ViewVC logotype

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 150 - (show annotations)
Tue Apr 17 08:22:40 2007 UTC (14 years, 2 months ago) by ph10
File size: 126531 byte(s)
Update HTML documentation.
1 ChangeLog for PCRE
2 ------------------
4 Version 7.1 12-Mar-07
5 ---------------------
7 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
8 that is more "standard", making use of automake and other Autotools. There
9 is some re-arrangement of the files and adjustment of comments consequent
10 on this.
12 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
13 for recursive directory scanning broke on some systems because the files
14 are not scanned in any specific order and on different systems the order
15 was different. A call to "sort" has been inserted into RunGrepTest for the
16 approprate test as a short-term fix. In the longer term there may be an
17 alternative.
19 3. I had an email from Eric Raymond about problems translating some of PCRE's
20 man pages to HTML (despite the fact that I distribute HTML pages, some
21 people do their own conversions for various reasons). The problems
22 concerned the use of low-level troff macros .br and .in. I have therefore
23 removed all such uses from the man pages (some were redundant, some could
24 be replaced by .nf/.fi pairs). The 132html script that I use to generate
25 HTML has been updated to handle .nf/.fi and to complain if it encounters
26 .br or .in.
28 4. Updated comments in configure.ac that get placed in config.h.in and also
29 arranged for config.h to be included in the distribution, with the name
30 config.h.generic, for the benefit of those who have to compile without
31 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
33 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
34 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
35 makevp.bat for the current PCRE, using the additional files
36 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
38 6. A Windows user reported a minor discrepancy with test 2, which turned out
39 to be caused by a trailing space on an input line that had got lost in his
40 copy. The trailing space was an accident, so I've just removed it.
42 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
43 that is needed.
45 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
46 as "const" (a) because they are and (b) because it helps the PHP
47 maintainers who have recently made a script to detect big data structures
48 in the php code that should be moved to the .rodata section. I remembered
49 to update Builducptable as well, so it won't revert if ucptable.h is ever
50 re-created.
52 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
53 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
54 order to be able to cut out the UTF-8 tables in the latter when UTF-8
55 support is not required. This saves 1.5-2K of code, which is important in
56 some applications.
58 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
59 so as not to refer to the tables, even though these functions will never be
60 called when UTF-8 support is disabled. Otherwise there are problems with a
61 shared library.
63 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
65 (a) It was defining its arguments as char * instead of void *.
67 (b) It was assuming that all moves were upwards in memory; this was true
68 a long time ago when I wrote it, but is no longer the case.
70 The emulated memove() is provided for those environments that have neither
71 memmove() nor bcopy(). I didn't think anyone used it these days, but that
72 is clearly not the case, as these two bugs were recently reported.
74 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
75 and Detrail to create the HTML documentation, the .txt form of the man
76 pages, and it removes trailing spaces from listed files. It also creates
77 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
78 case, it wraps all the #defines with #ifndefs. This script should be run
79 before "make dist".
81 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
82 with Unicode property support.
84 (a) For a maximizing quantifier, if the two different cases of the
85 character were of different lengths in their UTF-8 codings (there are
86 some cases like this - I found 11), and the matching function had to
87 back up over a mixture of the two cases, it incorrectly assumed they
88 were both the same length.
90 (b) When PCRE was configured to use the heap rather than the stack for
91 recursion during matching, it was not correctly preserving the data for
92 the other case of a UTF-8 character when checking ahead for a match
93 while processing a minimizing repeat. If the check also involved
94 matching a wide character, but failed, corruption could cause an
95 erroneous result when trying to check for a repeat of the original
96 character.
98 13. Some tidying changes to the testing mechanism:
100 (a) The RunTest script now detects the internal link size and whether there
101 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
102 values substituted by "configure". (The RunGrepTest script already did
103 this for UTF-8.) The configure.ac script no longer substitutes the
104 relevant variables.
106 (b) The debugging options /B and /D in pcretest show the compiled bytecode
107 with length and offset values. This means that the output is different
108 for different internal link sizes. Test 2 is skipped for link sizes
109 other than 2 because of this, bypassing the problem. Unfortunately,
110 there was also a test in test 3 (the locale tests) that used /B and
111 failed for link sizes other than 2. Rather than cut the whole test out,
112 I have added a new /Z option to pcretest that replaces the length and
113 offset values with spaces. This is now used to make test 3 independent
114 of link size. (Test 2 will be tidied up later.)
116 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
117 segmentation fault instead of returning the appropriate error message.
119 15. In multiline mode when the newline sequence was set to "any", the pattern
120 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
121 This doesn't seem right; it now treats the CRLF combination as the line
122 ending, and so does not match in that case. It's only a pattern such as ^$
123 that would hit this one: something like ^ABC$ would have failed after \r
124 and then tried again after \r\n.
126 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
127 in an attempt to make files that differ only in their line terminators
128 compare equal. This works on Linux.
130 17. Under certain error circumstances pcregrep might try to free random memory
131 as it exited. This is now fixed, thanks to valgrind.
133 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
134 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
135 was because its rules for how to advance for /g after matching an empty
136 string at the end of a line did not allow for this case. They now check for
137 it specially.
139 20. pcretest is supposed to handle patterns and data of any length, by
140 extending its buffers when necessary. It was getting this wrong when the
141 buffer for a data line had to be extended.
143 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
144 CRLF as a newline sequence.
147 Version 7.0 19-Dec-06
148 ---------------------
150 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
151 moving to gcc 4.1.1.
153 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
154 sys/time.h, which is documented as needed for this function. It doesn't
155 seem to matter on Linux, but it showed up on some releases of OS X.
157 3. It seems that there are systems where bytes whose values are greater than
158 127 match isprint() in the "C" locale. The "C" locale should be the
159 default when a C program starts up. In most systems, only ASCII printing
160 characters match isprint(). This difference caused the output from pcretest
161 to vary, making some of the tests fail. I have changed pcretest so that:
163 (a) When it is outputting text in the compiled version of a pattern, bytes
164 other than 32-126 are always shown as hex escapes.
166 (b) When it is outputting text that is a matched part of a subject string,
167 it does the same, unless a different locale has been set for the match
168 (using the /L modifier). In this case, it uses isprint() to decide.
170 4. Fixed a major bug that caused incorrect computation of the amount of memory
171 required for a compiled pattern when options that changed within the
172 pattern affected the logic of the preliminary scan that determines the
173 length. The relevant options are -x, and -i in UTF-8 mode. The result was
174 that the computed length was too small. The symptoms of this bug were
175 either the PCRE error "internal error: code overflow" from pcre_compile(),
176 or a glibc crash with a message such as "pcretest: free(): invalid next
177 size (fast)". Examples of patterns that provoked this bug (shown in
178 pcretest format) are:
180 /(?-x: )/x
181 /(?x)(?-x: \s*#\s*)/
182 /((?i)[\x{c0}])/8
183 /(?i:[\x{c0}])/8
185 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
186 is now done differently.
188 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
189 wrapper classes; (b) implement a new function in the C++ scanner that is
190 more efficient than the old way of doing things because it avoids levels of
191 recursion in the regex matching; (c) add a paragraph to the documentation
192 for the FullMatch() function.
194 6. The escape sequence \n was being treated as whatever was defined as
195 "newline". Not only was this contrary to the documentation, which states
196 that \n is character 10 (hex 0A), but it also went horribly wrong when
197 "newline" was defined as CRLF. This has been fixed.
199 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
200 was being set to -1 for the "end of line" case (supposedly a value that no
201 character can have). Though this value is never used (the check for end of
202 line is "zero bytes in current character"), it caused compiler complaints.
203 I've changed it to 0xffffffff.
205 8. In pcre_version.c, the version string was being built by a sequence of
206 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
207 string (as it is for production releases) called a macro with an empty
208 argument. The C standard says the result of this is undefined. The gcc
209 compiler treats it as an empty string (which was what was wanted) but it is
210 reported that Visual C gives an error. The source has been hacked around to
211 avoid this problem.
213 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
214 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
215 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
216 of them did).
218 10. Originally, pcretest opened its input and output without "b"; then I was
219 told that "b" was needed in some environments, so it was added for release
220 5.0 to both the input and output. (It makes no difference on Unix-like
221 systems.) Later I was told that it is wrong for the input on Windows. I've
222 now abstracted the modes into two macros, to make it easier to fiddle with
223 them, and removed "b" from the input mode under Windows.
225 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
227 12. Added -help and --help to pcretest as an official way of being reminded
228 of the options.
230 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
231 and pcrecpp.cc because they annoy compilers at high warning levels.
233 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
235 15. Fixed an occurrence of == in configure.ac that should have been = (shell
236 scripts are not C programs :-) and which was not noticed because it works
237 on Linux.
239 16. pcretest is supposed to handle any length of pattern and data line (as one
240 line or as a continued sequence of lines) by extending its input buffer if
241 necessary. This feature was broken for very long pattern lines, leading to
242 a string of junk being passed to pcre_compile() if the pattern was longer
243 than about 50K.
245 17. I have done a major re-factoring of the way pcre_compile() computes the
246 amount of memory needed for a compiled pattern. Previously, there was code
247 that made a preliminary scan of the pattern in order to do this. That was
248 OK when PCRE was new, but as the facilities have expanded, it has become
249 harder and harder to keep it in step with the real compile phase, and there
250 have been a number of bugs (see for example, 4 above). I have now found a
251 cunning way of running the real compile function in a "fake" mode that
252 enables it to compute how much memory it would need, while actually only
253 ever using a few hundred bytes of working memory and without too many
254 tests of the mode. This should make future maintenance and development
255 easier. A side effect of this work is that the limit of 200 on the nesting
256 depth of parentheses has been removed (though this was never a serious
257 limitation, I suspect). However, there is a downside: pcre_compile() now
258 runs more slowly than before (30% or more, depending on the pattern). I
259 hope this isn't a big issue. There is no effect on runtime performance.
261 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
262 newline (only possible for the last line of a file) and it was a
263 pattern that set a locale (followed by /Lsomething), pcretest crashed.
265 19. Added additional timing features to pcretest. (1) The -tm option now times
266 matching only, not compiling. (2) Both -t and -tm can be followed, as a
267 separate command line item, by a number that specifies the number of
268 repeats to use when timing. The default is 50000; this gives better
269 precision, but takes uncomfortably long for very large patterns.
271 20. Extended pcre_study() to be more clever in cases where a branch of a
272 subpattern has no definite first character. For example, (a*|b*)[cd] would
273 previously give no result from pcre_study(). Now it recognizes that the
274 first character must be a, b, c, or d.
276 21. There was an incorrect error "recursive call could loop indefinitely" if
277 a subpattern (or the entire pattern) that was being tested for matching an
278 empty string contained only one non-empty item after a nested subpattern.
279 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
280 incorrectly, because the \d was being skipped in the check.
282 22. The pcretest program now has a new pattern option /B and a command line
283 option -b, which is equivalent to adding /B to every pattern. This causes
284 it to show the compiled bytecode, without the additional information that
285 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
286 is the same as /B/I).
288 23. A new optimization is now able automatically to treat some sequences such
289 as a*b as a*+b. More specifically, if something simple (such as a character
290 or a simple class like \d) has an unlimited quantifier, and is followed by
291 something that cannot possibly match the quantified thing, the quantifier
292 is automatically "possessified".
294 24. A recursive reference to a subpattern whose number was greater than 39
295 went wrong under certain circumstances in UTF-8 mode. This bug could also
296 have affected the operation of pcre_study().
298 25. Realized that a little bit of performance could be had by replacing
299 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
301 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
303 27. Possessive quantifiers such as a++ were previously implemented by turning
304 them into atomic groups such as ($>a+). Now they have their own opcodes,
305 which improves performance. This includes the automatically created ones
306 from 23 above.
308 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
309 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
310 the first matched character to be a colon. This applied both to named and
311 numbered groups.
313 29. The ucpinternal.h header file was missing its idempotency #ifdef.
315 30. I was sent a "project" file called libpcre.a.dev which I understand makes
316 building PCRE on Windows easier, so I have included it in the distribution.
318 31. There is now a check in pcretest against a ridiculously large number being
319 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
320 loop, the loop is abandoned.
322 32. Forward references to subpatterns in conditions such as (?(2)...) where
323 subpattern 2 is defined later cause pcre_compile() to search forwards in
324 the pattern for the relevant set of parentheses. This search went wrong
325 when there were unescaped parentheses in a character class, parentheses
326 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
328 33. "Subroutine" calls and backreferences were previously restricted to
329 referencing subpatterns earlier in the regex. This restriction has now
330 been removed.
332 34. Added a number of extra features that are going to be in Perl 5.10. On the
333 whole, these are just syntactic alternatives for features that PCRE had
334 previously implemented using the Python syntax or my own invention. The
335 other formats are all retained for compatibility.
337 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
338 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
339 also .NET compatible.
341 (b) A recursion or subroutine call to a named group can now be defined as
342 (?&name) as well as (?P>name).
344 (c) A backreference to a named group can now be defined as \k<name> or
345 \k'name' as well as (?P=name). The new forms, as well as being in Perl
346 5.10, are also .NET compatible.
348 (d) A conditional reference to a named group can now use the syntax
349 (?(<name>) or (?('name') as well as (?(name).
351 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
352 groups (named and numbered) that are never evaluated inline, but can be
353 called as "subroutines" from elsewhere. In effect, the DEFINE condition
354 is always false. There may be only one alternative in such a group.
356 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
357 as the simple (?(R). The condition is true only if the most recent
358 recursion is that of the given number or name. It does not search out
359 through the entire recursion stack.
361 (g) The escape \gN or \g{N} has been added, where N is a positive or
362 negative number, specifying an absolute or relative reference.
364 35. Tidied to get rid of some further signed/unsigned compiler warnings and
365 some "unreachable code" warnings.
367 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
368 things, this adds five new scripts.
370 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
371 There were also incompatibilities regarding the handling of \Q..\E inside
372 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
373 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
375 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
376 matches an empty string, and forcibly breaks the loop. There were bugs in
377 this code in non-simple cases. For a pattern such as ^(a()*)* matched
378 against aaaa the result was just "a" rather than "aaaa", for example. Two
379 separate and independent bugs (that affected different cases) have been
380 fixed.
382 39. Refactored the code to abolish the use of different opcodes for small
383 capturing bracket numbers. This is a tidy that I avoided doing when I
384 removed the limit on the number of capturing brackets for 3.5 back in 2001.
385 The new approach is not only tidier, it makes it possible to reduce the
386 memory needed to fix the previous bug (38).
388 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
389 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
390 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
391 mode.
393 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
394 report.
396 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
397 copy construction and assignment in the C++ wrapper.
399 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
400 couple of bugs that could have given wrong results in the "--newline=crlf"
401 case.
403 44. Added a number of casts and did some reorganization of signed/unsigned int
404 variables following suggestions from Dair Grant. Also renamed the variable
405 "this" as "item" because it is a C++ keyword.
407 45. Arranged for dftables to add
409 #include "pcre_internal.h"
411 to pcre_chartables.c because without it, gcc 4.x may remove the array
412 definition from the final binary if PCRE is built into a static library and
413 dead code stripping is activated.
415 46. For an unanchored pattern, if a match attempt fails at the start of a
416 newline sequence, and the newline setting is CRLF or ANY, and the next two
417 characters are CRLF, advance by two characters instead of one.
420 Version 6.7 04-Jul-06
421 ---------------------
423 1. In order to handle tests when input lines are enormously long, pcretest has
424 been re-factored so that it automatically extends its buffers when
425 necessary. The code is crude, but this _is_ just a test program. The
426 default size has been increased from 32K to 50K.
428 2. The code in pcre_study() was using the value of the re argument before
429 testing it for NULL. (Of course, in any sensible call of the function, it
430 won't be NULL.)
432 3. The memmove() emulation function in pcre_internal.h, which is used on
433 systems that lack both memmove() and bcopy() - that is, hardly ever -
434 was missing a "static" storage class specifier.
436 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
437 containing an extended class (one that cannot be represented by a bitmap
438 because it contains high-valued characters or Unicode property items, e.g.
439 [\pZ]). Almost always one would set UTF-8 mode when processing such a
440 pattern, but PCRE should not loop if you do not (it no longer does).
441 [Detail: two cases were found: (a) a repeated subpattern containing an
442 extended class; (b) a recursive reference to a subpattern that followed a
443 previous extended class. It wasn't skipping over the extended class
444 correctly when UTF-8 mode was not set.]
446 5. A negated single-character class was not being recognized as fixed-length
447 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
448 compile error "lookbehind assertion is not fixed length".
450 6. The RunPerlTest auxiliary script was showing an unexpected difference
451 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
452 write a Perl script that can interpret lines of an input file either as
453 byte characters or as UTF-8, which is what "perltest" was being required to
454 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
455 can't do is switch easily at run time between having the "use utf8;" pragma
456 or not. In the end, I fudged it by using the RunPerlTest script to insert
457 "use utf8;" explicitly for the UTF-8 tests.
459 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
460 the end of the subject string, contrary to the documentation and to what
461 Perl does. This was true of both matching functions. Now it matches only at
462 the start of the subject and immediately after *internal* newlines.
464 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
465 a pointer to an int instead of a pointer to an unsigned long int. This
466 caused problems on 64-bit systems.
468 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
469 instance of the 'standard' template library not being so standard".
471 10. There was no check on the number of named subpatterns nor the maximum
472 length of a subpattern name. The product of these values is used to compute
473 the size of the memory block for a compiled pattern. By supplying a very
474 long subpattern name and a large number of named subpatterns, the size
475 computation could be caused to overflow. This is now prevented by limiting
476 the length of names to 32 characters, and the number of named subpatterns
477 to 10,000.
479 11. Subpatterns that are repeated with specific counts have to be replicated in
480 the compiled pattern. The size of memory for this was computed from the
481 length of the subpattern and the repeat count. The latter is limited to
482 65535, but there was no limit on the former, meaning that integer overflow
483 could in principle occur. The compiled length of a repeated subpattern is
484 now limited to 30,000 bytes in order to prevent this.
486 12. Added the optional facility to have named substrings with the same name.
488 13. Added the ability to use a named substring as a condition, using the
489 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
490 are numbers (not recommended). Forward references are permitted.
492 14. Added forward references in named backreferences (if you see what I mean).
494 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
495 pattern could run off the end of the subject. For example, the pattern
496 "(?s)(.{1,5})"8 did this with the subject "ab".
498 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
499 PCRE_CASELESS was set when matching characters that were quantified with ?
500 or *.
502 17. A character class other than a single negated character that had a minimum
503 but no maximum quantifier - for example [ab]{6,} - was not handled
504 correctly by pce_dfa_exec(). It would match only one character.
506 18. A valid (though odd) pattern that looked like a POSIX character
507 class but used an invalid character after [ (for example [[,abc,]]) caused
508 pcre_compile() to give the error "Failed: internal error: code overflow" or
509 in some cases to crash with a glibc free() error. This could even happen if
510 the pattern terminated after [[ but there just happened to be a sequence of
511 letters, a binary zero, and a closing ] in the memory that followed.
513 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
514 over the years. Originally (before any Unicode support), just the bottom 8
515 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
516 output from "man perlunicode" includes this:
518 The regular expression compiler produces polymorphic opcodes. That
519 is, the pattern adapts to the data and automatically switches to
520 the Unicode character scheme when presented with Unicode data--or
521 instead uses a traditional byte scheme when presented with byte
522 data.
524 Sadly, a wide octal escape does not cause a switch, and in a string with
525 no other multibyte characters, these octal escapes are treated as before.
526 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
527 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
528 Unicode string.
530 I have not perpetrated such confusion in PCRE. Up till now, it took just
531 the bottom 8 bits, as in old Perl. I have now made octal escapes with
532 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
533 translate to the appropriate multibyte character.
535 29. Applied some refactoring to reduce the number of warnings from Microsoft
536 and Borland compilers. This has included removing the fudge introduced
537 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
538 a warning about an unused variable.
540 21. PCRE has not included VT (character 0x0b) in the set of whitespace
541 characters since release 4.0, because Perl (from release 5.004) does not.
542 [Or at least, is documented not to: some releases seem to be in conflict
543 with the documentation.] However, when a pattern was studied with
544 pcre_study() and all its branches started with \s, PCRE still included VT
545 as a possible starting character. Of course, this did no harm; it just
546 caused an unnecessary match attempt.
548 22. Removed a now-redundant internal flag bit that recorded the fact that case
549 dependency changed within the pattern. This was once needed for "required
550 byte" processing, but is no longer used. This recovers a now-scarce options
551 bit. Also moved the least significant internal flag bit to the most-
552 significant bit of the word, which was not previously used (hangover from
553 the days when it was an int rather than a uint) to free up another bit for
554 the future.
556 23. Added support for CRLF line endings as well as CR and LF. As well as the
557 default being selectable at build time, it can now be changed at runtime
558 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
559 specify that it is scanning data with non-default line endings.
561 24. Changed the definition of CXXLINK to make it agree with the definition of
562 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
564 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
565 recursions. This makes a big different to stack usage for some patterns.
567 26. If a subpattern containing a named recursion or subroutine reference such
568 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
569 the space required for the compiled pattern went wrong and gave too small a
570 value. Depending on the environment, this could lead to "Failed: internal
571 error: code overflow at offset 49" or "glibc detected double free or
572 corruption" errors.
574 27. Applied patches from Google (a) to support the new newline modes and (b) to
575 advance over multibyte UTF-8 characters in GlobalReplace.
577 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
578 difference for some implementation of PCRE in some Windows version.
580 29. Added some extra testing facilities to pcretest:
582 \q<number> in a data line sets the "match limit" value
583 \Q<number> in a data line sets the "match recursion limt" value
584 -S <number> sets the stack size, where <number> is in megabytes
586 The -S option isn't available for Windows.
589 Version 6.6 06-Feb-06
590 ---------------------
592 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
593 in pcreposix.h. I have copied the definition from pcre.h.
595 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
596 because pcre.h is no longer a built file.
598 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
599 not normally included in the compiled code.
602 Version 6.5 01-Feb-06
603 ---------------------
605 1. When using the partial match feature with pcre_dfa_exec(), it was not
606 anchoring the second and subsequent partial matches at the new starting
607 point. This could lead to incorrect results. For example, with the pattern
608 /1234/, partially matching against "123" and then "a4" gave a match.
610 2. Changes to pcregrep:
612 (a) All non-match returns from pcre_exec() were being treated as failures
613 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
614 error message is output. Some extra information is given for the
616 probably the only errors that are likely to be caused by users (by
617 specifying a regex that has nested indefinite repeats, for instance).
618 If there are more than 20 of these errors, pcregrep is abandoned.
620 (b) A binary zero was treated as data while matching, but terminated the
621 output line if it was written out. This has been fixed: binary zeroes
622 are now no different to any other data bytes.
624 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
625 used to set a locale for matching. The --locale=xxxx long option has
626 been added (no short equivalent) to specify a locale explicitly on the
627 pcregrep command, overriding the environment variables.
629 (d) When -B was used with -n, some line numbers in the output were one less
630 than they should have been.
632 (e) Added the -o (--only-matching) option.
634 (f) If -A or -C was used with -c (count only), some lines of context were
635 accidentally printed for the final match.
637 (g) Added the -H (--with-filename) option.
639 (h) The combination of options -rh failed to suppress file names for files
640 that were found from directory arguments.
642 (i) Added the -D (--devices) and -d (--directories) options.
644 (j) Added the -F (--fixed-strings) option.
646 (k) Allow "-" to be used as a file name for -f as well as for a data file.
648 (l) Added the --colo(u)r option.
650 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
651 is not present by default.
653 3. A nasty bug was discovered in the handling of recursive patterns, that is,
654 items such as (?R) or (?1), when the recursion could match a number of
655 alternatives. If it matched one of the alternatives, but subsequently,
656 outside the recursion, there was a failure, the code tried to back up into
657 the recursion. However, because of the way PCRE is implemented, this is not
658 possible, and the result was an incorrect result from the match.
660 In order to prevent this happening, the specification of recursion has
661 been changed so that all such subpatterns are automatically treated as
662 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
664 4. I had overlooked the fact that, in some locales, there are characters for
665 which isalpha() is true but neither isupper() nor islower() are true. In
666 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
667 and ordfeminine) are like this. This affected the treatment of \w and \W
668 when they appeared in character classes, but not when they appeared outside
669 a character class. The bit map for "word" characters is now created
670 separately from the results of isalnum() instead of just taking it from the
671 upper, lower, and digit maps. (Plus the underscore character, of course.)
673 5. The above bug also affected the handling of POSIX character classes such as
674 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
675 permanent tables. Instead, the bit maps for such a class were previously
676 created as the appropriate unions of the upper, lower, and digit bitmaps.
677 Now they are created by subtraction from the [[:word:]] class, which has
678 its own bitmap.
680 6. The [[:blank:]] character class matches horizontal, but not vertical space.
681 It is created by subtracting the vertical space characters (\x09, \x0a,
682 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
683 subtraction was done in the overall bitmap for a character class, meaning
684 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
685 be recognized. This bug has been fixed.
687 7. Patches from the folks at Google:
689 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
690 real life, but is still worth protecting against".
692 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
693 regular expressions".
695 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
696 have it.
698 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
699 "configure" and the latter not, in order to fix a problem somebody had
700 with compiling the Arg class on HP-UX.
702 (e) Improve the error-handling of the C++ wrapper a little bit.
704 (f) New tests for checking recursion limiting.
706 8. The pcre_memmove() function, which is used only if the environment does not
707 have a standard memmove() function (and is therefore rarely compiled),
708 contained two bugs: (a) use of int instead of size_t, and (b) it was not
709 returning a result (though PCRE never actually uses the result).
711 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
712 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
713 returned instead of calling malloc() with an overflowing number that would
714 most likely cause subsequent chaos.
716 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
718 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
719 with this option is matched, the nmatch and pmatch options of regexec() are
720 ignored.
722 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
723 provided in case anyone wants to the the POSIX interface with UTF-8
724 strings.
726 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
727 C++ linking (needed for some HP-UX environments).
729 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
730 (unused parameter) and in the pcre_printint() function (omitted "default"
731 switch label when the default is to do nothing).
733 15. Added some code to make it possible, when PCRE is compiled as a C++
734 library, to replace subject pointers for pcre_exec() with a smart pointer
735 class, thus making it possible to process discontinuous strings.
737 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
738 much the same function. They were added by different people who were trying
739 to make PCRE easy to compile on non-Unix systems. It has been suggested
740 that PCRE_EXPORT be abolished now that there is more automatic apparatus
741 for compiling on Windows systems. I have therefore replaced it with
742 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
743 defaults to "extern" for C or "extern C" for C++, which works fine on
744 Unix-like systems. It is now possible to override the value of PCRE_DATA_
745 SCOPE with something explicit in config.h. In addition:
747 (a) pcreposix.h still had just "extern" instead of either of these macros;
748 I have replaced it with PCRE_DATA_SCOPE.
750 (b) Functions such as _pcre_xclass(), which are internal to the library,
751 but external in the C sense, all had PCRE_EXPORT in their definitions.
752 This is apparently wrong for the Windows case, so I have removed it.
753 (It makes no difference on Unix-like systems.)
755 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
756 of recursive calls to match(). This is different to MATCH_LIMIT because
757 that limits the total number of calls to match(), not all of which increase
758 the depth of recursion. Limiting the recursion depth limits the amount of
759 stack (or heap if NO_RECURSE is set) that is used. The default can be set
760 when PCRE is compiled, and changed at run time. A patch from Google adds
761 this functionality to the C++ interface.
763 18. Changes to the handling of Unicode character properties:
765 (a) Updated the table to Unicode 4.1.0.
767 (b) Recognize characters that are not in the table as "Cn" (undefined).
769 (c) I revised the way the table is implemented to a much improved format
770 which includes recognition of ranges. It now supports the ranges that
771 are defined in UnicodeData.txt, and it also amalgamates other
772 characters into ranges. This has reduced the number of entries in the
773 table from around 16,000 to around 3,000, thus reducing its size
774 considerably. I realized I did not need to use a tree structure after
775 all - a binary chop search is just as efficient. Having reduced the
776 number of entries, I extended their size from 6 bytes to 8 bytes to
777 allow for more data.
779 (d) Added support for Unicode script names via properties such as \p{Han}.
781 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
782 matching that character.
784 20. When matching a repeated Unicode property with a minimum greater than zero,
785 (for example \pL{2,}), PCRE could look past the end of the subject if it
786 reached it while seeking the minimum number of characters. This could
787 happen only if some of the characters were more than one byte long, because
788 there is a check for at least the minimum number of bytes.
790 21. Refactored the implementation of \p and \P so as to be more general, to
791 allow for more different types of property in future. This has changed the
792 compiled form incompatibly. Anybody with saved compiled patterns that use
793 \p or \P will have to recompile them.
795 22. Added "Any" and "L&" to the supported property types.
797 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
798 but give a compile time error if the value is greater than 0xff.
800 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
801 accidentally not being installed or uninstalled.
803 25. The pcre.h file was built from pcre.h.in, but the only changes that were
804 made were to insert the current release number. This seemed silly, because
805 it made things harder for people building PCRE on systems that don't run
806 "configure". I have turned pcre.h into a distributed file, no longer built
807 by "configure", with the version identification directly included. There is
808 no longer a pcre.h.in file.
810 However, this change necessitated a change to the pcre-config script as
811 well. It is built from pcre-config.in, and one of the substitutions was the
812 release number. I have updated configure.ac so that ./configure now finds
813 the release number by grepping pcre.h.
815 26. Added the ability to run the tests under valgrind.
818 Version 6.4 05-Sep-05
819 ---------------------
821 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
822 "--" to be printed when multiple files were scanned, even when none of the
823 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
824 consider it to be a bug, and have restored the previous behaviour.
826 2. A couple of code tidies to get rid of compiler warnings.
828 3. The pcretest program used to cheat by referring to symbols in the library
829 whose names begin with _pcre_. These are internal symbols that are not
830 really supposed to be visible externally, and in some environments it is
831 possible to suppress them. The cheating is now confined to including
832 certain files from the library's source, which is a bit cleaner.
834 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
835 file's purpose clearer.
837 5. Reorganized pcre_ucp_findchar().
840 Version 6.3 15-Aug-05
841 ---------------------
843 1. The file libpcre.pc.in did not have general read permission in the tarball.
845 2. There were some problems when building without C++ support:
847 (a) If C++ support was not built, "make install" and "make test" still
848 tried to test it.
850 (b) There were problems when the value of CXX was explicitly set. Some
851 changes have been made to try to fix these, and ...
853 (c) --disable-cpp can now be used to explicitly disable C++ support.
855 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
856 backslash in a target when C++ was disabled. This confuses some
857 versions of "make", apparently. Using an intermediate variable solves
858 this. (Same for CPP_LOBJ.)
860 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
861 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
862 necessary on certain architectures.
864 4. Added a setting of -export-symbols-regex to the link command to remove
865 those symbols that are exported in the C sense, but actually are local
866 within the library, and not documented. Their names all begin with
867 "_pcre_". This is not a perfect job, because (a) we have to except some
868 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
869 available (and never for static libraries). I have made a note to try to
870 find a way round (a) in the future.
873 Version 6.2 01-Aug-05
874 ---------------------
876 1. There was no test for integer overflow of quantifier values. A construction
877 such as {1111111111111111} would give undefined results. What is worse, if
878 a minimum quantifier for a parenthesized subpattern overflowed and became
879 negative, the calculation of the memory size went wrong. This could have
880 led to memory overwriting.
882 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
884 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
885 operating environments where this matters.
887 4. Applied Giuseppe Maxia's patch to add additional features for controlling
888 PCRE options from within the C++ wrapper.
890 5. Named capturing subpatterns were not being correctly counted when a pattern
891 was compiled. This caused two problems: (a) If there were more than 100
892 such subpatterns, the calculation of the memory needed for the whole
893 compiled pattern went wrong, leading to an overflow error. (b) Numerical
894 back references of the form \12, where the number was greater than 9, were
895 not recognized as back references, even though there were sufficient
896 previous subpatterns.
898 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
899 versions of gcc, e.g. 2.95.4.
902 Version 6.1 21-Jun-05
903 ---------------------
905 1. There was one reference to the variable "posix" in pcretest.c that was not
906 surrounded by "#if !defined NOPOSIX".
908 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
909 the cross-check on the old pcre_info() function, for the benefit of the
910 cut-down version of PCRE that is currently imported into Exim.
912 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
913 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
914 patterns that start (?i) but I don't think that matters. The use of (?i) is
915 just an example; this all applies to the other options as well.
917 4. Since libtool seems to echo the compile commands it is issuing, the output
918 from "make" can be reduced a bit by putting "@" in front of each libtool
919 compile command.
921 5. Patch from the folks at Google for configure.in to be a bit more thorough
922 in checking for a suitable C++ installation before trying to compile the
923 C++ stuff. This should fix a reported problem when a compiler was present,
924 but no suitable headers.
926 6. The man pages all had just "PCRE" as their title. I have changed them to
927 be the relevant file name. I have also arranged that these names are
928 retained in the file doc/pcre.txt, which is a concatenation in text format
929 of all the man pages except the little individual ones for each function.
931 7. The NON-UNIX-USE file had not been updated for the different set of source
932 files that come with release 6. I also added a few comments about the C++
933 wrapper.
936 Version 6.0 07-Jun-05
937 ---------------------
939 1. Some minor internal re-organization to help with my DFA experiments.
941 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
942 didn't matter for the library itself when fully configured, but did matter
943 when compiling without UCP support, or within Exim, where the ucp files are
944 not imported.
946 3. Refactoring of the library code to split up the various functions into
947 different source modules. The addition of the new DFA matching code (see
948 below) to a single monolithic source would have made it really too
949 unwieldy, quite apart from causing all the code to be include in a
950 statically linked application, when only some functions are used. This is
951 relevant even without the DFA addition now that patterns can be compiled in
952 one application and matched in another.
954 The downside of splitting up is that there have to be some external
955 functions and data tables that are used internally in different modules of
956 the library but which are not part of the API. These have all had their
957 names changed to start with "_pcre_" so that they are unlikely to clash
958 with other external names.
960 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
961 a different (DFA) algorithm. Although it is slower than the original
962 function, it does have some advantages for certain types of matching
963 problem.
965 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
966 including restarting after a partial match.
968 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
969 defined when compiling for Windows was sent to me. I have put it into the
970 code, though I have no means of testing or verifying it.
972 7. Added the pcre_refcount() auxiliary function.
974 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
975 match before or at the first newline in the subject string. In pcretest,
976 the /f option on a pattern can be used to set this.
978 9. A repeated \w when used in UTF-8 mode with characters greater than 256
979 would behave wrongly. This has been present in PCRE since release 4.0.
981 10. A number of changes to the pcregrep command:
983 (a) Refactored how -x works; insert ^(...)$ instead of setting
984 PCRE_ANCHORED and checking the length, in preparation for adding
985 something similar for -w.
987 (b) Added the -w (match as a word) option.
989 (c) Refactored the way lines are read and buffered so as to have more
990 than one at a time available.
992 (d) Implemented a pcregrep test script.
994 (e) Added the -M (multiline match) option. This allows patterns to match
995 over several lines of the subject. The buffering ensures that at least
996 8K, or the rest of the document (whichever is the shorter) is available
997 for matching (and similarly the previous 8K for lookbehind assertions).
999 (f) Changed the --help output so that it now says
1001 -w, --word-regex(p)
1003 instead of two lines, one with "regex" and the other with "regexp"
1004 because that confused at least one person since the short forms are the
1005 same. (This required a bit of code, as the output is generated
1006 automatically from a table. It wasn't just a text change.)
1008 (g) -- can be used to terminate pcregrep options if the next thing isn't an
1009 option but starts with a hyphen. Could be a pattern or a path name
1010 starting with a hyphen, for instance.
1012 (h) "-" can be given as a file name to represent stdin.
1014 (i) When file names are being printed, "(standard input)" is used for
1015 the standard input, for compatibility with GNU grep. Previously
1016 "<stdin>" was used.
1018 (j) The option --label=xxx can be used to supply a name to be used for
1019 stdin when file names are being printed. There is no short form.
1021 (k) Re-factored the options decoding logic because we are going to add
1022 two more options that take data. Such options can now be given in four
1023 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1025 (l) Added the -A, -B, and -C options for requesting that lines of context
1026 around matches be printed.
1028 (m) Added the -L option to print the names of files that do not contain
1029 any matching lines, that is, the complement of -l.
1031 (n) The return code is 2 if any file cannot be opened, but pcregrep does
1032 continue to scan other files.
1034 (o) The -s option was incorrectly implemented. For compatibility with other
1035 greps, it now suppresses the error message for a non-existent or non-
1036 accessible file (but not the return code). There is a new option called
1037 -q that suppresses the output of matching lines, which was what -s was
1038 previously doing.
1040 (p) Added --include and --exclude options to specify files for inclusion
1041 and exclusion when recursing.
1043 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1044 Hopefully, it now does.
1046 12. Missing cast in pcre_study().
1048 13. Added an "uninstall" target to the makefile.
1050 14. Replaced "extern" in the function prototypes in Makefile.in with
1051 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1052 world, but is set differently for Windows.
1054 15. Added a second compiling function called pcre_compile2(). The only
1055 difference is that it has an extra argument, which is a pointer to an
1056 integer error code. When there is a compile-time failure, this is set
1057 non-zero, in addition to the error test pointer being set to point to an
1058 error message. The new argument may be NULL if no error number is required
1059 (but then you may as well call pcre_compile(), which is now just a
1060 wrapper). This facility is provided because some applications need a
1061 numeric error indication, but it has also enabled me to tidy up the way
1062 compile-time errors are handled in the POSIX wrapper.
1064 16. Added VPATH=.libs to the makefile; this should help when building with one
1065 prefix path and installing with another. (Or so I'm told by someone who
1066 knows more about this stuff than I do.)
1068 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1069 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1070 match everything, including newlines. This is not POSIX-compatible, but
1071 somebody wanted the feature. From pcretest it can be activated by using
1072 both the P and the s flags.
1074 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1076 19. libpcre.pc was being incorrectly installed as executable.
1078 20. A couple of places in pcretest check for end-of-line by looking for '\n';
1079 it now also looks for '\r' so that it will work unmodified on Windows.
1081 21. Added Google's contributed C++ wrapper to the distribution.
1083 22. Added some untidy missing memory free() calls in pcretest, to keep
1084 Electric Fence happy when testing.
1088 Version 5.0 13-Sep-04
1089 ---------------------
1091 1. Internal change: literal characters are no longer packed up into items
1092 containing multiple characters in a single byte-string. Each character
1093 is now matched using a separate opcode. However, there may be more than one
1094 byte in the character in UTF-8 mode.
1096 2. The pcre_callout_block structure has two new fields: pattern_position and
1097 next_item_length. These contain the offset in the pattern to the next match
1098 item, and its length, respectively.
1100 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1101 insertion of callouts before each pattern item. Added the /C option to
1102 pcretest to make use of this.
1104 4. On the advice of a Windows user, the lines
1106 #if defined(_WIN32) || defined(WIN32)
1107 _setmode( _fileno( stdout ), 0x8000 );
1108 #endif /* defined(_WIN32) || defined(WIN32) */
1110 have been added to the source of pcretest. This apparently does useful
1111 magic in relation to line terminators.
1113 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1114 for the benefit of those environments where the "b" makes a difference.
1116 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1117 to know about it. I have put a hack into configure.in that adds in code
1118 to set GCC=yes if CC=icc. This seems to end up at a point in the
1119 generated configure script that is early enough to affect the setting of
1120 compiler options, which is what is needed, but I have no means of testing
1121 whether it really works. (The user who reported this had patched the
1122 generated configure script, which of course I cannot do.)
1124 LATER: After change 22 below (new libtool files), the configure script
1125 seems to know about icc (and also ecc). Therefore, I have commented out
1126 this hack in configure.in.
1128 7. Added support for pkg-config (2 patches were sent in).
1130 8. Negated POSIX character classes that used a combination of internal tables
1131 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1132 [[:^ascii]]. Typically, they would match almost any characters. The other
1133 POSIX classes were not broken in this way.
1135 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1136 to find the match, as PCRE was deluded into thinking that the match had to
1137 start at the start point or following a newline. The same bug applied to
1138 patterns with negative forward assertions or any backward assertions
1139 preceding ".*" at the start, unless the pattern required a fixed first
1140 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1142 10. In UTF-8 mode, when moving forwards in the subject after a failed match
1143 starting at the last subject character, bytes beyond the end of the subject
1144 string were read.
1146 11. Renamed the variable "class" as "classbits" to make life easier for C++
1147 users. (Previously there was a macro definition, but it apparently wasn't
1148 enough.)
1150 12. Added the new field "tables" to the extra data so that tables can be passed
1151 in at exec time, or the internal tables can be re-selected. This allows
1152 a compiled regex to be saved and re-used at a later time by a different
1153 program that might have everything at different addresses.
1155 13. Modified the pcre-config script so that, when run on Solaris, it shows a
1156 -R library as well as a -L library.
1158 14. The debugging options of pcretest (-d on the command line or D on a
1159 pattern) showed incorrect output for anything following an extended class
1160 that contained multibyte characters and which was followed by a quantifier.
1162 15. Added optional support for general category Unicode character properties
1163 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1164 support. It adds about 90K to the size of the library. The meanings of the
1165 inbuilt class escapes such as \d and \s have NOT been changed.
1167 16. Updated pcredemo.c to include calls to free() to release the memory for the
1168 compiled pattern.
1170 17. The generated file chartables.c was being created in the source directory
1171 instead of in the building directory. This caused the build to fail if the
1172 source directory was different from the building directory, and was
1173 read-only.
1175 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1176 file. No doubt somebody will tell me if they don't make sense... Also added
1177 Dan Mooney's comments about building on OpenVMS.
1179 19. Added support for partial matching via the PCRE_PARTIAL option for
1180 pcre_exec() and the \P data escape in pcretest.
1182 20. Extended pcretest with 3 new pattern features:
1184 (i) A pattern option of the form ">rest-of-line" causes pcretest to
1185 write the compiled pattern to the file whose name is "rest-of-line".
1186 This is a straight binary dump of the data, with the saved pointer to
1187 the character tables forced to be NULL. The study data, if any, is
1188 written too. After writing, pcretest reads a new pattern.
1190 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1191 compiled pattern from the given file. There must not be any
1192 occurrences of "<" in the file name (pretty unlikely); if there are,
1193 pcretest will instead treat the initial "<" as a pattern delimiter.
1194 After reading in the pattern, pcretest goes on to read data lines as
1195 usual.
1197 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1198 and 16-bit fields in a compiled pattern, to simulate a pattern that
1199 was compiled on a host of opposite endianness.
1201 21. The pcre-exec() function can now cope with patterns that were compiled on
1202 hosts of opposite endianness, with this restriction:
1204 As for any compiled expression that is saved and used later, the tables
1205 pointer field cannot be preserved; the extra_data field in the arguments
1206 to pcre_exec() should be used to pass in a tables address if a value
1207 other than the default internal tables were used at compile time.
1209 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1210 now diagnosed as an error. Previously, most of the time, a negative number
1211 would have been treated as zero, but if in addition "ovector" was passed as
1212 NULL, a crash could occur.
1214 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1215 new versions from the libtool 1.5 distribution (the last one is a copy of
1216 a file called libtool.m4). This seems to have fixed the need to patch
1217 "configure" to support Darwin 1.3 (which I used to do). However, I still
1218 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1219 workstation).
1221 24. Changed the PCRE licence to be the more standard "BSD" licence.
1224 Version 4.5 01-Dec-03
1225 ---------------------
1227 1. There has been some re-arrangement of the code for the match() function so
1228 that it can be compiled in a version that does not call itself recursively.
1229 Instead, it keeps those local variables that need separate instances for
1230 each "recursion" in a frame on the heap, and gets/frees frames whenever it
1231 needs to "recurse". Keeping track of where control must go is done by means
1232 of setjmp/longjmp. The whole thing is implemented by a set of macros that
1233 hide most of the details from the main code, and operates only if
1234 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1235 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1236 operating.
1238 To make it easier for callers to provide specially tailored get/free
1239 functions for this usage, two new functions, pcre_stack_malloc, and
1240 pcre_stack_free, are used. They are always called in strict stacking order,
1241 and the size of block requested is always the same.
1243 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1244 PCRE has been compiled to use the stack or the heap for recursion. The
1245 -C option of pcretest uses this to show which version is compiled.
1247 A new data escape \S, is added to pcretest; it causes the amounts of store
1248 obtained and freed by both kinds of malloc/free at match time to be added
1249 to the output.
1251 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1252 what's available on my current Linux desktop machine.
1254 3. When matching a UTF-8 string, the test for a valid string at the start has
1255 been extended. If start_offset is not zero, PCRE now checks that it points
1256 to a byte that is the start of a UTF-8 character. If not, it returns
1257 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1258 this is necessary because there may be backward assertions in the pattern.
1259 When matching the same subject several times, it may save resources to use
1260 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1262 4. The code for checking the validity of UTF-8 strings has been tightened so
1263 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1264 containing "overlong sequences".
1266 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1267 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1268 should have been "&", but it just so happened that all the cases this let
1269 through by mistake were picked up later in the function.
1271 6. I had used a variable called "isblank" - this is a C99 function, causing
1272 some compilers to warn. To avoid this, I renamed it (as "blankclass").
1274 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1275 prompting; (b) run "./pcretest /dev/null" at the start of the test script
1276 so the version is shown; (c) stop "make test" echoing "./RunTest".
1278 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1280 9. The prototype for memmove() for systems that don't have it was using
1281 size_t, but the inclusion of the header that defines size_t was later. I've
1282 moved the #includes for the C headers earlier to avoid this.
1284 10. Added some adjustments to the code to make it easier to compiler on certain
1285 special systems:
1287 (a) Some "const" qualifiers were missing.
1288 (b) Added the macro EXPORT before all exported functions; by default this
1289 is defined to be empty.
1290 (c) Changed the dftables auxiliary program (that builds chartables.c) so
1291 that it reads its output file name as an argument instead of writing
1292 to the standard output and assuming this can be redirected.
1294 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1295 class containing characters with values greater than 255, PCRE compilation
1296 went into a loop.
1298 12. A recursive reference to a subpattern that was within another subpattern
1299 that had a minimum quantifier of zero caused PCRE to crash. For example,
1300 (x(y(?2))z)? provoked this bug with a subject that got as far as the
1301 recursion. If the recursively-called subpattern itself had a zero repeat,
1302 that was OK.
1304 13. In pcretest, the buffer for reading a data line was set at 30K, but the
1305 buffer into which it was copied (for escape processing) was still set at
1306 1024, so long lines caused crashes.
1308 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1309 "internal error: code overflow...". This applied to any character class
1310 that was followed by a possessive quantifier.
1312 15. Modified the Makefile to add libpcre.la as a prerequisite for
1313 libpcreposix.la because I was told this is needed for a parallel build to
1314 work.
1316 16. If a pattern that contained .* following optional items at the start was
1317 studied, the wrong optimizing data was generated, leading to matching
1318 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1319 matching string must start with a or b or c. The correct conclusion for
1320 this pattern is that a match can start with any character.
1323 Version 4.4 13-Aug-03
1324 ---------------------
1326 1. In UTF-8 mode, a character class containing characters with values between
1327 127 and 255 was not handled correctly if the compiled pattern was studied.
1328 In fixing this, I have also improved the studying algorithm for such
1329 classes (slightly).
1331 2. Three internal functions had redundant arguments passed to them. Removal
1332 might give a very teeny performance improvement.
1334 3. Documentation bug: the value of the capture_top field in a callout is *one
1335 more than* the number of the hightest numbered captured substring.
1337 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1338 in incorrectly linking with a previously installed version. They now link
1339 explicitly with libpcre.la.
1341 5. configure.in no longer needs to recognize Cygwin specially.
1343 6. A problem in pcre.in for Windows platforms is fixed.
1345 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1346 pcretest, it used to include the size of the study block as part of its
1347 output. Unfortunately, the structure contains a field that has a different
1348 size on different hardware architectures. This meant that the tests that
1349 showed this size failed. As the block is currently always of a fixed size,
1350 this information isn't actually particularly useful in pcretest output, so
1351 I have just removed it.
1353 8. Three pre-processor statements accidentally did not start in column 1.
1354 Sadly, there are *still* compilers around that complain, even though
1355 standard C has not required this for well over a decade. Sigh.
1357 9. In pcretest, the code for checking callouts passed small integers in the
1358 callout_data field, which is a void * field. However, some picky compilers
1359 complained about the casts involved for this on 64-bit systems. Now
1360 pcretest passes the address of the small integer instead, which should get
1361 rid of the warnings.
1363 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1364 both compile and run time, and gives an error if an invalid UTF-8 sequence
1365 is found. There is a option for disabling this check in cases where the
1366 string is known to be correct and/or the maximum performance is wanted.
1368 11. In response to a bug report, I changed one line in Makefile.in from
1370 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1371 to
1372 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1374 to look similar to other lines, but I have no way of telling whether this
1375 is the right thing to do, as I do not use Windows. No doubt I'll get told
1376 if it's wrong...
1379 Version 4.3 21-May-03
1380 ---------------------
1382 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1383 Makefile.
1385 2. Some refactoring to improve the quality of the code:
1387 (i) The utf8_table... variables are now declared "const".
1389 (ii) The code for \cx, which used the "case flipping" table to upper case
1390 lower case letters, now just substracts 32. This is ASCII-specific,
1391 but the whole concept of \cx is ASCII-specific, so it seems
1392 reasonable.
1394 (iii) PCRE was using its character types table to recognize decimal and
1395 hexadecimal digits in the pattern. This is silly, because it handles
1396 only 0-9, a-f, and A-F, but the character types table is locale-
1397 specific, which means strange things might happen. A private
1398 table is now used for this - though it costs 256 bytes, a table is
1399 much faster than multiple explicit tests. Of course, the standard
1400 character types table is still used for matching digits in subject
1401 strings against \d.
1403 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1404 ending in _t are). So I've renamed it as ESC_tee.
1406 3. The first argument for regexec() in the POSIX wrapper should have been
1407 defined as "const".
1409 4. Changed pcretest to use malloc() for its buffers so that they can be
1410 Electric Fenced for debugging.
1412 5. There were several places in the code where, in UTF-8 mode, PCRE would try
1413 to read one or more bytes before the start of the subject string. Often this
1414 had no effect on PCRE's behaviour, but in some circumstances it could
1415 provoke a segmentation fault.
1417 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1418 to try to read one or more bytes before the start of the subject string.
1420 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1421 UTF-8 support could misbehave in various ways if the subject string
1422 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1423 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1424 back over UTF-8 characters.)
1427 Version 4.2 14-Apr-03
1428 ---------------------
1430 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1432 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1433 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1434 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1435 [WIN_PREFIX]: new variable, "cyg" for Cygwin
1436 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1438 Note: automatic setting of the BUILD variables is not yet working
1439 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1440 compile-time but not at link-time
1441 [LINK]: use for linking executables only
1442 make different versions for Windows and non-Windows
1443 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1444 libraries
1445 [LINK_FOR_BUILD]: new variable
1446 [OBJEXT]: use throughout
1447 [EXEEXT]: use throughout
1448 <winshared>: new target
1449 <wininstall>: new target
1450 <dftables.o>: use native compiler
1451 <dftables>: use native linker
1452 <install>: handle Windows platform correctly
1453 <clean>: ditto
1454 <check>: ditto
1455 copy DLL to top builddir before testing
1457 As part of these changes, -no-undefined was removed again. This was reported
1458 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1459 in any case.
1461 3. Some tidies to get rid of compiler warnings:
1463 . In the match_data structure, match_limit was an unsigned long int, whereas
1464 match_call_count was an int. I've made them both unsigned long ints.
1466 . In pcretest the fact that a const uschar * doesn't automatically cast to
1467 a void * provoked a warning.
1469 . Turning on some more compiler warnings threw up some "shadow" variables
1470 and a few more missing casts.
1472 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1473 option, a class that contained a single character with a value between 128
1474 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1476 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1477 option, a class that contained several characters, but with at least one
1478 whose value was between 128 and 255 caused PCRE to crash.
1481 Version 4.1 12-Mar-03
1482 ---------------------
1484 1. Compiling with gcc -pedantic found a couple of places where casts were
1485 needed, and a string in dftables.c that was longer than standard compilers are
1486 required to support.
1488 2. Compiling with Sun's compiler found a few more places where the code could
1489 be tidied up in order to avoid warnings.
1491 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1492 first of these names is deprecated in the latest Autoconf in favour of the name
1493 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1494 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1495 analogy I have changed it to CFLAGS_FOR_BUILD.
1497 4. Added -no-undefined to the linking command in the Makefile, because this is
1498 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1499 linking step for the pcreposix library.
1501 5. PCRE was failing to diagnose the case of two named groups with the same
1502 name.
1504 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1505 literal character that is needed in the subject for a match, and scans along to
1506 ensure that it is present before embarking on the full matching process. This
1507 saves time in cases of nested unlimited repeats that are never going to match.
1508 Problem: the scan can take a lot of time if the subject is very long (e.g.
1509 megabytes), thus penalizing straightforward matches. It is now done only if the
1510 amount of subject to be scanned is less than 1000 bytes.
1512 7. A lesser problem with the same optimization is that it was recording the
1513 first character of an anchored pattern as "needed", thus provoking a search
1514 right along the subject, even when the first match of the pattern was going to
1515 fail. The "needed" character is now not set for anchored patterns, unless it
1516 follows something in the pattern that is of non-fixed length. Thus, it still
1517 fulfils its original purpose of finding quick non-matches in cases of nested
1518 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1521 Version 4.0 17-Feb-03
1522 ---------------------
1524 1. If a comment in an extended regex that started immediately after a meta-item
1525 extended to the end of string, PCRE compiled incorrect data. This could lead to
1526 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1528 2. Moved to autoconf 2.53 and libtool 1.4.2.
1530 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1531 the special perltest8 script is no longer needed - all the tests can be run
1532 from a single perltest script.
1534 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1535 by \s. It has now been removed in PCRE. This means it isn't recognized as
1536 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1537 class [:space:] *does* include VT, thereby creating a mess.
1539 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1540 space and tab.
1542 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1543 its new features into the main test script, reducing the number of scripts.
1545 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1546 were backward compatible, and made the (?i) apply to the whole pattern, as if
1547 /i were given. Now it behaves more logically, and applies the option setting
1548 only to what follows. PCRE has been changed to follow suit. However, if it
1549 finds options settings right at the start of the pattern, it extracts them into
1550 the global options, as before. Thus, they show up in the info data.
1552 8. Added support for the \Q...\E escape sequence. Characters in between are
1553 treated as literals. This is slightly different from Perl in that $ and @ are
1554 also handled as literals inside the quotes. In Perl, they will cause variable
1555 interpolation. Note the following examples:
1557 Pattern PCRE matches Perl matches
1559 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
1560 \Qabc\$xyz\E abc\$xyz abc\$xyz
1561 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
1563 For compatibility with Perl, \Q...\E sequences are recognized inside character
1564 classes as well as outside them.
1566 9. Re-organized 3 code statements in pcretest to avoid "overflow in
1567 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1568 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1569 signed/unsigned warnings.
1571 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1572 option for pcretest, so I've replaced it by a simple function that does just
1573 that job.
1575 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1576 "pcregrep -".
1578 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1579 Java package. This provides some syntactic sugar for simple cases of what my
1580 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1581 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1582 item, you can use this simplified notation. Note that only makes sense with
1583 greedy quantifiers. Consequently, the use of the possessive quantifier forces
1584 greediness, whatever the setting of the PCRE_UNGREEDY option.
1586 13. A change of greediness default within a pattern was not taking effect at
1587 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1588 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1589 was abstracted outside.
1591 14. PCRE now supports the \G assertion. It is true when the current matching
1592 position is at the start point of the match. This differs from \A when the
1593 starting offset is non-zero. Used with the /g option of pcretest (or similar
1594 code), it works in the same way as it does for Perl's /g option. If all
1595 alternatives of a regex begin with \G, the expression is anchored to the start
1596 match position, and the "anchored" flag is set in the compiled expression.
1598 15. Some bugs concerning the handling of certain option changes within patterns
1599 have been fixed. These applied to options other than (?ims). For example,
1600 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1601 been the other way round. Some of this was related to change 7 above.
1603 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1604 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1605 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1606 POSIX classes only within a class (e.g. /[[:alpha:]]/).
1608 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1609 mode. Unlike ".", it always matches newline, whatever the setting of
1610 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1611 assertions. Perl allows it, but it doesn't (in general) work because it can't
1612 calculate the length of the lookbehind. At least, that's the case for Perl
1613 5.8.0 - I've been told they are going to document that it doesn't work in
1614 future.
1616 18. Added an error diagnosis for escapes that PCRE does not support: these are
1617 \L, \l, \N, \P, \p, \U, \u, and \X.
1619 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1620 reading past the end of the pattern in cases such as /[abcd/.
1622 20. PCRE was getting more memory than necessary for patterns with classes that
1623 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1625 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1626 compiling PCRE for use with Virtual Pascal.
1628 22. Small fix to the Makefile to make it work properly if the build is done
1629 outside the source tree.
1631 23. Added a new extension: a condition to go with recursion. If a conditional
1632 subpattern starts with (?(R) the "true" branch is used if recursion has
1633 happened, whereas the "false" branch is used only at the top level.
1635 24. When there was a very long string of literal characters (over 255 bytes
1636 without UTF support, over 250 bytes with UTF support), the computation of how
1637 much memory was required could be incorrect, leading to segfaults or other
1638 strange effects.
1640 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1641 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1642 there was a subsequent back reference to those brackets. This meant that, for
1643 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1644 possible to check for precisely this case. All we can do is abandon the
1645 optimization if .* occurs inside capturing brackets when there are any back
1646 references whatsoever. (See below for a better fix that came later.)
1648 26. The handling of the optimization for finding the first character of a
1649 non-anchored pattern, and for finding a character that is required later in the
1650 match were failing in some cases. This didn't break the matching; it just
1651 failed to optimize when it could. The way this is done has been re-implemented.
1653 27. Fixed typo in error message for invalid (?R item (it said "(?p").
1655 28. Added a new feature that provides some of the functionality that Perl
1656 provides with (?{...}). The facility is termed a "callout". The way it is done
1657 in PCRE is for the caller to provide an optional function, by setting
1658 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1659 global variable. By default it is unset, which disables all calling out. To get
1660 the function called, the regex must include (?C) at appropriate points. This
1661 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1662 This provides a means of identifying different callout points. When PCRE
1663 reaches such a point in the regex, if pcre_callout has been set, the external
1664 function is called. It is provided with data in a structure called
1665 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1666 matching continues; if it returns a non-zero value, the match at the current
1667 point fails. However, backtracking will occur if possible. [This was changed
1668 later and other features added - see item 49 below.]
1670 29. pcretest is upgraded to test the callout functionality. It provides a
1671 callout function that displays information. By default, it shows the start of
1672 the match and the current position in the text. There are some new data escapes
1673 to vary what happens:
1675 \C+ in addition, show current contents of captured substrings
1676 \C- do not supply a callout function
1677 \C!n return 1 when callout number n is reached
1678 \C!n!m return 1 when callout number n is reached for the mth time
1680 30. If pcregrep was called with the -l option and just a single file name, it
1681 output "<stdin>" if a match was found, instead of the file name.
1683 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1684 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1685 pcre_exec(). This saves a malloc/free per call. The default value of
1686 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1687 when configuring.
1689 32. The default maximum size of a compiled pattern is 64K. There have been a
1690 few cases of people hitting this limit. The code now uses macros to handle the
1691 storing of links as offsets within the compiled pattern. It defaults to 2-byte
1692 links, but this can be changed to 3 or 4 bytes by --with-link-size when
1693 configuring. Tests 2 and 5 work only with 2-byte links because they output
1694 debugging information about compiled patterns.
1696 33. Internal code re-arrangements:
1698 (a) Moved the debugging function for printing out a compiled regex into
1699 its own source file (printint.c) and used #include to pull it into
1700 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1701 separate copies.
1703 (b) Defined the list of op-code names for debugging as a macro in
1704 internal.h so that it is next to the definition of the opcodes.
1706 (c) Defined a table of op-code lengths for simpler skipping along compiled
1707 code. This is again a macro in internal.h so that it is next to the
1708 definition of the opcodes.
1710 34. Added support for recursive calls to individual subpatterns, along the
1711 lines of Robin Houston's patch (but implemented somewhat differently).
1713 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1714 allow it to read and process whole directories in Win32. This code was
1715 contributed by Lionel Fourquaux; it has not been tested by me.
1717 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1718 used to name a group. Names consist of alphanumerics and underscores, and must
1719 be unique. Back references use the syntax (?P=name) and recursive calls use
1720 (?P>name) which is a PCRE extension to the Python extension. Groups still have
1721 numbers. The function pcre_fullinfo() can be used after compilation to extract
1722 a name/number map. There are three relevant calls:
1724 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
1725 PCRE_INFO_NAMECOUNT yields the number of entries
1726 PCRE_INFO_NAMETABLE yields a pointer to the map.
1728 The map is a vector of fixed-size entries. The size of each entry depends on
1729 the length of the longest name used. The first two bytes of each entry are the
1730 group number, most significant byte first. There follows the corresponding
1731 name, zero terminated. The names are in alphabetical order.
1733 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1734 case instead of 255. Making it the same both with and without UTF-8 support
1735 means that the same test output works with both.
1737 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1738 calling malloc() with a zero argument.
1740 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1741 optimization. I've improved things by keeping a bitmap of backreferences with
1742 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1743 fact referenced, the optimization can be applied. It is unlikely that a
1744 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1745 the match to follow \n) will appear inside brackets with a number greater than
1746 31, but if it does, any back reference > 31 suppresses the optimization.
1748 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1749 of disabling numbered capturing parentheses. Any opening parenthesis that is
1750 not followed by ? behaves as if it were followed by ?: but named parentheses
1751 can still be used for capturing (and they will acquire numbers in the usual
1752 way).
1754 41. Redesigned the return codes from the match() function into yes/no/error so
1755 that errors can be passed back from deep inside the nested calls. A malloc
1756 failure while inside a recursive subpattern call now causes the
1757 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
1759 42. It is now possible to set a limit on the number of times the match()
1760 function is called in a call to pcre_exec(). This facility makes it possible to
1761 limit the amount of recursion and backtracking, though not in a directly
1762 obvious way, because the match() function is used in a number of different
1763 circumstances. The count starts from zero for each position in the subject
1764 string (for non-anchored patterns). The default limit is, for compatibility, a
1765 large number, namely 10 000 000. You can change this in two ways:
1767 (a) When configuring PCRE before making, you can use --with-match-limit=n
1768 to set a default value for the compiled library.
1770 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
1771 a different value is set. See 45 below.
1773 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1775 43. Added a new function pcre_config(int, void *) to enable run-time extraction
1776 of things that can be changed at compile time. The first argument specifies
1777 what is wanted and the second points to where the information is to be placed.
1778 The current list of available information is:
1782 The output is an integer that is set to one if UTF-8 support is available;
1783 otherwise it is set to zero.
1787 The output is an integer that it set to the value of the code that is used for
1788 newline. It is either LF (10) or CR (13).
1792 The output is an integer that contains the number of bytes used for internal
1793 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
1797 The output is an integer that contains the threshold above which the POSIX
1798 interface uses malloc() for output vectors. See item 31 above.
1802 The output is an unsigned integer that contains the default limit of the number
1803 of match() calls in a pcre_exec() execution. See 42 above.
1805 44. pcretest has been upgraded by the addition of the -C option. This causes it
1806 to extract all the available output from the new pcre_config() function, and to
1807 output it. The program then exits immediately.
1809 45. A need has arisen to pass over additional data with calls to pcre_exec() in
1810 order to support additional features. One way would have been to define
1811 pcre_exec2() (for example) with extra arguments, but this would not have been
1812 extensible, and would also have required all calls to the original function to
1813 be mapped to the new one. Instead, I have chosen to extend the mechanism that
1814 is used for passing in "extra" data from pcre_study().
1816 The pcre_extra structure is now exposed and defined in pcre.h. It currently
1817 contains the following fields:
1819 flags a bitmap indicating which of the following fields are set
1820 study_data opaque data from pcre_study()
1821 match_limit a way of specifying a limit on match() calls for a specific
1822 call to pcre_exec()
1823 callout_data data for callouts (see 49 below)
1825 The flag bits are also defined in pcre.h, and are
1831 The pcre_study() function now returns one of these new pcre_extra blocks, with
1832 the actual study data pointed to by the study_data field, and the
1833 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
1834 before. That is, this change is entirely upwards-compatible and requires no
1835 change to existing code.
1837 If you want to pass in additional data to pcre_exec(), you can either place it
1838 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
1839 block.
1841 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
1842 data string contains the escape sequence \M, pcretest calls pcre_exec() several
1843 times with different match limits, until it finds the minimum value needed for
1844 pcre_exec() to complete. The value is then output. This can be instructive; for
1845 most simple matches the number is quite small, but for pathological cases it
1846 gets very large very quickly.
1848 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
1849 returns the size of the data block pointed to by the study_data field in a
1850 pcre_extra block, that is, the value that was passed as the argument to
1851 pcre_malloc() when PCRE was getting memory in which to place the information
1852 created by pcre_study(). The fourth argument should point to a size_t variable.
1853 pcretest has been extended so that this information is shown after a successful
1854 pcre_study() call when information about the compiled regex is being displayed.
1856 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
1857 because what follows is always an absolute path. (Later: it turns out that this
1858 is more than cosmetic for MinGW, because it doesn't like empty path
1859 components.)
1861 49. Some changes have been made to the callout feature (see 28 above):
1863 (i) A callout function now has three choices for what it returns:
1865 0 => success, carry on matching
1866 > 0 => failure at this point, but backtrack if possible
1867 < 0 => serious error, return this value from pcre_exec()
1869 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
1870 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
1871 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
1872 use by callout functions. It will never be used by PCRE itself.
1874 (ii) The pcre_extra structure (see 45 above) has a void * field called
1875 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
1876 pcre_callout_block structure has a field of the same name. The contents of
1877 the field passed in the pcre_extra structure are passed to the callout
1878 function in the corresponding field in the callout block. This makes it
1879 easier to use the same callout-containing regex from multiple threads. For
1880 testing, the pcretest program has a new data escape
1882 \C*n pass the number n (may be negative) as callout_data
1884 If the callout function in pcretest receives a non-zero value as
1885 callout_data, it returns that value.
1887 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
1888 there were some redundant $(CFLAGS) in commands that are now specified as
1889 $(LINK), which already includes $(CFLAGS).
1891 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
1892 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
1893 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
1894 one-byte characters throughout. Note that case-insensitive matching applies
1895 only to characters whose values are less than 256. PCRE doesn't support the
1896 notion of cases for higher-valued characters.
1898 (i) A character class whose characters are all within 0-255 is handled as
1899 a bit map, and the map is inverted for negative classes. Previously, a
1900 character > 255 always failed to match such a class; however it should
1901 match if the class was a negative one (e.g. [^ab]). This has been fixed.
1903 (ii) A negated character class with a single character < 255 is coded as
1904 "not this character" (OP_NOT). This wasn't working properly when the test
1905 character was multibyte, either singly or repeated.
1907 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
1908 mode, for example: \x{100}{2,3}.
1910 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
1911 singly or repeated) now correctly test multibyte characters. However,
1912 PCRE doesn't recognize any characters with values greater than 255 as
1913 digits, spaces, or word characters. Such characters always match \D, \S,
1914 and \W, and never match \d, \s, or \w.
1916 (v) Classes may now contain characters and character ranges with values
1917 greater than 255. For example: [ab\x{100}-\x{400}].
1919 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
1920 PCRE in UTF-8 mode.
1922 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
1923 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
1924 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
1925 value.)
1927 53. The single man page has become too large. I have therefore split it up into
1928 a number of separate man pages. These also give rise to individual HTML pages;
1929 these are now put in a separate directory, and there is an index.html page that
1930 lists them all. Some hyperlinking between the pages has been installed.
1932 54. Added convenience functions for handling named capturing parentheses.
1934 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
1935 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
1936 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
1937 are faulted.
1939 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
1940 calling configure. These values are used when compiling the dftables.c program
1941 which is run to generate the source of the default character tables. They
1942 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
1943 you will need to set these values.
1945 57. Updated the building process for Windows DLL, as provided by Fred Cox.
1948 Version 3.9 02-Jan-02
1949 ---------------------
1951 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
1953 2. If --disable-static was given, the building process failed when trying to
1954 build pcretest and pcregrep. (For some reason it was using libtool to compile
1955 them, which is not right, as they aren't part of the library.)
1958 Version 3.8 18-Dec-01
1959 ---------------------
1961 1. The experimental UTF-8 code was completely screwed up. It was packing the
1962 bytes in the wrong order. How dumb can you get?
1965 Version 3.7 29-Oct-01
1966 ---------------------
1968 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
1969 This caused pcretest, when used on the test data, to segfault. Unfortunately,
1970 this didn't happen under Solaris 8, where I normally test things.
1972 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
1973 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
1974 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
1975 this fix an hour or so after the initial 3.7 release.)
1978 Version 3.6 23-Oct-01
1979 ---------------------
1981 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
1982 offsets passed as NULL with zero offset count.
1984 2. The config.guess and config.sub files had not been updated when I moved to
1985 the latest autoconf.
1988 Version 3.5 15-Aug-01
1989 ---------------------
1991 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
1992 had been forgotten.
1994 2. By using declared but undefined structures, we can avoid using "void"
1995 definitions in pcre.h while keeping the internal definitions of the structures
1996 private.
1998 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
1999 user point of view, this means that both static and shared libraries are built
2000 by default, but this can be individually controlled. More of the work of
2001 handling this static/shared cases is now inside libtool instead of PCRE's make
2002 file.
2004 4. The pcretest utility is now installed along with pcregrep because it is
2005 useful for users (to test regexs) and by doing this, it automatically gets
2006 relinked by libtool. The documentation has been turned into a man page, so
2007 there are now .1, .txt, and .html versions in /doc.
2009 5. Upgrades to pcregrep:
2010 (i) Added long-form option names like gnu grep.
2011 (ii) Added --help to list all options with an explanatory phrase.
2012 (iii) Added -r, --recursive to recurse into sub-directories.
2013 (iv) Added -f, --file to read patterns from a file.
2015 6. pcre_exec() was referring to its "code" argument before testing that
2016 argument for NULL (and giving an error if it was NULL).
2018 7. Upgraded Makefile.in to allow for compiling in a different directory from
2019 the source directory.
2021 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2022 options bits, the pointer it was passed was to an int instead of to an unsigned
2023 long int. This mattered only on 64-bit systems.
2025 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2026 generated) instead of pcre.in, which it its source. Also made the same change
2027 in several of the .c files.
2029 10. A new release of gcc defines printf() as a macro, which broke pcretest
2030 because it had an ifdef in the middle of a string argument for printf(). Fixed
2031 by using separate calls to printf().
2033 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2034 script, to force use of CR or LF instead of \n in the source. On non-Unix
2035 systems, the value can be set in config.h.
2037 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2038 absolute limit. Changed the text of the error message to make this clear, and
2039 likewise updated the man page.
2041 13. The limit of 99 on the number of capturing subpatterns has been removed.
2042 The new limit is 65535, which I hope will not be a "real" limit.
2045 Version 3.4 22-Aug-00
2046 ---------------------
2048 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2050 2. Diagnose condition (?(0) as an error instead of crashing on matching.
2053 Version 3.3 01-Aug-00
2054 ---------------------
2056 1. If an octal character was given, but the value was greater than \377, it
2057 was not getting masked to the least significant bits, as documented. This could
2058 lead to crashes in some systems.
2060 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2061 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2063 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2064 These just pass their arguments on to (pcre_free)(), but they are provided
2065 because some uses of PCRE bind it to non-C systems that can call its functions,
2066 but cannot call free() or pcre_free() directly.
2068 4. Add "make test" as a synonym for "make check". Corrected some comments in
2069 the Makefile.
2071 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2072 Makefile.
2074 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2075 command called pgrep for grepping around the active processes.
2077 7. Added the beginnings of support for UTF-8 character strings.
2079 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2080 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2081 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2082 out for the ar command.)
2085 Version 3.2 12-May-00
2086 ---------------------
2088 This is purely a bug fixing release.
2090 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2091 of ZA. This was just one example of several cases that could provoke this bug,
2092 which was introduced by change 9 of version 2.00. The code for breaking
2093 infinite loops after an iteration that matches an empty string was't working
2094 correctly.
2096 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2097 when matched against abbab (for example). After matching an empty string, it
2098 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2099 caused it to match further down the string than it should.
2101 3. The code contained an inclusion of sys/types.h. It isn't clear why this
2102 was there because it doesn't seem to be needed, and it causes trouble on some
2103 systems, as it is not a Standard C header. It has been removed.
2105 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2106 were reported on the Macintosh. The changes were from
2108 while ((c = *(++ptr)) != 0 && c != '\n');
2109 to
2110 while ((c = *(++ptr)) != 0 && c != '\n') ;
2112 Totally extraordinary, but if that's what it takes...
2114 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2115 available. Added HAVE_BCOPY and an autoconf test for it; if neither
2116 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2117 assumes the way PCRE uses memmove() (always moving upwards).
2119 6. PCRE is being used in one environment where strchr() is not available. There
2120 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2121 faster code anyway.
2124 Version 3.1 09-Feb-00
2125 ---------------------
2127 The only change in this release is the fixing of some bugs in Makefile.in for
2128 the "install" target:
2130 (1) It was failing to install pcreposix.h.
2132 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2135 Version 3.0 01-Feb-00
2136 ---------------------
2138 1. Add support for the /+ modifier to perltest (to output $` like it does in
2139 pcretest).
2141 2. Add support for the /g modifier to perltest.
2143 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2144 matches null strings.
2146 4. Fix perltest so that it doesn't do unwanted things when fed an empty
2147 pattern. Perl treats empty patterns specially - it reuses the most recent
2148 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2149 effect.
2151 5. The POSIX interface was broken in that it was just handing over the POSIX
2152 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2153 required a bigger vector, with some working space on the end. This means that
2154 the POSIX wrapper now has to get and free some memory, and copy the results.
2156 6. Added some simple autoconf support, placing the test data and the
2157 documentation in separate directories, re-organizing some of the
2158 information files, and making it build pcre-config (a GNU standard). Also added
2159 libtool support for building PCRE as a shared library, which is now the
2160 default.
2162 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2163 09 are not valid octal constants. Single digits will be used for minor values
2164 less than 10.
2166 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2167 existing programs that set these in the POSIX interface can use PCRE without
2168 modification.
2170 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2171 return all that pcre_info() returns, plus additional data. The pcre_info()
2172 function is retained for compatibility, but is considered to be obsolete.
2174 10. Added experimental recursion feature (?R) to handle one common case that
2175 Perl 5.6 will be able to do with (?p{...}).
2177 11. Added support for POSIX character classes like [:alpha:], which Perl is
2178 adopting.
2181 Version 2.08 31-Aug-99
2182 ----------------------
2184 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2185 trying to match at the startoffset position, but instead was moving forward to
2186 the next newline as if a previous match had failed.
2188 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2189 and could get into a loop if a null string was matched other than at the start
2190 of the subject.
2192 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2193 be distinguished at compile time, and for completeness also added PCRE_DATE.
2195 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2196 in GnuWin32 environments.
2199 Version 2.07 29-Jul-99
2200 ----------------------
2202 1. The documentation is now supplied in plain text form and HTML as well as in
2203 the form of man page sources.
2205 2. C++ compilers don't like assigning (void *) values to other pointer types.
2206 In particular this affects malloc(). Although there is no problem in Standard
2207 C, I've put in casts to keep C++ compilers happy.
2209 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2210 should be (const char *).
2212 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2213 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2214 However, I haven't made this a standard facility. The documentation doesn't
2215 mention it, and the Makefile doesn't support it.
2217 5. The Makefile now contains an "install" target, with editable destinations at
2218 the top of the file. The pcretest program is not installed.
2220 6. pgrep -V now gives the PCRE version number and date.
2222 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2223 causing the entire string to be ignored, instead of just the last character.
2225 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2226 non-matching string, it can take a very, very long time, even for strings of
2227 quite modest length, because of the nested recursion. PCRE now does better in
2228 some of these cases. It does this by remembering the last required literal
2229 character in the pattern, and pre-searching the subject to ensure it is present
2230 before running the real match. In other words, it applies a heuristic to detect
2231 some types of certain failure quickly, and in the above example, if presented
2232 with a string that has no trailing " it gives "no match" very quickly.
2234 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2235 other alternatives are tried instead.
2238 Version 2.06 09-Jun-99
2239 ----------------------
2241 1. Change pcretest's output for amount of store used to show just the code
2242 space, because the remainder (the data block) varies in size between 32-bit and
2243 64-bit systems.
2245 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2246 start matching at. This allows lookbehinds to work when searching for multiple
2247 occurrences in a string.
2249 3. Added additional options to pcretest for testing multiple occurrences:
2251 /+ outputs the rest of the string that follows a match
2252 /g loops for multiple occurrences, using the new startoffset argument
2253 /G loops for multiple occurrences by passing an incremented pointer
2255 4. PCRE wasn't doing the "first character" optimization for patterns starting
2256 with \b or \B, though it was doing it for other lookbehind assertions. That is,
2257 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2258 the letter 'x'. On long subject strings, this gives a significant speed-up.
2261 Version 2.05 21-Apr-99
2262 ----------------------
2264 1. Changed the type of magic_number from int to long int so that it works
2265 properly on 16-bit systems.
2267 2. Fixed a bug which caused patterns starting with .* not to work correctly
2268 when the subject string contained newline characters. PCRE was assuming
2269 anchoring for such patterns in all cases, which is not correct because .* will
2270 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2271 DOTALL is set at top level; otherwise it knows that patterns starting with .*
2272 must be retried after every newline in the subject.
2275 Version 2.04 18-Feb-99
2276 ----------------------
2278 1. For parenthesized subpatterns with repeats whose minimum was zero, the
2279 computation of the store needed to hold the pattern was incorrect (too large).
2280 If such patterns were nested a few deep, this could multiply and become a real
2281 problem.
2283 2. Added /M option to pcretest to show the memory requirement of a specific
2284 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2286 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2287 compiled in such a way that the backtracking after subsequent failure was
2288 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2289 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2292 Version 2.03 02-Feb-99
2293 ----------------------
2295 1. Fixed typo and small mistake in man page.
2297 2. Added 4th condition (GPL supersedes if conflict) and created separate
2298 LICENCE file containing the conditions.
2300 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2301 Perl, that is the internal \ allows the delimiter to be included in the
2302 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2303 the final delimiter, add \ to the end of the pattern (to test the error).
2305 4. Added the convenience functions for extracting substrings after a successful
2306 match. Updated pcretest to make it able to test these functions.
2309 Version 2.02 14-Jan-99
2310 ----------------------
2312 1. Initialized the working variables associated with each extraction so that
2313 their saving and restoring doesn't refer to uninitialized store.
2315 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2316 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2317 fix the problem.
2319 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2320 calls, and wasn't printing the correct value for compiling calls. Increased the
2321 default value of LOOPREPEAT, and the number of significant figures in the
2322 times.
2324 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2326 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2327 a building problem on Windows NT with a FAT file system.
2330 Version 2.01 21-Oct-98
2331 ----------------------
2333 1. Changed the API for pcre_compile() to allow for the provision of a pointer
2334 to character tables built by pcre_maketables() in the current locale. If NULL
2335 is passed, the default tables are used.
2338 Version 2.00 24-Sep-98
2339 ----------------------
2341 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2342 it any more.
2344 2. Allow quantification of (?>) groups, and make it work correctly.
2346 3. The first character computation wasn't working for (?>) groups.
2348 4. Correct the implementation of \Z (it is permitted to match on the \n at the
2349 end of the subject) and add 5.005's \z, which really does match only at the
2350 very end of the subject.
2352 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2354 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2355 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2356 localized options. All options to pcre_study() were also removed.
2358 7. Add other new features from 5.005:
2360 $(?<= positive lookbehind
2361 $(?<! negative lookbehind
2362 (?imsx-imsx) added the unsetting capability
2363 such a setting is global if at outer level; local otherwise
2364 (?imsx-imsx:) non-capturing groups with option setting
2365 (?(cond)re|re) conditional pattern matching
2367 A backreference to itself in a repeated group matches the previous
2368 captured string.
2370 8. General tidying up of studying (both automatic and via "study")
2371 consequential on the addition of new assertions.
2373 9. As in 5.005, unlimited repeated groups that could match an empty substring
2374 are no longer faulted at compile time. Instead, the loop is forcibly broken at
2375 runtime if any iteration does actually match an empty substring.
2377 10. Include the RunTest script in the distribution.
2379 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2380 discrepancies, some of which were old and were also with respect to 5.004. They
2381 have now been fixed.
2384 Version 1.09 28-Apr-98
2385 ----------------------
2387 1. A negated single character class followed by a quantifier with a minimum
2388 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
2389 program crashes, or just wrong answers. This did not apply to negated classes
2390 containing more than one character, or to minima other than one.
2393 Version 1.08 27-Mar-98
2394 ----------------------
2396 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
2398 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
2399 latter must appear before anything that relies on it in the pattern.
2402 Version 1.07 16-Feb-98
2403 ----------------------
2405 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
2406 repeat of a potentially empty string).
2409 Version 1.06 23-Jan-98
2410 ----------------------
2412 1. Added Markus Oberhumer's little patches for C++.
2414 2. Literal strings longer than 255 characters were broken.
2417 Version 1.05 23-Dec-97
2418 ----------------------
2420 1. Negated character classes containing more than one character were failing if
2421 PCRE_CASELESS was set at run time.
2424 Version 1.04 19-Dec-97
2425 ----------------------
2427 1. Corrected the man page, where some "const" qualifiers had been omitted.
2429 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
2430 input syntax.
2432 3. Fixed memory leak which occurred when a regex with back references was
2433 matched with an offsets vector that wasn't big enough. The temporary memory
2434 that is used in this case wasn't being freed if the match failed.
2436 4. Tidied pcretest to ensure it frees memory that it gets.
2438 5. Temporary memory was being obtained in the case where the passed offsets
2439 vector was exactly big enough.
2441 6. Corrected definition of offsetof() from change 5 below.
2443 7. I had screwed up change 6 below and broken the rules for the use of
2444 setjmp(). Now fixed.
2447 Version 1.03 18-Dec-97
2448 ----------------------
2450 1. A erroneous regex with a missing opening parenthesis was correctly
2451 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
2452 on some systems.
2454 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
2455 it was reported that one broken compiler failed on the former because "code" is
2456 also an independent variable.
2458 3. The erroneous regex a[]b caused an array overrun reference.
2460 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
2461 fail on data ending with that character. (It was going on too far, and checking
2462 the next character, typically a binary zero.) This was specific to the
2463 optimized code for single-character negative classes.
2465 5. Added a contributed patch from the TIN world which does the following:
2467 + Add an undef for memmove, in case the the system defines a macro for it.
2469 + Add a definition of offsetof(), in case there isn't one. (I don't know
2470 the reason behind this - offsetof() is part of the ANSI standard - but
2471 it does no harm).
2473 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
2474 most of the places where whitespace preceded '#'. I have given up and
2475 allowed the remaining 2 cases to be at the margin.
2477 + Rename some variables in pcre to eliminate shadowing. This seems very
2478 pedantic, but does no harm, of course.
2480 6. Moved the call to setjmp() into its own function, to get rid of warnings
2481 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
2483 7. Constructs such as \d{8,} were compiling into the equivalent of
2484 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
2485 outcome, but in this particular case used more store than had been allocated,
2486 which caused the bug to be discovered because it threw up an internal error.
2488 8. The debugging code in both pcre and pcretest for outputting the compiled
2489 form of a regex was going wrong in the case of back references followed by
2490 curly-bracketed repeats.
2493 Version 1.02 12-Dec-97
2494 ----------------------
2496 1. Typos in pcre.3 and comments in the source fixed.
2498 2. Applied a contributed patch to get rid of places where it used to remove
2499 'const' from variables, and fixed some signed/unsigned and uninitialized
2500 variable warnings.
2502 3. Added the "runtest" target to Makefile.
2504 4. Set default compiler flag to -O2 rather than just -O.
2507 Version 1.01 19-Nov-97
2508 ----------------------
2510 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
2511 like /([ab]*)*/, that is, for classes with more than one character in them.
2513 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
2514 as /((?>a*))*/ (a PCRE_EXTRA facility).
2517 Version 1.00 18-Nov-97
2518 ----------------------
2520 1. Added compile-time macros to support systems such as SunOS4 which don't have
2521 memmove() or strerror() but have other things that can be used instead.
2523 2. Arranged that "make clean" removes the executables.
2526 Version 0.99 27-Oct-97
2527 ----------------------
2529 1. Fixed bug in code for optimizing classes with only one character. It was
2530 initializing a 32-byte map regardless, which could cause it to run off the end
2531 of the memory it had got.
2533 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
2536 Version 0.98 22-Oct-97
2537 ----------------------
2539 1. Fixed bug in code for handling temporary memory usage when there are more
2540 back references than supplied space in the ovector. This could cause segfaults.
2543 Version 0.97 21-Oct-97
2544 ----------------------
2546 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
2548 2. Optimized negated single characters not to use a bit map.
2550 3. Brought error texts together as macro definitions; clarified some of them;
2551 fixed one that was wrong - it said "range out of order" when it meant "invalid
2552 escape sequence".
2554 4. Changed some char * arguments to const char *.
2556 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
2558 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
2559 pcretest.
2562 Version 0.96 16-Oct-97
2563 ----------------------
2565 1. Added a simple "pgrep" utility to the distribution.
2567 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
2568 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
2569 where "ddd" means "one or more decimal digits".
2571 3. Fixed serious bug. If a pattern had a back reference, but the call to
2572 pcre_exec() didn't supply a large enough ovector to record the related
2573 identifying subpattern, the match always failed. PCRE now remembers the number
2574 of the largest back reference, and gets some temporary memory in which to save
2575 the offsets during matching if necessary, in order to ensure that
2576 backreferences always work.
2578 4. Increased the compatibility with Perl in a number of ways:
2580 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
2581 to request this handling. The option can be set at compile or exec time.
2583 (b) $ matches before a terminating newline by default; an option
2584 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
2585 mode). The option can be set at compile or exec time.
2587 (c) The handling of \ followed by a digit other than 0 is now supposed to be
2588 the same as Perl's. If the decimal number it represents is less than 10
2589 or there aren't that many previous left capturing parentheses, an octal
2590 escape is read. Inside a character class, it's always an octal escape,
2591 even if it is a single digit.
2593 (d) An escaped but undefined alphabetic character is taken as a literal,
2594 unless PCRE_EXTRA is set. Currently this just reserves the remaining
2595 escapes.
2597 (e) {0} is now permitted. (The previous item is removed from the compiled
2598 pattern).
2600 5. Changed all the names of code files so that the basic parts are no longer
2601 than 10 characters, and abolished the teeny "globals.c" file.
2603 6. Changed the handling of character classes; they are now done with a 32-byte
2604 bit map always.
2606 7. Added the -d and /D options to pcretest to make it possible to look at the
2607 internals of compilation without having to recompile pcre.
2610 Version 0.95 23-Sep-97
2611 ----------------------
2613 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
2614 \x20 at the start of a run of normal characters. These were being treated as
2615 real characters, instead of the source characters being re-checked.
2618 Version 0.94 18-Sep-97
2619 ----------------------
2621 1. The functions are now thread-safe, with the caveat that the global variables
2622 containing pointers to malloc() and free() or alternative functions are the
2623 same for all threads.
2625 2. Get pcre_study() to generate a bitmap of initial characters for non-
2626 anchored patterns when this is possible, and use it if passed to pcre_exec().
2629 Version 0.93 15-Sep-97
2630 ----------------------
2632 1. /(b)|(:+)/ was computing an incorrect first character.
2634 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
2635 but not actually doing anything yet.
2637 3. Treat "-" characters in classes that cannot be part of ranges as literals,
2638 as Perl does (e.g. [-az] or [az-]).
2640 4. Set the anchored flag if a branch starts with .* or .*? because that tests
2641 all possible positions.
2643 5. Split up into different modules to avoid including unneeded functions in a
2644 compiled binary. However, compile and exec are still in one module. The "study"
2645 function is split off.
2647 6. The character tables are now in a separate module whose source is generated
2648 by an auxiliary program - but can then be edited by hand if required. There are
2649 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
2650 toupper() in the code.
2652 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
2653 make them global. Abolish the function for setting them, as the caller can now
2654 set them directly.
2657 Version 0.92 11-Sep-97
2658 ----------------------
2660 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
2661 (e.g. /a{1,3}/) was broken (I mis-optimized it).
2663 2. Caseless matching was not working in character classes if the characters in
2664 the pattern were in upper case.
2666 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
2668 4. Make PCRE_ANCHORED public and accept as a compile option.
2670 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
2671 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
2672 pass them.
2674 6. Give an error if bad option bits passed at compile or run time.
2676 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
2677 pcretest to cause it to pass that flag.
2679 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
2680 options, and the first character, if set.
2682 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
2685 Version 0.91 10-Sep-97
2686 ----------------------
2688 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
2689 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
2691 2. PCRE was looping on encountering an indefinitely repeated back reference to
2692 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
2693 Perl does - treats the match as successful.
2695 ****


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

  ViewVC Help
Powered by ViewVC 1.1.5