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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 143 - (show annotations) (download)
Mon Apr 2 10:08:14 2007 UTC (7 years, 6 months ago) by ph10
File size: 126224 byte(s)
Previous fix for pcretest was buggy. Try again.

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

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