ViewVC logotype

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 203 - (show annotations)
Fri Aug 3 10:09:57 2007 UTC (14 years, 1 month ago) by ph10
File size: 133780 byte(s)
Fix doc bug for duplicate named subpatterns.
1 ChangeLog for PCRE
2 ------------------
4 Version 7.3 05-Jul-07
5 ---------------------
7 1. In the rejigging of the build system that eventually resulted in 7.1, the
8 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
9 brackets there is not right, since it causes compilers to look for an
10 installed pcre.h, not the version that is in the source that is being
11 compiled (which of course may be different). I have changed it back to:
13 #include "pcre.h"
15 I have a vague recollection that the change was concerned with compiling in
16 different directories, but in the new build system, that is taken care of
17 by the VPATH setting the Makefile.
19 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
20 when the subject happened to end in the byte 0x85 (e.g. if the last
21 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
22 characters but of course it shouldn't be taken as a newline when it is part
23 of another character. The bug was that, for an unlimited repeat of . in
24 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
25 characters when looking for a newline.
27 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
29 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
30 in debug output.
32 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
33 long printing in the pcrecpp unittest when running under MinGW.
35 6. ESC_K was left out of the EBCDIC table.
37 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
38 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
39 limit also applies to "virtual nesting" when a pattern is recursive, and in
40 this case 1000 isn't so big. I have been able to remove this limit at the
41 expense of backing off one optimization in certain circumstances. Normally,
42 when pcre_exec() would call its internal match() function recursively and
43 immediately return the result unconditionally, it uses a "tail recursion"
44 feature to save stack. However, when a subpattern that can match an empty
45 string has an unlimited repetition quantifier, it no longer makes this
46 optimization. That gives it a stack frame in which to save the data for
47 checking that an empty string has been matched. Previously this was taken
48 from the 1000-entry workspace that had been reserved. So now there is no
49 explicit limit, but more stack is used.
51 8. Applied Daniel's patches to solve problems with the import/export magic
52 syntax that is required for Windows, and which was going wrong for the
53 pcreposix and pcrecpp parts of the library. These were overlooked when this
54 problem was solved for the main library.
56 9. There were some crude static tests to avoid integer overflow when computing
57 the size of patterns that contain repeated groups with explicit upper
58 limits. As the maximum quantifier is 65535, the maximum group length was
59 set at 30,000 so that the product of these two numbers did not overflow a
60 32-bit integer. However, it turns out that people want to use groups that
61 are longer than 30,000 bytes (though not repeat them that many times).
62 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
63 made it possible to implement the integer overflow checks in a much more
64 dynamic way, which I have now done. The artificial limitation on group
65 length has been removed - we now have only the limit on the total length of
66 the compiled pattern, which depends on the LINK_SIZE setting.
68 10. Fixed a bug in the documentation for get/copy named substring when
69 duplicate names are permitted. If none of the named substrings are set, the
70 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
71 empty string.
74 Version 7.2 19-Jun-07
75 ---------------------
77 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
78 which is apparently normally available under Windows.
80 2. Re-jig the pcregrep tests with different newline settings in an attempt
81 to make them independent of the local environment's newline setting.
83 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
85 4. Some of the "internals" tests were previously cut out when the link size
86 was not 2, because the output contained actual offsets. The recent new
87 "Z" feature of pcretest means that these can be cut out, making the tests
88 usable with all link sizes.
90 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
91 stack recursion. This gives a massive performance boost under BSD, but just
92 a small improvement under Linux. However, it saves one field in the frame
93 in all cases.
95 6. Added more features from the forthcoming Perl 5.10:
97 (a) (?-n) (where n is a string of digits) is a relative subroutine or
98 recursion call. It refers to the nth most recently opened parentheses.
100 (b) (?+n) is also a relative subroutine call; it refers to the nth next
101 to be opened parentheses.
103 (c) Conditions that refer to capturing parentheses can be specified
104 relatively, for example, (?(-2)... or (?(+3)...
106 (d) \K resets the start of the current match so that everything before
107 is not part of it.
109 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
111 (f) \g{name} is another synonym - part of Perl 5.10's unification of
112 reference syntax.
114 (g) (?| introduces a group in which the numbering of parentheses in each
115 alternative starts with the same number.
117 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
119 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
122 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
123 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
124 for detecting groups that can match an empty string.
126 9. A pattern with a very large number of alternatives (more than several
127 hundred) was running out of internal workspace during the pre-compile
128 phase, where pcre_compile() figures out how much memory will be needed. A
129 bit of new cunning has reduced the workspace needed for groups with
130 alternatives. The 1000-alternative test pattern now uses 12 bytes of
131 workspace instead of running out of the 4096 that are available.
133 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
135 11. Applied patch from Google to remove an optimization that didn't quite work.
136 The report of the bug said:
138 pcrecpp::RE("a*").FullMatch("aaa") matches, while
139 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
140 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
142 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
143 it matched the wrong number of bytes.
146 Version 7.1 24-Apr-07
147 ---------------------
149 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
150 that is more "standard", making use of automake and other Autotools. There
151 is some re-arrangement of the files and adjustment of comments consequent
152 on this.
154 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
155 for recursive directory scanning broke on some systems because the files
156 are not scanned in any specific order and on different systems the order
157 was different. A call to "sort" has been inserted into RunGrepTest for the
158 approprate test as a short-term fix. In the longer term there may be an
159 alternative.
161 3. I had an email from Eric Raymond about problems translating some of PCRE's
162 man pages to HTML (despite the fact that I distribute HTML pages, some
163 people do their own conversions for various reasons). The problems
164 concerned the use of low-level troff macros .br and .in. I have therefore
165 removed all such uses from the man pages (some were redundant, some could
166 be replaced by .nf/.fi pairs). The 132html script that I use to generate
167 HTML has been updated to handle .nf/.fi and to complain if it encounters
168 .br or .in.
170 4. Updated comments in configure.ac that get placed in config.h.in and also
171 arranged for config.h to be included in the distribution, with the name
172 config.h.generic, for the benefit of those who have to compile without
173 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
175 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
176 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
177 makevp.bat for the current PCRE, using the additional files
178 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
180 6. A Windows user reported a minor discrepancy with test 2, which turned out
181 to be caused by a trailing space on an input line that had got lost in his
182 copy. The trailing space was an accident, so I've just removed it.
184 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
185 that is needed.
187 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
188 as "const" (a) because they are and (b) because it helps the PHP
189 maintainers who have recently made a script to detect big data structures
190 in the php code that should be moved to the .rodata section. I remembered
191 to update Builducptable as well, so it won't revert if ucptable.h is ever
192 re-created.
194 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
195 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
196 order to be able to cut out the UTF-8 tables in the latter when UTF-8
197 support is not required. This saves 1.5-2K of code, which is important in
198 some applications.
200 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
201 so as not to refer to the tables, even though these functions will never be
202 called when UTF-8 support is disabled. Otherwise there are problems with a
203 shared library.
205 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
207 (a) It was defining its arguments as char * instead of void *.
209 (b) It was assuming that all moves were upwards in memory; this was true
210 a long time ago when I wrote it, but is no longer the case.
212 The emulated memove() is provided for those environments that have neither
213 memmove() nor bcopy(). I didn't think anyone used it these days, but that
214 is clearly not the case, as these two bugs were recently reported.
216 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
217 and Detrail to create the HTML documentation, the .txt form of the man
218 pages, and it removes trailing spaces from listed files. It also creates
219 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
220 case, it wraps all the #defines with #ifndefs. This script should be run
221 before "make dist".
223 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
224 with Unicode property support.
226 (a) For a maximizing quantifier, if the two different cases of the
227 character were of different lengths in their UTF-8 codings (there are
228 some cases like this - I found 11), and the matching function had to
229 back up over a mixture of the two cases, it incorrectly assumed they
230 were both the same length.
232 (b) When PCRE was configured to use the heap rather than the stack for
233 recursion during matching, it was not correctly preserving the data for
234 the other case of a UTF-8 character when checking ahead for a match
235 while processing a minimizing repeat. If the check also involved
236 matching a wide character, but failed, corruption could cause an
237 erroneous result when trying to check for a repeat of the original
238 character.
240 13. Some tidying changes to the testing mechanism:
242 (a) The RunTest script now detects the internal link size and whether there
243 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
244 values substituted by "configure". (The RunGrepTest script already did
245 this for UTF-8.) The configure.ac script no longer substitutes the
246 relevant variables.
248 (b) The debugging options /B and /D in pcretest show the compiled bytecode
249 with length and offset values. This means that the output is different
250 for different internal link sizes. Test 2 is skipped for link sizes
251 other than 2 because of this, bypassing the problem. Unfortunately,
252 there was also a test in test 3 (the locale tests) that used /B and
253 failed for link sizes other than 2. Rather than cut the whole test out,
254 I have added a new /Z option to pcretest that replaces the length and
255 offset values with spaces. This is now used to make test 3 independent
256 of link size. (Test 2 will be tidied up later.)
258 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
259 segmentation fault instead of returning the appropriate error message.
261 15. In multiline mode when the newline sequence was set to "any", the pattern
262 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
263 This doesn't seem right; it now treats the CRLF combination as the line
264 ending, and so does not match in that case. It's only a pattern such as ^$
265 that would hit this one: something like ^ABC$ would have failed after \r
266 and then tried again after \r\n.
268 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
269 in an attempt to make files that differ only in their line terminators
270 compare equal. This works on Linux.
272 17. Under certain error circumstances pcregrep might try to free random memory
273 as it exited. This is now fixed, thanks to valgrind.
275 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
276 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
277 was because its rules for how to advance for /g after matching an empty
278 string at the end of a line did not allow for this case. They now check for
279 it specially.
281 20. pcretest is supposed to handle patterns and data of any length, by
282 extending its buffers when necessary. It was getting this wrong when the
283 buffer for a data line had to be extended.
285 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
286 CRLF as a newline sequence.
288 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
289 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
290 I have nevertheless tidied it up.
292 23. Added some casts to kill warnings from HP-UX ia64 compiler.
294 24. Added a man page for pcre-config.
297 Version 7.0 19-Dec-06
298 ---------------------
300 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
301 moving to gcc 4.1.1.
303 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
304 sys/time.h, which is documented as needed for this function. It doesn't
305 seem to matter on Linux, but it showed up on some releases of OS X.
307 3. It seems that there are systems where bytes whose values are greater than
308 127 match isprint() in the "C" locale. The "C" locale should be the
309 default when a C program starts up. In most systems, only ASCII printing
310 characters match isprint(). This difference caused the output from pcretest
311 to vary, making some of the tests fail. I have changed pcretest so that:
313 (a) When it is outputting text in the compiled version of a pattern, bytes
314 other than 32-126 are always shown as hex escapes.
316 (b) When it is outputting text that is a matched part of a subject string,
317 it does the same, unless a different locale has been set for the match
318 (using the /L modifier). In this case, it uses isprint() to decide.
320 4. Fixed a major bug that caused incorrect computation of the amount of memory
321 required for a compiled pattern when options that changed within the
322 pattern affected the logic of the preliminary scan that determines the
323 length. The relevant options are -x, and -i in UTF-8 mode. The result was
324 that the computed length was too small. The symptoms of this bug were
325 either the PCRE error "internal error: code overflow" from pcre_compile(),
326 or a glibc crash with a message such as "pcretest: free(): invalid next
327 size (fast)". Examples of patterns that provoked this bug (shown in
328 pcretest format) are:
330 /(?-x: )/x
331 /(?x)(?-x: \s*#\s*)/
332 /((?i)[\x{c0}])/8
333 /(?i:[\x{c0}])/8
335 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
336 is now done differently.
338 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
339 wrapper classes; (b) implement a new function in the C++ scanner that is
340 more efficient than the old way of doing things because it avoids levels of
341 recursion in the regex matching; (c) add a paragraph to the documentation
342 for the FullMatch() function.
344 6. The escape sequence \n was being treated as whatever was defined as
345 "newline". Not only was this contrary to the documentation, which states
346 that \n is character 10 (hex 0A), but it also went horribly wrong when
347 "newline" was defined as CRLF. This has been fixed.
349 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
350 was being set to -1 for the "end of line" case (supposedly a value that no
351 character can have). Though this value is never used (the check for end of
352 line is "zero bytes in current character"), it caused compiler complaints.
353 I've changed it to 0xffffffff.
355 8. In pcre_version.c, the version string was being built by a sequence of
356 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
357 string (as it is for production releases) called a macro with an empty
358 argument. The C standard says the result of this is undefined. The gcc
359 compiler treats it as an empty string (which was what was wanted) but it is
360 reported that Visual C gives an error. The source has been hacked around to
361 avoid this problem.
363 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
364 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
365 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
366 of them did).
368 10. Originally, pcretest opened its input and output without "b"; then I was
369 told that "b" was needed in some environments, so it was added for release
370 5.0 to both the input and output. (It makes no difference on Unix-like
371 systems.) Later I was told that it is wrong for the input on Windows. I've
372 now abstracted the modes into two macros, to make it easier to fiddle with
373 them, and removed "b" from the input mode under Windows.
375 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
377 12. Added -help and --help to pcretest as an official way of being reminded
378 of the options.
380 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
381 and pcrecpp.cc because they annoy compilers at high warning levels.
383 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
385 15. Fixed an occurrence of == in configure.ac that should have been = (shell
386 scripts are not C programs :-) and which was not noticed because it works
387 on Linux.
389 16. pcretest is supposed to handle any length of pattern and data line (as one
390 line or as a continued sequence of lines) by extending its input buffer if
391 necessary. This feature was broken for very long pattern lines, leading to
392 a string of junk being passed to pcre_compile() if the pattern was longer
393 than about 50K.
395 17. I have done a major re-factoring of the way pcre_compile() computes the
396 amount of memory needed for a compiled pattern. Previously, there was code
397 that made a preliminary scan of the pattern in order to do this. That was
398 OK when PCRE was new, but as the facilities have expanded, it has become
399 harder and harder to keep it in step with the real compile phase, and there
400 have been a number of bugs (see for example, 4 above). I have now found a
401 cunning way of running the real compile function in a "fake" mode that
402 enables it to compute how much memory it would need, while actually only
403 ever using a few hundred bytes of working memory and without too many
404 tests of the mode. This should make future maintenance and development
405 easier. A side effect of this work is that the limit of 200 on the nesting
406 depth of parentheses has been removed (though this was never a serious
407 limitation, I suspect). However, there is a downside: pcre_compile() now
408 runs more slowly than before (30% or more, depending on the pattern). I
409 hope this isn't a big issue. There is no effect on runtime performance.
411 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
412 newline (only possible for the last line of a file) and it was a
413 pattern that set a locale (followed by /Lsomething), pcretest crashed.
415 19. Added additional timing features to pcretest. (1) The -tm option now times
416 matching only, not compiling. (2) Both -t and -tm can be followed, as a
417 separate command line item, by a number that specifies the number of
418 repeats to use when timing. The default is 50000; this gives better
419 precision, but takes uncomfortably long for very large patterns.
421 20. Extended pcre_study() to be more clever in cases where a branch of a
422 subpattern has no definite first character. For example, (a*|b*)[cd] would
423 previously give no result from pcre_study(). Now it recognizes that the
424 first character must be a, b, c, or d.
426 21. There was an incorrect error "recursive call could loop indefinitely" if
427 a subpattern (or the entire pattern) that was being tested for matching an
428 empty string contained only one non-empty item after a nested subpattern.
429 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
430 incorrectly, because the \d was being skipped in the check.
432 22. The pcretest program now has a new pattern option /B and a command line
433 option -b, which is equivalent to adding /B to every pattern. This causes
434 it to show the compiled bytecode, without the additional information that
435 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
436 is the same as /B/I).
438 23. A new optimization is now able automatically to treat some sequences such
439 as a*b as a*+b. More specifically, if something simple (such as a character
440 or a simple class like \d) has an unlimited quantifier, and is followed by
441 something that cannot possibly match the quantified thing, the quantifier
442 is automatically "possessified".
444 24. A recursive reference to a subpattern whose number was greater than 39
445 went wrong under certain circumstances in UTF-8 mode. This bug could also
446 have affected the operation of pcre_study().
448 25. Realized that a little bit of performance could be had by replacing
449 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
451 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
453 27. Possessive quantifiers such as a++ were previously implemented by turning
454 them into atomic groups such as ($>a+). Now they have their own opcodes,
455 which improves performance. This includes the automatically created ones
456 from 23 above.
458 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
459 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
460 the first matched character to be a colon. This applied both to named and
461 numbered groups.
463 29. The ucpinternal.h header file was missing its idempotency #ifdef.
465 30. I was sent a "project" file called libpcre.a.dev which I understand makes
466 building PCRE on Windows easier, so I have included it in the distribution.
468 31. There is now a check in pcretest against a ridiculously large number being
469 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
470 loop, the loop is abandoned.
472 32. Forward references to subpatterns in conditions such as (?(2)...) where
473 subpattern 2 is defined later cause pcre_compile() to search forwards in
474 the pattern for the relevant set of parentheses. This search went wrong
475 when there were unescaped parentheses in a character class, parentheses
476 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
478 33. "Subroutine" calls and backreferences were previously restricted to
479 referencing subpatterns earlier in the regex. This restriction has now
480 been removed.
482 34. Added a number of extra features that are going to be in Perl 5.10. On the
483 whole, these are just syntactic alternatives for features that PCRE had
484 previously implemented using the Python syntax or my own invention. The
485 other formats are all retained for compatibility.
487 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
488 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
489 also .NET compatible.
491 (b) A recursion or subroutine call to a named group can now be defined as
492 (?&name) as well as (?P>name).
494 (c) A backreference to a named group can now be defined as \k<name> or
495 \k'name' as well as (?P=name). The new forms, as well as being in Perl
496 5.10, are also .NET compatible.
498 (d) A conditional reference to a named group can now use the syntax
499 (?(<name>) or (?('name') as well as (?(name).
501 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
502 groups (named and numbered) that are never evaluated inline, but can be
503 called as "subroutines" from elsewhere. In effect, the DEFINE condition
504 is always false. There may be only one alternative in such a group.
506 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
507 as the simple (?(R). The condition is true only if the most recent
508 recursion is that of the given number or name. It does not search out
509 through the entire recursion stack.
511 (g) The escape \gN or \g{N} has been added, where N is a positive or
512 negative number, specifying an absolute or relative reference.
514 35. Tidied to get rid of some further signed/unsigned compiler warnings and
515 some "unreachable code" warnings.
517 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
518 things, this adds five new scripts.
520 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
521 There were also incompatibilities regarding the handling of \Q..\E inside
522 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
523 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
525 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
526 matches an empty string, and forcibly breaks the loop. There were bugs in
527 this code in non-simple cases. For a pattern such as ^(a()*)* matched
528 against aaaa the result was just "a" rather than "aaaa", for example. Two
529 separate and independent bugs (that affected different cases) have been
530 fixed.
532 39. Refactored the code to abolish the use of different opcodes for small
533 capturing bracket numbers. This is a tidy that I avoided doing when I
534 removed the limit on the number of capturing brackets for 3.5 back in 2001.
535 The new approach is not only tidier, it makes it possible to reduce the
536 memory needed to fix the previous bug (38).
538 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
539 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
540 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
541 mode.
543 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
544 report.
546 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
547 copy construction and assignment in the C++ wrapper.
549 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
550 couple of bugs that could have given wrong results in the "--newline=crlf"
551 case.
553 44. Added a number of casts and did some reorganization of signed/unsigned int
554 variables following suggestions from Dair Grant. Also renamed the variable
555 "this" as "item" because it is a C++ keyword.
557 45. Arranged for dftables to add
559 #include "pcre_internal.h"
561 to pcre_chartables.c because without it, gcc 4.x may remove the array
562 definition from the final binary if PCRE is built into a static library and
563 dead code stripping is activated.
565 46. For an unanchored pattern, if a match attempt fails at the start of a
566 newline sequence, and the newline setting is CRLF or ANY, and the next two
567 characters are CRLF, advance by two characters instead of one.
570 Version 6.7 04-Jul-06
571 ---------------------
573 1. In order to handle tests when input lines are enormously long, pcretest has
574 been re-factored so that it automatically extends its buffers when
575 necessary. The code is crude, but this _is_ just a test program. The
576 default size has been increased from 32K to 50K.
578 2. The code in pcre_study() was using the value of the re argument before
579 testing it for NULL. (Of course, in any sensible call of the function, it
580 won't be NULL.)
582 3. The memmove() emulation function in pcre_internal.h, which is used on
583 systems that lack both memmove() and bcopy() - that is, hardly ever -
584 was missing a "static" storage class specifier.
586 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
587 containing an extended class (one that cannot be represented by a bitmap
588 because it contains high-valued characters or Unicode property items, e.g.
589 [\pZ]). Almost always one would set UTF-8 mode when processing such a
590 pattern, but PCRE should not loop if you do not (it no longer does).
591 [Detail: two cases were found: (a) a repeated subpattern containing an
592 extended class; (b) a recursive reference to a subpattern that followed a
593 previous extended class. It wasn't skipping over the extended class
594 correctly when UTF-8 mode was not set.]
596 5. A negated single-character class was not being recognized as fixed-length
597 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
598 compile error "lookbehind assertion is not fixed length".
600 6. The RunPerlTest auxiliary script was showing an unexpected difference
601 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
602 write a Perl script that can interpret lines of an input file either as
603 byte characters or as UTF-8, which is what "perltest" was being required to
604 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
605 can't do is switch easily at run time between having the "use utf8;" pragma
606 or not. In the end, I fudged it by using the RunPerlTest script to insert
607 "use utf8;" explicitly for the UTF-8 tests.
609 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
610 the end of the subject string, contrary to the documentation and to what
611 Perl does. This was true of both matching functions. Now it matches only at
612 the start of the subject and immediately after *internal* newlines.
614 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
615 a pointer to an int instead of a pointer to an unsigned long int. This
616 caused problems on 64-bit systems.
618 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
619 instance of the 'standard' template library not being so standard".
621 10. There was no check on the number of named subpatterns nor the maximum
622 length of a subpattern name. The product of these values is used to compute
623 the size of the memory block for a compiled pattern. By supplying a very
624 long subpattern name and a large number of named subpatterns, the size
625 computation could be caused to overflow. This is now prevented by limiting
626 the length of names to 32 characters, and the number of named subpatterns
627 to 10,000.
629 11. Subpatterns that are repeated with specific counts have to be replicated in
630 the compiled pattern. The size of memory for this was computed from the
631 length of the subpattern and the repeat count. The latter is limited to
632 65535, but there was no limit on the former, meaning that integer overflow
633 could in principle occur. The compiled length of a repeated subpattern is
634 now limited to 30,000 bytes in order to prevent this.
636 12. Added the optional facility to have named substrings with the same name.
638 13. Added the ability to use a named substring as a condition, using the
639 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
640 are numbers (not recommended). Forward references are permitted.
642 14. Added forward references in named backreferences (if you see what I mean).
644 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
645 pattern could run off the end of the subject. For example, the pattern
646 "(?s)(.{1,5})"8 did this with the subject "ab".
648 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
649 PCRE_CASELESS was set when matching characters that were quantified with ?
650 or *.
652 17. A character class other than a single negated character that had a minimum
653 but no maximum quantifier - for example [ab]{6,} - was not handled
654 correctly by pce_dfa_exec(). It would match only one character.
656 18. A valid (though odd) pattern that looked like a POSIX character
657 class but used an invalid character after [ (for example [[,abc,]]) caused
658 pcre_compile() to give the error "Failed: internal error: code overflow" or
659 in some cases to crash with a glibc free() error. This could even happen if
660 the pattern terminated after [[ but there just happened to be a sequence of
661 letters, a binary zero, and a closing ] in the memory that followed.
663 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
664 over the years. Originally (before any Unicode support), just the bottom 8
665 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
666 output from "man perlunicode" includes this:
668 The regular expression compiler produces polymorphic opcodes. That
669 is, the pattern adapts to the data and automatically switches to
670 the Unicode character scheme when presented with Unicode data--or
671 instead uses a traditional byte scheme when presented with byte
672 data.
674 Sadly, a wide octal escape does not cause a switch, and in a string with
675 no other multibyte characters, these octal escapes are treated as before.
676 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
677 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
678 Unicode string.
680 I have not perpetrated such confusion in PCRE. Up till now, it took just
681 the bottom 8 bits, as in old Perl. I have now made octal escapes with
682 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
683 translate to the appropriate multibyte character.
685 29. Applied some refactoring to reduce the number of warnings from Microsoft
686 and Borland compilers. This has included removing the fudge introduced
687 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
688 a warning about an unused variable.
690 21. PCRE has not included VT (character 0x0b) in the set of whitespace
691 characters since release 4.0, because Perl (from release 5.004) does not.
692 [Or at least, is documented not to: some releases seem to be in conflict
693 with the documentation.] However, when a pattern was studied with
694 pcre_study() and all its branches started with \s, PCRE still included VT
695 as a possible starting character. Of course, this did no harm; it just
696 caused an unnecessary match attempt.
698 22. Removed a now-redundant internal flag bit that recorded the fact that case
699 dependency changed within the pattern. This was once needed for "required
700 byte" processing, but is no longer used. This recovers a now-scarce options
701 bit. Also moved the least significant internal flag bit to the most-
702 significant bit of the word, which was not previously used (hangover from
703 the days when it was an int rather than a uint) to free up another bit for
704 the future.
706 23. Added support for CRLF line endings as well as CR and LF. As well as the
707 default being selectable at build time, it can now be changed at runtime
708 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
709 specify that it is scanning data with non-default line endings.
711 24. Changed the definition of CXXLINK to make it agree with the definition of
712 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
714 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
715 recursions. This makes a big different to stack usage for some patterns.
717 26. If a subpattern containing a named recursion or subroutine reference such
718 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
719 the space required for the compiled pattern went wrong and gave too small a
720 value. Depending on the environment, this could lead to "Failed: internal
721 error: code overflow at offset 49" or "glibc detected double free or
722 corruption" errors.
724 27. Applied patches from Google (a) to support the new newline modes and (b) to
725 advance over multibyte UTF-8 characters in GlobalReplace.
727 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
728 difference for some implementation of PCRE in some Windows version.
730 29. Added some extra testing facilities to pcretest:
732 \q<number> in a data line sets the "match limit" value
733 \Q<number> in a data line sets the "match recursion limt" value
734 -S <number> sets the stack size, where <number> is in megabytes
736 The -S option isn't available for Windows.
739 Version 6.6 06-Feb-06
740 ---------------------
742 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
743 in pcreposix.h. I have copied the definition from pcre.h.
745 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
746 because pcre.h is no longer a built file.
748 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
749 not normally included in the compiled code.
752 Version 6.5 01-Feb-06
753 ---------------------
755 1. When using the partial match feature with pcre_dfa_exec(), it was not
756 anchoring the second and subsequent partial matches at the new starting
757 point. This could lead to incorrect results. For example, with the pattern
758 /1234/, partially matching against "123" and then "a4" gave a match.
760 2. Changes to pcregrep:
762 (a) All non-match returns from pcre_exec() were being treated as failures
763 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
764 error message is output. Some extra information is given for the
766 probably the only errors that are likely to be caused by users (by
767 specifying a regex that has nested indefinite repeats, for instance).
768 If there are more than 20 of these errors, pcregrep is abandoned.
770 (b) A binary zero was treated as data while matching, but terminated the
771 output line if it was written out. This has been fixed: binary zeroes
772 are now no different to any other data bytes.
774 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
775 used to set a locale for matching. The --locale=xxxx long option has
776 been added (no short equivalent) to specify a locale explicitly on the
777 pcregrep command, overriding the environment variables.
779 (d) When -B was used with -n, some line numbers in the output were one less
780 than they should have been.
782 (e) Added the -o (--only-matching) option.
784 (f) If -A or -C was used with -c (count only), some lines of context were
785 accidentally printed for the final match.
787 (g) Added the -H (--with-filename) option.
789 (h) The combination of options -rh failed to suppress file names for files
790 that were found from directory arguments.
792 (i) Added the -D (--devices) and -d (--directories) options.
794 (j) Added the -F (--fixed-strings) option.
796 (k) Allow "-" to be used as a file name for -f as well as for a data file.
798 (l) Added the --colo(u)r option.
800 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
801 is not present by default.
803 3. A nasty bug was discovered in the handling of recursive patterns, that is,
804 items such as (?R) or (?1), when the recursion could match a number of
805 alternatives. If it matched one of the alternatives, but subsequently,
806 outside the recursion, there was a failure, the code tried to back up into
807 the recursion. However, because of the way PCRE is implemented, this is not
808 possible, and the result was an incorrect result from the match.
810 In order to prevent this happening, the specification of recursion has
811 been changed so that all such subpatterns are automatically treated as
812 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
814 4. I had overlooked the fact that, in some locales, there are characters for
815 which isalpha() is true but neither isupper() nor islower() are true. In
816 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
817 and ordfeminine) are like this. This affected the treatment of \w and \W
818 when they appeared in character classes, but not when they appeared outside
819 a character class. The bit map for "word" characters is now created
820 separately from the results of isalnum() instead of just taking it from the
821 upper, lower, and digit maps. (Plus the underscore character, of course.)
823 5. The above bug also affected the handling of POSIX character classes such as
824 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
825 permanent tables. Instead, the bit maps for such a class were previously
826 created as the appropriate unions of the upper, lower, and digit bitmaps.
827 Now they are created by subtraction from the [[:word:]] class, which has
828 its own bitmap.
830 6. The [[:blank:]] character class matches horizontal, but not vertical space.
831 It is created by subtracting the vertical space characters (\x09, \x0a,
832 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
833 subtraction was done in the overall bitmap for a character class, meaning
834 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
835 be recognized. This bug has been fixed.
837 7. Patches from the folks at Google:
839 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
840 real life, but is still worth protecting against".
842 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
843 regular expressions".
845 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
846 have it.
848 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
849 "configure" and the latter not, in order to fix a problem somebody had
850 with compiling the Arg class on HP-UX.
852 (e) Improve the error-handling of the C++ wrapper a little bit.
854 (f) New tests for checking recursion limiting.
856 8. The pcre_memmove() function, which is used only if the environment does not
857 have a standard memmove() function (and is therefore rarely compiled),
858 contained two bugs: (a) use of int instead of size_t, and (b) it was not
859 returning a result (though PCRE never actually uses the result).
861 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
862 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
863 returned instead of calling malloc() with an overflowing number that would
864 most likely cause subsequent chaos.
866 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
868 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
869 with this option is matched, the nmatch and pmatch options of regexec() are
870 ignored.
872 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
873 provided in case anyone wants to the the POSIX interface with UTF-8
874 strings.
876 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
877 C++ linking (needed for some HP-UX environments).
879 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
880 (unused parameter) and in the pcre_printint() function (omitted "default"
881 switch label when the default is to do nothing).
883 15. Added some code to make it possible, when PCRE is compiled as a C++
884 library, to replace subject pointers for pcre_exec() with a smart pointer
885 class, thus making it possible to process discontinuous strings.
887 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
888 much the same function. They were added by different people who were trying
889 to make PCRE easy to compile on non-Unix systems. It has been suggested
890 that PCRE_EXPORT be abolished now that there is more automatic apparatus
891 for compiling on Windows systems. I have therefore replaced it with
892 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
893 defaults to "extern" for C or "extern C" for C++, which works fine on
894 Unix-like systems. It is now possible to override the value of PCRE_DATA_
895 SCOPE with something explicit in config.h. In addition:
897 (a) pcreposix.h still had just "extern" instead of either of these macros;
898 I have replaced it with PCRE_DATA_SCOPE.
900 (b) Functions such as _pcre_xclass(), which are internal to the library,
901 but external in the C sense, all had PCRE_EXPORT in their definitions.
902 This is apparently wrong for the Windows case, so I have removed it.
903 (It makes no difference on Unix-like systems.)
905 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
906 of recursive calls to match(). This is different to MATCH_LIMIT because
907 that limits the total number of calls to match(), not all of which increase
908 the depth of recursion. Limiting the recursion depth limits the amount of
909 stack (or heap if NO_RECURSE is set) that is used. The default can be set
910 when PCRE is compiled, and changed at run time. A patch from Google adds
911 this functionality to the C++ interface.
913 18. Changes to the handling of Unicode character properties:
915 (a) Updated the table to Unicode 4.1.0.
917 (b) Recognize characters that are not in the table as "Cn" (undefined).
919 (c) I revised the way the table is implemented to a much improved format
920 which includes recognition of ranges. It now supports the ranges that
921 are defined in UnicodeData.txt, and it also amalgamates other
922 characters into ranges. This has reduced the number of entries in the
923 table from around 16,000 to around 3,000, thus reducing its size
924 considerably. I realized I did not need to use a tree structure after
925 all - a binary chop search is just as efficient. Having reduced the
926 number of entries, I extended their size from 6 bytes to 8 bytes to
927 allow for more data.
929 (d) Added support for Unicode script names via properties such as \p{Han}.
931 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
932 matching that character.
934 20. When matching a repeated Unicode property with a minimum greater than zero,
935 (for example \pL{2,}), PCRE could look past the end of the subject if it
936 reached it while seeking the minimum number of characters. This could
937 happen only if some of the characters were more than one byte long, because
938 there is a check for at least the minimum number of bytes.
940 21. Refactored the implementation of \p and \P so as to be more general, to
941 allow for more different types of property in future. This has changed the
942 compiled form incompatibly. Anybody with saved compiled patterns that use
943 \p or \P will have to recompile them.
945 22. Added "Any" and "L&" to the supported property types.
947 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
948 but give a compile time error if the value is greater than 0xff.
950 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
951 accidentally not being installed or uninstalled.
953 25. The pcre.h file was built from pcre.h.in, but the only changes that were
954 made were to insert the current release number. This seemed silly, because
955 it made things harder for people building PCRE on systems that don't run
956 "configure". I have turned pcre.h into a distributed file, no longer built
957 by "configure", with the version identification directly included. There is
958 no longer a pcre.h.in file.
960 However, this change necessitated a change to the pcre-config script as
961 well. It is built from pcre-config.in, and one of the substitutions was the
962 release number. I have updated configure.ac so that ./configure now finds
963 the release number by grepping pcre.h.
965 26. Added the ability to run the tests under valgrind.
968 Version 6.4 05-Sep-05
969 ---------------------
971 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
972 "--" to be printed when multiple files were scanned, even when none of the
973 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
974 consider it to be a bug, and have restored the previous behaviour.
976 2. A couple of code tidies to get rid of compiler warnings.
978 3. The pcretest program used to cheat by referring to symbols in the library
979 whose names begin with _pcre_. These are internal symbols that are not
980 really supposed to be visible externally, and in some environments it is
981 possible to suppress them. The cheating is now confined to including
982 certain files from the library's source, which is a bit cleaner.
984 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
985 file's purpose clearer.
987 5. Reorganized pcre_ucp_findchar().
990 Version 6.3 15-Aug-05
991 ---------------------
993 1. The file libpcre.pc.in did not have general read permission in the tarball.
995 2. There were some problems when building without C++ support:
997 (a) If C++ support was not built, "make install" and "make test" still
998 tried to test it.
1000 (b) There were problems when the value of CXX was explicitly set. Some
1001 changes have been made to try to fix these, and ...
1003 (c) --disable-cpp can now be used to explicitly disable C++ support.
1005 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1006 backslash in a target when C++ was disabled. This confuses some
1007 versions of "make", apparently. Using an intermediate variable solves
1008 this. (Same for CPP_LOBJ.)
1010 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1011 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1012 necessary on certain architectures.
1014 4. Added a setting of -export-symbols-regex to the link command to remove
1015 those symbols that are exported in the C sense, but actually are local
1016 within the library, and not documented. Their names all begin with
1017 "_pcre_". This is not a perfect job, because (a) we have to except some
1018 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1019 available (and never for static libraries). I have made a note to try to
1020 find a way round (a) in the future.
1023 Version 6.2 01-Aug-05
1024 ---------------------
1026 1. There was no test for integer overflow of quantifier values. A construction
1027 such as {1111111111111111} would give undefined results. What is worse, if
1028 a minimum quantifier for a parenthesized subpattern overflowed and became
1029 negative, the calculation of the memory size went wrong. This could have
1030 led to memory overwriting.
1032 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1034 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1035 operating environments where this matters.
1037 4. Applied Giuseppe Maxia's patch to add additional features for controlling
1038 PCRE options from within the C++ wrapper.
1040 5. Named capturing subpatterns were not being correctly counted when a pattern
1041 was compiled. This caused two problems: (a) If there were more than 100
1042 such subpatterns, the calculation of the memory needed for the whole
1043 compiled pattern went wrong, leading to an overflow error. (b) Numerical
1044 back references of the form \12, where the number was greater than 9, were
1045 not recognized as back references, even though there were sufficient
1046 previous subpatterns.
1048 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1049 versions of gcc, e.g. 2.95.4.
1052 Version 6.1 21-Jun-05
1053 ---------------------
1055 1. There was one reference to the variable "posix" in pcretest.c that was not
1056 surrounded by "#if !defined NOPOSIX".
1058 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1059 the cross-check on the old pcre_info() function, for the benefit of the
1060 cut-down version of PCRE that is currently imported into Exim.
1062 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1063 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1064 patterns that start (?i) but I don't think that matters. The use of (?i) is
1065 just an example; this all applies to the other options as well.
1067 4. Since libtool seems to echo the compile commands it is issuing, the output
1068 from "make" can be reduced a bit by putting "@" in front of each libtool
1069 compile command.
1071 5. Patch from the folks at Google for configure.in to be a bit more thorough
1072 in checking for a suitable C++ installation before trying to compile the
1073 C++ stuff. This should fix a reported problem when a compiler was present,
1074 but no suitable headers.
1076 6. The man pages all had just "PCRE" as their title. I have changed them to
1077 be the relevant file name. I have also arranged that these names are
1078 retained in the file doc/pcre.txt, which is a concatenation in text format
1079 of all the man pages except the little individual ones for each function.
1081 7. The NON-UNIX-USE file had not been updated for the different set of source
1082 files that come with release 6. I also added a few comments about the C++
1083 wrapper.
1086 Version 6.0 07-Jun-05
1087 ---------------------
1089 1. Some minor internal re-organization to help with my DFA experiments.
1091 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1092 didn't matter for the library itself when fully configured, but did matter
1093 when compiling without UCP support, or within Exim, where the ucp files are
1094 not imported.
1096 3. Refactoring of the library code to split up the various functions into
1097 different source modules. The addition of the new DFA matching code (see
1098 below) to a single monolithic source would have made it really too
1099 unwieldy, quite apart from causing all the code to be include in a
1100 statically linked application, when only some functions are used. This is
1101 relevant even without the DFA addition now that patterns can be compiled in
1102 one application and matched in another.
1104 The downside of splitting up is that there have to be some external
1105 functions and data tables that are used internally in different modules of
1106 the library but which are not part of the API. These have all had their
1107 names changed to start with "_pcre_" so that they are unlikely to clash
1108 with other external names.
1110 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1111 a different (DFA) algorithm. Although it is slower than the original
1112 function, it does have some advantages for certain types of matching
1113 problem.
1115 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1116 including restarting after a partial match.
1118 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1119 defined when compiling for Windows was sent to me. I have put it into the
1120 code, though I have no means of testing or verifying it.
1122 7. Added the pcre_refcount() auxiliary function.
1124 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1125 match before or at the first newline in the subject string. In pcretest,
1126 the /f option on a pattern can be used to set this.
1128 9. A repeated \w when used in UTF-8 mode with characters greater than 256
1129 would behave wrongly. This has been present in PCRE since release 4.0.
1131 10. A number of changes to the pcregrep command:
1133 (a) Refactored how -x works; insert ^(...)$ instead of setting
1134 PCRE_ANCHORED and checking the length, in preparation for adding
1135 something similar for -w.
1137 (b) Added the -w (match as a word) option.
1139 (c) Refactored the way lines are read and buffered so as to have more
1140 than one at a time available.
1142 (d) Implemented a pcregrep test script.
1144 (e) Added the -M (multiline match) option. This allows patterns to match
1145 over several lines of the subject. The buffering ensures that at least
1146 8K, or the rest of the document (whichever is the shorter) is available
1147 for matching (and similarly the previous 8K for lookbehind assertions).
1149 (f) Changed the --help output so that it now says
1151 -w, --word-regex(p)
1153 instead of two lines, one with "regex" and the other with "regexp"
1154 because that confused at least one person since the short forms are the
1155 same. (This required a bit of code, as the output is generated
1156 automatically from a table. It wasn't just a text change.)
1158 (g) -- can be used to terminate pcregrep options if the next thing isn't an
1159 option but starts with a hyphen. Could be a pattern or a path name
1160 starting with a hyphen, for instance.
1162 (h) "-" can be given as a file name to represent stdin.
1164 (i) When file names are being printed, "(standard input)" is used for
1165 the standard input, for compatibility with GNU grep. Previously
1166 "<stdin>" was used.
1168 (j) The option --label=xxx can be used to supply a name to be used for
1169 stdin when file names are being printed. There is no short form.
1171 (k) Re-factored the options decoding logic because we are going to add
1172 two more options that take data. Such options can now be given in four
1173 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1175 (l) Added the -A, -B, and -C options for requesting that lines of context
1176 around matches be printed.
1178 (m) Added the -L option to print the names of files that do not contain
1179 any matching lines, that is, the complement of -l.
1181 (n) The return code is 2 if any file cannot be opened, but pcregrep does
1182 continue to scan other files.
1184 (o) The -s option was incorrectly implemented. For compatibility with other
1185 greps, it now suppresses the error message for a non-existent or non-
1186 accessible file (but not the return code). There is a new option called
1187 -q that suppresses the output of matching lines, which was what -s was
1188 previously doing.
1190 (p) Added --include and --exclude options to specify files for inclusion
1191 and exclusion when recursing.
1193 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1194 Hopefully, it now does.
1196 12. Missing cast in pcre_study().
1198 13. Added an "uninstall" target to the makefile.
1200 14. Replaced "extern" in the function prototypes in Makefile.in with
1201 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1202 world, but is set differently for Windows.
1204 15. Added a second compiling function called pcre_compile2(). The only
1205 difference is that it has an extra argument, which is a pointer to an
1206 integer error code. When there is a compile-time failure, this is set
1207 non-zero, in addition to the error test pointer being set to point to an
1208 error message. The new argument may be NULL if no error number is required
1209 (but then you may as well call pcre_compile(), which is now just a
1210 wrapper). This facility is provided because some applications need a
1211 numeric error indication, but it has also enabled me to tidy up the way
1212 compile-time errors are handled in the POSIX wrapper.
1214 16. Added VPATH=.libs to the makefile; this should help when building with one
1215 prefix path and installing with another. (Or so I'm told by someone who
1216 knows more about this stuff than I do.)
1218 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1219 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1220 match everything, including newlines. This is not POSIX-compatible, but
1221 somebody wanted the feature. From pcretest it can be activated by using
1222 both the P and the s flags.
1224 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1226 19. libpcre.pc was being incorrectly installed as executable.
1228 20. A couple of places in pcretest check for end-of-line by looking for '\n';
1229 it now also looks for '\r' so that it will work unmodified on Windows.
1231 21. Added Google's contributed C++ wrapper to the distribution.
1233 22. Added some untidy missing memory free() calls in pcretest, to keep
1234 Electric Fence happy when testing.
1238 Version 5.0 13-Sep-04
1239 ---------------------
1241 1. Internal change: literal characters are no longer packed up into items
1242 containing multiple characters in a single byte-string. Each character
1243 is now matched using a separate opcode. However, there may be more than one
1244 byte in the character in UTF-8 mode.
1246 2. The pcre_callout_block structure has two new fields: pattern_position and
1247 next_item_length. These contain the offset in the pattern to the next match
1248 item, and its length, respectively.
1250 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1251 insertion of callouts before each pattern item. Added the /C option to
1252 pcretest to make use of this.
1254 4. On the advice of a Windows user, the lines
1256 #if defined(_WIN32) || defined(WIN32)
1257 _setmode( _fileno( stdout ), 0x8000 );
1258 #endif /* defined(_WIN32) || defined(WIN32) */
1260 have been added to the source of pcretest. This apparently does useful
1261 magic in relation to line terminators.
1263 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1264 for the benefit of those environments where the "b" makes a difference.
1266 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1267 to know about it. I have put a hack into configure.in that adds in code
1268 to set GCC=yes if CC=icc. This seems to end up at a point in the
1269 generated configure script that is early enough to affect the setting of
1270 compiler options, which is what is needed, but I have no means of testing
1271 whether it really works. (The user who reported this had patched the
1272 generated configure script, which of course I cannot do.)
1274 LATER: After change 22 below (new libtool files), the configure script
1275 seems to know about icc (and also ecc). Therefore, I have commented out
1276 this hack in configure.in.
1278 7. Added support for pkg-config (2 patches were sent in).
1280 8. Negated POSIX character classes that used a combination of internal tables
1281 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1282 [[:^ascii]]. Typically, they would match almost any characters. The other
1283 POSIX classes were not broken in this way.
1285 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1286 to find the match, as PCRE was deluded into thinking that the match had to
1287 start at the start point or following a newline. The same bug applied to
1288 patterns with negative forward assertions or any backward assertions
1289 preceding ".*" at the start, unless the pattern required a fixed first
1290 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1292 10. In UTF-8 mode, when moving forwards in the subject after a failed match
1293 starting at the last subject character, bytes beyond the end of the subject
1294 string were read.
1296 11. Renamed the variable "class" as "classbits" to make life easier for C++
1297 users. (Previously there was a macro definition, but it apparently wasn't
1298 enough.)
1300 12. Added the new field "tables" to the extra data so that tables can be passed
1301 in at exec time, or the internal tables can be re-selected. This allows
1302 a compiled regex to be saved and re-used at a later time by a different
1303 program that might have everything at different addresses.
1305 13. Modified the pcre-config script so that, when run on Solaris, it shows a
1306 -R library as well as a -L library.
1308 14. The debugging options of pcretest (-d on the command line or D on a
1309 pattern) showed incorrect output for anything following an extended class
1310 that contained multibyte characters and which was followed by a quantifier.
1312 15. Added optional support for general category Unicode character properties
1313 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1314 support. It adds about 90K to the size of the library. The meanings of the
1315 inbuilt class escapes such as \d and \s have NOT been changed.
1317 16. Updated pcredemo.c to include calls to free() to release the memory for the
1318 compiled pattern.
1320 17. The generated file chartables.c was being created in the source directory
1321 instead of in the building directory. This caused the build to fail if the
1322 source directory was different from the building directory, and was
1323 read-only.
1325 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1326 file. No doubt somebody will tell me if they don't make sense... Also added
1327 Dan Mooney's comments about building on OpenVMS.
1329 19. Added support for partial matching via the PCRE_PARTIAL option for
1330 pcre_exec() and the \P data escape in pcretest.
1332 20. Extended pcretest with 3 new pattern features:
1334 (i) A pattern option of the form ">rest-of-line" causes pcretest to
1335 write the compiled pattern to the file whose name is "rest-of-line".
1336 This is a straight binary dump of the data, with the saved pointer to
1337 the character tables forced to be NULL. The study data, if any, is
1338 written too. After writing, pcretest reads a new pattern.
1340 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1341 compiled pattern from the given file. There must not be any
1342 occurrences of "<" in the file name (pretty unlikely); if there are,
1343 pcretest will instead treat the initial "<" as a pattern delimiter.
1344 After reading in the pattern, pcretest goes on to read data lines as
1345 usual.
1347 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1348 and 16-bit fields in a compiled pattern, to simulate a pattern that
1349 was compiled on a host of opposite endianness.
1351 21. The pcre-exec() function can now cope with patterns that were compiled on
1352 hosts of opposite endianness, with this restriction:
1354 As for any compiled expression that is saved and used later, the tables
1355 pointer field cannot be preserved; the extra_data field in the arguments
1356 to pcre_exec() should be used to pass in a tables address if a value
1357 other than the default internal tables were used at compile time.
1359 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1360 now diagnosed as an error. Previously, most of the time, a negative number
1361 would have been treated as zero, but if in addition "ovector" was passed as
1362 NULL, a crash could occur.
1364 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1365 new versions from the libtool 1.5 distribution (the last one is a copy of
1366 a file called libtool.m4). This seems to have fixed the need to patch
1367 "configure" to support Darwin 1.3 (which I used to do). However, I still
1368 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1369 workstation).
1371 24. Changed the PCRE licence to be the more standard "BSD" licence.
1374 Version 4.5 01-Dec-03
1375 ---------------------
1377 1. There has been some re-arrangement of the code for the match() function so
1378 that it can be compiled in a version that does not call itself recursively.
1379 Instead, it keeps those local variables that need separate instances for
1380 each "recursion" in a frame on the heap, and gets/frees frames whenever it
1381 needs to "recurse". Keeping track of where control must go is done by means
1382 of setjmp/longjmp. The whole thing is implemented by a set of macros that
1383 hide most of the details from the main code, and operates only if
1384 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1385 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1386 operating.
1388 To make it easier for callers to provide specially tailored get/free
1389 functions for this usage, two new functions, pcre_stack_malloc, and
1390 pcre_stack_free, are used. They are always called in strict stacking order,
1391 and the size of block requested is always the same.
1393 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1394 PCRE has been compiled to use the stack or the heap for recursion. The
1395 -C option of pcretest uses this to show which version is compiled.
1397 A new data escape \S, is added to pcretest; it causes the amounts of store
1398 obtained and freed by both kinds of malloc/free at match time to be added
1399 to the output.
1401 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1402 what's available on my current Linux desktop machine.
1404 3. When matching a UTF-8 string, the test for a valid string at the start has
1405 been extended. If start_offset is not zero, PCRE now checks that it points
1406 to a byte that is the start of a UTF-8 character. If not, it returns
1407 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1408 this is necessary because there may be backward assertions in the pattern.
1409 When matching the same subject several times, it may save resources to use
1410 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1412 4. The code for checking the validity of UTF-8 strings has been tightened so
1413 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1414 containing "overlong sequences".
1416 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1417 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1418 should have been "&", but it just so happened that all the cases this let
1419 through by mistake were picked up later in the function.
1421 6. I had used a variable called "isblank" - this is a C99 function, causing
1422 some compilers to warn. To avoid this, I renamed it (as "blankclass").
1424 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1425 prompting; (b) run "./pcretest /dev/null" at the start of the test script
1426 so the version is shown; (c) stop "make test" echoing "./RunTest".
1428 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1430 9. The prototype for memmove() for systems that don't have it was using
1431 size_t, but the inclusion of the header that defines size_t was later. I've
1432 moved the #includes for the C headers earlier to avoid this.
1434 10. Added some adjustments to the code to make it easier to compiler on certain
1435 special systems:
1437 (a) Some "const" qualifiers were missing.
1438 (b) Added the macro EXPORT before all exported functions; by default this
1439 is defined to be empty.
1440 (c) Changed the dftables auxiliary program (that builds chartables.c) so
1441 that it reads its output file name as an argument instead of writing
1442 to the standard output and assuming this can be redirected.
1444 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1445 class containing characters with values greater than 255, PCRE compilation
1446 went into a loop.
1448 12. A recursive reference to a subpattern that was within another subpattern
1449 that had a minimum quantifier of zero caused PCRE to crash. For example,
1450 (x(y(?2))z)? provoked this bug with a subject that got as far as the
1451 recursion. If the recursively-called subpattern itself had a zero repeat,
1452 that was OK.
1454 13. In pcretest, the buffer for reading a data line was set at 30K, but the
1455 buffer into which it was copied (for escape processing) was still set at
1456 1024, so long lines caused crashes.
1458 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1459 "internal error: code overflow...". This applied to any character class
1460 that was followed by a possessive quantifier.
1462 15. Modified the Makefile to add libpcre.la as a prerequisite for
1463 libpcreposix.la because I was told this is needed for a parallel build to
1464 work.
1466 16. If a pattern that contained .* following optional items at the start was
1467 studied, the wrong optimizing data was generated, leading to matching
1468 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1469 matching string must start with a or b or c. The correct conclusion for
1470 this pattern is that a match can start with any character.
1473 Version 4.4 13-Aug-03
1474 ---------------------
1476 1. In UTF-8 mode, a character class containing characters with values between
1477 127 and 255 was not handled correctly if the compiled pattern was studied.
1478 In fixing this, I have also improved the studying algorithm for such
1479 classes (slightly).
1481 2. Three internal functions had redundant arguments passed to them. Removal
1482 might give a very teeny performance improvement.
1484 3. Documentation bug: the value of the capture_top field in a callout is *one
1485 more than* the number of the hightest numbered captured substring.
1487 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1488 in incorrectly linking with a previously installed version. They now link
1489 explicitly with libpcre.la.
1491 5. configure.in no longer needs to recognize Cygwin specially.
1493 6. A problem in pcre.in for Windows platforms is fixed.
1495 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1496 pcretest, it used to include the size of the study block as part of its
1497 output. Unfortunately, the structure contains a field that has a different
1498 size on different hardware architectures. This meant that the tests that
1499 showed this size failed. As the block is currently always of a fixed size,
1500 this information isn't actually particularly useful in pcretest output, so
1501 I have just removed it.
1503 8. Three pre-processor statements accidentally did not start in column 1.
1504 Sadly, there are *still* compilers around that complain, even though
1505 standard C has not required this for well over a decade. Sigh.
1507 9. In pcretest, the code for checking callouts passed small integers in the
1508 callout_data field, which is a void * field. However, some picky compilers
1509 complained about the casts involved for this on 64-bit systems. Now
1510 pcretest passes the address of the small integer instead, which should get
1511 rid of the warnings.
1513 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1514 both compile and run time, and gives an error if an invalid UTF-8 sequence
1515 is found. There is a option for disabling this check in cases where the
1516 string is known to be correct and/or the maximum performance is wanted.
1518 11. In response to a bug report, I changed one line in Makefile.in from
1520 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1521 to
1522 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1524 to look similar to other lines, but I have no way of telling whether this
1525 is the right thing to do, as I do not use Windows. No doubt I'll get told
1526 if it's wrong...
1529 Version 4.3 21-May-03
1530 ---------------------
1532 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1533 Makefile.
1535 2. Some refactoring to improve the quality of the code:
1537 (i) The utf8_table... variables are now declared "const".
1539 (ii) The code for \cx, which used the "case flipping" table to upper case
1540 lower case letters, now just substracts 32. This is ASCII-specific,
1541 but the whole concept of \cx is ASCII-specific, so it seems
1542 reasonable.
1544 (iii) PCRE was using its character types table to recognize decimal and
1545 hexadecimal digits in the pattern. This is silly, because it handles
1546 only 0-9, a-f, and A-F, but the character types table is locale-
1547 specific, which means strange things might happen. A private
1548 table is now used for this - though it costs 256 bytes, a table is
1549 much faster than multiple explicit tests. Of course, the standard
1550 character types table is still used for matching digits in subject
1551 strings against \d.
1553 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1554 ending in _t are). So I've renamed it as ESC_tee.
1556 3. The first argument for regexec() in the POSIX wrapper should have been
1557 defined as "const".
1559 4. Changed pcretest to use malloc() for its buffers so that they can be
1560 Electric Fenced for debugging.
1562 5. There were several places in the code where, in UTF-8 mode, PCRE would try
1563 to read one or more bytes before the start of the subject string. Often this
1564 had no effect on PCRE's behaviour, but in some circumstances it could
1565 provoke a segmentation fault.
1567 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1568 to try to read one or more bytes before the start of the subject string.
1570 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1571 UTF-8 support could misbehave in various ways if the subject string
1572 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1573 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1574 back over UTF-8 characters.)
1577 Version 4.2 14-Apr-03
1578 ---------------------
1580 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1582 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1583 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1584 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1585 [WIN_PREFIX]: new variable, "cyg" for Cygwin
1586 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1588 Note: automatic setting of the BUILD variables is not yet working
1589 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1590 compile-time but not at link-time
1591 [LINK]: use for linking executables only
1592 make different versions for Windows and non-Windows
1593 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1594 libraries
1595 [LINK_FOR_BUILD]: new variable
1596 [OBJEXT]: use throughout
1597 [EXEEXT]: use throughout
1598 <winshared>: new target
1599 <wininstall>: new target
1600 <dftables.o>: use native compiler
1601 <dftables>: use native linker
1602 <install>: handle Windows platform correctly
1603 <clean>: ditto
1604 <check>: ditto
1605 copy DLL to top builddir before testing
1607 As part of these changes, -no-undefined was removed again. This was reported
1608 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1609 in any case.
1611 3. Some tidies to get rid of compiler warnings:
1613 . In the match_data structure, match_limit was an unsigned long int, whereas
1614 match_call_count was an int. I've made them both unsigned long ints.
1616 . In pcretest the fact that a const uschar * doesn't automatically cast to
1617 a void * provoked a warning.
1619 . Turning on some more compiler warnings threw up some "shadow" variables
1620 and a few more missing casts.
1622 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1623 option, a class that contained a single character with a value between 128
1624 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1626 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1627 option, a class that contained several characters, but with at least one
1628 whose value was between 128 and 255 caused PCRE to crash.
1631 Version 4.1 12-Mar-03
1632 ---------------------
1634 1. Compiling with gcc -pedantic found a couple of places where casts were
1635 needed, and a string in dftables.c that was longer than standard compilers are
1636 required to support.
1638 2. Compiling with Sun's compiler found a few more places where the code could
1639 be tidied up in order to avoid warnings.
1641 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1642 first of these names is deprecated in the latest Autoconf in favour of the name
1643 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1644 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1645 analogy I have changed it to CFLAGS_FOR_BUILD.
1647 4. Added -no-undefined to the linking command in the Makefile, because this is
1648 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1649 linking step for the pcreposix library.
1651 5. PCRE was failing to diagnose the case of two named groups with the same
1652 name.
1654 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1655 literal character that is needed in the subject for a match, and scans along to
1656 ensure that it is present before embarking on the full matching process. This
1657 saves time in cases of nested unlimited repeats that are never going to match.
1658 Problem: the scan can take a lot of time if the subject is very long (e.g.
1659 megabytes), thus penalizing straightforward matches. It is now done only if the
1660 amount of subject to be scanned is less than 1000 bytes.
1662 7. A lesser problem with the same optimization is that it was recording the
1663 first character of an anchored pattern as "needed", thus provoking a search
1664 right along the subject, even when the first match of the pattern was going to
1665 fail. The "needed" character is now not set for anchored patterns, unless it
1666 follows something in the pattern that is of non-fixed length. Thus, it still
1667 fulfils its original purpose of finding quick non-matches in cases of nested
1668 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1671 Version 4.0 17-Feb-03
1672 ---------------------
1674 1. If a comment in an extended regex that started immediately after a meta-item
1675 extended to the end of string, PCRE compiled incorrect data. This could lead to
1676 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1678 2. Moved to autoconf 2.53 and libtool 1.4.2.
1680 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1681 the special perltest8 script is no longer needed - all the tests can be run
1682 from a single perltest script.
1684 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1685 by \s. It has now been removed in PCRE. This means it isn't recognized as
1686 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1687 class [:space:] *does* include VT, thereby creating a mess.
1689 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1690 space and tab.
1692 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1693 its new features into the main test script, reducing the number of scripts.
1695 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1696 were backward compatible, and made the (?i) apply to the whole pattern, as if
1697 /i were given. Now it behaves more logically, and applies the option setting
1698 only to what follows. PCRE has been changed to follow suit. However, if it
1699 finds options settings right at the start of the pattern, it extracts them into
1700 the global options, as before. Thus, they show up in the info data.
1702 8. Added support for the \Q...\E escape sequence. Characters in between are
1703 treated as literals. This is slightly different from Perl in that $ and @ are
1704 also handled as literals inside the quotes. In Perl, they will cause variable
1705 interpolation. Note the following examples:
1707 Pattern PCRE matches Perl matches
1709 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
1710 \Qabc\$xyz\E abc\$xyz abc\$xyz
1711 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
1713 For compatibility with Perl, \Q...\E sequences are recognized inside character
1714 classes as well as outside them.
1716 9. Re-organized 3 code statements in pcretest to avoid "overflow in
1717 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1718 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1719 signed/unsigned warnings.
1721 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1722 option for pcretest, so I've replaced it by a simple function that does just
1723 that job.
1725 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1726 "pcregrep -".
1728 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1729 Java package. This provides some syntactic sugar for simple cases of what my
1730 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1731 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1732 item, you can use this simplified notation. Note that only makes sense with
1733 greedy quantifiers. Consequently, the use of the possessive quantifier forces
1734 greediness, whatever the setting of the PCRE_UNGREEDY option.
1736 13. A change of greediness default within a pattern was not taking effect at
1737 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1738 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1739 was abstracted outside.
1741 14. PCRE now supports the \G assertion. It is true when the current matching
1742 position is at the start point of the match. This differs from \A when the
1743 starting offset is non-zero. Used with the /g option of pcretest (or similar
1744 code), it works in the same way as it does for Perl's /g option. If all
1745 alternatives of a regex begin with \G, the expression is anchored to the start
1746 match position, and the "anchored" flag is set in the compiled expression.
1748 15. Some bugs concerning the handling of certain option changes within patterns
1749 have been fixed. These applied to options other than (?ims). For example,
1750 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1751 been the other way round. Some of this was related to change 7 above.
1753 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1754 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1755 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1756 POSIX classes only within a class (e.g. /[[:alpha:]]/).
1758 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1759 mode. Unlike ".", it always matches newline, whatever the setting of
1760 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1761 assertions. Perl allows it, but it doesn't (in general) work because it can't
1762 calculate the length of the lookbehind. At least, that's the case for Perl
1763 5.8.0 - I've been told they are going to document that it doesn't work in
1764 future.
1766 18. Added an error diagnosis for escapes that PCRE does not support: these are
1767 \L, \l, \N, \P, \p, \U, \u, and \X.
1769 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1770 reading past the end of the pattern in cases such as /[abcd/.
1772 20. PCRE was getting more memory than necessary for patterns with classes that
1773 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1775 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1776 compiling PCRE for use with Virtual Pascal.
1778 22. Small fix to the Makefile to make it work properly if the build is done
1779 outside the source tree.
1781 23. Added a new extension: a condition to go with recursion. If a conditional
1782 subpattern starts with (?(R) the "true" branch is used if recursion has
1783 happened, whereas the "false" branch is used only at the top level.
1785 24. When there was a very long string of literal characters (over 255 bytes
1786 without UTF support, over 250 bytes with UTF support), the computation of how
1787 much memory was required could be incorrect, leading to segfaults or other
1788 strange effects.
1790 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1791 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1792 there was a subsequent back reference to those brackets. This meant that, for
1793 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1794 possible to check for precisely this case. All we can do is abandon the
1795 optimization if .* occurs inside capturing brackets when there are any back
1796 references whatsoever. (See below for a better fix that came later.)
1798 26. The handling of the optimization for finding the first character of a
1799 non-anchored pattern, and for finding a character that is required later in the
1800 match were failing in some cases. This didn't break the matching; it just
1801 failed to optimize when it could. The way this is done has been re-implemented.
1803 27. Fixed typo in error message for invalid (?R item (it said "(?p").
1805 28. Added a new feature that provides some of the functionality that Perl
1806 provides with (?{...}). The facility is termed a "callout". The way it is done
1807 in PCRE is for the caller to provide an optional function, by setting
1808 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1809 global variable. By default it is unset, which disables all calling out. To get
1810 the function called, the regex must include (?C) at appropriate points. This
1811 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1812 This provides a means of identifying different callout points. When PCRE
1813 reaches such a point in the regex, if pcre_callout has been set, the external
1814 function is called. It is provided with data in a structure called
1815 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1816 matching continues; if it returns a non-zero value, the match at the current
1817 point fails. However, backtracking will occur if possible. [This was changed
1818 later and other features added - see item 49 below.]
1820 29. pcretest is upgraded to test the callout functionality. It provides a
1821 callout function that displays information. By default, it shows the start of
1822 the match and the current position in the text. There are some new data escapes
1823 to vary what happens:
1825 \C+ in addition, show current contents of captured substrings
1826 \C- do not supply a callout function
1827 \C!n return 1 when callout number n is reached
1828 \C!n!m return 1 when callout number n is reached for the mth time
1830 30. If pcregrep was called with the -l option and just a single file name, it
1831 output "<stdin>" if a match was found, instead of the file name.
1833 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1834 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1835 pcre_exec(). This saves a malloc/free per call. The default value of
1836 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1837 when configuring.
1839 32. The default maximum size of a compiled pattern is 64K. There have been a
1840 few cases of people hitting this limit. The code now uses macros to handle the
1841 storing of links as offsets within the compiled pattern. It defaults to 2-byte
1842 links, but this can be changed to 3 or 4 bytes by --with-link-size when
1843 configuring. Tests 2 and 5 work only with 2-byte links because they output
1844 debugging information about compiled patterns.
1846 33. Internal code re-arrangements:
1848 (a) Moved the debugging function for printing out a compiled regex into
1849 its own source file (printint.c) and used #include to pull it into
1850 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1851 separate copies.
1853 (b) Defined the list of op-code names for debugging as a macro in
1854 internal.h so that it is next to the definition of the opcodes.
1856 (c) Defined a table of op-code lengths for simpler skipping along compiled
1857 code. This is again a macro in internal.h so that it is next to the
1858 definition of the opcodes.
1860 34. Added support for recursive calls to individual subpatterns, along the
1861 lines of Robin Houston's patch (but implemented somewhat differently).
1863 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1864 allow it to read and process whole directories in Win32. This code was
1865 contributed by Lionel Fourquaux; it has not been tested by me.
1867 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1868 used to name a group. Names consist of alphanumerics and underscores, and must
1869 be unique. Back references use the syntax (?P=name) and recursive calls use
1870 (?P>name) which is a PCRE extension to the Python extension. Groups still have
1871 numbers. The function pcre_fullinfo() can be used after compilation to extract
1872 a name/number map. There are three relevant calls:
1874 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
1875 PCRE_INFO_NAMECOUNT yields the number of entries
1876 PCRE_INFO_NAMETABLE yields a pointer to the map.
1878 The map is a vector of fixed-size entries. The size of each entry depends on
1879 the length of the longest name used. The first two bytes of each entry are the
1880 group number, most significant byte first. There follows the corresponding
1881 name, zero terminated. The names are in alphabetical order.
1883 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1884 case instead of 255. Making it the same both with and without UTF-8 support
1885 means that the same test output works with both.
1887 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1888 calling malloc() with a zero argument.
1890 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1891 optimization. I've improved things by keeping a bitmap of backreferences with
1892 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1893 fact referenced, the optimization can be applied. It is unlikely that a
1894 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1895 the match to follow \n) will appear inside brackets with a number greater than
1896 31, but if it does, any back reference > 31 suppresses the optimization.
1898 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1899 of disabling numbered capturing parentheses. Any opening parenthesis that is
1900 not followed by ? behaves as if it were followed by ?: but named parentheses
1901 can still be used for capturing (and they will acquire numbers in the usual
1902 way).
1904 41. Redesigned the return codes from the match() function into yes/no/error so
1905 that errors can be passed back from deep inside the nested calls. A malloc
1906 failure while inside a recursive subpattern call now causes the
1907 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
1909 42. It is now possible to set a limit on the number of times the match()
1910 function is called in a call to pcre_exec(). This facility makes it possible to
1911 limit the amount of recursion and backtracking, though not in a directly
1912 obvious way, because the match() function is used in a number of different
1913 circumstances. The count starts from zero for each position in the subject
1914 string (for non-anchored patterns). The default limit is, for compatibility, a
1915 large number, namely 10 000 000. You can change this in two ways:
1917 (a) When configuring PCRE before making, you can use --with-match-limit=n
1918 to set a default value for the compiled library.
1920 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
1921 a different value is set. See 45 below.
1923 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1925 43. Added a new function pcre_config(int, void *) to enable run-time extraction
1926 of things that can be changed at compile time. The first argument specifies
1927 what is wanted and the second points to where the information is to be placed.
1928 The current list of available information is:
1932 The output is an integer that is set to one if UTF-8 support is available;
1933 otherwise it is set to zero.
1937 The output is an integer that it set to the value of the code that is used for
1938 newline. It is either LF (10) or CR (13).
1942 The output is an integer that contains the number of bytes used for internal
1943 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
1947 The output is an integer that contains the threshold above which the POSIX
1948 interface uses malloc() for output vectors. See item 31 above.
1952 The output is an unsigned integer that contains the default limit of the number
1953 of match() calls in a pcre_exec() execution. See 42 above.
1955 44. pcretest has been upgraded by the addition of the -C option. This causes it
1956 to extract all the available output from the new pcre_config() function, and to
1957 output it. The program then exits immediately.
1959 45. A need has arisen to pass over additional data with calls to pcre_exec() in
1960 order to support additional features. One way would have been to define
1961 pcre_exec2() (for example) with extra arguments, but this would not have been
1962 extensible, and would also have required all calls to the original function to
1963 be mapped to the new one. Instead, I have chosen to extend the mechanism that
1964 is used for passing in "extra" data from pcre_study().
1966 The pcre_extra structure is now exposed and defined in pcre.h. It currently
1967 contains the following fields:
1969 flags a bitmap indicating which of the following fields are set
1970 study_data opaque data from pcre_study()
1971 match_limit a way of specifying a limit on match() calls for a specific
1972 call to pcre_exec()
1973 callout_data data for callouts (see 49 below)
1975 The flag bits are also defined in pcre.h, and are
1981 The pcre_study() function now returns one of these new pcre_extra blocks, with
1982 the actual study data pointed to by the study_data field, and the
1983 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
1984 before. That is, this change is entirely upwards-compatible and requires no
1985 change to existing code.
1987 If you want to pass in additional data to pcre_exec(), you can either place it
1988 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
1989 block.
1991 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
1992 data string contains the escape sequence \M, pcretest calls pcre_exec() several
1993 times with different match limits, until it finds the minimum value needed for
1994 pcre_exec() to complete. The value is then output. This can be instructive; for
1995 most simple matches the number is quite small, but for pathological cases it
1996 gets very large very quickly.
1998 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
1999 returns the size of the data block pointed to by the study_data field in a
2000 pcre_extra block, that is, the value that was passed as the argument to
2001 pcre_malloc() when PCRE was getting memory in which to place the information
2002 created by pcre_study(). The fourth argument should point to a size_t variable.
2003 pcretest has been extended so that this information is shown after a successful
2004 pcre_study() call when information about the compiled regex is being displayed.
2006 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2007 because what follows is always an absolute path. (Later: it turns out that this
2008 is more than cosmetic for MinGW, because it doesn't like empty path
2009 components.)
2011 49. Some changes have been made to the callout feature (see 28 above):
2013 (i) A callout function now has three choices for what it returns:
2015 0 => success, carry on matching
2016 > 0 => failure at this point, but backtrack if possible
2017 < 0 => serious error, return this value from pcre_exec()
2019 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2020 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2021 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2022 use by callout functions. It will never be used by PCRE itself.
2024 (ii) The pcre_extra structure (see 45 above) has a void * field called
2025 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2026 pcre_callout_block structure has a field of the same name. The contents of
2027 the field passed in the pcre_extra structure are passed to the callout
2028 function in the corresponding field in the callout block. This makes it
2029 easier to use the same callout-containing regex from multiple threads. For
2030 testing, the pcretest program has a new data escape
2032 \C*n pass the number n (may be negative) as callout_data
2034 If the callout function in pcretest receives a non-zero value as
2035 callout_data, it returns that value.
2037 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2038 there were some redundant $(CFLAGS) in commands that are now specified as
2039 $(LINK), which already includes $(CFLAGS).
2041 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2042 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2043 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2044 one-byte characters throughout. Note that case-insensitive matching applies
2045 only to characters whose values are less than 256. PCRE doesn't support the
2046 notion of cases for higher-valued characters.
2048 (i) A character class whose characters are all within 0-255 is handled as
2049 a bit map, and the map is inverted for negative classes. Previously, a
2050 character > 255 always failed to match such a class; however it should
2051 match if the class was a negative one (e.g. [^ab]). This has been fixed.
2053 (ii) A negated character class with a single character < 255 is coded as
2054 "not this character" (OP_NOT). This wasn't working properly when the test
2055 character was multibyte, either singly or repeated.
2057 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2058 mode, for example: \x{100}{2,3}.
2060 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2061 singly or repeated) now correctly test multibyte characters. However,
2062 PCRE doesn't recognize any characters with values greater than 255 as
2063 digits, spaces, or word characters. Such characters always match \D, \S,
2064 and \W, and never match \d, \s, or \w.
2066 (v) Classes may now contain characters and character ranges with values
2067 greater than 255. For example: [ab\x{100}-\x{400}].
2069 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
2070 PCRE in UTF-8 mode.
2072 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2073 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2074 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2075 value.)
2077 53. The single man page has become too large. I have therefore split it up into
2078 a number of separate man pages. These also give rise to individual HTML pages;
2079 these are now put in a separate directory, and there is an index.html page that
2080 lists them all. Some hyperlinking between the pages has been installed.
2082 54. Added convenience functions for handling named capturing parentheses.
2084 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2085 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2086 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2087 are faulted.
2089 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2090 calling configure. These values are used when compiling the dftables.c program
2091 which is run to generate the source of the default character tables. They
2092 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2093 you will need to set these values.
2095 57. Updated the building process for Windows DLL, as provided by Fred Cox.
2098 Version 3.9 02-Jan-02
2099 ---------------------
2101 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2103 2. If --disable-static was given, the building process failed when trying to
2104 build pcretest and pcregrep. (For some reason it was using libtool to compile
2105 them, which is not right, as they aren't part of the library.)
2108 Version 3.8 18-Dec-01
2109 ---------------------
2111 1. The experimental UTF-8 code was completely screwed up. It was packing the
2112 bytes in the wrong order. How dumb can you get?
2115 Version 3.7 29-Oct-01
2116 ---------------------
2118 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2119 This caused pcretest, when used on the test data, to segfault. Unfortunately,
2120 this didn't happen under Solaris 8, where I normally test things.
2122 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2123 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2124 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2125 this fix an hour or so after the initial 3.7 release.)
2128 Version 3.6 23-Oct-01
2129 ---------------------
2131 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2132 offsets passed as NULL with zero offset count.
2134 2. The config.guess and config.sub files had not been updated when I moved to
2135 the latest autoconf.
2138 Version 3.5 15-Aug-01
2139 ---------------------
2141 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2142 had been forgotten.
2144 2. By using declared but undefined structures, we can avoid using "void"
2145 definitions in pcre.h while keeping the internal definitions of the structures
2146 private.
2148 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2149 user point of view, this means that both static and shared libraries are built
2150 by default, but this can be individually controlled. More of the work of
2151 handling this static/shared cases is now inside libtool instead of PCRE's make
2152 file.
2154 4. The pcretest utility is now installed along with pcregrep because it is
2155 useful for users (to test regexs) and by doing this, it automatically gets
2156 relinked by libtool. The documentation has been turned into a man page, so
2157 there are now .1, .txt, and .html versions in /doc.
2159 5. Upgrades to pcregrep:
2160 (i) Added long-form option names like gnu grep.
2161 (ii) Added --help to list all options with an explanatory phrase.
2162 (iii) Added -r, --recursive to recurse into sub-directories.
2163 (iv) Added -f, --file to read patterns from a file.
2165 6. pcre_exec() was referring to its "code" argument before testing that
2166 argument for NULL (and giving an error if it was NULL).
2168 7. Upgraded Makefile.in to allow for compiling in a different directory from
2169 the source directory.
2171 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2172 options bits, the pointer it was passed was to an int instead of to an unsigned
2173 long int. This mattered only on 64-bit systems.
2175 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2176 generated) instead of pcre.in, which it its source. Also made the same change
2177 in several of the .c files.
2179 10. A new release of gcc defines printf() as a macro, which broke pcretest
2180 because it had an ifdef in the middle of a string argument for printf(). Fixed
2181 by using separate calls to printf().
2183 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2184 script, to force use of CR or LF instead of \n in the source. On non-Unix
2185 systems, the value can be set in config.h.
2187 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2188 absolute limit. Changed the text of the error message to make this clear, and
2189 likewise updated the man page.
2191 13. The limit of 99 on the number of capturing subpatterns has been removed.
2192 The new limit is 65535, which I hope will not be a "real" limit.
2195 Version 3.4 22-Aug-00
2196 ---------------------
2198 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2200 2. Diagnose condition (?(0) as an error instead of crashing on matching.
2203 Version 3.3 01-Aug-00
2204 ---------------------
2206 1. If an octal character was given, but the value was greater than \377, it
2207 was not getting masked to the least significant bits, as documented. This could
2208 lead to crashes in some systems.
2210 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2211 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2213 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2214 These just pass their arguments on to (pcre_free)(), but they are provided
2215 because some uses of PCRE bind it to non-C systems that can call its functions,
2216 but cannot call free() or pcre_free() directly.
2218 4. Add "make test" as a synonym for "make check". Corrected some comments in
2219 the Makefile.
2221 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2222 Makefile.
2224 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2225 command called pgrep for grepping around the active processes.
2227 7. Added the beginnings of support for UTF-8 character strings.
2229 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2230 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2231 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2232 out for the ar command.)
2235 Version 3.2 12-May-00
2236 ---------------------
2238 This is purely a bug fixing release.
2240 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2241 of ZA. This was just one example of several cases that could provoke this bug,
2242 which was introduced by change 9 of version 2.00. The code for breaking
2243 infinite loops after an iteration that matches an empty string was't working
2244 correctly.
2246 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2247 when matched against abbab (for example). After matching an empty string, it
2248 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2249 caused it to match further down the string than it should.
2251 3. The code contained an inclusion of sys/types.h. It isn't clear why this
2252 was there because it doesn't seem to be needed, and it causes trouble on some
2253 systems, as it is not a Standard C header. It has been removed.
2255 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2256 were reported on the Macintosh. The changes were from
2258 while ((c = *(++ptr)) != 0 && c != '\n');
2259 to
2260 while ((c = *(++ptr)) != 0 && c != '\n') ;
2262 Totally extraordinary, but if that's what it takes...
2264 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2265 available. Added HAVE_BCOPY and an autoconf test for it; if neither
2266 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2267 assumes the way PCRE uses memmove() (always moving upwards).
2269 6. PCRE is being used in one environment where strchr() is not available. There
2270 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2271 faster code anyway.
2274 Version 3.1 09-Feb-00
2275 ---------------------
2277 The only change in this release is the fixing of some bugs in Makefile.in for
2278 the "install" target:
2280 (1) It was failing to install pcreposix.h.
2282 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2285 Version 3.0 01-Feb-00
2286 ---------------------
2288 1. Add support for the /+ modifier to perltest (to output $` like it does in
2289 pcretest).
2291 2. Add support for the /g modifier to perltest.
2293 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2294 matches null strings.
2296 4. Fix perltest so that it doesn't do unwanted things when fed an empty
2297 pattern. Perl treats empty patterns specially - it reuses the most recent
2298 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2299 effect.
2301 5. The POSIX interface was broken in that it was just handing over the POSIX
2302 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2303 required a bigger vector, with some working space on the end. This means that
2304 the POSIX wrapper now has to get and free some memory, and copy the results.
2306 6. Added some simple autoconf support, placing the test data and the
2307 documentation in separate directories, re-organizing some of the
2308 information files, and making it build pcre-config (a GNU standard). Also added
2309 libtool support for building PCRE as a shared library, which is now the
2310 default.
2312 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2313 09 are not valid octal constants. Single digits will be used for minor values
2314 less than 10.
2316 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2317 existing programs that set these in the POSIX interface can use PCRE without
2318 modification.
2320 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2321 return all that pcre_info() returns, plus additional data. The pcre_info()
2322 function is retained for compatibility, but is considered to be obsolete.
2324 10. Added experimental recursion feature (?R) to handle one common case that
2325 Perl 5.6 will be able to do with (?p{...}).
2327 11. Added support for POSIX character classes like [:alpha:], which Perl is
2328 adopting.
2331 Version 2.08 31-Aug-99
2332 ----------------------
2334 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2335 trying to match at the startoffset position, but instead was moving forward to
2336 the next newline as if a previous match had failed.
2338 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2339 and could get into a loop if a null string was matched other than at the start
2340 of the subject.
2342 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2343 be distinguished at compile time, and for completeness also added PCRE_DATE.
2345 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2346 in GnuWin32 environments.
2349 Version 2.07 29-Jul-99
2350 ----------------------
2352 1. The documentation is now supplied in plain text form and HTML as well as in
2353 the form of man page sources.
2355 2. C++ compilers don't like assigning (void *) values to other pointer types.
2356 In particular this affects malloc(). Although there is no problem in Standard
2357 C, I've put in casts to keep C++ compilers happy.
2359 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2360 should be (const char *).
2362 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2363 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2364 However, I haven't made this a standard facility. The documentation doesn't
2365 mention it, and the Makefile doesn't support it.
2367 5. The Makefile now contains an "install" target, with editable destinations at
2368 the top of the file. The pcretest program is not installed.
2370 6. pgrep -V now gives the PCRE version number and date.
2372 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2373 causing the entire string to be ignored, instead of just the last character.
2375 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2376 non-matching string, it can take a very, very long time, even for strings of
2377 quite modest length, because of the nested recursion. PCRE now does better in
2378 some of these cases. It does this by remembering the last required literal
2379 character in the pattern, and pre-searching the subject to ensure it is present
2380 before running the real match. In other words, it applies a heuristic to detect
2381 some types of certain failure quickly, and in the above example, if presented
2382 with a string that has no trailing " it gives "no match" very quickly.
2384 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2385 other alternatives are tried instead.
2388 Version 2.06 09-Jun-99
2389 ----------------------
2391 1. Change pcretest's output for amount of store used to show just the code
2392 space, because the remainder (the data block) varies in size between 32-bit and
2393 64-bit systems.
2395 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2396 start matching at. This allows lookbehinds to work when searching for multiple
2397 occurrences in a string.
2399 3. Added additional options to pcretest for testing multiple occurrences:
2401 /+ outputs the rest of the string that follows a match
2402 /g loops for multiple occurrences, using the new startoffset argument
2403 /G loops for multiple occurrences by passing an incremented pointer
2405 4. PCRE wasn't doing the "first character" optimization for patterns starting
2406 with \b or \B, though it was doing it for other lookbehind assertions. That is,
2407 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2408 the letter 'x'. On long subject strings, this gives a significant speed-up.
2411 Version 2.05 21-Apr-99
2412 ----------------------
2414 1. Changed the type of magic_number from int to long int so that it works
2415 properly on 16-bit systems.
2417 2. Fixed a bug which caused patterns starting with .* not to work correctly
2418 when the subject string contained newline characters. PCRE was assuming
2419 anchoring for such patterns in all cases, which is not correct because .* will
2420 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2421 DOTALL is set at top level; otherwise it knows that patterns starting with .*
2422 must be retried after every newline in the subject.
2425 Version 2.04 18-Feb-99
2426 ----------------------
2428 1. For parenthesized subpatterns with repeats whose minimum was zero, the
2429 computation of the store needed to hold the pattern was incorrect (too large).
2430 If such patterns were nested a few deep, this could multiply and become a real
2431 problem.
2433 2. Added /M option to pcretest to show the memory requirement of a specific
2434 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2436 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2437 compiled in such a way that the backtracking after subsequent failure was
2438 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2439 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2442 Version 2.03 02-Feb-99
2443 ----------------------
2445 1. Fixed typo and small mistake in man page.
2447 2. Added 4th condition (GPL supersedes if conflict) and created separate
2448 LICENCE file containing the conditions.
2450 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2451 Perl, that is the internal \ allows the delimiter to be included in the
2452 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2453 the final delimiter, add \ to the end of the pattern (to test the error).
2455 4. Added the convenience functions for extracting substrings after a successful
2456 match. Updated pcretest to make it able to test these functions.
2459 Version 2.02 14-Jan-99
2460 ----------------------
2462 1. Initialized the working variables associated with each extraction so that
2463 their saving and restoring doesn't refer to uninitialized store.
2465 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2466 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2467 fix the problem.
2469 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2470 calls, and wasn't printing the correct value for compiling calls. Increased the
2471 default value of LOOPREPEAT, and the number of significant figures in the
2472 times.
2474 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2476 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2477 a building problem on Windows NT with a FAT file system.
2480 Version 2.01 21-Oct-98
2481 ----------------------
2483 1. Changed the API for pcre_compile() to allow for the provision of a pointer
2484 to character tables built by pcre_maketables() in the current locale. If NULL
2485 is passed, the default tables are used.
2488 Version 2.00 24-Sep-98
2489 ----------------------
2491 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2492 it any more.
2494 2. Allow quantification of (?>) groups, and make it work correctly.
2496 3. The first character computation wasn't working for (?>) groups.
2498 4. Correct the implementation of \Z (it is permitted to match on the \n at the
2499 end of the subject) and add 5.005's \z, which really does match only at the
2500 very end of the subject.
2502 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2504 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2505 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2506 localized options. All options to pcre_study() were also removed.
2508 7. Add other new features from 5.005:
2510 $(?<= positive lookbehind
2511 $(?<! negative lookbehind
2512 (?imsx-imsx) added the unsetting capability
2513 such a setting is global if at outer level; local otherwise
2514 (?imsx-imsx:) non-capturing groups with option setting
2515 (?(cond)re|re) conditional pattern matching
2517 A backreference to itself in a repeated group matches the previous
2518 captured string.
2520 8. General tidying up of studying (both automatic and via "study")
2521 consequential on the addition of new assertions.
2523 9. As in 5.005, unlimited repeated groups that could match an empty substring
2524 are no longer faulted at compile time. Instead, the loop is forcibly broken at
2525 runtime if any iteration does actually match an empty substring.
2527 10. Include the RunTest script in the distribution.
2529 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2530 discrepancies, some of which were old and were also with respect to 5.004. They
2531 have now been fixed.
2534 Version 1.09 28-Apr-98
2535 ----------------------
2537 1. A negated single character class followed by a quantifier with a minimum
2538 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
2539 program crashes, or just wrong answers. This did not apply to negated classes
2540 containing more than one character, or to minima other than one.
2543 Version 1.08 27-Mar-98
2544 ----------------------
2546 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
2548 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
2549 latter must appear before anything that relies on it in the pattern.
2552 Version 1.07 16-Feb-98
2553 ----------------------
2555 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
2556 repeat of a potentially empty string).
2559 Version 1.06 23-Jan-98
2560 ----------------------
2562 1. Added Markus Oberhumer's little patches for C++.
2564 2. Literal strings longer than 255 characters were broken.
2567 Version 1.05 23-Dec-97
2568 ----------------------
2570 1. Negated character classes containing more than one character were failing if
2571 PCRE_CASELESS was set at run time.
2574 Version 1.04 19-Dec-97
2575 ----------------------
2577 1. Corrected the man page, where some "const" qualifiers had been omitted.
2579 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
2580 input syntax.
2582 3. Fixed memory leak which occurred when a regex with back references was
2583 matched with an offsets vector that wasn't big enough. The temporary memory
2584 that is used in this case wasn't being freed if the match failed.
2586 4. Tidied pcretest to ensure it frees memory that it gets.
2588 5. Temporary memory was being obtained in the case where the passed offsets
2589 vector was exactly big enough.
2591 6. Corrected definition of offsetof() from change 5 below.
2593 7. I had screwed up change 6 below and broken the rules for the use of
2594 setjmp(). Now fixed.
2597 Version 1.03 18-Dec-97
2598 ----------------------
2600 1. A erroneous regex with a missing opening parenthesis was correctly
2601 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
2602 on some systems.
2604 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
2605 it was reported that one broken compiler failed on the former because "code" is
2606 also an independent variable.
2608 3. The erroneous regex a[]b caused an array overrun reference.
2610 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
2611 fail on data ending with that character. (It was going on too far, and checking
2612 the next character, typically a binary zero.) This was specific to the
2613 optimized code for single-character negative classes.
2615 5. Added a contributed patch from the TIN world which does the following:
2617 + Add an undef for memmove, in case the the system defines a macro for it.
2619 + Add a definition of offsetof(), in case there isn't one. (I don't know
2620 the reason behind this - offsetof() is part of the ANSI standard - but
2621 it does no harm).
2623 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
2624 most of the places where whitespace preceded '#'. I have given up and
2625 allowed the remaining 2 cases to be at the margin.
2627 + Rename some variables in pcre to eliminate shadowing. This seems very
2628 pedantic, but does no harm, of course.
2630 6. Moved the call to setjmp() into its own function, to get rid of warnings
2631 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
2633 7. Constructs such as \d{8,} were compiling into the equivalent of
2634 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
2635 outcome, but in this particular case used more store than had been allocated,
2636 which caused the bug to be discovered because it threw up an internal error.
2638 8. The debugging code in both pcre and pcretest for outputting the compiled
2639 form of a regex was going wrong in the case of back references followed by
2640 curly-bracketed repeats.
2643 Version 1.02 12-Dec-97
2644 ----------------------
2646 1. Typos in pcre.3 and comments in the source fixed.
2648 2. Applied a contributed patch to get rid of places where it used to remove
2649 'const' from variables, and fixed some signed/unsigned and uninitialized
2650 variable warnings.
2652 3. Added the "runtest" target to Makefile.
2654 4. Set default compiler flag to -O2 rather than just -O.
2657 Version 1.01 19-Nov-97
2658 ----------------------
2660 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
2661 like /([ab]*)*/, that is, for classes with more than one character in them.
2663 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
2664 as /((?>a*))*/ (a PCRE_EXTRA facility).
2667 Version 1.00 18-Nov-97
2668 ----------------------
2670 1. Added compile-time macros to support systems such as SunOS4 which don't have
2671 memmove() or strerror() but have other things that can be used instead.
2673 2. Arranged that "make clean" removes the executables.
2676 Version 0.99 27-Oct-97
2677 ----------------------
2679 1. Fixed bug in code for optimizing classes with only one character. It was
2680 initializing a 32-byte map regardless, which could cause it to run off the end
2681 of the memory it had got.
2683 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
2686 Version 0.98 22-Oct-97
2687 ----------------------
2689 1. Fixed bug in code for handling temporary memory usage when there are more
2690 back references than supplied space in the ovector. This could cause segfaults.
2693 Version 0.97 21-Oct-97
2694 ----------------------
2696 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
2698 2. Optimized negated single characters not to use a bit map.
2700 3. Brought error texts together as macro definitions; clarified some of them;
2701 fixed one that was wrong - it said "range out of order" when it meant "invalid
2702 escape sequence".
2704 4. Changed some char * arguments to const char *.
2706 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
2708 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
2709 pcretest.
2712 Version 0.96 16-Oct-97
2713 ----------------------
2715 1. Added a simple "pgrep" utility to the distribution.
2717 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
2718 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
2719 where "ddd" means "one or more decimal digits".
2721 3. Fixed serious bug. If a pattern had a back reference, but the call to
2722 pcre_exec() didn't supply a large enough ovector to record the related
2723 identifying subpattern, the match always failed. PCRE now remembers the number
2724 of the largest back reference, and gets some temporary memory in which to save
2725 the offsets during matching if necessary, in order to ensure that
2726 backreferences always work.
2728 4. Increased the compatibility with Perl in a number of ways:
2730 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
2731 to request this handling. The option can be set at compile or exec time.
2733 (b) $ matches before a terminating newline by default; an option
2734 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
2735 mode). The option can be set at compile or exec time.
2737 (c) The handling of \ followed by a digit other than 0 is now supposed to be
2738 the same as Perl's. If the decimal number it represents is less than 10
2739 or there aren't that many previous left capturing parentheses, an octal
2740 escape is read. Inside a character class, it's always an octal escape,
2741 even if it is a single digit.
2743 (d) An escaped but undefined alphabetic character is taken as a literal,
2744 unless PCRE_EXTRA is set. Currently this just reserves the remaining
2745 escapes.
2747 (e) {0} is now permitted. (The previous item is removed from the compiled
2748 pattern).
2750 5. Changed all the names of code files so that the basic parts are no longer
2751 than 10 characters, and abolished the teeny "globals.c" file.
2753 6. Changed the handling of character classes; they are now done with a 32-byte
2754 bit map always.
2756 7. Added the -d and /D options to pcretest to make it possible to look at the
2757 internals of compilation without having to recompile pcre.
2760 Version 0.95 23-Sep-97
2761 ----------------------
2763 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
2764 \x20 at the start of a run of normal characters. These were being treated as
2765 real characters, instead of the source characters being re-checked.
2768 Version 0.94 18-Sep-97
2769 ----------------------
2771 1. The functions are now thread-safe, with the caveat that the global variables
2772 containing pointers to malloc() and free() or alternative functions are the
2773 same for all threads.
2775 2. Get pcre_study() to generate a bitmap of initial characters for non-
2776 anchored patterns when this is possible, and use it if passed to pcre_exec().
2779 Version 0.93 15-Sep-97
2780 ----------------------
2782 1. /(b)|(:+)/ was computing an incorrect first character.
2784 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
2785 but not actually doing anything yet.
2787 3. Treat "-" characters in classes that cannot be part of ranges as literals,
2788 as Perl does (e.g. [-az] or [az-]).
2790 4. Set the anchored flag if a branch starts with .* or .*? because that tests
2791 all possible positions.
2793 5. Split up into different modules to avoid including unneeded functions in a
2794 compiled binary. However, compile and exec are still in one module. The "study"
2795 function is split off.
2797 6. The character tables are now in a separate module whose source is generated
2798 by an auxiliary program - but can then be edited by hand if required. There are
2799 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
2800 toupper() in the code.
2802 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
2803 make them global. Abolish the function for setting them, as the caller can now
2804 set them directly.
2807 Version 0.92 11-Sep-97
2808 ----------------------
2810 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
2811 (e.g. /a{1,3}/) was broken (I mis-optimized it).
2813 2. Caseless matching was not working in character classes if the characters in
2814 the pattern were in upper case.
2816 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
2818 4. Make PCRE_ANCHORED public and accept as a compile option.
2820 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
2821 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
2822 pass them.
2824 6. Give an error if bad option bits passed at compile or run time.
2826 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
2827 pcretest to cause it to pass that flag.
2829 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
2830 options, and the first character, if set.
2832 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
2835 Version 0.91 10-Sep-97
2836 ----------------------
2838 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
2839 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
2841 2. PCRE was looping on encountering an indefinitely repeated back reference to
2842 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
2843 Perl does - treats the match as successful.
2845 ****


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

  ViewVC Help
Powered by ViewVC 1.1.5