2620 Version 2.08 31-Aug-99
2621 ----------------------
2622
2623 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2624 trying to match at the startoffset position, but instead was moving forward to
2625 the next newline as if a previous match had failed.
2626
2627 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2628 and could get into a loop if a null string was matched other than at the start
2629 of the subject.
2630
2631 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2632 be distinguished at compile time, and for completeness also added PCRE_DATE.
2633
2634 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2635 in GnuWin32 environments.
2636
2637
2638 Version 2.07 29-Jul-99
2639 ----------------------
2640
2641 1. The documentation is now supplied in plain text form and HTML as well as in
2642 the form of man page sources.
2643
2644 2. C++ compilers don't like assigning (void *) values to other pointer types.
2645 In particular this affects malloc(). Although there is no problem in Standard
2646 C, I've put in casts to keep C++ compilers happy.
2647
2648 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2649 should be (const char *).
2650
2651 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2652 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2653 However, I haven't made this a standard facility. The documentation doesn't
2654 mention it, and the Makefile doesn't support it.
2655
2656 5. The Makefile now contains an "install" target, with editable destinations at
2657 the top of the file. The pcretest program is not installed.
2658
2659 6. pgrep -V now gives the PCRE version number and date.
2660
2661 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2662 causing the entire string to be ignored, instead of just the last character.
2663
2664 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2665 non-matching string, it can take a very, very long time, even for strings of
2666 quite modest length, because of the nested recursion. PCRE now does better in
2667 some of these cases. It does this by remembering the last required literal
2668 character in the pattern, and pre-searching the subject to ensure it is present
2669 before running the real match. In other words, it applies a heuristic to detect
2670 some types of certain failure quickly, and in the above example, if presented
2671 with a string that has no trailing " it gives "no match" very quickly.
2672
2673 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2674 other alternatives are tried instead.
2675
2676
2677 Version 2.06 09-Jun-99
2678 ----------------------
2679
2680 1. Change pcretest's output for amount of store used to show just the code
2681 space, because the remainder (the data block) varies in size between 32-bit and
2682 64-bit systems.
2683
2684 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2685 start matching at. This allows lookbehinds to work when searching for multiple
2686 occurrences in a string.
2687
2688 3. Added additional options to pcretest for testing multiple occurrences:
2689
2690 /+ outputs the rest of the string that follows a match
2691 /g loops for multiple occurrences, using the new startoffset argument
2692 /G loops for multiple occurrences by passing an incremented pointer
2693
2694 4. PCRE wasn't doing the "first character" optimization for patterns starting
2695 with \b or \B, though it was doing it for other lookbehind assertions. That is,
2696 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2697 the letter 'x'. On long subject strings, this gives a significant speed-up.
2698
2699
2700 Version 2.05 21-Apr-99
2701 ----------------------
2702
2703 1. Changed the type of magic_number from int to long int so that it works
2704 properly on 16-bit systems.
2705
2706 2. Fixed a bug which caused patterns starting with .* not to work correctly
2707 when the subject string contained newline characters. PCRE was assuming
2708 anchoring for such patterns in all cases, which is not correct because .* will
2709 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2710 DOTALL is set at top level; otherwise it knows that patterns starting with .*
2711 must be retried after every newline in the subject.
2712
2713
2714 Version 2.04 18-Feb-99
2715 ----------------------
2716
2717 1. For parenthesized subpatterns with repeats whose minimum was zero, the
2718 computation of the store needed to hold the pattern was incorrect (too large).
2719 If such patterns were nested a few deep, this could multiply and become a real
2720 problem.
2721
2722 2. Added /M option to pcretest to show the memory requirement of a specific
2723 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2724
2725 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2726 compiled in such a way that the backtracking after subsequent failure was
2727 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2728 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2729
2730
2731 Version 2.03 02-Feb-99
2732 ----------------------
2733
2734 1. Fixed typo and small mistake in man page.
2735
2736 2. Added 4th condition (GPL supersedes if conflict) and created separate
2737 LICENCE file containing the conditions.
2738
2739 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2740 Perl, that is the internal \ allows the delimiter to be included in the
2741 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2742 the final delimiter, add \ to the end of the pattern (to test the error).
2743
2744 4. Added the convenience functions for extracting substrings after a successful
2745 match. Updated pcretest to make it able to test these functions.
2746
2747
2748 Version 2.02 14-Jan-99
2749 ----------------------
2750
2751 1. Initialized the working variables associated with each extraction so that
2752 their saving and restoring doesn't refer to uninitialized store.
2753
2754 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2755 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2756 fix the problem.
2757
2758 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2759 calls, and wasn't printing the correct value for compiling calls. Increased the
2760 default value of LOOPREPEAT, and the number of significant figures in the
2761 times.
2762
2763 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2764
2765 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2766 a building problem on Windows NT with a FAT file system.
2767
2768
2769 Version 2.01 21-Oct-98
2770 ----------------------
2771
2772 1. Changed the API for pcre_compile() to allow for the provision of a pointer
2773 to character tables built by pcre_maketables() in the current locale. If NULL
2774 is passed, the default tables are used.
2775
2776
2777 Version 2.00 24-Sep-98
2778 ----------------------
2779
2780 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2781 it any more.
2782
2783 2. Allow quantification of (?>) groups, and make it work correctly.
2784
2785 3. The first character computation wasn't working for (?>) groups.
2786
2787 4. Correct the implementation of \Z (it is permitted to match on the \n at the
2788 end of the subject) and add 5.005's \z, which really does match only at the
2789 very end of the subject.
2790
2791 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2792
2793 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2794 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2795 localized options. All options to pcre_study() were also removed.
2796
2797 7. Add other new features from 5.005:
2798
2799 $(?<= positive lookbehind
2800 $(?<! negative lookbehind
2801 (?imsx-imsx) added the unsetting capability
2802 such a setting is global if at outer level; local otherwise
2803 (?imsx-imsx:) non-capturing groups with option setting
2804 (?(cond)re|re) conditional pattern matching
2805
2806 A backreference to itself in a repeated group matches the previous
2807 captured string.
2808
2809 8. General tidying up of studying (both automatic and via "study")
2810 consequential on the addition of new assertions.
2811
2812 9. As in 5.005, unlimited repeated groups that could match an empty substring
2813 are no longer faulted at compile time. Instead, the loop is forcibly broken at
2814 runtime if any iteration does actually match an empty substring.
2815
2816 10. Include the RunTest script in the distribution.
2817
2818 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2819 discrepancies, some of which were old and were also with respect to 5.004. They
2820 have now been fixed.
2821
2822
2823 Version 1.09 28-Apr-98
2824 ----------------------
2825
2826 1. A negated single character class followed by a quantifier with a minimum
2827 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
2828 program crashes, or just wrong answers. This did not apply to negated classes
2829 containing more than one character, or to minima other than one.
2830
2831
2832 Version 1.08 27-Mar-98
2833 ----------------------
2834
2835 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
2836
2837 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
2838 latter must appear before anything that relies on it in the pattern.
2839
2840
2841 Version 1.07 16-Feb-98
2842 ----------------------
2843
2844 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
2845 repeat of a potentially empty string).
2846
2847
2848 Version 1.06 23-Jan-98
2849 ----------------------
2850
2851 1. Added Markus Oberhumer's little patches for C++.
2852
2853 2. Literal strings longer than 255 characters were broken.
2854
2855
2856 Version 1.05 23-Dec-97
2857 ----------------------
2858
2859 1. Negated character classes containing more than one character were failing if
2860 PCRE_CASELESS was set at run time.
2861
2862
2863 Version 1.04 19-Dec-97
2864 ----------------------
2865
2866 1. Corrected the man page, where some "const" qualifiers had been omitted.
2867
2868 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
2869 input syntax.
2870
2871 3. Fixed memory leak which occurred when a regex with back references was
2872 matched with an offsets vector that wasn't big enough. The temporary memory
2873 that is used in this case wasn't being freed if the match failed.
2874
2875 4. Tidied pcretest to ensure it frees memory that it gets.
2876
2877 5. Temporary memory was being obtained in the case where the passed offsets
2878 vector was exactly big enough.
2879
2880 6. Corrected definition of offsetof() from change 5 below.
2881
2882 7. I had screwed up change 6 below and broken the rules for the use of
2883 setjmp(). Now fixed.
2884
2885
2886 Version 1.03 18-Dec-97
2887 ----------------------
2888
2889 1. A erroneous regex with a missing opening parenthesis was correctly
2890 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
2891 on some systems.
2892
2893 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
2894 it was reported that one broken compiler failed on the former because "code" is
2895 also an independent variable.
2896
2897 3. The erroneous regex a[]b caused an array overrun reference.
2898
2899 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
2900 fail on data ending with that character. (It was going on too far, and checking
2901 the next character, typically a binary zero.) This was specific to the
2902 optimized code for single-character negative classes.
2903
2904 5. Added a contributed patch from the TIN world which does the following:
2905
2906 + Add an undef for memmove, in case the the system defines a macro for it.
2907
2908 + Add a definition of offsetof(), in case there isn't one. (I don't know
2909 the reason behind this - offsetof() is part of the ANSI standard - but
2910 it does no harm).
2911
2912 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
2913 most of the places where whitespace preceded '#'. I have given up and
2914 allowed the remaining 2 cases to be at the margin.
2915
2916 + Rename some variables in pcre to eliminate shadowing. This seems very
2917 pedantic, but does no harm, of course.
2918
2919 6. Moved the call to setjmp() into its own function, to get rid of warnings
2920 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
2921
2922 7. Constructs such as \d{8,} were compiling into the equivalent of
2923 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
2924 outcome, but in this particular case used more store than had been allocated,
2925 which caused the bug to be discovered because it threw up an internal error.
2926
2927 8. The debugging code in both pcre and pcretest for outputting the compiled
2928 form of a regex was going wrong in the case of back references followed by
2929 curly-bracketed repeats.
2930
2931
2932 Version 1.02 12-Dec-97
2933 ----------------------
2934
2935 1. Typos in pcre.3 and comments in the source fixed.
2936
2937 2. Applied a contributed patch to get rid of places where it used to remove
2938 'const' from variables, and fixed some signed/unsigned and uninitialized
2939 variable warnings.
2940
2941 3. Added the "runtest" target to Makefile.
2942
2943 4. Set default compiler flag to -O2 rather than just -O.
2944
2945
2946 Version 1.01 19-Nov-97
2947 ----------------------
2948
2949 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
2950 like /([ab]*)*/, that is, for classes with more than one character in them.
2951
2952 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
2953 as /((?>a*))*/ (a PCRE_EXTRA facility).
2954
2955
2956 Version 1.00 18-Nov-97
2957 ----------------------
2958
2959 1. Added compile-time macros to support systems such as SunOS4 which don't have
2960 memmove() or strerror() but have other things that can be used instead.
2961
2962 2. Arranged that "make clean" removes the executables.
2963
2964
2965 Version 0.99 27-Oct-97
2966 ----------------------
2967
2968 1. Fixed bug in code for optimizing classes with only one character. It was
2969 initializing a 32-byte map regardless, which could cause it to run off the end
2970 of the memory it had got.
2971
2972 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
2973
2974
2975 Version 0.98 22-Oct-97
2976 ----------------------
2977
2978 1. Fixed bug in code for handling temporary memory usage when there are more
2979 back references than supplied space in the ovector. This could cause segfaults.
2980
2981
2982 Version 0.97 21-Oct-97
2983 ----------------------
2984
2985 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
2986
2987 2. Optimized negated single characters not to use a bit map.
2988
2989 3. Brought error texts together as macro definitions; clarified some of them;
2990 fixed one that was wrong - it said "range out of order" when it meant "invalid
2991 escape sequence".
2992
2993 4. Changed some char * arguments to const char *.
2994
2995 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
2996
2997 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
2998 pcretest.
2999
3000
3001 Version 0.96 16-Oct-97
3002 ----------------------
3003
3004 1. Added a simple "pgrep" utility to the distribution.
3005
3006 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
3007 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
3008 where "ddd" means "one or more decimal digits".
3009
3010 3. Fixed serious bug. If a pattern had a back reference, but the call to
3011 pcre_exec() didn't supply a large enough ovector to record the related
3012 identifying subpattern, the match always failed. PCRE now remembers the number
3013 of the largest back reference, and gets some temporary memory in which to save
3014 the offsets during matching if necessary, in order to ensure that
3015 backreferences always work.
3016
3017 4. Increased the compatibility with Perl in a number of ways:
3018
3019 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
3020 to request this handling. The option can be set at compile or exec time.
3021
3022 (b) $ matches before a terminating newline by default; an option
3023 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
3024 mode). The option can be set at compile or exec time.
3025
3026 (c) The handling of \ followed by a digit other than 0 is now supposed to be
3027 the same as Perl's. If the decimal number it represents is less than 10
3028 or there aren't that many previous left capturing parentheses, an octal
3029 escape is read. Inside a character class, it's always an octal escape,
3030 even if it is a single digit.
3031
3032 (d) An escaped but undefined alphabetic character is taken as a literal,
3033 unless PCRE_EXTRA is set. Currently this just reserves the remaining
3034 escapes.
3035
3036 (e) {0} is now permitted. (The previous item is removed from the compiled
3037 pattern).
3038
3039 5. Changed all the names of code files so that the basic parts are no longer
3040 than 10 characters, and abolished the teeny "globals.c" file.
3041
3042 6. Changed the handling of character classes; they are now done with a 32-byte
3043 bit map always.
3044
3045 7. Added the -d and /D options to pcretest to make it possible to look at the
3046 internals of compilation without having to recompile pcre.
3047
3048
3049 Version 0.95 23-Sep-97
3050 ----------------------
3051
3052 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
3053 \x20 at the start of a run of normal characters. These were being treated as
3054 real characters, instead of the source characters being re-checked.
3055
3056
3057 Version 0.94 18-Sep-97
3058 ----------------------
3059
3060 1. The functions are now thread-safe, with the caveat that the global variables
3061 containing pointers to malloc() and free() or alternative functions are the
3062 same for all threads.
3063
3064 2. Get pcre_study() to generate a bitmap of initial characters for non-
3065 anchored patterns when this is possible, and use it if passed to pcre_exec().
3066
3067
3068 Version 0.93 15-Sep-97
3069 ----------------------
3070
3071 1. /(b)|(:+)/ was computing an incorrect first character.
3072
3073 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
3074 but not actually doing anything yet.
3075
3076 3. Treat "-" characters in classes that cannot be part of ranges as literals,
3077 as Perl does (e.g. [-az] or [az-]).
3078
3079 4. Set the anchored flag if a branch starts with .* or .*? because that tests
3080 all possible positions.
3081
3082 5. Split up into different modules to avoid including unneeded functions in a
3083 compiled binary. However, compile and exec are still in one module. The "study"
3084 function is split off.
3085
3086 6. The character tables are now in a separate module whose source is generated
3087 by an auxiliary program - but can then be edited by hand if required. There are
3088 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
3089 toupper() in the code.
3090
3091 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
3092 make them global. Abolish the function for setting them, as the caller can now
3093 set them directly.
3094
3095
3096 Version 0.92 11-Sep-97
3097 ----------------------
3098
3099 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
3100 (e.g. /a{1,3}/) was broken (I mis-optimized it).
3101
3102 2. Caseless matching was not working in character classes if the characters in
3103 the pattern were in upper case.
3104
3105 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
3106
3107 4. Make PCRE_ANCHORED public and accept as a compile option.
3108
3109 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
3110 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
3111 pass them.
3112
3113 6. Give an error if bad option bits passed at compile or run time.
3114
3115 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
3116 pcretest to cause it to pass that flag.
3117
3118 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
3119 options, and the first character, if set.
3120
3121 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
3122
3123
3124 Version 0.91 10-Sep-97
3125 ----------------------
3126
3127 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
3128 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
3129
3130 2. PCRE was looping on encountering an indefinitely repeated back reference to
3131 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
3132 Perl does - treats the match as successful.
3133
3134 ****

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