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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 223 - (show annotations) (download)
Mon Aug 20 11:07:53 2007 UTC (7 years, 3 months ago) by ph10
File size: 137546 byte(s)
Fix loop for classes containing \p or \P and just one ascii character.

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

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