ViewVC logotype

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 157 - (show annotations)
Tue May 1 09:19:56 2007 UTC (8 years, 5 months ago) by ph10
File size: 127019 byte(s)
Try "french" locale if "fr_FR" not present. Update NEWS re Win32 libs.
1 ChangeLog for PCRE
2 ------------------
4 Version 7.2 01-May-07
5 ---------------------
7 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
8 which is apparently normally available under Windows.
11 Version 7.1 24-Apr-07
12 ---------------------
14 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
15 that is more "standard", making use of automake and other Autotools. There
16 is some re-arrangement of the files and adjustment of comments consequent
17 on this.
19 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
20 for recursive directory scanning broke on some systems because the files
21 are not scanned in any specific order and on different systems the order
22 was different. A call to "sort" has been inserted into RunGrepTest for the
23 approprate test as a short-term fix. In the longer term there may be an
24 alternative.
26 3. I had an email from Eric Raymond about problems translating some of PCRE's
27 man pages to HTML (despite the fact that I distribute HTML pages, some
28 people do their own conversions for various reasons). The problems
29 concerned the use of low-level troff macros .br and .in. I have therefore
30 removed all such uses from the man pages (some were redundant, some could
31 be replaced by .nf/.fi pairs). The 132html script that I use to generate
32 HTML has been updated to handle .nf/.fi and to complain if it encounters
33 .br or .in.
35 4. Updated comments in configure.ac that get placed in config.h.in and also
36 arranged for config.h to be included in the distribution, with the name
37 config.h.generic, for the benefit of those who have to compile without
38 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
40 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
41 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
42 makevp.bat for the current PCRE, using the additional files
43 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
45 6. A Windows user reported a minor discrepancy with test 2, which turned out
46 to be caused by a trailing space on an input line that had got lost in his
47 copy. The trailing space was an accident, so I've just removed it.
49 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
50 that is needed.
52 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
53 as "const" (a) because they are and (b) because it helps the PHP
54 maintainers who have recently made a script to detect big data structures
55 in the php code that should be moved to the .rodata section. I remembered
56 to update Builducptable as well, so it won't revert if ucptable.h is ever
57 re-created.
59 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
60 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
61 order to be able to cut out the UTF-8 tables in the latter when UTF-8
62 support is not required. This saves 1.5-2K of code, which is important in
63 some applications.
65 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
66 so as not to refer to the tables, even though these functions will never be
67 called when UTF-8 support is disabled. Otherwise there are problems with a
68 shared library.
70 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
72 (a) It was defining its arguments as char * instead of void *.
74 (b) It was assuming that all moves were upwards in memory; this was true
75 a long time ago when I wrote it, but is no longer the case.
77 The emulated memove() is provided for those environments that have neither
78 memmove() nor bcopy(). I didn't think anyone used it these days, but that
79 is clearly not the case, as these two bugs were recently reported.
81 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
82 and Detrail to create the HTML documentation, the .txt form of the man
83 pages, and it removes trailing spaces from listed files. It also creates
84 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
85 case, it wraps all the #defines with #ifndefs. This script should be run
86 before "make dist".
88 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
89 with Unicode property support.
91 (a) For a maximizing quantifier, if the two different cases of the
92 character were of different lengths in their UTF-8 codings (there are
93 some cases like this - I found 11), and the matching function had to
94 back up over a mixture of the two cases, it incorrectly assumed they
95 were both the same length.
97 (b) When PCRE was configured to use the heap rather than the stack for
98 recursion during matching, it was not correctly preserving the data for
99 the other case of a UTF-8 character when checking ahead for a match
100 while processing a minimizing repeat. If the check also involved
101 matching a wide character, but failed, corruption could cause an
102 erroneous result when trying to check for a repeat of the original
103 character.
105 13. Some tidying changes to the testing mechanism:
107 (a) The RunTest script now detects the internal link size and whether there
108 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
109 values substituted by "configure". (The RunGrepTest script already did
110 this for UTF-8.) The configure.ac script no longer substitutes the
111 relevant variables.
113 (b) The debugging options /B and /D in pcretest show the compiled bytecode
114 with length and offset values. This means that the output is different
115 for different internal link sizes. Test 2 is skipped for link sizes
116 other than 2 because of this, bypassing the problem. Unfortunately,
117 there was also a test in test 3 (the locale tests) that used /B and
118 failed for link sizes other than 2. Rather than cut the whole test out,
119 I have added a new /Z option to pcretest that replaces the length and
120 offset values with spaces. This is now used to make test 3 independent
121 of link size. (Test 2 will be tidied up later.)
123 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
124 segmentation fault instead of returning the appropriate error message.
126 15. In multiline mode when the newline sequence was set to "any", the pattern
127 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
128 This doesn't seem right; it now treats the CRLF combination as the line
129 ending, and so does not match in that case. It's only a pattern such as ^$
130 that would hit this one: something like ^ABC$ would have failed after \r
131 and then tried again after \r\n.
133 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
134 in an attempt to make files that differ only in their line terminators
135 compare equal. This works on Linux.
137 17. Under certain error circumstances pcregrep might try to free random memory
138 as it exited. This is now fixed, thanks to valgrind.
140 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
141 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
142 was because its rules for how to advance for /g after matching an empty
143 string at the end of a line did not allow for this case. They now check for
144 it specially.
146 20. pcretest is supposed to handle patterns and data of any length, by
147 extending its buffers when necessary. It was getting this wrong when the
148 buffer for a data line had to be extended.
150 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
151 CRLF as a newline sequence.
153 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
154 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
155 I have nevertheless tidied it up.
157 23. Added some casts to kill warnings from HP-UX ia64 compiler.
159 24. Added a man page for pcre-config.
162 Version 7.0 19-Dec-06
163 ---------------------
165 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
166 moving to gcc 4.1.1.
168 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
169 sys/time.h, which is documented as needed for this function. It doesn't
170 seem to matter on Linux, but it showed up on some releases of OS X.
172 3. It seems that there are systems where bytes whose values are greater than
173 127 match isprint() in the "C" locale. The "C" locale should be the
174 default when a C program starts up. In most systems, only ASCII printing
175 characters match isprint(). This difference caused the output from pcretest
176 to vary, making some of the tests fail. I have changed pcretest so that:
178 (a) When it is outputting text in the compiled version of a pattern, bytes
179 other than 32-126 are always shown as hex escapes.
181 (b) When it is outputting text that is a matched part of a subject string,
182 it does the same, unless a different locale has been set for the match
183 (using the /L modifier). In this case, it uses isprint() to decide.
185 4. Fixed a major bug that caused incorrect computation of the amount of memory
186 required for a compiled pattern when options that changed within the
187 pattern affected the logic of the preliminary scan that determines the
188 length. The relevant options are -x, and -i in UTF-8 mode. The result was
189 that the computed length was too small. The symptoms of this bug were
190 either the PCRE error "internal error: code overflow" from pcre_compile(),
191 or a glibc crash with a message such as "pcretest: free(): invalid next
192 size (fast)". Examples of patterns that provoked this bug (shown in
193 pcretest format) are:
195 /(?-x: )/x
196 /(?x)(?-x: \s*#\s*)/
197 /((?i)[\x{c0}])/8
198 /(?i:[\x{c0}])/8
200 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
201 is now done differently.
203 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
204 wrapper classes; (b) implement a new function in the C++ scanner that is
205 more efficient than the old way of doing things because it avoids levels of
206 recursion in the regex matching; (c) add a paragraph to the documentation
207 for the FullMatch() function.
209 6. The escape sequence \n was being treated as whatever was defined as
210 "newline". Not only was this contrary to the documentation, which states
211 that \n is character 10 (hex 0A), but it also went horribly wrong when
212 "newline" was defined as CRLF. This has been fixed.
214 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
215 was being set to -1 for the "end of line" case (supposedly a value that no
216 character can have). Though this value is never used (the check for end of
217 line is "zero bytes in current character"), it caused compiler complaints.
218 I've changed it to 0xffffffff.
220 8. In pcre_version.c, the version string was being built by a sequence of
221 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
222 string (as it is for production releases) called a macro with an empty
223 argument. The C standard says the result of this is undefined. The gcc
224 compiler treats it as an empty string (which was what was wanted) but it is
225 reported that Visual C gives an error. The source has been hacked around to
226 avoid this problem.
228 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
229 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
230 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
231 of them did).
233 10. Originally, pcretest opened its input and output without "b"; then I was
234 told that "b" was needed in some environments, so it was added for release
235 5.0 to both the input and output. (It makes no difference on Unix-like
236 systems.) Later I was told that it is wrong for the input on Windows. I've
237 now abstracted the modes into two macros, to make it easier to fiddle with
238 them, and removed "b" from the input mode under Windows.
240 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
242 12. Added -help and --help to pcretest as an official way of being reminded
243 of the options.
245 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
246 and pcrecpp.cc because they annoy compilers at high warning levels.
248 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
250 15. Fixed an occurrence of == in configure.ac that should have been = (shell
251 scripts are not C programs :-) and which was not noticed because it works
252 on Linux.
254 16. pcretest is supposed to handle any length of pattern and data line (as one
255 line or as a continued sequence of lines) by extending its input buffer if
256 necessary. This feature was broken for very long pattern lines, leading to
257 a string of junk being passed to pcre_compile() if the pattern was longer
258 than about 50K.
260 17. I have done a major re-factoring of the way pcre_compile() computes the
261 amount of memory needed for a compiled pattern. Previously, there was code
262 that made a preliminary scan of the pattern in order to do this. That was
263 OK when PCRE was new, but as the facilities have expanded, it has become
264 harder and harder to keep it in step with the real compile phase, and there
265 have been a number of bugs (see for example, 4 above). I have now found a
266 cunning way of running the real compile function in a "fake" mode that
267 enables it to compute how much memory it would need, while actually only
268 ever using a few hundred bytes of working memory and without too many
269 tests of the mode. This should make future maintenance and development
270 easier. A side effect of this work is that the limit of 200 on the nesting
271 depth of parentheses has been removed (though this was never a serious
272 limitation, I suspect). However, there is a downside: pcre_compile() now
273 runs more slowly than before (30% or more, depending on the pattern). I
274 hope this isn't a big issue. There is no effect on runtime performance.
276 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
277 newline (only possible for the last line of a file) and it was a
278 pattern that set a locale (followed by /Lsomething), pcretest crashed.
280 19. Added additional timing features to pcretest. (1) The -tm option now times
281 matching only, not compiling. (2) Both -t and -tm can be followed, as a
282 separate command line item, by a number that specifies the number of
283 repeats to use when timing. The default is 50000; this gives better
284 precision, but takes uncomfortably long for very large patterns.
286 20. Extended pcre_study() to be more clever in cases where a branch of a
287 subpattern has no definite first character. For example, (a*|b*)[cd] would
288 previously give no result from pcre_study(). Now it recognizes that the
289 first character must be a, b, c, or d.
291 21. There was an incorrect error "recursive call could loop indefinitely" if
292 a subpattern (or the entire pattern) that was being tested for matching an
293 empty string contained only one non-empty item after a nested subpattern.
294 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
295 incorrectly, because the \d was being skipped in the check.
297 22. The pcretest program now has a new pattern option /B and a command line
298 option -b, which is equivalent to adding /B to every pattern. This causes
299 it to show the compiled bytecode, without the additional information that
300 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
301 is the same as /B/I).
303 23. A new optimization is now able automatically to treat some sequences such
304 as a*b as a*+b. More specifically, if something simple (such as a character
305 or a simple class like \d) has an unlimited quantifier, and is followed by
306 something that cannot possibly match the quantified thing, the quantifier
307 is automatically "possessified".
309 24. A recursive reference to a subpattern whose number was greater than 39
310 went wrong under certain circumstances in UTF-8 mode. This bug could also
311 have affected the operation of pcre_study().
313 25. Realized that a little bit of performance could be had by replacing
314 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
316 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
318 27. Possessive quantifiers such as a++ were previously implemented by turning
319 them into atomic groups such as ($>a+). Now they have their own opcodes,
320 which improves performance. This includes the automatically created ones
321 from 23 above.
323 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
324 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
325 the first matched character to be a colon. This applied both to named and
326 numbered groups.
328 29. The ucpinternal.h header file was missing its idempotency #ifdef.
330 30. I was sent a "project" file called libpcre.a.dev which I understand makes
331 building PCRE on Windows easier, so I have included it in the distribution.
333 31. There is now a check in pcretest against a ridiculously large number being
334 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
335 loop, the loop is abandoned.
337 32. Forward references to subpatterns in conditions such as (?(2)...) where
338 subpattern 2 is defined later cause pcre_compile() to search forwards in
339 the pattern for the relevant set of parentheses. This search went wrong
340 when there were unescaped parentheses in a character class, parentheses
341 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
343 33. "Subroutine" calls and backreferences were previously restricted to
344 referencing subpatterns earlier in the regex. This restriction has now
345 been removed.
347 34. Added a number of extra features that are going to be in Perl 5.10. On the
348 whole, these are just syntactic alternatives for features that PCRE had
349 previously implemented using the Python syntax or my own invention. The
350 other formats are all retained for compatibility.
352 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
353 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
354 also .NET compatible.
356 (b) A recursion or subroutine call to a named group can now be defined as
357 (?&name) as well as (?P>name).
359 (c) A backreference to a named group can now be defined as \k<name> or
360 \k'name' as well as (?P=name). The new forms, as well as being in Perl
361 5.10, are also .NET compatible.
363 (d) A conditional reference to a named group can now use the syntax
364 (?(<name>) or (?('name') as well as (?(name).
366 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
367 groups (named and numbered) that are never evaluated inline, but can be
368 called as "subroutines" from elsewhere. In effect, the DEFINE condition
369 is always false. There may be only one alternative in such a group.
371 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
372 as the simple (?(R). The condition is true only if the most recent
373 recursion is that of the given number or name. It does not search out
374 through the entire recursion stack.
376 (g) The escape \gN or \g{N} has been added, where N is a positive or
377 negative number, specifying an absolute or relative reference.
379 35. Tidied to get rid of some further signed/unsigned compiler warnings and
380 some "unreachable code" warnings.
382 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
383 things, this adds five new scripts.
385 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
386 There were also incompatibilities regarding the handling of \Q..\E inside
387 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
388 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
390 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
391 matches an empty string, and forcibly breaks the loop. There were bugs in
392 this code in non-simple cases. For a pattern such as ^(a()*)* matched
393 against aaaa the result was just "a" rather than "aaaa", for example. Two
394 separate and independent bugs (that affected different cases) have been
395 fixed.
397 39. Refactored the code to abolish the use of different opcodes for small
398 capturing bracket numbers. This is a tidy that I avoided doing when I
399 removed the limit on the number of capturing brackets for 3.5 back in 2001.
400 The new approach is not only tidier, it makes it possible to reduce the
401 memory needed to fix the previous bug (38).
403 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
404 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
405 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
406 mode.
408 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
409 report.
411 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
412 copy construction and assignment in the C++ wrapper.
414 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
415 couple of bugs that could have given wrong results in the "--newline=crlf"
416 case.
418 44. Added a number of casts and did some reorganization of signed/unsigned int
419 variables following suggestions from Dair Grant. Also renamed the variable
420 "this" as "item" because it is a C++ keyword.
422 45. Arranged for dftables to add
424 #include "pcre_internal.h"
426 to pcre_chartables.c because without it, gcc 4.x may remove the array
427 definition from the final binary if PCRE is built into a static library and
428 dead code stripping is activated.
430 46. For an unanchored pattern, if a match attempt fails at the start of a
431 newline sequence, and the newline setting is CRLF or ANY, and the next two
432 characters are CRLF, advance by two characters instead of one.
435 Version 6.7 04-Jul-06
436 ---------------------
438 1. In order to handle tests when input lines are enormously long, pcretest has
439 been re-factored so that it automatically extends its buffers when
440 necessary. The code is crude, but this _is_ just a test program. The
441 default size has been increased from 32K to 50K.
443 2. The code in pcre_study() was using the value of the re argument before
444 testing it for NULL. (Of course, in any sensible call of the function, it
445 won't be NULL.)
447 3. The memmove() emulation function in pcre_internal.h, which is used on
448 systems that lack both memmove() and bcopy() - that is, hardly ever -
449 was missing a "static" storage class specifier.
451 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
452 containing an extended class (one that cannot be represented by a bitmap
453 because it contains high-valued characters or Unicode property items, e.g.
454 [\pZ]). Almost always one would set UTF-8 mode when processing such a
455 pattern, but PCRE should not loop if you do not (it no longer does).
456 [Detail: two cases were found: (a) a repeated subpattern containing an
457 extended class; (b) a recursive reference to a subpattern that followed a
458 previous extended class. It wasn't skipping over the extended class
459 correctly when UTF-8 mode was not set.]
461 5. A negated single-character class was not being recognized as fixed-length
462 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
463 compile error "lookbehind assertion is not fixed length".
465 6. The RunPerlTest auxiliary script was showing an unexpected difference
466 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
467 write a Perl script that can interpret lines of an input file either as
468 byte characters or as UTF-8, which is what "perltest" was being required to
469 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
470 can't do is switch easily at run time between having the "use utf8;" pragma
471 or not. In the end, I fudged it by using the RunPerlTest script to insert
472 "use utf8;" explicitly for the UTF-8 tests.
474 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
475 the end of the subject string, contrary to the documentation and to what
476 Perl does. This was true of both matching functions. Now it matches only at
477 the start of the subject and immediately after *internal* newlines.
479 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
480 a pointer to an int instead of a pointer to an unsigned long int. This
481 caused problems on 64-bit systems.
483 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
484 instance of the 'standard' template library not being so standard".
486 10. There was no check on the number of named subpatterns nor the maximum
487 length of a subpattern name. The product of these values is used to compute
488 the size of the memory block for a compiled pattern. By supplying a very
489 long subpattern name and a large number of named subpatterns, the size
490 computation could be caused to overflow. This is now prevented by limiting
491 the length of names to 32 characters, and the number of named subpatterns
492 to 10,000.
494 11. Subpatterns that are repeated with specific counts have to be replicated in
495 the compiled pattern. The size of memory for this was computed from the
496 length of the subpattern and the repeat count. The latter is limited to
497 65535, but there was no limit on the former, meaning that integer overflow
498 could in principle occur. The compiled length of a repeated subpattern is
499 now limited to 30,000 bytes in order to prevent this.
501 12. Added the optional facility to have named substrings with the same name.
503 13. Added the ability to use a named substring as a condition, using the
504 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
505 are numbers (not recommended). Forward references are permitted.
507 14. Added forward references in named backreferences (if you see what I mean).
509 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
510 pattern could run off the end of the subject. For example, the pattern
511 "(?s)(.{1,5})"8 did this with the subject "ab".
513 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
514 PCRE_CASELESS was set when matching characters that were quantified with ?
515 or *.
517 17. A character class other than a single negated character that had a minimum
518 but no maximum quantifier - for example [ab]{6,} - was not handled
519 correctly by pce_dfa_exec(). It would match only one character.
521 18. A valid (though odd) pattern that looked like a POSIX character
522 class but used an invalid character after [ (for example [[,abc,]]) caused
523 pcre_compile() to give the error "Failed: internal error: code overflow" or
524 in some cases to crash with a glibc free() error. This could even happen if
525 the pattern terminated after [[ but there just happened to be a sequence of
526 letters, a binary zero, and a closing ] in the memory that followed.
528 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
529 over the years. Originally (before any Unicode support), just the bottom 8
530 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
531 output from "man perlunicode" includes this:
533 The regular expression compiler produces polymorphic opcodes. That
534 is, the pattern adapts to the data and automatically switches to
535 the Unicode character scheme when presented with Unicode data--or
536 instead uses a traditional byte scheme when presented with byte
537 data.
539 Sadly, a wide octal escape does not cause a switch, and in a string with
540 no other multibyte characters, these octal escapes are treated as before.
541 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
542 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
543 Unicode string.
545 I have not perpetrated such confusion in PCRE. Up till now, it took just
546 the bottom 8 bits, as in old Perl. I have now made octal escapes with
547 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
548 translate to the appropriate multibyte character.
550 29. Applied some refactoring to reduce the number of warnings from Microsoft
551 and Borland compilers. This has included removing the fudge introduced
552 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
553 a warning about an unused variable.
555 21. PCRE has not included VT (character 0x0b) in the set of whitespace
556 characters since release 4.0, because Perl (from release 5.004) does not.
557 [Or at least, is documented not to: some releases seem to be in conflict
558 with the documentation.] However, when a pattern was studied with
559 pcre_study() and all its branches started with \s, PCRE still included VT
560 as a possible starting character. Of course, this did no harm; it just
561 caused an unnecessary match attempt.
563 22. Removed a now-redundant internal flag bit that recorded the fact that case
564 dependency changed within the pattern. This was once needed for "required
565 byte" processing, but is no longer used. This recovers a now-scarce options
566 bit. Also moved the least significant internal flag bit to the most-
567 significant bit of the word, which was not previously used (hangover from
568 the days when it was an int rather than a uint) to free up another bit for
569 the future.
571 23. Added support for CRLF line endings as well as CR and LF. As well as the
572 default being selectable at build time, it can now be changed at runtime
573 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
574 specify that it is scanning data with non-default line endings.
576 24. Changed the definition of CXXLINK to make it agree with the definition of
577 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
579 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
580 recursions. This makes a big different to stack usage for some patterns.
582 26. If a subpattern containing a named recursion or subroutine reference such
583 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
584 the space required for the compiled pattern went wrong and gave too small a
585 value. Depending on the environment, this could lead to "Failed: internal
586 error: code overflow at offset 49" or "glibc detected double free or
587 corruption" errors.
589 27. Applied patches from Google (a) to support the new newline modes and (b) to
590 advance over multibyte UTF-8 characters in GlobalReplace.
592 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
593 difference for some implementation of PCRE in some Windows version.
595 29. Added some extra testing facilities to pcretest:
597 \q<number> in a data line sets the "match limit" value
598 \Q<number> in a data line sets the "match recursion limt" value
599 -S <number> sets the stack size, where <number> is in megabytes
601 The -S option isn't available for Windows.
604 Version 6.6 06-Feb-06
605 ---------------------
607 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
608 in pcreposix.h. I have copied the definition from pcre.h.
610 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
611 because pcre.h is no longer a built file.
613 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
614 not normally included in the compiled code.
617 Version 6.5 01-Feb-06
618 ---------------------
620 1. When using the partial match feature with pcre_dfa_exec(), it was not
621 anchoring the second and subsequent partial matches at the new starting
622 point. This could lead to incorrect results. For example, with the pattern
623 /1234/, partially matching against "123" and then "a4" gave a match.
625 2. Changes to pcregrep:
627 (a) All non-match returns from pcre_exec() were being treated as failures
628 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
629 error message is output. Some extra information is given for the
631 probably the only errors that are likely to be caused by users (by
632 specifying a regex that has nested indefinite repeats, for instance).
633 If there are more than 20 of these errors, pcregrep is abandoned.
635 (b) A binary zero was treated as data while matching, but terminated the
636 output line if it was written out. This has been fixed: binary zeroes
637 are now no different to any other data bytes.
639 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
640 used to set a locale for matching. The --locale=xxxx long option has
641 been added (no short equivalent) to specify a locale explicitly on the
642 pcregrep command, overriding the environment variables.
644 (d) When -B was used with -n, some line numbers in the output were one less
645 than they should have been.
647 (e) Added the -o (--only-matching) option.
649 (f) If -A or -C was used with -c (count only), some lines of context were
650 accidentally printed for the final match.
652 (g) Added the -H (--with-filename) option.
654 (h) The combination of options -rh failed to suppress file names for files
655 that were found from directory arguments.
657 (i) Added the -D (--devices) and -d (--directories) options.
659 (j) Added the -F (--fixed-strings) option.
661 (k) Allow "-" to be used as a file name for -f as well as for a data file.
663 (l) Added the --colo(u)r option.
665 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
666 is not present by default.
668 3. A nasty bug was discovered in the handling of recursive patterns, that is,
669 items such as (?R) or (?1), when the recursion could match a number of
670 alternatives. If it matched one of the alternatives, but subsequently,
671 outside the recursion, there was a failure, the code tried to back up into
672 the recursion. However, because of the way PCRE is implemented, this is not
673 possible, and the result was an incorrect result from the match.
675 In order to prevent this happening, the specification of recursion has
676 been changed so that all such subpatterns are automatically treated as
677 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
679 4. I had overlooked the fact that, in some locales, there are characters for
680 which isalpha() is true but neither isupper() nor islower() are true. In
681 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
682 and ordfeminine) are like this. This affected the treatment of \w and \W
683 when they appeared in character classes, but not when they appeared outside
684 a character class. The bit map for "word" characters is now created
685 separately from the results of isalnum() instead of just taking it from the
686 upper, lower, and digit maps. (Plus the underscore character, of course.)
688 5. The above bug also affected the handling of POSIX character classes such as
689 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
690 permanent tables. Instead, the bit maps for such a class were previously
691 created as the appropriate unions of the upper, lower, and digit bitmaps.
692 Now they are created by subtraction from the [[:word:]] class, which has
693 its own bitmap.
695 6. The [[:blank:]] character class matches horizontal, but not vertical space.
696 It is created by subtracting the vertical space characters (\x09, \x0a,
697 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
698 subtraction was done in the overall bitmap for a character class, meaning
699 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
700 be recognized. This bug has been fixed.
702 7. Patches from the folks at Google:
704 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
705 real life, but is still worth protecting against".
707 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
708 regular expressions".
710 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
711 have it.
713 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
714 "configure" and the latter not, in order to fix a problem somebody had
715 with compiling the Arg class on HP-UX.
717 (e) Improve the error-handling of the C++ wrapper a little bit.
719 (f) New tests for checking recursion limiting.
721 8. The pcre_memmove() function, which is used only if the environment does not
722 have a standard memmove() function (and is therefore rarely compiled),
723 contained two bugs: (a) use of int instead of size_t, and (b) it was not
724 returning a result (though PCRE never actually uses the result).
726 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
727 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
728 returned instead of calling malloc() with an overflowing number that would
729 most likely cause subsequent chaos.
731 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
733 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
734 with this option is matched, the nmatch and pmatch options of regexec() are
735 ignored.
737 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
738 provided in case anyone wants to the the POSIX interface with UTF-8
739 strings.
741 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
742 C++ linking (needed for some HP-UX environments).
744 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
745 (unused parameter) and in the pcre_printint() function (omitted "default"
746 switch label when the default is to do nothing).
748 15. Added some code to make it possible, when PCRE is compiled as a C++
749 library, to replace subject pointers for pcre_exec() with a smart pointer
750 class, thus making it possible to process discontinuous strings.
752 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
753 much the same function. They were added by different people who were trying
754 to make PCRE easy to compile on non-Unix systems. It has been suggested
755 that PCRE_EXPORT be abolished now that there is more automatic apparatus
756 for compiling on Windows systems. I have therefore replaced it with
757 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
758 defaults to "extern" for C or "extern C" for C++, which works fine on
759 Unix-like systems. It is now possible to override the value of PCRE_DATA_
760 SCOPE with something explicit in config.h. In addition:
762 (a) pcreposix.h still had just "extern" instead of either of these macros;
763 I have replaced it with PCRE_DATA_SCOPE.
765 (b) Functions such as _pcre_xclass(), which are internal to the library,
766 but external in the C sense, all had PCRE_EXPORT in their definitions.
767 This is apparently wrong for the Windows case, so I have removed it.
768 (It makes no difference on Unix-like systems.)
770 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
771 of recursive calls to match(). This is different to MATCH_LIMIT because
772 that limits the total number of calls to match(), not all of which increase
773 the depth of recursion. Limiting the recursion depth limits the amount of
774 stack (or heap if NO_RECURSE is set) that is used. The default can be set
775 when PCRE is compiled, and changed at run time. A patch from Google adds
776 this functionality to the C++ interface.
778 18. Changes to the handling of Unicode character properties:
780 (a) Updated the table to Unicode 4.1.0.
782 (b) Recognize characters that are not in the table as "Cn" (undefined).
784 (c) I revised the way the table is implemented to a much improved format
785 which includes recognition of ranges. It now supports the ranges that
786 are defined in UnicodeData.txt, and it also amalgamates other
787 characters into ranges. This has reduced the number of entries in the
788 table from around 16,000 to around 3,000, thus reducing its size
789 considerably. I realized I did not need to use a tree structure after
790 all - a binary chop search is just as efficient. Having reduced the
791 number of entries, I extended their size from 6 bytes to 8 bytes to
792 allow for more data.
794 (d) Added support for Unicode script names via properties such as \p{Han}.
796 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
797 matching that character.
799 20. When matching a repeated Unicode property with a minimum greater than zero,
800 (for example \pL{2,}), PCRE could look past the end of the subject if it
801 reached it while seeking the minimum number of characters. This could
802 happen only if some of the characters were more than one byte long, because
803 there is a check for at least the minimum number of bytes.
805 21. Refactored the implementation of \p and \P so as to be more general, to
806 allow for more different types of property in future. This has changed the
807 compiled form incompatibly. Anybody with saved compiled patterns that use
808 \p or \P will have to recompile them.
810 22. Added "Any" and "L&" to the supported property types.
812 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
813 but give a compile time error if the value is greater than 0xff.
815 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
816 accidentally not being installed or uninstalled.
818 25. The pcre.h file was built from pcre.h.in, but the only changes that were
819 made were to insert the current release number. This seemed silly, because
820 it made things harder for people building PCRE on systems that don't run
821 "configure". I have turned pcre.h into a distributed file, no longer built
822 by "configure", with the version identification directly included. There is
823 no longer a pcre.h.in file.
825 However, this change necessitated a change to the pcre-config script as
826 well. It is built from pcre-config.in, and one of the substitutions was the
827 release number. I have updated configure.ac so that ./configure now finds
828 the release number by grepping pcre.h.
830 26. Added the ability to run the tests under valgrind.
833 Version 6.4 05-Sep-05
834 ---------------------
836 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
837 "--" to be printed when multiple files were scanned, even when none of the
838 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
839 consider it to be a bug, and have restored the previous behaviour.
841 2. A couple of code tidies to get rid of compiler warnings.
843 3. The pcretest program used to cheat by referring to symbols in the library
844 whose names begin with _pcre_. These are internal symbols that are not
845 really supposed to be visible externally, and in some environments it is
846 possible to suppress them. The cheating is now confined to including
847 certain files from the library's source, which is a bit cleaner.
849 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
850 file's purpose clearer.
852 5. Reorganized pcre_ucp_findchar().
855 Version 6.3 15-Aug-05
856 ---------------------
858 1. The file libpcre.pc.in did not have general read permission in the tarball.
860 2. There were some problems when building without C++ support:
862 (a) If C++ support was not built, "make install" and "make test" still
863 tried to test it.
865 (b) There were problems when the value of CXX was explicitly set. Some
866 changes have been made to try to fix these, and ...
868 (c) --disable-cpp can now be used to explicitly disable C++ support.
870 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
871 backslash in a target when C++ was disabled. This confuses some
872 versions of "make", apparently. Using an intermediate variable solves
873 this. (Same for CPP_LOBJ.)
875 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
876 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
877 necessary on certain architectures.
879 4. Added a setting of -export-symbols-regex to the link command to remove
880 those symbols that are exported in the C sense, but actually are local
881 within the library, and not documented. Their names all begin with
882 "_pcre_". This is not a perfect job, because (a) we have to except some
883 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
884 available (and never for static libraries). I have made a note to try to
885 find a way round (a) in the future.
888 Version 6.2 01-Aug-05
889 ---------------------
891 1. There was no test for integer overflow of quantifier values. A construction
892 such as {1111111111111111} would give undefined results. What is worse, if
893 a minimum quantifier for a parenthesized subpattern overflowed and became
894 negative, the calculation of the memory size went wrong. This could have
895 led to memory overwriting.
897 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
899 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
900 operating environments where this matters.
902 4. Applied Giuseppe Maxia's patch to add additional features for controlling
903 PCRE options from within the C++ wrapper.
905 5. Named capturing subpatterns were not being correctly counted when a pattern
906 was compiled. This caused two problems: (a) If there were more than 100
907 such subpatterns, the calculation of the memory needed for the whole
908 compiled pattern went wrong, leading to an overflow error. (b) Numerical
909 back references of the form \12, where the number was greater than 9, were
910 not recognized as back references, even though there were sufficient
911 previous subpatterns.
913 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
914 versions of gcc, e.g. 2.95.4.
917 Version 6.1 21-Jun-05
918 ---------------------
920 1. There was one reference to the variable "posix" in pcretest.c that was not
921 surrounded by "#if !defined NOPOSIX".
923 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
924 the cross-check on the old pcre_info() function, for the benefit of the
925 cut-down version of PCRE that is currently imported into Exim.
927 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
928 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
929 patterns that start (?i) but I don't think that matters. The use of (?i) is
930 just an example; this all applies to the other options as well.
932 4. Since libtool seems to echo the compile commands it is issuing, the output
933 from "make" can be reduced a bit by putting "@" in front of each libtool
934 compile command.
936 5. Patch from the folks at Google for configure.in to be a bit more thorough
937 in checking for a suitable C++ installation before trying to compile the
938 C++ stuff. This should fix a reported problem when a compiler was present,
939 but no suitable headers.
941 6. The man pages all had just "PCRE" as their title. I have changed them to
942 be the relevant file name. I have also arranged that these names are
943 retained in the file doc/pcre.txt, which is a concatenation in text format
944 of all the man pages except the little individual ones for each function.
946 7. The NON-UNIX-USE file had not been updated for the different set of source
947 files that come with release 6. I also added a few comments about the C++
948 wrapper.
951 Version 6.0 07-Jun-05
952 ---------------------
954 1. Some minor internal re-organization to help with my DFA experiments.
956 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
957 didn't matter for the library itself when fully configured, but did matter
958 when compiling without UCP support, or within Exim, where the ucp files are
959 not imported.
961 3. Refactoring of the library code to split up the various functions into
962 different source modules. The addition of the new DFA matching code (see
963 below) to a single monolithic source would have made it really too
964 unwieldy, quite apart from causing all the code to be include in a
965 statically linked application, when only some functions are used. This is
966 relevant even without the DFA addition now that patterns can be compiled in
967 one application and matched in another.
969 The downside of splitting up is that there have to be some external
970 functions and data tables that are used internally in different modules of
971 the library but which are not part of the API. These have all had their
972 names changed to start with "_pcre_" so that they are unlikely to clash
973 with other external names.
975 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
976 a different (DFA) algorithm. Although it is slower than the original
977 function, it does have some advantages for certain types of matching
978 problem.
980 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
981 including restarting after a partial match.
983 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
984 defined when compiling for Windows was sent to me. I have put it into the
985 code, though I have no means of testing or verifying it.
987 7. Added the pcre_refcount() auxiliary function.
989 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
990 match before or at the first newline in the subject string. In pcretest,
991 the /f option on a pattern can be used to set this.
993 9. A repeated \w when used in UTF-8 mode with characters greater than 256
994 would behave wrongly. This has been present in PCRE since release 4.0.
996 10. A number of changes to the pcregrep command:
998 (a) Refactored how -x works; insert ^(...)$ instead of setting
999 PCRE_ANCHORED and checking the length, in preparation for adding
1000 something similar for -w.
1002 (b) Added the -w (match as a word) option.
1004 (c) Refactored the way lines are read and buffered so as to have more
1005 than one at a time available.
1007 (d) Implemented a pcregrep test script.
1009 (e) Added the -M (multiline match) option. This allows patterns to match
1010 over several lines of the subject. The buffering ensures that at least
1011 8K, or the rest of the document (whichever is the shorter) is available
1012 for matching (and similarly the previous 8K for lookbehind assertions).
1014 (f) Changed the --help output so that it now says
1016 -w, --word-regex(p)
1018 instead of two lines, one with "regex" and the other with "regexp"
1019 because that confused at least one person since the short forms are the
1020 same. (This required a bit of code, as the output is generated
1021 automatically from a table. It wasn't just a text change.)
1023 (g) -- can be used to terminate pcregrep options if the next thing isn't an
1024 option but starts with a hyphen. Could be a pattern or a path name
1025 starting with a hyphen, for instance.
1027 (h) "-" can be given as a file name to represent stdin.
1029 (i) When file names are being printed, "(standard input)" is used for
1030 the standard input, for compatibility with GNU grep. Previously
1031 "<stdin>" was used.
1033 (j) The option --label=xxx can be used to supply a name to be used for
1034 stdin when file names are being printed. There is no short form.
1036 (k) Re-factored the options decoding logic because we are going to add
1037 two more options that take data. Such options can now be given in four
1038 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1040 (l) Added the -A, -B, and -C options for requesting that lines of context
1041 around matches be printed.
1043 (m) Added the -L option to print the names of files that do not contain
1044 any matching lines, that is, the complement of -l.
1046 (n) The return code is 2 if any file cannot be opened, but pcregrep does
1047 continue to scan other files.
1049 (o) The -s option was incorrectly implemented. For compatibility with other
1050 greps, it now suppresses the error message for a non-existent or non-
1051 accessible file (but not the return code). There is a new option called
1052 -q that suppresses the output of matching lines, which was what -s was
1053 previously doing.
1055 (p) Added --include and --exclude options to specify files for inclusion
1056 and exclusion when recursing.
1058 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1059 Hopefully, it now does.
1061 12. Missing cast in pcre_study().
1063 13. Added an "uninstall" target to the makefile.
1065 14. Replaced "extern" in the function prototypes in Makefile.in with
1066 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1067 world, but is set differently for Windows.
1069 15. Added a second compiling function called pcre_compile2(). The only
1070 difference is that it has an extra argument, which is a pointer to an
1071 integer error code. When there is a compile-time failure, this is set
1072 non-zero, in addition to the error test pointer being set to point to an
1073 error message. The new argument may be NULL if no error number is required
1074 (but then you may as well call pcre_compile(), which is now just a
1075 wrapper). This facility is provided because some applications need a
1076 numeric error indication, but it has also enabled me to tidy up the way
1077 compile-time errors are handled in the POSIX wrapper.
1079 16. Added VPATH=.libs to the makefile; this should help when building with one
1080 prefix path and installing with another. (Or so I'm told by someone who
1081 knows more about this stuff than I do.)
1083 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1084 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1085 match everything, including newlines. This is not POSIX-compatible, but
1086 somebody wanted the feature. From pcretest it can be activated by using
1087 both the P and the s flags.
1089 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1091 19. libpcre.pc was being incorrectly installed as executable.
1093 20. A couple of places in pcretest check for end-of-line by looking for '\n';
1094 it now also looks for '\r' so that it will work unmodified on Windows.
1096 21. Added Google's contributed C++ wrapper to the distribution.
1098 22. Added some untidy missing memory free() calls in pcretest, to keep
1099 Electric Fence happy when testing.
1103 Version 5.0 13-Sep-04
1104 ---------------------
1106 1. Internal change: literal characters are no longer packed up into items
1107 containing multiple characters in a single byte-string. Each character
1108 is now matched using a separate opcode. However, there may be more than one
1109 byte in the character in UTF-8 mode.
1111 2. The pcre_callout_block structure has two new fields: pattern_position and
1112 next_item_length. These contain the offset in the pattern to the next match
1113 item, and its length, respectively.
1115 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1116 insertion of callouts before each pattern item. Added the /C option to
1117 pcretest to make use of this.
1119 4. On the advice of a Windows user, the lines
1121 #if defined(_WIN32) || defined(WIN32)
1122 _setmode( _fileno( stdout ), 0x8000 );
1123 #endif /* defined(_WIN32) || defined(WIN32) */
1125 have been added to the source of pcretest. This apparently does useful
1126 magic in relation to line terminators.
1128 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1129 for the benefit of those environments where the "b" makes a difference.
1131 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1132 to know about it. I have put a hack into configure.in that adds in code
1133 to set GCC=yes if CC=icc. This seems to end up at a point in the
1134 generated configure script that is early enough to affect the setting of
1135 compiler options, which is what is needed, but I have no means of testing
1136 whether it really works. (The user who reported this had patched the
1137 generated configure script, which of course I cannot do.)
1139 LATER: After change 22 below (new libtool files), the configure script
1140 seems to know about icc (and also ecc). Therefore, I have commented out
1141 this hack in configure.in.
1143 7. Added support for pkg-config (2 patches were sent in).
1145 8. Negated POSIX character classes that used a combination of internal tables
1146 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1147 [[:^ascii]]. Typically, they would match almost any characters. The other
1148 POSIX classes were not broken in this way.
1150 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1151 to find the match, as PCRE was deluded into thinking that the match had to
1152 start at the start point or following a newline. The same bug applied to
1153 patterns with negative forward assertions or any backward assertions
1154 preceding ".*" at the start, unless the pattern required a fixed first
1155 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1157 10. In UTF-8 mode, when moving forwards in the subject after a failed match
1158 starting at the last subject character, bytes beyond the end of the subject
1159 string were read.
1161 11. Renamed the variable "class" as "classbits" to make life easier for C++
1162 users. (Previously there was a macro definition, but it apparently wasn't
1163 enough.)
1165 12. Added the new field "tables" to the extra data so that tables can be passed
1166 in at exec time, or the internal tables can be re-selected. This allows
1167 a compiled regex to be saved and re-used at a later time by a different
1168 program that might have everything at different addresses.
1170 13. Modified the pcre-config script so that, when run on Solaris, it shows a
1171 -R library as well as a -L library.
1173 14. The debugging options of pcretest (-d on the command line or D on a
1174 pattern) showed incorrect output for anything following an extended class
1175 that contained multibyte characters and which was followed by a quantifier.
1177 15. Added optional support for general category Unicode character properties
1178 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1179 support. It adds about 90K to the size of the library. The meanings of the
1180 inbuilt class escapes such as \d and \s have NOT been changed.
1182 16. Updated pcredemo.c to include calls to free() to release the memory for the
1183 compiled pattern.
1185 17. The generated file chartables.c was being created in the source directory
1186 instead of in the building directory. This caused the build to fail if the
1187 source directory was different from the building directory, and was
1188 read-only.
1190 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1191 file. No doubt somebody will tell me if they don't make sense... Also added
1192 Dan Mooney's comments about building on OpenVMS.
1194 19. Added support for partial matching via the PCRE_PARTIAL option for
1195 pcre_exec() and the \P data escape in pcretest.
1197 20. Extended pcretest with 3 new pattern features:
1199 (i) A pattern option of the form ">rest-of-line" causes pcretest to
1200 write the compiled pattern to the file whose name is "rest-of-line".
1201 This is a straight binary dump of the data, with the saved pointer to
1202 the character tables forced to be NULL. The study data, if any, is
1203 written too. After writing, pcretest reads a new pattern.
1205 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1206 compiled pattern from the given file. There must not be any
1207 occurrences of "<" in the file name (pretty unlikely); if there are,
1208 pcretest will instead treat the initial "<" as a pattern delimiter.
1209 After reading in the pattern, pcretest goes on to read data lines as
1210 usual.
1212 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1213 and 16-bit fields in a compiled pattern, to simulate a pattern that
1214 was compiled on a host of opposite endianness.
1216 21. The pcre-exec() function can now cope with patterns that were compiled on
1217 hosts of opposite endianness, with this restriction:
1219 As for any compiled expression that is saved and used later, the tables
1220 pointer field cannot be preserved; the extra_data field in the arguments
1221 to pcre_exec() should be used to pass in a tables address if a value
1222 other than the default internal tables were used at compile time.
1224 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1225 now diagnosed as an error. Previously, most of the time, a negative number
1226 would have been treated as zero, but if in addition "ovector" was passed as
1227 NULL, a crash could occur.
1229 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1230 new versions from the libtool 1.5 distribution (the last one is a copy of
1231 a file called libtool.m4). This seems to have fixed the need to patch
1232 "configure" to support Darwin 1.3 (which I used to do). However, I still
1233 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1234 workstation).
1236 24. Changed the PCRE licence to be the more standard "BSD" licence.
1239 Version 4.5 01-Dec-03
1240 ---------------------
1242 1. There has been some re-arrangement of the code for the match() function so
1243 that it can be compiled in a version that does not call itself recursively.
1244 Instead, it keeps those local variables that need separate instances for
1245 each "recursion" in a frame on the heap, and gets/frees frames whenever it
1246 needs to "recurse". Keeping track of where control must go is done by means
1247 of setjmp/longjmp. The whole thing is implemented by a set of macros that
1248 hide most of the details from the main code, and operates only if
1249 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1250 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1251 operating.
1253 To make it easier for callers to provide specially tailored get/free
1254 functions for this usage, two new functions, pcre_stack_malloc, and
1255 pcre_stack_free, are used. They are always called in strict stacking order,
1256 and the size of block requested is always the same.
1258 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1259 PCRE has been compiled to use the stack or the heap for recursion. The
1260 -C option of pcretest uses this to show which version is compiled.
1262 A new data escape \S, is added to pcretest; it causes the amounts of store
1263 obtained and freed by both kinds of malloc/free at match time to be added
1264 to the output.
1266 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1267 what's available on my current Linux desktop machine.
1269 3. When matching a UTF-8 string, the test for a valid string at the start has
1270 been extended. If start_offset is not zero, PCRE now checks that it points
1271 to a byte that is the start of a UTF-8 character. If not, it returns
1272 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1273 this is necessary because there may be backward assertions in the pattern.
1274 When matching the same subject several times, it may save resources to use
1275 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1277 4. The code for checking the validity of UTF-8 strings has been tightened so
1278 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1279 containing "overlong sequences".
1281 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1282 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1283 should have been "&", but it just so happened that all the cases this let
1284 through by mistake were picked up later in the function.
1286 6. I had used a variable called "isblank" - this is a C99 function, causing
1287 some compilers to warn. To avoid this, I renamed it (as "blankclass").
1289 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1290 prompting; (b) run "./pcretest /dev/null" at the start of the test script
1291 so the version is shown; (c) stop "make test" echoing "./RunTest".
1293 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1295 9. The prototype for memmove() for systems that don't have it was using
1296 size_t, but the inclusion of the header that defines size_t was later. I've
1297 moved the #includes for the C headers earlier to avoid this.
1299 10. Added some adjustments to the code to make it easier to compiler on certain
1300 special systems:
1302 (a) Some "const" qualifiers were missing.
1303 (b) Added the macro EXPORT before all exported functions; by default this
1304 is defined to be empty.
1305 (c) Changed the dftables auxiliary program (that builds chartables.c) so
1306 that it reads its output file name as an argument instead of writing
1307 to the standard output and assuming this can be redirected.
1309 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1310 class containing characters with values greater than 255, PCRE compilation
1311 went into a loop.
1313 12. A recursive reference to a subpattern that was within another subpattern
1314 that had a minimum quantifier of zero caused PCRE to crash. For example,
1315 (x(y(?2))z)? provoked this bug with a subject that got as far as the
1316 recursion. If the recursively-called subpattern itself had a zero repeat,
1317 that was OK.
1319 13. In pcretest, the buffer for reading a data line was set at 30K, but the
1320 buffer into which it was copied (for escape processing) was still set at
1321 1024, so long lines caused crashes.
1323 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1324 "internal error: code overflow...". This applied to any character class
1325 that was followed by a possessive quantifier.
1327 15. Modified the Makefile to add libpcre.la as a prerequisite for
1328 libpcreposix.la because I was told this is needed for a parallel build to
1329 work.
1331 16. If a pattern that contained .* following optional items at the start was
1332 studied, the wrong optimizing data was generated, leading to matching
1333 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1334 matching string must start with a or b or c. The correct conclusion for
1335 this pattern is that a match can start with any character.
1338 Version 4.4 13-Aug-03
1339 ---------------------
1341 1. In UTF-8 mode, a character class containing characters with values between
1342 127 and 255 was not handled correctly if the compiled pattern was studied.
1343 In fixing this, I have also improved the studying algorithm for such
1344 classes (slightly).
1346 2. Three internal functions had redundant arguments passed to them. Removal
1347 might give a very teeny performance improvement.
1349 3. Documentation bug: the value of the capture_top field in a callout is *one
1350 more than* the number of the hightest numbered captured substring.
1352 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1353 in incorrectly linking with a previously installed version. They now link
1354 explicitly with libpcre.la.
1356 5. configure.in no longer needs to recognize Cygwin specially.
1358 6. A problem in pcre.in for Windows platforms is fixed.
1360 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1361 pcretest, it used to include the size of the study block as part of its
1362 output. Unfortunately, the structure contains a field that has a different
1363 size on different hardware architectures. This meant that the tests that
1364 showed this size failed. As the block is currently always of a fixed size,
1365 this information isn't actually particularly useful in pcretest output, so
1366 I have just removed it.
1368 8. Three pre-processor statements accidentally did not start in column 1.
1369 Sadly, there are *still* compilers around that complain, even though
1370 standard C has not required this for well over a decade. Sigh.
1372 9. In pcretest, the code for checking callouts passed small integers in the
1373 callout_data field, which is a void * field. However, some picky compilers
1374 complained about the casts involved for this on 64-bit systems. Now
1375 pcretest passes the address of the small integer instead, which should get
1376 rid of the warnings.
1378 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1379 both compile and run time, and gives an error if an invalid UTF-8 sequence
1380 is found. There is a option for disabling this check in cases where the
1381 string is known to be correct and/or the maximum performance is wanted.
1383 11. In response to a bug report, I changed one line in Makefile.in from
1385 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1386 to
1387 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1389 to look similar to other lines, but I have no way of telling whether this
1390 is the right thing to do, as I do not use Windows. No doubt I'll get told
1391 if it's wrong...
1394 Version 4.3 21-May-03
1395 ---------------------
1397 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1398 Makefile.
1400 2. Some refactoring to improve the quality of the code:
1402 (i) The utf8_table... variables are now declared "const".
1404 (ii) The code for \cx, which used the "case flipping" table to upper case
1405 lower case letters, now just substracts 32. This is ASCII-specific,
1406 but the whole concept of \cx is ASCII-specific, so it seems
1407 reasonable.
1409 (iii) PCRE was using its character types table to recognize decimal and
1410 hexadecimal digits in the pattern. This is silly, because it handles
1411 only 0-9, a-f, and A-F, but the character types table is locale-
1412 specific, which means strange things might happen. A private
1413 table is now used for this - though it costs 256 bytes, a table is
1414 much faster than multiple explicit tests. Of course, the standard
1415 character types table is still used for matching digits in subject
1416 strings against \d.
1418 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1419 ending in _t are). So I've renamed it as ESC_tee.
1421 3. The first argument for regexec() in the POSIX wrapper should have been
1422 defined as "const".
1424 4. Changed pcretest to use malloc() for its buffers so that they can be
1425 Electric Fenced for debugging.
1427 5. There were several places in the code where, in UTF-8 mode, PCRE would try
1428 to read one or more bytes before the start of the subject string. Often this
1429 had no effect on PCRE's behaviour, but in some circumstances it could
1430 provoke a segmentation fault.
1432 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1433 to try to read one or more bytes before the start of the subject string.
1435 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1436 UTF-8 support could misbehave in various ways if the subject string
1437 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1438 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1439 back over UTF-8 characters.)
1442 Version 4.2 14-Apr-03
1443 ---------------------
1445 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1447 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1448 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1449 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1450 [WIN_PREFIX]: new variable, "cyg" for Cygwin
1451 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1453 Note: automatic setting of the BUILD variables is not yet working
1454 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1455 compile-time but not at link-time
1456 [LINK]: use for linking executables only
1457 make different versions for Windows and non-Windows
1458 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1459 libraries
1460 [LINK_FOR_BUILD]: new variable
1461 [OBJEXT]: use throughout
1462 [EXEEXT]: use throughout
1463 <winshared>: new target
1464 <wininstall>: new target
1465 <dftables.o>: use native compiler
1466 <dftables>: use native linker
1467 <install>: handle Windows platform correctly
1468 <clean>: ditto
1469 <check>: ditto
1470 copy DLL to top builddir before testing
1472 As part of these changes, -no-undefined was removed again. This was reported
1473 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1474 in any case.
1476 3. Some tidies to get rid of compiler warnings:
1478 . In the match_data structure, match_limit was an unsigned long int, whereas
1479 match_call_count was an int. I've made them both unsigned long ints.
1481 . In pcretest the fact that a const uschar * doesn't automatically cast to
1482 a void * provoked a warning.
1484 . Turning on some more compiler warnings threw up some "shadow" variables
1485 and a few more missing casts.
1487 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1488 option, a class that contained a single character with a value between 128
1489 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1491 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1492 option, a class that contained several characters, but with at least one
1493 whose value was between 128 and 255 caused PCRE to crash.
1496 Version 4.1 12-Mar-03
1497 ---------------------
1499 1. Compiling with gcc -pedantic found a couple of places where casts were
1500 needed, and a string in dftables.c that was longer than standard compilers are
1501 required to support.
1503 2. Compiling with Sun's compiler found a few more places where the code could
1504 be tidied up in order to avoid warnings.
1506 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1507 first of these names is deprecated in the latest Autoconf in favour of the name
1508 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1509 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1510 analogy I have changed it to CFLAGS_FOR_BUILD.
1512 4. Added -no-undefined to the linking command in the Makefile, because this is
1513 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1514 linking step for the pcreposix library.
1516 5. PCRE was failing to diagnose the case of two named groups with the same
1517 name.
1519 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1520 literal character that is needed in the subject for a match, and scans along to
1521 ensure that it is present before embarking on the full matching process. This
1522 saves time in cases of nested unlimited repeats that are never going to match.
1523 Problem: the scan can take a lot of time if the subject is very long (e.g.
1524 megabytes), thus penalizing straightforward matches. It is now done only if the
1525 amount of subject to be scanned is less than 1000 bytes.
1527 7. A lesser problem with the same optimization is that it was recording the
1528 first character of an anchored pattern as "needed", thus provoking a search
1529 right along the subject, even when the first match of the pattern was going to
1530 fail. The "needed" character is now not set for anchored patterns, unless it
1531 follows something in the pattern that is of non-fixed length. Thus, it still
1532 fulfils its original purpose of finding quick non-matches in cases of nested
1533 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1536 Version 4.0 17-Feb-03
1537 ---------------------
1539 1. If a comment in an extended regex that started immediately after a meta-item
1540 extended to the end of string, PCRE compiled incorrect data. This could lead to
1541 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1543 2. Moved to autoconf 2.53 and libtool 1.4.2.
1545 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1546 the special perltest8 script is no longer needed - all the tests can be run
1547 from a single perltest script.
1549 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1550 by \s. It has now been removed in PCRE. This means it isn't recognized as
1551 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1552 class [:space:] *does* include VT, thereby creating a mess.
1554 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1555 space and tab.
1557 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1558 its new features into the main test script, reducing the number of scripts.
1560 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1561 were backward compatible, and made the (?i) apply to the whole pattern, as if
1562 /i were given. Now it behaves more logically, and applies the option setting
1563 only to what follows. PCRE has been changed to follow suit. However, if it
1564 finds options settings right at the start of the pattern, it extracts them into
1565 the global options, as before. Thus, they show up in the info data.
1567 8. Added support for the \Q...\E escape sequence. Characters in between are
1568 treated as literals. This is slightly different from Perl in that $ and @ are
1569 also handled as literals inside the quotes. In Perl, they will cause variable
1570 interpolation. Note the following examples:
1572 Pattern PCRE matches Perl matches
1574 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
1575 \Qabc\$xyz\E abc\$xyz abc\$xyz
1576 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
1578 For compatibility with Perl, \Q...\E sequences are recognized inside character
1579 classes as well as outside them.
1581 9. Re-organized 3 code statements in pcretest to avoid "overflow in
1582 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1583 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1584 signed/unsigned warnings.
1586 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1587 option for pcretest, so I've replaced it by a simple function that does just
1588 that job.
1590 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1591 "pcregrep -".
1593 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1594 Java package. This provides some syntactic sugar for simple cases of what my
1595 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1596 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1597 item, you can use this simplified notation. Note that only makes sense with
1598 greedy quantifiers. Consequently, the use of the possessive quantifier forces
1599 greediness, whatever the setting of the PCRE_UNGREEDY option.
1601 13. A change of greediness default within a pattern was not taking effect at
1602 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1603 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1604 was abstracted outside.
1606 14. PCRE now supports the \G assertion. It is true when the current matching
1607 position is at the start point of the match. This differs from \A when the
1608 starting offset is non-zero. Used with the /g option of pcretest (or similar
1609 code), it works in the same way as it does for Perl's /g option. If all
1610 alternatives of a regex begin with \G, the expression is anchored to the start
1611 match position, and the "anchored" flag is set in the compiled expression.
1613 15. Some bugs concerning the handling of certain option changes within patterns
1614 have been fixed. These applied to options other than (?ims). For example,
1615 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1616 been the other way round. Some of this was related to change 7 above.
1618 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1619 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1620 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1621 POSIX classes only within a class (e.g. /[[:alpha:]]/).
1623 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1624 mode. Unlike ".", it always matches newline, whatever the setting of
1625 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1626 assertions. Perl allows it, but it doesn't (in general) work because it can't
1627 calculate the length of the lookbehind. At least, that's the case for Perl
1628 5.8.0 - I've been told they are going to document that it doesn't work in
1629 future.
1631 18. Added an error diagnosis for escapes that PCRE does not support: these are
1632 \L, \l, \N, \P, \p, \U, \u, and \X.
1634 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1635 reading past the end of the pattern in cases such as /[abcd/.
1637 20. PCRE was getting more memory than necessary for patterns with classes that
1638 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1640 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1641 compiling PCRE for use with Virtual Pascal.
1643 22. Small fix to the Makefile to make it work properly if the build is done
1644 outside the source tree.
1646 23. Added a new extension: a condition to go with recursion. If a conditional
1647 subpattern starts with (?(R) the "true" branch is used if recursion has
1648 happened, whereas the "false" branch is used only at the top level.
1650 24. When there was a very long string of literal characters (over 255 bytes
1651 without UTF support, over 250 bytes with UTF support), the computation of how
1652 much memory was required could be incorrect, leading to segfaults or other
1653 strange effects.
1655 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1656 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1657 there was a subsequent back reference to those brackets. This meant that, for
1658 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1659 possible to check for precisely this case. All we can do is abandon the
1660 optimization if .* occurs inside capturing brackets when there are any back
1661 references whatsoever. (See below for a better fix that came later.)
1663 26. The handling of the optimization for finding the first character of a
1664 non-anchored pattern, and for finding a character that is required later in the
1665 match were failing in some cases. This didn't break the matching; it just
1666 failed to optimize when it could. The way this is done has been re-implemented.
1668 27. Fixed typo in error message for invalid (?R item (it said "(?p").
1670 28. Added a new feature that provides some of the functionality that Perl
1671 provides with (?{...}). The facility is termed a "callout". The way it is done
1672 in PCRE is for the caller to provide an optional function, by setting
1673 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1674 global variable. By default it is unset, which disables all calling out. To get
1675 the function called, the regex must include (?C) at appropriate points. This
1676 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1677 This provides a means of identifying different callout points. When PCRE
1678 reaches such a point in the regex, if pcre_callout has been set, the external
1679 function is called. It is provided with data in a structure called
1680 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1681 matching continues; if it returns a non-zero value, the match at the current
1682 point fails. However, backtracking will occur if possible. [This was changed
1683 later and other features added - see item 49 below.]
1685 29. pcretest is upgraded to test the callout functionality. It provides a
1686 callout function that displays information. By default, it shows the start of
1687 the match and the current position in the text. There are some new data escapes
1688 to vary what happens:
1690 \C+ in addition, show current contents of captured substrings
1691 \C- do not supply a callout function
1692 \C!n return 1 when callout number n is reached
1693 \C!n!m return 1 when callout number n is reached for the mth time
1695 30. If pcregrep was called with the -l option and just a single file name, it
1696 output "<stdin>" if a match was found, instead of the file name.
1698 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1699 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1700 pcre_exec(). This saves a malloc/free per call. The default value of
1701 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1702 when configuring.
1704 32. The default maximum size of a compiled pattern is 64K. There have been a
1705 few cases of people hitting this limit. The code now uses macros to handle the
1706 storing of links as offsets within the compiled pattern. It defaults to 2-byte
1707 links, but this can be changed to 3 or 4 bytes by --with-link-size when
1708 configuring. Tests 2 and 5 work only with 2-byte links because they output
1709 debugging information about compiled patterns.
1711 33. Internal code re-arrangements:
1713 (a) Moved the debugging function for printing out a compiled regex into
1714 its own source file (printint.c) and used #include to pull it into
1715 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1716 separate copies.
1718 (b) Defined the list of op-code names for debugging as a macro in
1719 internal.h so that it is next to the definition of the opcodes.
1721 (c) Defined a table of op-code lengths for simpler skipping along compiled
1722 code. This is again a macro in internal.h so that it is next to the
1723 definition of the opcodes.
1725 34. Added support for recursive calls to individual subpatterns, along the
1726 lines of Robin Houston's patch (but implemented somewhat differently).
1728 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1729 allow it to read and process whole directories in Win32. This code was
1730 contributed by Lionel Fourquaux; it has not been tested by me.
1732 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1733 used to name a group. Names consist of alphanumerics and underscores, and must
1734 be unique. Back references use the syntax (?P=name) and recursive calls use
1735 (?P>name) which is a PCRE extension to the Python extension. Groups still have
1736 numbers. The function pcre_fullinfo() can be used after compilation to extract
1737 a name/number map. There are three relevant calls:
1739 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
1740 PCRE_INFO_NAMECOUNT yields the number of entries
1741 PCRE_INFO_NAMETABLE yields a pointer to the map.
1743 The map is a vector of fixed-size entries. The size of each entry depends on
1744 the length of the longest name used. The first two bytes of each entry are the
1745 group number, most significant byte first. There follows the corresponding
1746 name, zero terminated. The names are in alphabetical order.
1748 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1749 case instead of 255. Making it the same both with and without UTF-8 support
1750 means that the same test output works with both.
1752 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1753 calling malloc() with a zero argument.
1755 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1756 optimization. I've improved things by keeping a bitmap of backreferences with
1757 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1758 fact referenced, the optimization can be applied. It is unlikely that a
1759 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1760 the match to follow \n) will appear inside brackets with a number greater than
1761 31, but if it does, any back reference > 31 suppresses the optimization.
1763 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1764 of disabling numbered capturing parentheses. Any opening parenthesis that is
1765 not followed by ? behaves as if it were followed by ?: but named parentheses
1766 can still be used for capturing (and they will acquire numbers in the usual
1767 way).
1769 41. Redesigned the return codes from the match() function into yes/no/error so
1770 that errors can be passed back from deep inside the nested calls. A malloc
1771 failure while inside a recursive subpattern call now causes the
1772 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
1774 42. It is now possible to set a limit on the number of times the match()
1775 function is called in a call to pcre_exec(). This facility makes it possible to
1776 limit the amount of recursion and backtracking, though not in a directly
1777 obvious way, because the match() function is used in a number of different
1778 circumstances. The count starts from zero for each position in the subject
1779 string (for non-anchored patterns). The default limit is, for compatibility, a
1780 large number, namely 10 000 000. You can change this in two ways:
1782 (a) When configuring PCRE before making, you can use --with-match-limit=n
1783 to set a default value for the compiled library.
1785 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
1786 a different value is set. See 45 below.
1788 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1790 43. Added a new function pcre_config(int, void *) to enable run-time extraction
1791 of things that can be changed at compile time. The first argument specifies
1792 what is wanted and the second points to where the information is to be placed.
1793 The current list of available information is:
1797 The output is an integer that is set to one if UTF-8 support is available;
1798 otherwise it is set to zero.
1802 The output is an integer that it set to the value of the code that is used for
1803 newline. It is either LF (10) or CR (13).
1807 The output is an integer that contains the number of bytes used for internal
1808 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
1812 The output is an integer that contains the threshold above which the POSIX
1813 interface uses malloc() for output vectors. See item 31 above.
1817 The output is an unsigned integer that contains the default limit of the number
1818 of match() calls in a pcre_exec() execution. See 42 above.
1820 44. pcretest has been upgraded by the addition of the -C option. This causes it
1821 to extract all the available output from the new pcre_config() function, and to
1822 output it. The program then exits immediately.
1824 45. A need has arisen to pass over additional data with calls to pcre_exec() in
1825 order to support additional features. One way would have been to define
1826 pcre_exec2() (for example) with extra arguments, but this would not have been
1827 extensible, and would also have required all calls to the original function to
1828 be mapped to the new one. Instead, I have chosen to extend the mechanism that
1829 is used for passing in "extra" data from pcre_study().
1831 The pcre_extra structure is now exposed and defined in pcre.h. It currently
1832 contains the following fields:
1834 flags a bitmap indicating which of the following fields are set
1835 study_data opaque data from pcre_study()
1836 match_limit a way of specifying a limit on match() calls for a specific
1837 call to pcre_exec()
1838 callout_data data for callouts (see 49 below)
1840 The flag bits are also defined in pcre.h, and are
1846 The pcre_study() function now returns one of these new pcre_extra blocks, with
1847 the actual study data pointed to by the study_data field, and the
1848 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
1849 before. That is, this change is entirely upwards-compatible and requires no
1850 change to existing code.
1852 If you want to pass in additional data to pcre_exec(), you can either place it
1853 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
1854 block.
1856 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
1857 data string contains the escape sequence \M, pcretest calls pcre_exec() several
1858 times with different match limits, until it finds the minimum value needed for
1859 pcre_exec() to complete. The value is then output. This can be instructive; for
1860 most simple matches the number is quite small, but for pathological cases it
1861 gets very large very quickly.
1863 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
1864 returns the size of the data block pointed to by the study_data field in a
1865 pcre_extra block, that is, the value that was passed as the argument to
1866 pcre_malloc() when PCRE was getting memory in which to place the information
1867 created by pcre_study(). The fourth argument should point to a size_t variable.
1868 pcretest has been extended so that this information is shown after a successful
1869 pcre_study() call when information about the compiled regex is being displayed.
1871 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
1872 because what follows is always an absolute path. (Later: it turns out that this
1873 is more than cosmetic for MinGW, because it doesn't like empty path
1874 components.)
1876 49. Some changes have been made to the callout feature (see 28 above):
1878 (i) A callout function now has three choices for what it returns:
1880 0 => success, carry on matching
1881 > 0 => failure at this point, but backtrack if possible
1882 < 0 => serious error, return this value from pcre_exec()
1884 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
1885 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
1886 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
1887 use by callout functions. It will never be used by PCRE itself.
1889 (ii) The pcre_extra structure (see 45 above) has a void * field called
1890 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
1891 pcre_callout_block structure has a field of the same name. The contents of
1892 the field passed in the pcre_extra structure are passed to the callout
1893 function in the corresponding field in the callout block. This makes it
1894 easier to use the same callout-containing regex from multiple threads. For
1895 testing, the pcretest program has a new data escape
1897 \C*n pass the number n (may be negative) as callout_data
1899 If the callout function in pcretest receives a non-zero value as
1900 callout_data, it returns that value.
1902 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
1903 there were some redundant $(CFLAGS) in commands that are now specified as
1904 $(LINK), which already includes $(CFLAGS).
1906 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
1907 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
1908 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
1909 one-byte characters throughout. Note that case-insensitive matching applies
1910 only to characters whose values are less than 256. PCRE doesn't support the
1911 notion of cases for higher-valued characters.
1913 (i) A character class whose characters are all within 0-255 is handled as
1914 a bit map, and the map is inverted for negative classes. Previously, a
1915 character > 255 always failed to match such a class; however it should
1916 match if the class was a negative one (e.g. [^ab]). This has been fixed.
1918 (ii) A negated character class with a single character < 255 is coded as
1919 "not this character" (OP_NOT). This wasn't working properly when the test
1920 character was multibyte, either singly or repeated.
1922 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
1923 mode, for example: \x{100}{2,3}.
1925 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
1926 singly or repeated) now correctly test multibyte characters. However,
1927 PCRE doesn't recognize any characters with values greater than 255 as
1928 digits, spaces, or word characters. Such characters always match \D, \S,
1929 and \W, and never match \d, \s, or \w.
1931 (v) Classes may now contain characters and character ranges with values
1932 greater than 255. For example: [ab\x{100}-\x{400}].
1934 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
1935 PCRE in UTF-8 mode.
1937 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
1938 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
1939 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
1940 value.)
1942 53. The single man page has become too large. I have therefore split it up into
1943 a number of separate man pages. These also give rise to individual HTML pages;
1944 these are now put in a separate directory, and there is an index.html page that
1945 lists them all. Some hyperlinking between the pages has been installed.
1947 54. Added convenience functions for handling named capturing parentheses.
1949 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
1950 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
1951 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
1952 are faulted.
1954 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
1955 calling configure. These values are used when compiling the dftables.c program
1956 which is run to generate the source of the default character tables. They
1957 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
1958 you will need to set these values.
1960 57. Updated the building process for Windows DLL, as provided by Fred Cox.
1963 Version 3.9 02-Jan-02
1964 ---------------------
1966 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
1968 2. If --disable-static was given, the building process failed when trying to
1969 build pcretest and pcregrep. (For some reason it was using libtool to compile
1970 them, which is not right, as they aren't part of the library.)
1973 Version 3.8 18-Dec-01
1974 ---------------------
1976 1. The experimental UTF-8 code was completely screwed up. It was packing the
1977 bytes in the wrong order. How dumb can you get?
1980 Version 3.7 29-Oct-01
1981 ---------------------
1983 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
1984 This caused pcretest, when used on the test data, to segfault. Unfortunately,
1985 this didn't happen under Solaris 8, where I normally test things.
1987 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
1988 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
1989 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
1990 this fix an hour or so after the initial 3.7 release.)
1993 Version 3.6 23-Oct-01
1994 ---------------------
1996 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
1997 offsets passed as NULL with zero offset count.
1999 2. The config.guess and config.sub files had not been updated when I moved to
2000 the latest autoconf.
2003 Version 3.5 15-Aug-01
2004 ---------------------
2006 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2007 had been forgotten.
2009 2. By using declared but undefined structures, we can avoid using "void"
2010 definitions in pcre.h while keeping the internal definitions of the structures
2011 private.
2013 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2014 user point of view, this means that both static and shared libraries are built
2015 by default, but this can be individually controlled. More of the work of
2016 handling this static/shared cases is now inside libtool instead of PCRE's make
2017 file.
2019 4. The pcretest utility is now installed along with pcregrep because it is
2020 useful for users (to test regexs) and by doing this, it automatically gets
2021 relinked by libtool. The documentation has been turned into a man page, so
2022 there are now .1, .txt, and .html versions in /doc.
2024 5. Upgrades to pcregrep:
2025 (i) Added long-form option names like gnu grep.
2026 (ii) Added --help to list all options with an explanatory phrase.
2027 (iii) Added -r, --recursive to recurse into sub-directories.
2028 (iv) Added -f, --file to read patterns from a file.
2030 6. pcre_exec() was referring to its "code" argument before testing that
2031 argument for NULL (and giving an error if it was NULL).
2033 7. Upgraded Makefile.in to allow for compiling in a different directory from
2034 the source directory.
2036 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2037 options bits, the pointer it was passed was to an int instead of to an unsigned
2038 long int. This mattered only on 64-bit systems.
2040 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2041 generated) instead of pcre.in, which it its source. Also made the same change
2042 in several of the .c files.
2044 10. A new release of gcc defines printf() as a macro, which broke pcretest
2045 because it had an ifdef in the middle of a string argument for printf(). Fixed
2046 by using separate calls to printf().
2048 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2049 script, to force use of CR or LF instead of \n in the source. On non-Unix
2050 systems, the value can be set in config.h.
2052 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2053 absolute limit. Changed the text of the error message to make this clear, and
2054 likewise updated the man page.
2056 13. The limit of 99 on the number of capturing subpatterns has been removed.
2057 The new limit is 65535, which I hope will not be a "real" limit.
2060 Version 3.4 22-Aug-00
2061 ---------------------
2063 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2065 2. Diagnose condition (?(0) as an error instead of crashing on matching.
2068 Version 3.3 01-Aug-00
2069 ---------------------
2071 1. If an octal character was given, but the value was greater than \377, it
2072 was not getting masked to the least significant bits, as documented. This could
2073 lead to crashes in some systems.
2075 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2076 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2078 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2079 These just pass their arguments on to (pcre_free)(), but they are provided
2080 because some uses of PCRE bind it to non-C systems that can call its functions,
2081 but cannot call free() or pcre_free() directly.
2083 4. Add "make test" as a synonym for "make check". Corrected some comments in
2084 the Makefile.
2086 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2087 Makefile.
2089 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2090 command called pgrep for grepping around the active processes.
2092 7. Added the beginnings of support for UTF-8 character strings.
2094 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2095 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2096 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2097 out for the ar command.)
2100 Version 3.2 12-May-00
2101 ---------------------
2103 This is purely a bug fixing release.
2105 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2106 of ZA. This was just one example of several cases that could provoke this bug,
2107 which was introduced by change 9 of version 2.00. The code for breaking
2108 infinite loops after an iteration that matches an empty string was't working
2109 correctly.
2111 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2112 when matched against abbab (for example). After matching an empty string, it
2113 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2114 caused it to match further down the string than it should.
2116 3. The code contained an inclusion of sys/types.h. It isn't clear why this
2117 was there because it doesn't seem to be needed, and it causes trouble on some
2118 systems, as it is not a Standard C header. It has been removed.
2120 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2121 were reported on the Macintosh. The changes were from
2123 while ((c = *(++ptr)) != 0 && c != '\n');
2124 to
2125 while ((c = *(++ptr)) != 0 && c != '\n') ;
2127 Totally extraordinary, but if that's what it takes...
2129 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2130 available. Added HAVE_BCOPY and an autoconf test for it; if neither
2131 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2132 assumes the way PCRE uses memmove() (always moving upwards).
2134 6. PCRE is being used in one environment where strchr() is not available. There
2135 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2136 faster code anyway.
2139 Version 3.1 09-Feb-00
2140 ---------------------
2142 The only change in this release is the fixing of some bugs in Makefile.in for
2143 the "install" target:
2145 (1) It was failing to install pcreposix.h.
2147 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2150 Version 3.0 01-Feb-00
2151 ---------------------
2153 1. Add support for the /+ modifier to perltest (to output $` like it does in
2154 pcretest).
2156 2. Add support for the /g modifier to perltest.
2158 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2159 matches null strings.
2161 4. Fix perltest so that it doesn't do unwanted things when fed an empty
2162 pattern. Perl treats empty patterns specially - it reuses the most recent
2163 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2164 effect.
2166 5. The POSIX interface was broken in that it was just handing over the POSIX
2167 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2168 required a bigger vector, with some working space on the end. This means that
2169 the POSIX wrapper now has to get and free some memory, and copy the results.
2171 6. Added some simple autoconf support, placing the test data and the
2172 documentation in separate directories, re-organizing some of the
2173 information files, and making it build pcre-config (a GNU standard). Also added
2174 libtool support for building PCRE as a shared library, which is now the
2175 default.
2177 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2178 09 are not valid octal constants. Single digits will be used for minor values
2179 less than 10.
2181 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2182 existing programs that set these in the POSIX interface can use PCRE without
2183 modification.
2185 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2186 return all that pcre_info() returns, plus additional data. The pcre_info()
2187 function is retained for compatibility, but is considered to be obsolete.
2189 10. Added experimental recursion feature (?R) to handle one common case that
2190 Perl 5.6 will be able to do with (?p{...}).
2192 11. Added support for POSIX character classes like [:alpha:], which Perl is
2193 adopting.
2196 Version 2.08 31-Aug-99
2197 ----------------------
2199 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2200 trying to match at the startoffset position, but instead was moving forward to
2201 the next newline as if a previous match had failed.
2203 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2204 and could get into a loop if a null string was matched other than at the start
2205 of the subject.
2207 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2208 be distinguished at compile time, and for completeness also added PCRE_DATE.
2210 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2211 in GnuWin32 environments.
2214 Version 2.07 29-Jul-99
2215 ----------------------
2217 1. The documentation is now supplied in plain text form and HTML as well as in
2218 the form of man page sources.
2220 2. C++ compilers don't like assigning (void *) values to other pointer types.
2221 In particular this affects malloc(). Although there is no problem in Standard
2222 C, I've put in casts to keep C++ compilers happy.
2224 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2225 should be (const char *).
2227 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2228 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2229 However, I haven't made this a standard facility. The documentation doesn't
2230 mention it, and the Makefile doesn't support it.
2232 5. The Makefile now contains an "install" target, with editable destinations at
2233 the top of the file. The pcretest program is not installed.
2235 6. pgrep -V now gives the PCRE version number and date.
2237 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2238 causing the entire string to be ignored, instead of just the last character.
2240 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2241 non-matching string, it can take a very, very long time, even for strings of
2242 quite modest length, because of the nested recursion. PCRE now does better in
2243 some of these cases. It does this by remembering the last required literal
2244 character in the pattern, and pre-searching the subject to ensure it is present
2245 before running the real match. In other words, it applies a heuristic to detect
2246 some types of certain failure quickly, and in the above example, if presented
2247 with a string that has no trailing " it gives "no match" very quickly.
2249 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2250 other alternatives are tried instead.
2253 Version 2.06 09-Jun-99
2254 ----------------------
2256 1. Change pcretest's output for amount of store used to show just the code
2257 space, because the remainder (the data block) varies in size between 32-bit and
2258 64-bit systems.
2260 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2261 start matching at. This allows lookbehinds to work when searching for multiple
2262 occurrences in a string.
2264 3. Added additional options to pcretest for testing multiple occurrences:
2266 /+ outputs the rest of the string that follows a match
2267 /g loops for multiple occurrences, using the new startoffset argument
2268 /G loops for multiple occurrences by passing an incremented pointer
2270 4. PCRE wasn't doing the "first character" optimization for patterns starting
2271 with \b or \B, though it was doing it for other lookbehind assertions. That is,
2272 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2273 the letter 'x'. On long subject strings, this gives a significant speed-up.
2276 Version 2.05 21-Apr-99
2277 ----------------------
2279 1. Changed the type of magic_number from int to long int so that it works
2280 properly on 16-bit systems.
2282 2. Fixed a bug which caused patterns starting with .* not to work correctly
2283 when the subject string contained newline characters. PCRE was assuming
2284 anchoring for such patterns in all cases, which is not correct because .* will
2285 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2286 DOTALL is set at top level; otherwise it knows that patterns starting with .*
2287 must be retried after every newline in the subject.
2290 Version 2.04 18-Feb-99
2291 ----------------------
2293 1. For parenthesized subpatterns with repeats whose minimum was zero, the
2294 computation of the store needed to hold the pattern was incorrect (too large).
2295 If such patterns were nested a few deep, this could multiply and become a real
2296 problem.
2298 2. Added /M option to pcretest to show the memory requirement of a specific
2299 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2301 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2302 compiled in such a way that the backtracking after subsequent failure was
2303 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2304 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2307 Version 2.03 02-Feb-99
2308 ----------------------
2310 1. Fixed typo and small mistake in man page.
2312 2. Added 4th condition (GPL supersedes if conflict) and created separate
2313 LICENCE file containing the conditions.
2315 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2316 Perl, that is the internal \ allows the delimiter to be included in the
2317 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2318 the final delimiter, add \ to the end of the pattern (to test the error).
2320 4. Added the convenience functions for extracting substrings after a successful
2321 match. Updated pcretest to make it able to test these functions.
2324 Version 2.02 14-Jan-99
2325 ----------------------
2327 1. Initialized the working variables associated with each extraction so that
2328 their saving and restoring doesn't refer to uninitialized store.
2330 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2331 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2332 fix the problem.
2334 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2335 calls, and wasn't printing the correct value for compiling calls. Increased the
2336 default value of LOOPREPEAT, and the number of significant figures in the
2337 times.
2339 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2341 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2342 a building problem on Windows NT with a FAT file system.
2345 Version 2.01 21-Oct-98
2346 ----------------------
2348 1. Changed the API for pcre_compile() to allow for the provision of a pointer
2349 to character tables built by pcre_maketables() in the current locale. If NULL
2350 is passed, the default tables are used.
2353 Version 2.00 24-Sep-98
2354 ----------------------
2356 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2357 it any more.
2359 2. Allow quantification of (?>) groups, and make it work correctly.
2361 3. The first character computation wasn't working for (?>) groups.
2363 4. Correct the implementation of \Z (it is permitted to match on the \n at the
2364 end of the subject) and add 5.005's \z, which really does match only at the
2365 very end of the subject.
2367 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2369 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2370 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2371 localized options. All options to pcre_study() were also removed.
2373 7. Add other new features from 5.005:
2375 $(?<= positive lookbehind
2376 $(?<! negative lookbehind
2377 (?imsx-imsx) added the unsetting capability
2378 such a setting is global if at outer level; local otherwise
2379 (?imsx-imsx:) non-capturing groups with option setting
2380 (?(cond)re|re) conditional pattern matching
2382 A backreference to itself in a repeated group matches the previous
2383 captured string.
2385 8. General tidying up of studying (both automatic and via "study")
2386 consequential on the addition of new assertions.
2388 9. As in 5.005, unlimited repeated groups that could match an empty substring
2389 are no longer faulted at compile time. Instead, the loop is forcibly broken at
2390 runtime if any iteration does actually match an empty substring.
2392 10. Include the RunTest script in the distribution.
2394 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2395 discrepancies, some of which were old and were also with respect to 5.004. They
2396 have now been fixed.
2399 Version 1.09 28-Apr-98
2400 ----------------------
2402 1. A negated single character class followed by a quantifier with a minimum
2403 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
2404 program crashes, or just wrong answers. This did not apply to negated classes
2405 containing more than one character, or to minima other than one.
2408 Version 1.08 27-Mar-98
2409 ----------------------
2411 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
2413 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
2414 latter must appear before anything that relies on it in the pattern.
2417 Version 1.07 16-Feb-98
2418 ----------------------
2420 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
2421 repeat of a potentially empty string).
2424 Version 1.06 23-Jan-98
2425 ----------------------
2427 1. Added Markus Oberhumer's little patches for C++.
2429 2. Literal strings longer than 255 characters were broken.
2432 Version 1.05 23-Dec-97
2433 ----------------------
2435 1. Negated character classes containing more than one character were failing if
2436 PCRE_CASELESS was set at run time.
2439 Version 1.04 19-Dec-97
2440 ----------------------
2442 1. Corrected the man page, where some "const" qualifiers had been omitted.
2444 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
2445 input syntax.
2447 3. Fixed memory leak which occurred when a regex with back references was
2448 matched with an offsets vector that wasn't big enough. The temporary memory
2449 that is used in this case wasn't being freed if the match failed.
2451 4. Tidied pcretest to ensure it frees memory that it gets.
2453 5. Temporary memory was being obtained in the case where the passed offsets
2454 vector was exactly big enough.
2456 6. Corrected definition of offsetof() from change 5 below.
2458 7. I had screwed up change 6 below and broken the rules for the use of
2459 setjmp(). Now fixed.
2462 Version 1.03 18-Dec-97
2463 ----------------------
2465 1. A erroneous regex with a missing opening parenthesis was correctly
2466 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
2467 on some systems.
2469 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
2470 it was reported that one broken compiler failed on the former because "code" is
2471 also an independent variable.
2473 3. The erroneous regex a[]b caused an array overrun reference.
2475 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
2476 fail on data ending with that character. (It was going on too far, and checking
2477 the next character, typically a binary zero.) This was specific to the
2478 optimized code for single-character negative classes.
2480 5. Added a contributed patch from the TIN world which does the following:
2482 + Add an undef for memmove, in case the the system defines a macro for it.
2484 + Add a definition of offsetof(), in case there isn't one. (I don't know
2485 the reason behind this - offsetof() is part of the ANSI standard - but
2486 it does no harm).
2488 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
2489 most of the places where whitespace preceded '#'. I have given up and
2490 allowed the remaining 2 cases to be at the margin.
2492 + Rename some variables in pcre to eliminate shadowing. This seems very
2493 pedantic, but does no harm, of course.
2495 6. Moved the call to setjmp() into its own function, to get rid of warnings
2496 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
2498 7. Constructs such as \d{8,} were compiling into the equivalent of
2499 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
2500 outcome, but in this particular case used more store than had been allocated,
2501 which caused the bug to be discovered because it threw up an internal error.
2503 8. The debugging code in both pcre and pcretest for outputting the compiled
2504 form of a regex was going wrong in the case of back references followed by
2505 curly-bracketed repeats.
2508 Version 1.02 12-Dec-97
2509 ----------------------
2511 1. Typos in pcre.3 and comments in the source fixed.
2513 2. Applied a contributed patch to get rid of places where it used to remove
2514 'const' from variables, and fixed some signed/unsigned and uninitialized
2515 variable warnings.
2517 3. Added the "runtest" target to Makefile.
2519 4. Set default compiler flag to -O2 rather than just -O.
2522 Version 1.01 19-Nov-97
2523 ----------------------
2525 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
2526 like /([ab]*)*/, that is, for classes with more than one character in them.
2528 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
2529 as /((?>a*))*/ (a PCRE_EXTRA facility).
2532 Version 1.00 18-Nov-97
2533 ----------------------
2535 1. Added compile-time macros to support systems such as SunOS4 which don't have
2536 memmove() or strerror() but have other things that can be used instead.
2538 2. Arranged that "make clean" removes the executables.
2541 Version 0.99 27-Oct-97
2542 ----------------------
2544 1. Fixed bug in code for optimizing classes with only one character. It was
2545 initializing a 32-byte map regardless, which could cause it to run off the end
2546 of the memory it had got.
2548 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
2551 Version 0.98 22-Oct-97
2552 ----------------------
2554 1. Fixed bug in code for handling temporary memory usage when there are more
2555 back references than supplied space in the ovector. This could cause segfaults.
2558 Version 0.97 21-Oct-97
2559 ----------------------
2561 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
2563 2. Optimized negated single characters not to use a bit map.
2565 3. Brought error texts together as macro definitions; clarified some of them;
2566 fixed one that was wrong - it said "range out of order" when it meant "invalid
2567 escape sequence".
2569 4. Changed some char * arguments to const char *.
2571 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
2573 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
2574 pcretest.
2577 Version 0.96 16-Oct-97
2578 ----------------------
2580 1. Added a simple "pgrep" utility to the distribution.
2582 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
2583 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
2584 where "ddd" means "one or more decimal digits".
2586 3. Fixed serious bug. If a pattern had a back reference, but the call to
2587 pcre_exec() didn't supply a large enough ovector to record the related
2588 identifying subpattern, the match always failed. PCRE now remembers the number
2589 of the largest back reference, and gets some temporary memory in which to save
2590 the offsets during matching if necessary, in order to ensure that
2591 backreferences always work.
2593 4. Increased the compatibility with Perl in a number of ways:
2595 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
2596 to request this handling. The option can be set at compile or exec time.
2598 (b) $ matches before a terminating newline by default; an option
2599 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
2600 mode). The option can be set at compile or exec time.
2602 (c) The handling of \ followed by a digit other than 0 is now supposed to be
2603 the same as Perl's. If the decimal number it represents is less than 10
2604 or there aren't that many previous left capturing parentheses, an octal
2605 escape is read. Inside a character class, it's always an octal escape,
2606 even if it is a single digit.
2608 (d) An escaped but undefined alphabetic character is taken as a literal,
2609 unless PCRE_EXTRA is set. Currently this just reserves the remaining
2610 escapes.
2612 (e) {0} is now permitted. (The previous item is removed from the compiled
2613 pattern).
2615 5. Changed all the names of code files so that the basic parts are no longer
2616 than 10 characters, and abolished the teeny "globals.c" file.
2618 6. Changed the handling of character classes; they are now done with a 32-byte
2619 bit map always.
2621 7. Added the -d and /D options to pcretest to make it possible to look at the
2622 internals of compilation without having to recompile pcre.
2625 Version 0.95 23-Sep-97
2626 ----------------------
2628 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
2629 \x20 at the start of a run of normal characters. These were being treated as
2630 real characters, instead of the source characters being re-checked.
2633 Version 0.94 18-Sep-97
2634 ----------------------
2636 1. The functions are now thread-safe, with the caveat that the global variables
2637 containing pointers to malloc() and free() or alternative functions are the
2638 same for all threads.
2640 2. Get pcre_study() to generate a bitmap of initial characters for non-
2641 anchored patterns when this is possible, and use it if passed to pcre_exec().
2644 Version 0.93 15-Sep-97
2645 ----------------------
2647 1. /(b)|(:+)/ was computing an incorrect first character.
2649 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
2650 but not actually doing anything yet.
2652 3. Treat "-" characters in classes that cannot be part of ranges as literals,
2653 as Perl does (e.g. [-az] or [az-]).
2655 4. Set the anchored flag if a branch starts with .* or .*? because that tests
2656 all possible positions.
2658 5. Split up into different modules to avoid including unneeded functions in a
2659 compiled binary. However, compile and exec are still in one module. The "study"
2660 function is split off.
2662 6. The character tables are now in a separate module whose source is generated
2663 by an auxiliary program - but can then be edited by hand if required. There are
2664 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
2665 toupper() in the code.
2667 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
2668 make them global. Abolish the function for setting them, as the caller can now
2669 set them directly.
2672 Version 0.92 11-Sep-97
2673 ----------------------
2675 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
2676 (e.g. /a{1,3}/) was broken (I mis-optimized it).
2678 2. Caseless matching was not working in character classes if the characters in
2679 the pattern were in upper case.
2681 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
2683 4. Make PCRE_ANCHORED public and accept as a compile option.
2685 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
2686 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
2687 pass them.
2689 6. Give an error if bad option bits passed at compile or run time.
2691 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
2692 pcretest to cause it to pass that flag.
2694 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
2695 options, and the first character, if set.
2697 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
2700 Version 0.91 10-Sep-97
2701 ----------------------
2703 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
2704 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
2706 2. PCRE was looping on encountering an indefinitely repeated back reference to
2707 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
2708 Perl does - treats the match as successful.
2710 ****


Name Value
svn:eol-style native
svn:keywords "Author Date Id Revision Url"

  ViewVC Help
Powered by ViewVC 1.1.5