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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 108 - (show annotations) (download)
Wed Mar 7 11:11:23 2007 UTC (7 years, 4 months ago) by ph10
File size: 121994 byte(s)
Arguments for the emulated memmove() should be void *, not char *.

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

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