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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 292 - (show annotations) (download)
Thu Dec 27 09:27:23 2007 UTC (6 years, 7 months ago) by ph10
File size: 146852 byte(s)
Tidies for 7.5-RC2 release.

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

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