/[pcre]/code/tags/pcre-8.01/ChangeLog
ViewVC logotype

Contents of /code/tags/pcre-8.01/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 100 - (show annotations) (download)
Tue Mar 6 14:53:10 2007 UTC (7 years, 6 months ago) by ph10
Original Path: code/trunk/ChangeLog
File size: 120448 byte(s)
Updated comments in configure.ac that end up in config.h.in, and arrange for 
config.h to be distributed (compare pcre.h).

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

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