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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 256 - (show annotations) (download)
Wed Sep 19 08:57:53 2007 UTC (7 years, 1 month ago) by ph10
File size: 141531 byte(s)
Patches to C++ code to avoid compiler warnings.

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

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