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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 99 - (show annotations) (download)
Tue Mar 6 12:27:42 2007 UTC (7 years, 1 month ago) by ph10
File size: 119802 byte(s)
1. Move the comment about version numbers from pcre.h.in to configure.ac 
because that's where they are now set.
2. Update all the man pages to remove the use of .br and .in because this
causes trouble for some HTML converters. Also standardised the final sections 
giving author information and revision date.
3. Update the maintain/132html man page converter to handle .nf/.fi and to barf 
at .br/.in.

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

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