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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 110 - (show annotations) (download)
Wed Mar 7 16:31:04 2007 UTC (7 years, 1 month ago) by ph10
File size: 122258 byte(s)
Some more #ifdefs for SUPPORT_UTF8.

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

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