ViewVC logotype

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 202 - (show annotations)
Fri Aug 3 09:44:26 2007 UTC (14 years, 1 month ago) by ph10
File size: 133523 byte(s)
Refactor the integer overflow testing so as to avoid imposing an artificial 
limit on the size of 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.
69 Version 7.2 19-Jun-07
70 ---------------------
72 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
73 which is apparently normally available under Windows.
75 2. Re-jig the pcregrep tests with different newline settings in an attempt
76 to make them independent of the local environment's newline setting.
78 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
80 4. Some of the "internals" tests were previously cut out when the link size
81 was not 2, because the output contained actual offsets. The recent new
82 "Z" feature of pcretest means that these can be cut out, making the tests
83 usable with all link sizes.
85 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
86 stack recursion. This gives a massive performance boost under BSD, but just
87 a small improvement under Linux. However, it saves one field in the frame
88 in all cases.
90 6. Added more features from the forthcoming Perl 5.10:
92 (a) (?-n) (where n is a string of digits) is a relative subroutine or
93 recursion call. It refers to the nth most recently opened parentheses.
95 (b) (?+n) is also a relative subroutine call; it refers to the nth next
96 to be opened parentheses.
98 (c) Conditions that refer to capturing parentheses can be specified
99 relatively, for example, (?(-2)... or (?(+3)...
101 (d) \K resets the start of the current match so that everything before
102 is not part of it.
104 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
106 (f) \g{name} is another synonym - part of Perl 5.10's unification of
107 reference syntax.
109 (g) (?| introduces a group in which the numbering of parentheses in each
110 alternative starts with the same number.
112 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
114 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
117 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
118 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
119 for detecting groups that can match an empty string.
121 9. A pattern with a very large number of alternatives (more than several
122 hundred) was running out of internal workspace during the pre-compile
123 phase, where pcre_compile() figures out how much memory will be needed. A
124 bit of new cunning has reduced the workspace needed for groups with
125 alternatives. The 1000-alternative test pattern now uses 12 bytes of
126 workspace instead of running out of the 4096 that are available.
128 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
130 11. Applied patch from Google to remove an optimization that didn't quite work.
131 The report of the bug said:
133 pcrecpp::RE("a*").FullMatch("aaa") matches, while
134 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
135 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
137 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
138 it matched the wrong number of bytes.
141 Version 7.1 24-Apr-07
142 ---------------------
144 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
145 that is more "standard", making use of automake and other Autotools. There
146 is some re-arrangement of the files and adjustment of comments consequent
147 on this.
149 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
150 for recursive directory scanning broke on some systems because the files
151 are not scanned in any specific order and on different systems the order
152 was different. A call to "sort" has been inserted into RunGrepTest for the
153 approprate test as a short-term fix. In the longer term there may be an
154 alternative.
156 3. I had an email from Eric Raymond about problems translating some of PCRE's
157 man pages to HTML (despite the fact that I distribute HTML pages, some
158 people do their own conversions for various reasons). The problems
159 concerned the use of low-level troff macros .br and .in. I have therefore
160 removed all such uses from the man pages (some were redundant, some could
161 be replaced by .nf/.fi pairs). The 132html script that I use to generate
162 HTML has been updated to handle .nf/.fi and to complain if it encounters
163 .br or .in.
165 4. Updated comments in configure.ac that get placed in config.h.in and also
166 arranged for config.h to be included in the distribution, with the name
167 config.h.generic, for the benefit of those who have to compile without
168 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
170 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
171 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
172 makevp.bat for the current PCRE, using the additional files
173 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
175 6. A Windows user reported a minor discrepancy with test 2, which turned out
176 to be caused by a trailing space on an input line that had got lost in his
177 copy. The trailing space was an accident, so I've just removed it.
179 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
180 that is needed.
182 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
183 as "const" (a) because they are and (b) because it helps the PHP
184 maintainers who have recently made a script to detect big data structures
185 in the php code that should be moved to the .rodata section. I remembered
186 to update Builducptable as well, so it won't revert if ucptable.h is ever
187 re-created.
189 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
190 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
191 order to be able to cut out the UTF-8 tables in the latter when UTF-8
192 support is not required. This saves 1.5-2K of code, which is important in
193 some applications.
195 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
196 so as not to refer to the tables, even though these functions will never be
197 called when UTF-8 support is disabled. Otherwise there are problems with a
198 shared library.
200 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
202 (a) It was defining its arguments as char * instead of void *.
204 (b) It was assuming that all moves were upwards in memory; this was true
205 a long time ago when I wrote it, but is no longer the case.
207 The emulated memove() is provided for those environments that have neither
208 memmove() nor bcopy(). I didn't think anyone used it these days, but that
209 is clearly not the case, as these two bugs were recently reported.
211 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
212 and Detrail to create the HTML documentation, the .txt form of the man
213 pages, and it removes trailing spaces from listed files. It also creates
214 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
215 case, it wraps all the #defines with #ifndefs. This script should be run
216 before "make dist".
218 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
219 with Unicode property support.
221 (a) For a maximizing quantifier, if the two different cases of the
222 character were of different lengths in their UTF-8 codings (there are
223 some cases like this - I found 11), and the matching function had to
224 back up over a mixture of the two cases, it incorrectly assumed they
225 were both the same length.
227 (b) When PCRE was configured to use the heap rather than the stack for
228 recursion during matching, it was not correctly preserving the data for
229 the other case of a UTF-8 character when checking ahead for a match
230 while processing a minimizing repeat. If the check also involved
231 matching a wide character, but failed, corruption could cause an
232 erroneous result when trying to check for a repeat of the original
233 character.
235 13. Some tidying changes to the testing mechanism:
237 (a) The RunTest script now detects the internal link size and whether there
238 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
239 values substituted by "configure". (The RunGrepTest script already did
240 this for UTF-8.) The configure.ac script no longer substitutes the
241 relevant variables.
243 (b) The debugging options /B and /D in pcretest show the compiled bytecode
244 with length and offset values. This means that the output is different
245 for different internal link sizes. Test 2 is skipped for link sizes
246 other than 2 because of this, bypassing the problem. Unfortunately,
247 there was also a test in test 3 (the locale tests) that used /B and
248 failed for link sizes other than 2. Rather than cut the whole test out,
249 I have added a new /Z option to pcretest that replaces the length and
250 offset values with spaces. This is now used to make test 3 independent
251 of link size. (Test 2 will be tidied up later.)
253 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
254 segmentation fault instead of returning the appropriate error message.
256 15. In multiline mode when the newline sequence was set to "any", the pattern
257 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
258 This doesn't seem right; it now treats the CRLF combination as the line
259 ending, and so does not match in that case. It's only a pattern such as ^$
260 that would hit this one: something like ^ABC$ would have failed after \r
261 and then tried again after \r\n.
263 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
264 in an attempt to make files that differ only in their line terminators
265 compare equal. This works on Linux.
267 17. Under certain error circumstances pcregrep might try to free random memory
268 as it exited. This is now fixed, thanks to valgrind.
270 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
271 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
272 was because its rules for how to advance for /g after matching an empty
273 string at the end of a line did not allow for this case. They now check for
274 it specially.
276 20. pcretest is supposed to handle patterns and data of any length, by
277 extending its buffers when necessary. It was getting this wrong when the
278 buffer for a data line had to be extended.
280 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
281 CRLF as a newline sequence.
283 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
284 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
285 I have nevertheless tidied it up.
287 23. Added some casts to kill warnings from HP-UX ia64 compiler.
289 24. Added a man page for pcre-config.
292 Version 7.0 19-Dec-06
293 ---------------------
295 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
296 moving to gcc 4.1.1.
298 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
299 sys/time.h, which is documented as needed for this function. It doesn't
300 seem to matter on Linux, but it showed up on some releases of OS X.
302 3. It seems that there are systems where bytes whose values are greater than
303 127 match isprint() in the "C" locale. The "C" locale should be the
304 default when a C program starts up. In most systems, only ASCII printing
305 characters match isprint(). This difference caused the output from pcretest
306 to vary, making some of the tests fail. I have changed pcretest so that:
308 (a) When it is outputting text in the compiled version of a pattern, bytes
309 other than 32-126 are always shown as hex escapes.
311 (b) When it is outputting text that is a matched part of a subject string,
312 it does the same, unless a different locale has been set for the match
313 (using the /L modifier). In this case, it uses isprint() to decide.
315 4. Fixed a major bug that caused incorrect computation of the amount of memory
316 required for a compiled pattern when options that changed within the
317 pattern affected the logic of the preliminary scan that determines the
318 length. The relevant options are -x, and -i in UTF-8 mode. The result was
319 that the computed length was too small. The symptoms of this bug were
320 either the PCRE error "internal error: code overflow" from pcre_compile(),
321 or a glibc crash with a message such as "pcretest: free(): invalid next
322 size (fast)". Examples of patterns that provoked this bug (shown in
323 pcretest format) are:
325 /(?-x: )/x
326 /(?x)(?-x: \s*#\s*)/
327 /((?i)[\x{c0}])/8
328 /(?i:[\x{c0}])/8
330 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
331 is now done differently.
333 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
334 wrapper classes; (b) implement a new function in the C++ scanner that is
335 more efficient than the old way of doing things because it avoids levels of
336 recursion in the regex matching; (c) add a paragraph to the documentation
337 for the FullMatch() function.
339 6. The escape sequence \n was being treated as whatever was defined as
340 "newline". Not only was this contrary to the documentation, which states
341 that \n is character 10 (hex 0A), but it also went horribly wrong when
342 "newline" was defined as CRLF. This has been fixed.
344 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
345 was being set to -1 for the "end of line" case (supposedly a value that no
346 character can have). Though this value is never used (the check for end of
347 line is "zero bytes in current character"), it caused compiler complaints.
348 I've changed it to 0xffffffff.
350 8. In pcre_version.c, the version string was being built by a sequence of
351 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
352 string (as it is for production releases) called a macro with an empty
353 argument. The C standard says the result of this is undefined. The gcc
354 compiler treats it as an empty string (which was what was wanted) but it is
355 reported that Visual C gives an error. The source has been hacked around to
356 avoid this problem.
358 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
359 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
360 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
361 of them did).
363 10. Originally, pcretest opened its input and output without "b"; then I was
364 told that "b" was needed in some environments, so it was added for release
365 5.0 to both the input and output. (It makes no difference on Unix-like
366 systems.) Later I was told that it is wrong for the input on Windows. I've
367 now abstracted the modes into two macros, to make it easier to fiddle with
368 them, and removed "b" from the input mode under Windows.
370 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
372 12. Added -help and --help to pcretest as an official way of being reminded
373 of the options.
375 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
376 and pcrecpp.cc because they annoy compilers at high warning levels.
378 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
380 15. Fixed an occurrence of == in configure.ac that should have been = (shell
381 scripts are not C programs :-) and which was not noticed because it works
382 on Linux.
384 16. pcretest is supposed to handle any length of pattern and data line (as one
385 line or as a continued sequence of lines) by extending its input buffer if
386 necessary. This feature was broken for very long pattern lines, leading to
387 a string of junk being passed to pcre_compile() if the pattern was longer
388 than about 50K.
390 17. I have done a major re-factoring of the way pcre_compile() computes the
391 amount of memory needed for a compiled pattern. Previously, there was code
392 that made a preliminary scan of the pattern in order to do this. That was
393 OK when PCRE was new, but as the facilities have expanded, it has become
394 harder and harder to keep it in step with the real compile phase, and there
395 have been a number of bugs (see for example, 4 above). I have now found a
396 cunning way of running the real compile function in a "fake" mode that
397 enables it to compute how much memory it would need, while actually only
398 ever using a few hundred bytes of working memory and without too many
399 tests of the mode. This should make future maintenance and development
400 easier. A side effect of this work is that the limit of 200 on the nesting
401 depth of parentheses has been removed (though this was never a serious
402 limitation, I suspect). However, there is a downside: pcre_compile() now
403 runs more slowly than before (30% or more, depending on the pattern). I
404 hope this isn't a big issue. There is no effect on runtime performance.
406 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
407 newline (only possible for the last line of a file) and it was a
408 pattern that set a locale (followed by /Lsomething), pcretest crashed.
410 19. Added additional timing features to pcretest. (1) The -tm option now times
411 matching only, not compiling. (2) Both -t and -tm can be followed, as a
412 separate command line item, by a number that specifies the number of
413 repeats to use when timing. The default is 50000; this gives better
414 precision, but takes uncomfortably long for very large patterns.
416 20. Extended pcre_study() to be more clever in cases where a branch of a
417 subpattern has no definite first character. For example, (a*|b*)[cd] would
418 previously give no result from pcre_study(). Now it recognizes that the
419 first character must be a, b, c, or d.
421 21. There was an incorrect error "recursive call could loop indefinitely" if
422 a subpattern (or the entire pattern) that was being tested for matching an
423 empty string contained only one non-empty item after a nested subpattern.
424 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
425 incorrectly, because the \d was being skipped in the check.
427 22. The pcretest program now has a new pattern option /B and a command line
428 option -b, which is equivalent to adding /B to every pattern. This causes
429 it to show the compiled bytecode, without the additional information that
430 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
431 is the same as /B/I).
433 23. A new optimization is now able automatically to treat some sequences such
434 as a*b as a*+b. More specifically, if something simple (such as a character
435 or a simple class like \d) has an unlimited quantifier, and is followed by
436 something that cannot possibly match the quantified thing, the quantifier
437 is automatically "possessified".
439 24. A recursive reference to a subpattern whose number was greater than 39
440 went wrong under certain circumstances in UTF-8 mode. This bug could also
441 have affected the operation of pcre_study().
443 25. Realized that a little bit of performance could be had by replacing
444 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
446 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
448 27. Possessive quantifiers such as a++ were previously implemented by turning
449 them into atomic groups such as ($>a+). Now they have their own opcodes,
450 which improves performance. This includes the automatically created ones
451 from 23 above.
453 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
454 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
455 the first matched character to be a colon. This applied both to named and
456 numbered groups.
458 29. The ucpinternal.h header file was missing its idempotency #ifdef.
460 30. I was sent a "project" file called libpcre.a.dev which I understand makes
461 building PCRE on Windows easier, so I have included it in the distribution.
463 31. There is now a check in pcretest against a ridiculously large number being
464 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
465 loop, the loop is abandoned.
467 32. Forward references to subpatterns in conditions such as (?(2)...) where
468 subpattern 2 is defined later cause pcre_compile() to search forwards in
469 the pattern for the relevant set of parentheses. This search went wrong
470 when there were unescaped parentheses in a character class, parentheses
471 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
473 33. "Subroutine" calls and backreferences were previously restricted to
474 referencing subpatterns earlier in the regex. This restriction has now
475 been removed.
477 34. Added a number of extra features that are going to be in Perl 5.10. On the
478 whole, these are just syntactic alternatives for features that PCRE had
479 previously implemented using the Python syntax or my own invention. The
480 other formats are all retained for compatibility.
482 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
483 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
484 also .NET compatible.
486 (b) A recursion or subroutine call to a named group can now be defined as
487 (?&name) as well as (?P>name).
489 (c) A backreference to a named group can now be defined as \k<name> or
490 \k'name' as well as (?P=name). The new forms, as well as being in Perl
491 5.10, are also .NET compatible.
493 (d) A conditional reference to a named group can now use the syntax
494 (?(<name>) or (?('name') as well as (?(name).
496 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
497 groups (named and numbered) that are never evaluated inline, but can be
498 called as "subroutines" from elsewhere. In effect, the DEFINE condition
499 is always false. There may be only one alternative in such a group.
501 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
502 as the simple (?(R). The condition is true only if the most recent
503 recursion is that of the given number or name. It does not search out
504 through the entire recursion stack.
506 (g) The escape \gN or \g{N} has been added, where N is a positive or
507 negative number, specifying an absolute or relative reference.
509 35. Tidied to get rid of some further signed/unsigned compiler warnings and
510 some "unreachable code" warnings.
512 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
513 things, this adds five new scripts.
515 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
516 There were also incompatibilities regarding the handling of \Q..\E inside
517 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
518 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
520 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
521 matches an empty string, and forcibly breaks the loop. There were bugs in
522 this code in non-simple cases. For a pattern such as ^(a()*)* matched
523 against aaaa the result was just "a" rather than "aaaa", for example. Two
524 separate and independent bugs (that affected different cases) have been
525 fixed.
527 39. Refactored the code to abolish the use of different opcodes for small
528 capturing bracket numbers. This is a tidy that I avoided doing when I
529 removed the limit on the number of capturing brackets for 3.5 back in 2001.
530 The new approach is not only tidier, it makes it possible to reduce the
531 memory needed to fix the previous bug (38).
533 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
534 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
535 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
536 mode.
538 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
539 report.
541 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
542 copy construction and assignment in the C++ wrapper.
544 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
545 couple of bugs that could have given wrong results in the "--newline=crlf"
546 case.
548 44. Added a number of casts and did some reorganization of signed/unsigned int
549 variables following suggestions from Dair Grant. Also renamed the variable
550 "this" as "item" because it is a C++ keyword.
552 45. Arranged for dftables to add
554 #include "pcre_internal.h"
556 to pcre_chartables.c because without it, gcc 4.x may remove the array
557 definition from the final binary if PCRE is built into a static library and
558 dead code stripping is activated.
560 46. For an unanchored pattern, if a match attempt fails at the start of a
561 newline sequence, and the newline setting is CRLF or ANY, and the next two
562 characters are CRLF, advance by two characters instead of one.
565 Version 6.7 04-Jul-06
566 ---------------------
568 1. In order to handle tests when input lines are enormously long, pcretest has
569 been re-factored so that it automatically extends its buffers when
570 necessary. The code is crude, but this _is_ just a test program. The
571 default size has been increased from 32K to 50K.
573 2. The code in pcre_study() was using the value of the re argument before
574 testing it for NULL. (Of course, in any sensible call of the function, it
575 won't be NULL.)
577 3. The memmove() emulation function in pcre_internal.h, which is used on
578 systems that lack both memmove() and bcopy() - that is, hardly ever -
579 was missing a "static" storage class specifier.
581 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
582 containing an extended class (one that cannot be represented by a bitmap
583 because it contains high-valued characters or Unicode property items, e.g.
584 [\pZ]). Almost always one would set UTF-8 mode when processing such a
585 pattern, but PCRE should not loop if you do not (it no longer does).
586 [Detail: two cases were found: (a) a repeated subpattern containing an
587 extended class; (b) a recursive reference to a subpattern that followed a
588 previous extended class. It wasn't skipping over the extended class
589 correctly when UTF-8 mode was not set.]
591 5. A negated single-character class was not being recognized as fixed-length
592 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
593 compile error "lookbehind assertion is not fixed length".
595 6. The RunPerlTest auxiliary script was showing an unexpected difference
596 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
597 write a Perl script that can interpret lines of an input file either as
598 byte characters or as UTF-8, which is what "perltest" was being required to
599 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
600 can't do is switch easily at run time between having the "use utf8;" pragma
601 or not. In the end, I fudged it by using the RunPerlTest script to insert
602 "use utf8;" explicitly for the UTF-8 tests.
604 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
605 the end of the subject string, contrary to the documentation and to what
606 Perl does. This was true of both matching functions. Now it matches only at
607 the start of the subject and immediately after *internal* newlines.
609 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
610 a pointer to an int instead of a pointer to an unsigned long int. This
611 caused problems on 64-bit systems.
613 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
614 instance of the 'standard' template library not being so standard".
616 10. There was no check on the number of named subpatterns nor the maximum
617 length of a subpattern name. The product of these values is used to compute
618 the size of the memory block for a compiled pattern. By supplying a very
619 long subpattern name and a large number of named subpatterns, the size
620 computation could be caused to overflow. This is now prevented by limiting
621 the length of names to 32 characters, and the number of named subpatterns
622 to 10,000.
624 11. Subpatterns that are repeated with specific counts have to be replicated in
625 the compiled pattern. The size of memory for this was computed from the
626 length of the subpattern and the repeat count. The latter is limited to
627 65535, but there was no limit on the former, meaning that integer overflow
628 could in principle occur. The compiled length of a repeated subpattern is
629 now limited to 30,000 bytes in order to prevent this.
631 12. Added the optional facility to have named substrings with the same name.
633 13. Added the ability to use a named substring as a condition, using the
634 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
635 are numbers (not recommended). Forward references are permitted.
637 14. Added forward references in named backreferences (if you see what I mean).
639 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
640 pattern could run off the end of the subject. For example, the pattern
641 "(?s)(.{1,5})"8 did this with the subject "ab".
643 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
644 PCRE_CASELESS was set when matching characters that were quantified with ?
645 or *.
647 17. A character class other than a single negated character that had a minimum
648 but no maximum quantifier - for example [ab]{6,} - was not handled
649 correctly by pce_dfa_exec(). It would match only one character.
651 18. A valid (though odd) pattern that looked like a POSIX character
652 class but used an invalid character after [ (for example [[,abc,]]) caused
653 pcre_compile() to give the error "Failed: internal error: code overflow" or
654 in some cases to crash with a glibc free() error. This could even happen if
655 the pattern terminated after [[ but there just happened to be a sequence of
656 letters, a binary zero, and a closing ] in the memory that followed.
658 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
659 over the years. Originally (before any Unicode support), just the bottom 8
660 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
661 output from "man perlunicode" includes this:
663 The regular expression compiler produces polymorphic opcodes. That
664 is, the pattern adapts to the data and automatically switches to
665 the Unicode character scheme when presented with Unicode data--or
666 instead uses a traditional byte scheme when presented with byte
667 data.
669 Sadly, a wide octal escape does not cause a switch, and in a string with
670 no other multibyte characters, these octal escapes are treated as before.
671 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
672 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
673 Unicode string.
675 I have not perpetrated such confusion in PCRE. Up till now, it took just
676 the bottom 8 bits, as in old Perl. I have now made octal escapes with
677 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
678 translate to the appropriate multibyte character.
680 29. Applied some refactoring to reduce the number of warnings from Microsoft
681 and Borland compilers. This has included removing the fudge introduced
682 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
683 a warning about an unused variable.
685 21. PCRE has not included VT (character 0x0b) in the set of whitespace
686 characters since release 4.0, because Perl (from release 5.004) does not.
687 [Or at least, is documented not to: some releases seem to be in conflict
688 with the documentation.] However, when a pattern was studied with
689 pcre_study() and all its branches started with \s, PCRE still included VT
690 as a possible starting character. Of course, this did no harm; it just
691 caused an unnecessary match attempt.
693 22. Removed a now-redundant internal flag bit that recorded the fact that case
694 dependency changed within the pattern. This was once needed for "required
695 byte" processing, but is no longer used. This recovers a now-scarce options
696 bit. Also moved the least significant internal flag bit to the most-
697 significant bit of the word, which was not previously used (hangover from
698 the days when it was an int rather than a uint) to free up another bit for
699 the future.
701 23. Added support for CRLF line endings as well as CR and LF. As well as the
702 default being selectable at build time, it can now be changed at runtime
703 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
704 specify that it is scanning data with non-default line endings.
706 24. Changed the definition of CXXLINK to make it agree with the definition of
707 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
709 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
710 recursions. This makes a big different to stack usage for some patterns.
712 26. If a subpattern containing a named recursion or subroutine reference such
713 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
714 the space required for the compiled pattern went wrong and gave too small a
715 value. Depending on the environment, this could lead to "Failed: internal
716 error: code overflow at offset 49" or "glibc detected double free or
717 corruption" errors.
719 27. Applied patches from Google (a) to support the new newline modes and (b) to
720 advance over multibyte UTF-8 characters in GlobalReplace.
722 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
723 difference for some implementation of PCRE in some Windows version.
725 29. Added some extra testing facilities to pcretest:
727 \q<number> in a data line sets the "match limit" value
728 \Q<number> in a data line sets the "match recursion limt" value
729 -S <number> sets the stack size, where <number> is in megabytes
731 The -S option isn't available for Windows.
734 Version 6.6 06-Feb-06
735 ---------------------
737 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
738 in pcreposix.h. I have copied the definition from pcre.h.
740 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
741 because pcre.h is no longer a built file.
743 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
744 not normally included in the compiled code.
747 Version 6.5 01-Feb-06
748 ---------------------
750 1. When using the partial match feature with pcre_dfa_exec(), it was not
751 anchoring the second and subsequent partial matches at the new starting
752 point. This could lead to incorrect results. For example, with the pattern
753 /1234/, partially matching against "123" and then "a4" gave a match.
755 2. Changes to pcregrep:
757 (a) All non-match returns from pcre_exec() were being treated as failures
758 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
759 error message is output. Some extra information is given for the
761 probably the only errors that are likely to be caused by users (by
762 specifying a regex that has nested indefinite repeats, for instance).
763 If there are more than 20 of these errors, pcregrep is abandoned.
765 (b) A binary zero was treated as data while matching, but terminated the
766 output line if it was written out. This has been fixed: binary zeroes
767 are now no different to any other data bytes.
769 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
770 used to set a locale for matching. The --locale=xxxx long option has
771 been added (no short equivalent) to specify a locale explicitly on the
772 pcregrep command, overriding the environment variables.
774 (d) When -B was used with -n, some line numbers in the output were one less
775 than they should have been.
777 (e) Added the -o (--only-matching) option.
779 (f) If -A or -C was used with -c (count only), some lines of context were
780 accidentally printed for the final match.
782 (g) Added the -H (--with-filename) option.
784 (h) The combination of options -rh failed to suppress file names for files
785 that were found from directory arguments.
787 (i) Added the -D (--devices) and -d (--directories) options.
789 (j) Added the -F (--fixed-strings) option.
791 (k) Allow "-" to be used as a file name for -f as well as for a data file.
793 (l) Added the --colo(u)r option.
795 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
796 is not present by default.
798 3. A nasty bug was discovered in the handling of recursive patterns, that is,
799 items such as (?R) or (?1), when the recursion could match a number of
800 alternatives. If it matched one of the alternatives, but subsequently,
801 outside the recursion, there was a failure, the code tried to back up into
802 the recursion. However, because of the way PCRE is implemented, this is not
803 possible, and the result was an incorrect result from the match.
805 In order to prevent this happening, the specification of recursion has
806 been changed so that all such subpatterns are automatically treated as
807 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
809 4. I had overlooked the fact that, in some locales, there are characters for
810 which isalpha() is true but neither isupper() nor islower() are true. In
811 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
812 and ordfeminine) are like this. This affected the treatment of \w and \W
813 when they appeared in character classes, but not when they appeared outside
814 a character class. The bit map for "word" characters is now created
815 separately from the results of isalnum() instead of just taking it from the
816 upper, lower, and digit maps. (Plus the underscore character, of course.)
818 5. The above bug also affected the handling of POSIX character classes such as
819 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
820 permanent tables. Instead, the bit maps for such a class were previously
821 created as the appropriate unions of the upper, lower, and digit bitmaps.
822 Now they are created by subtraction from the [[:word:]] class, which has
823 its own bitmap.
825 6. The [[:blank:]] character class matches horizontal, but not vertical space.
826 It is created by subtracting the vertical space characters (\x09, \x0a,
827 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
828 subtraction was done in the overall bitmap for a character class, meaning
829 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
830 be recognized. This bug has been fixed.
832 7. Patches from the folks at Google:
834 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
835 real life, but is still worth protecting against".
837 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
838 regular expressions".
840 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
841 have it.
843 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
844 "configure" and the latter not, in order to fix a problem somebody had
845 with compiling the Arg class on HP-UX.
847 (e) Improve the error-handling of the C++ wrapper a little bit.
849 (f) New tests for checking recursion limiting.
851 8. The pcre_memmove() function, which is used only if the environment does not
852 have a standard memmove() function (and is therefore rarely compiled),
853 contained two bugs: (a) use of int instead of size_t, and (b) it was not
854 returning a result (though PCRE never actually uses the result).
856 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
857 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
858 returned instead of calling malloc() with an overflowing number that would
859 most likely cause subsequent chaos.
861 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
863 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
864 with this option is matched, the nmatch and pmatch options of regexec() are
865 ignored.
867 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
868 provided in case anyone wants to the the POSIX interface with UTF-8
869 strings.
871 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
872 C++ linking (needed for some HP-UX environments).
874 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
875 (unused parameter) and in the pcre_printint() function (omitted "default"
876 switch label when the default is to do nothing).
878 15. Added some code to make it possible, when PCRE is compiled as a C++
879 library, to replace subject pointers for pcre_exec() with a smart pointer
880 class, thus making it possible to process discontinuous strings.
882 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
883 much the same function. They were added by different people who were trying
884 to make PCRE easy to compile on non-Unix systems. It has been suggested
885 that PCRE_EXPORT be abolished now that there is more automatic apparatus
886 for compiling on Windows systems. I have therefore replaced it with
887 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
888 defaults to "extern" for C or "extern C" for C++, which works fine on
889 Unix-like systems. It is now possible to override the value of PCRE_DATA_
890 SCOPE with something explicit in config.h. In addition:
892 (a) pcreposix.h still had just "extern" instead of either of these macros;
893 I have replaced it with PCRE_DATA_SCOPE.
895 (b) Functions such as _pcre_xclass(), which are internal to the library,
896 but external in the C sense, all had PCRE_EXPORT in their definitions.
897 This is apparently wrong for the Windows case, so I have removed it.
898 (It makes no difference on Unix-like systems.)
900 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
901 of recursive calls to match(). This is different to MATCH_LIMIT because
902 that limits the total number of calls to match(), not all of which increase
903 the depth of recursion. Limiting the recursion depth limits the amount of
904 stack (or heap if NO_RECURSE is set) that is used. The default can be set
905 when PCRE is compiled, and changed at run time. A patch from Google adds
906 this functionality to the C++ interface.
908 18. Changes to the handling of Unicode character properties:
910 (a) Updated the table to Unicode 4.1.0.
912 (b) Recognize characters that are not in the table as "Cn" (undefined).
914 (c) I revised the way the table is implemented to a much improved format
915 which includes recognition of ranges. It now supports the ranges that
916 are defined in UnicodeData.txt, and it also amalgamates other
917 characters into ranges. This has reduced the number of entries in the
918 table from around 16,000 to around 3,000, thus reducing its size
919 considerably. I realized I did not need to use a tree structure after
920 all - a binary chop search is just as efficient. Having reduced the
921 number of entries, I extended their size from 6 bytes to 8 bytes to
922 allow for more data.
924 (d) Added support for Unicode script names via properties such as \p{Han}.
926 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
927 matching that character.
929 20. When matching a repeated Unicode property with a minimum greater than zero,
930 (for example \pL{2,}), PCRE could look past the end of the subject if it
931 reached it while seeking the minimum number of characters. This could
932 happen only if some of the characters were more than one byte long, because
933 there is a check for at least the minimum number of bytes.
935 21. Refactored the implementation of \p and \P so as to be more general, to
936 allow for more different types of property in future. This has changed the
937 compiled form incompatibly. Anybody with saved compiled patterns that use
938 \p or \P will have to recompile them.
940 22. Added "Any" and "L&" to the supported property types.
942 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
943 but give a compile time error if the value is greater than 0xff.
945 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
946 accidentally not being installed or uninstalled.
948 25. The pcre.h file was built from pcre.h.in, but the only changes that were
949 made were to insert the current release number. This seemed silly, because
950 it made things harder for people building PCRE on systems that don't run
951 "configure". I have turned pcre.h into a distributed file, no longer built
952 by "configure", with the version identification directly included. There is
953 no longer a pcre.h.in file.
955 However, this change necessitated a change to the pcre-config script as
956 well. It is built from pcre-config.in, and one of the substitutions was the
957 release number. I have updated configure.ac so that ./configure now finds
958 the release number by grepping pcre.h.
960 26. Added the ability to run the tests under valgrind.
963 Version 6.4 05-Sep-05
964 ---------------------
966 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
967 "--" to be printed when multiple files were scanned, even when none of the
968 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
969 consider it to be a bug, and have restored the previous behaviour.
971 2. A couple of code tidies to get rid of compiler warnings.
973 3. The pcretest program used to cheat by referring to symbols in the library
974 whose names begin with _pcre_. These are internal symbols that are not
975 really supposed to be visible externally, and in some environments it is
976 possible to suppress them. The cheating is now confined to including
977 certain files from the library's source, which is a bit cleaner.
979 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
980 file's purpose clearer.
982 5. Reorganized pcre_ucp_findchar().
985 Version 6.3 15-Aug-05
986 ---------------------
988 1. The file libpcre.pc.in did not have general read permission in the tarball.
990 2. There were some problems when building without C++ support:
992 (a) If C++ support was not built, "make install" and "make test" still
993 tried to test it.
995 (b) There were problems when the value of CXX was explicitly set. Some
996 changes have been made to try to fix these, and ...
998 (c) --disable-cpp can now be used to explicitly disable C++ support.
1000 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1001 backslash in a target when C++ was disabled. This confuses some
1002 versions of "make", apparently. Using an intermediate variable solves
1003 this. (Same for CPP_LOBJ.)
1005 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1006 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1007 necessary on certain architectures.
1009 4. Added a setting of -export-symbols-regex to the link command to remove
1010 those symbols that are exported in the C sense, but actually are local
1011 within the library, and not documented. Their names all begin with
1012 "_pcre_". This is not a perfect job, because (a) we have to except some
1013 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1014 available (and never for static libraries). I have made a note to try to
1015 find a way round (a) in the future.
1018 Version 6.2 01-Aug-05
1019 ---------------------
1021 1. There was no test for integer overflow of quantifier values. A construction
1022 such as {1111111111111111} would give undefined results. What is worse, if
1023 a minimum quantifier for a parenthesized subpattern overflowed and became
1024 negative, the calculation of the memory size went wrong. This could have
1025 led to memory overwriting.
1027 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1029 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1030 operating environments where this matters.
1032 4. Applied Giuseppe Maxia's patch to add additional features for controlling
1033 PCRE options from within the C++ wrapper.
1035 5. Named capturing subpatterns were not being correctly counted when a pattern
1036 was compiled. This caused two problems: (a) If there were more than 100
1037 such subpatterns, the calculation of the memory needed for the whole
1038 compiled pattern went wrong, leading to an overflow error. (b) Numerical
1039 back references of the form \12, where the number was greater than 9, were
1040 not recognized as back references, even though there were sufficient
1041 previous subpatterns.
1043 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1044 versions of gcc, e.g. 2.95.4.
1047 Version 6.1 21-Jun-05
1048 ---------------------
1050 1. There was one reference to the variable "posix" in pcretest.c that was not
1051 surrounded by "#if !defined NOPOSIX".
1053 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1054 the cross-check on the old pcre_info() function, for the benefit of the
1055 cut-down version of PCRE that is currently imported into Exim.
1057 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1058 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1059 patterns that start (?i) but I don't think that matters. The use of (?i) is
1060 just an example; this all applies to the other options as well.
1062 4. Since libtool seems to echo the compile commands it is issuing, the output
1063 from "make" can be reduced a bit by putting "@" in front of each libtool
1064 compile command.
1066 5. Patch from the folks at Google for configure.in to be a bit more thorough
1067 in checking for a suitable C++ installation before trying to compile the
1068 C++ stuff. This should fix a reported problem when a compiler was present,
1069 but no suitable headers.
1071 6. The man pages all had just "PCRE" as their title. I have changed them to
1072 be the relevant file name. I have also arranged that these names are
1073 retained in the file doc/pcre.txt, which is a concatenation in text format
1074 of all the man pages except the little individual ones for each function.
1076 7. The NON-UNIX-USE file had not been updated for the different set of source
1077 files that come with release 6. I also added a few comments about the C++
1078 wrapper.
1081 Version 6.0 07-Jun-05
1082 ---------------------
1084 1. Some minor internal re-organization to help with my DFA experiments.
1086 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1087 didn't matter for the library itself when fully configured, but did matter
1088 when compiling without UCP support, or within Exim, where the ucp files are
1089 not imported.
1091 3. Refactoring of the library code to split up the various functions into
1092 different source modules. The addition of the new DFA matching code (see
1093 below) to a single monolithic source would have made it really too
1094 unwieldy, quite apart from causing all the code to be include in a
1095 statically linked application, when only some functions are used. This is
1096 relevant even without the DFA addition now that patterns can be compiled in
1097 one application and matched in another.
1099 The downside of splitting up is that there have to be some external
1100 functions and data tables that are used internally in different modules of
1101 the library but which are not part of the API. These have all had their
1102 names changed to start with "_pcre_" so that they are unlikely to clash
1103 with other external names.
1105 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1106 a different (DFA) algorithm. Although it is slower than the original
1107 function, it does have some advantages for certain types of matching
1108 problem.
1110 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1111 including restarting after a partial match.
1113 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1114 defined when compiling for Windows was sent to me. I have put it into the
1115 code, though I have no means of testing or verifying it.
1117 7. Added the pcre_refcount() auxiliary function.
1119 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1120 match before or at the first newline in the subject string. In pcretest,
1121 the /f option on a pattern can be used to set this.
1123 9. A repeated \w when used in UTF-8 mode with characters greater than 256
1124 would behave wrongly. This has been present in PCRE since release 4.0.
1126 10. A number of changes to the pcregrep command:
1128 (a) Refactored how -x works; insert ^(...)$ instead of setting
1129 PCRE_ANCHORED and checking the length, in preparation for adding
1130 something similar for -w.
1132 (b) Added the -w (match as a word) option.
1134 (c) Refactored the way lines are read and buffered so as to have more
1135 than one at a time available.
1137 (d) Implemented a pcregrep test script.
1139 (e) Added the -M (multiline match) option. This allows patterns to match
1140 over several lines of the subject. The buffering ensures that at least
1141 8K, or the rest of the document (whichever is the shorter) is available
1142 for matching (and similarly the previous 8K for lookbehind assertions).
1144 (f) Changed the --help output so that it now says
1146 -w, --word-regex(p)
1148 instead of two lines, one with "regex" and the other with "regexp"
1149 because that confused at least one person since the short forms are the
1150 same. (This required a bit of code, as the output is generated
1151 automatically from a table. It wasn't just a text change.)
1153 (g) -- can be used to terminate pcregrep options if the next thing isn't an
1154 option but starts with a hyphen. Could be a pattern or a path name
1155 starting with a hyphen, for instance.
1157 (h) "-" can be given as a file name to represent stdin.
1159 (i) When file names are being printed, "(standard input)" is used for
1160 the standard input, for compatibility with GNU grep. Previously
1161 "<stdin>" was used.
1163 (j) The option --label=xxx can be used to supply a name to be used for
1164 stdin when file names are being printed. There is no short form.
1166 (k) Re-factored the options decoding logic because we are going to add
1167 two more options that take data. Such options can now be given in four
1168 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1170 (l) Added the -A, -B, and -C options for requesting that lines of context
1171 around matches be printed.
1173 (m) Added the -L option to print the names of files that do not contain
1174 any matching lines, that is, the complement of -l.
1176 (n) The return code is 2 if any file cannot be opened, but pcregrep does
1177 continue to scan other files.
1179 (o) The -s option was incorrectly implemented. For compatibility with other
1180 greps, it now suppresses the error message for a non-existent or non-
1181 accessible file (but not the return code). There is a new option called
1182 -q that suppresses the output of matching lines, which was what -s was
1183 previously doing.
1185 (p) Added --include and --exclude options to specify files for inclusion
1186 and exclusion when recursing.
1188 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1189 Hopefully, it now does.
1191 12. Missing cast in pcre_study().
1193 13. Added an "uninstall" target to the makefile.
1195 14. Replaced "extern" in the function prototypes in Makefile.in with
1196 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1197 world, but is set differently for Windows.
1199 15. Added a second compiling function called pcre_compile2(). The only
1200 difference is that it has an extra argument, which is a pointer to an
1201 integer error code. When there is a compile-time failure, this is set
1202 non-zero, in addition to the error test pointer being set to point to an
1203 error message. The new argument may be NULL if no error number is required
1204 (but then you may as well call pcre_compile(), which is now just a
1205 wrapper). This facility is provided because some applications need a
1206 numeric error indication, but it has also enabled me to tidy up the way
1207 compile-time errors are handled in the POSIX wrapper.
1209 16. Added VPATH=.libs to the makefile; this should help when building with one
1210 prefix path and installing with another. (Or so I'm told by someone who
1211 knows more about this stuff than I do.)
1213 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1214 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1215 match everything, including newlines. This is not POSIX-compatible, but
1216 somebody wanted the feature. From pcretest it can be activated by using
1217 both the P and the s flags.
1219 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1221 19. libpcre.pc was being incorrectly installed as executable.
1223 20. A couple of places in pcretest check for end-of-line by looking for '\n';
1224 it now also looks for '\r' so that it will work unmodified on Windows.
1226 21. Added Google's contributed C++ wrapper to the distribution.
1228 22. Added some untidy missing memory free() calls in pcretest, to keep
1229 Electric Fence happy when testing.
1233 Version 5.0 13-Sep-04
1234 ---------------------
1236 1. Internal change: literal characters are no longer packed up into items
1237 containing multiple characters in a single byte-string. Each character
1238 is now matched using a separate opcode. However, there may be more than one
1239 byte in the character in UTF-8 mode.
1241 2. The pcre_callout_block structure has two new fields: pattern_position and
1242 next_item_length. These contain the offset in the pattern to the next match
1243 item, and its length, respectively.
1245 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1246 insertion of callouts before each pattern item. Added the /C option to
1247 pcretest to make use of this.
1249 4. On the advice of a Windows user, the lines
1251 #if defined(_WIN32) || defined(WIN32)
1252 _setmode( _fileno( stdout ), 0x8000 );
1253 #endif /* defined(_WIN32) || defined(WIN32) */
1255 have been added to the source of pcretest. This apparently does useful
1256 magic in relation to line terminators.
1258 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1259 for the benefit of those environments where the "b" makes a difference.
1261 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1262 to know about it. I have put a hack into configure.in that adds in code
1263 to set GCC=yes if CC=icc. This seems to end up at a point in the
1264 generated configure script that is early enough to affect the setting of
1265 compiler options, which is what is needed, but I have no means of testing
1266 whether it really works. (The user who reported this had patched the
1267 generated configure script, which of course I cannot do.)
1269 LATER: After change 22 below (new libtool files), the configure script
1270 seems to know about icc (and also ecc). Therefore, I have commented out
1271 this hack in configure.in.
1273 7. Added support for pkg-config (2 patches were sent in).
1275 8. Negated POSIX character classes that used a combination of internal tables
1276 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1277 [[:^ascii]]. Typically, they would match almost any characters. The other
1278 POSIX classes were not broken in this way.
1280 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1281 to find the match, as PCRE was deluded into thinking that the match had to
1282 start at the start point or following a newline. The same bug applied to
1283 patterns with negative forward assertions or any backward assertions
1284 preceding ".*" at the start, unless the pattern required a fixed first
1285 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1287 10. In UTF-8 mode, when moving forwards in the subject after a failed match
1288 starting at the last subject character, bytes beyond the end of the subject
1289 string were read.
1291 11. Renamed the variable "class" as "classbits" to make life easier for C++
1292 users. (Previously there was a macro definition, but it apparently wasn't
1293 enough.)
1295 12. Added the new field "tables" to the extra data so that tables can be passed
1296 in at exec time, or the internal tables can be re-selected. This allows
1297 a compiled regex to be saved and re-used at a later time by a different
1298 program that might have everything at different addresses.
1300 13. Modified the pcre-config script so that, when run on Solaris, it shows a
1301 -R library as well as a -L library.
1303 14. The debugging options of pcretest (-d on the command line or D on a
1304 pattern) showed incorrect output for anything following an extended class
1305 that contained multibyte characters and which was followed by a quantifier.
1307 15. Added optional support for general category Unicode character properties
1308 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1309 support. It adds about 90K to the size of the library. The meanings of the
1310 inbuilt class escapes such as \d and \s have NOT been changed.
1312 16. Updated pcredemo.c to include calls to free() to release the memory for the
1313 compiled pattern.
1315 17. The generated file chartables.c was being created in the source directory
1316 instead of in the building directory. This caused the build to fail if the
1317 source directory was different from the building directory, and was
1318 read-only.
1320 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1321 file. No doubt somebody will tell me if they don't make sense... Also added
1322 Dan Mooney's comments about building on OpenVMS.
1324 19. Added support for partial matching via the PCRE_PARTIAL option for
1325 pcre_exec() and the \P data escape in pcretest.
1327 20. Extended pcretest with 3 new pattern features:
1329 (i) A pattern option of the form ">rest-of-line" causes pcretest to
1330 write the compiled pattern to the file whose name is "rest-of-line".
1331 This is a straight binary dump of the data, with the saved pointer to
1332 the character tables forced to be NULL. The study data, if any, is
1333 written too. After writing, pcretest reads a new pattern.
1335 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1336 compiled pattern from the given file. There must not be any
1337 occurrences of "<" in the file name (pretty unlikely); if there are,
1338 pcretest will instead treat the initial "<" as a pattern delimiter.
1339 After reading in the pattern, pcretest goes on to read data lines as
1340 usual.
1342 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1343 and 16-bit fields in a compiled pattern, to simulate a pattern that
1344 was compiled on a host of opposite endianness.
1346 21. The pcre-exec() function can now cope with patterns that were compiled on
1347 hosts of opposite endianness, with this restriction:
1349 As for any compiled expression that is saved and used later, the tables
1350 pointer field cannot be preserved; the extra_data field in the arguments
1351 to pcre_exec() should be used to pass in a tables address if a value
1352 other than the default internal tables were used at compile time.
1354 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1355 now diagnosed as an error. Previously, most of the time, a negative number
1356 would have been treated as zero, but if in addition "ovector" was passed as
1357 NULL, a crash could occur.
1359 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1360 new versions from the libtool 1.5 distribution (the last one is a copy of
1361 a file called libtool.m4). This seems to have fixed the need to patch
1362 "configure" to support Darwin 1.3 (which I used to do). However, I still
1363 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1364 workstation).
1366 24. Changed the PCRE licence to be the more standard "BSD" licence.
1369 Version 4.5 01-Dec-03
1370 ---------------------
1372 1. There has been some re-arrangement of the code for the match() function so
1373 that it can be compiled in a version that does not call itself recursively.
1374 Instead, it keeps those local variables that need separate instances for
1375 each "recursion" in a frame on the heap, and gets/frees frames whenever it
1376 needs to "recurse". Keeping track of where control must go is done by means
1377 of setjmp/longjmp. The whole thing is implemented by a set of macros that
1378 hide most of the details from the main code, and operates only if
1379 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1380 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1381 operating.
1383 To make it easier for callers to provide specially tailored get/free
1384 functions for this usage, two new functions, pcre_stack_malloc, and
1385 pcre_stack_free, are used. They are always called in strict stacking order,
1386 and the size of block requested is always the same.
1388 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1389 PCRE has been compiled to use the stack or the heap for recursion. The
1390 -C option of pcretest uses this to show which version is compiled.
1392 A new data escape \S, is added to pcretest; it causes the amounts of store
1393 obtained and freed by both kinds of malloc/free at match time to be added
1394 to the output.
1396 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1397 what's available on my current Linux desktop machine.
1399 3. When matching a UTF-8 string, the test for a valid string at the start has
1400 been extended. If start_offset is not zero, PCRE now checks that it points
1401 to a byte that is the start of a UTF-8 character. If not, it returns
1402 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1403 this is necessary because there may be backward assertions in the pattern.
1404 When matching the same subject several times, it may save resources to use
1405 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1407 4. The code for checking the validity of UTF-8 strings has been tightened so
1408 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1409 containing "overlong sequences".
1411 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1412 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1413 should have been "&", but it just so happened that all the cases this let
1414 through by mistake were picked up later in the function.
1416 6. I had used a variable called "isblank" - this is a C99 function, causing
1417 some compilers to warn. To avoid this, I renamed it (as "blankclass").
1419 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1420 prompting; (b) run "./pcretest /dev/null" at the start of the test script
1421 so the version is shown; (c) stop "make test" echoing "./RunTest".
1423 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1425 9. The prototype for memmove() for systems that don't have it was using
1426 size_t, but the inclusion of the header that defines size_t was later. I've
1427 moved the #includes for the C headers earlier to avoid this.
1429 10. Added some adjustments to the code to make it easier to compiler on certain
1430 special systems:
1432 (a) Some "const" qualifiers were missing.
1433 (b) Added the macro EXPORT before all exported functions; by default this
1434 is defined to be empty.
1435 (c) Changed the dftables auxiliary program (that builds chartables.c) so
1436 that it reads its output file name as an argument instead of writing
1437 to the standard output and assuming this can be redirected.
1439 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1440 class containing characters with values greater than 255, PCRE compilation
1441 went into a loop.
1443 12. A recursive reference to a subpattern that was within another subpattern
1444 that had a minimum quantifier of zero caused PCRE to crash. For example,
1445 (x(y(?2))z)? provoked this bug with a subject that got as far as the
1446 recursion. If the recursively-called subpattern itself had a zero repeat,
1447 that was OK.
1449 13. In pcretest, the buffer for reading a data line was set at 30K, but the
1450 buffer into which it was copied (for escape processing) was still set at
1451 1024, so long lines caused crashes.
1453 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1454 "internal error: code overflow...". This applied to any character class
1455 that was followed by a possessive quantifier.
1457 15. Modified the Makefile to add libpcre.la as a prerequisite for
1458 libpcreposix.la because I was told this is needed for a parallel build to
1459 work.
1461 16. If a pattern that contained .* following optional items at the start was
1462 studied, the wrong optimizing data was generated, leading to matching
1463 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1464 matching string must start with a or b or c. The correct conclusion for
1465 this pattern is that a match can start with any character.
1468 Version 4.4 13-Aug-03
1469 ---------------------
1471 1. In UTF-8 mode, a character class containing characters with values between
1472 127 and 255 was not handled correctly if the compiled pattern was studied.
1473 In fixing this, I have also improved the studying algorithm for such
1474 classes (slightly).
1476 2. Three internal functions had redundant arguments passed to them. Removal
1477 might give a very teeny performance improvement.
1479 3. Documentation bug: the value of the capture_top field in a callout is *one
1480 more than* the number of the hightest numbered captured substring.
1482 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1483 in incorrectly linking with a previously installed version. They now link
1484 explicitly with libpcre.la.
1486 5. configure.in no longer needs to recognize Cygwin specially.
1488 6. A problem in pcre.in for Windows platforms is fixed.
1490 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1491 pcretest, it used to include the size of the study block as part of its
1492 output. Unfortunately, the structure contains a field that has a different
1493 size on different hardware architectures. This meant that the tests that
1494 showed this size failed. As the block is currently always of a fixed size,
1495 this information isn't actually particularly useful in pcretest output, so
1496 I have just removed it.
1498 8. Three pre-processor statements accidentally did not start in column 1.
1499 Sadly, there are *still* compilers around that complain, even though
1500 standard C has not required this for well over a decade. Sigh.
1502 9. In pcretest, the code for checking callouts passed small integers in the
1503 callout_data field, which is a void * field. However, some picky compilers
1504 complained about the casts involved for this on 64-bit systems. Now
1505 pcretest passes the address of the small integer instead, which should get
1506 rid of the warnings.
1508 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1509 both compile and run time, and gives an error if an invalid UTF-8 sequence
1510 is found. There is a option for disabling this check in cases where the
1511 string is known to be correct and/or the maximum performance is wanted.
1513 11. In response to a bug report, I changed one line in Makefile.in from
1515 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1516 to
1517 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1519 to look similar to other lines, but I have no way of telling whether this
1520 is the right thing to do, as I do not use Windows. No doubt I'll get told
1521 if it's wrong...
1524 Version 4.3 21-May-03
1525 ---------------------
1527 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1528 Makefile.
1530 2. Some refactoring to improve the quality of the code:
1532 (i) The utf8_table... variables are now declared "const".
1534 (ii) The code for \cx, which used the "case flipping" table to upper case
1535 lower case letters, now just substracts 32. This is ASCII-specific,
1536 but the whole concept of \cx is ASCII-specific, so it seems
1537 reasonable.
1539 (iii) PCRE was using its character types table to recognize decimal and
1540 hexadecimal digits in the pattern. This is silly, because it handles
1541 only 0-9, a-f, and A-F, but the character types table is locale-
1542 specific, which means strange things might happen. A private
1543 table is now used for this - though it costs 256 bytes, a table is
1544 much faster than multiple explicit tests. Of course, the standard
1545 character types table is still used for matching digits in subject
1546 strings against \d.
1548 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1549 ending in _t are). So I've renamed it as ESC_tee.
1551 3. The first argument for regexec() in the POSIX wrapper should have been
1552 defined as "const".
1554 4. Changed pcretest to use malloc() for its buffers so that they can be
1555 Electric Fenced for debugging.
1557 5. There were several places in the code where, in UTF-8 mode, PCRE would try
1558 to read one or more bytes before the start of the subject string. Often this
1559 had no effect on PCRE's behaviour, but in some circumstances it could
1560 provoke a segmentation fault.
1562 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1563 to try to read one or more bytes before the start of the subject string.
1565 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1566 UTF-8 support could misbehave in various ways if the subject string
1567 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1568 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1569 back over UTF-8 characters.)
1572 Version 4.2 14-Apr-03
1573 ---------------------
1575 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1577 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1578 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1579 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1580 [WIN_PREFIX]: new variable, "cyg" for Cygwin
1581 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1583 Note: automatic setting of the BUILD variables is not yet working
1584 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1585 compile-time but not at link-time
1586 [LINK]: use for linking executables only
1587 make different versions for Windows and non-Windows
1588 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1589 libraries
1590 [LINK_FOR_BUILD]: new variable
1591 [OBJEXT]: use throughout
1592 [EXEEXT]: use throughout
1593 <winshared>: new target
1594 <wininstall>: new target
1595 <dftables.o>: use native compiler
1596 <dftables>: use native linker
1597 <install>: handle Windows platform correctly
1598 <clean>: ditto
1599 <check>: ditto
1600 copy DLL to top builddir before testing
1602 As part of these changes, -no-undefined was removed again. This was reported
1603 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1604 in any case.
1606 3. Some tidies to get rid of compiler warnings:
1608 . In the match_data structure, match_limit was an unsigned long int, whereas
1609 match_call_count was an int. I've made them both unsigned long ints.
1611 . In pcretest the fact that a const uschar * doesn't automatically cast to
1612 a void * provoked a warning.
1614 . Turning on some more compiler warnings threw up some "shadow" variables
1615 and a few more missing casts.
1617 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1618 option, a class that contained a single character with a value between 128
1619 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1621 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1622 option, a class that contained several characters, but with at least one
1623 whose value was between 128 and 255 caused PCRE to crash.
1626 Version 4.1 12-Mar-03
1627 ---------------------
1629 1. Compiling with gcc -pedantic found a couple of places where casts were
1630 needed, and a string in dftables.c that was longer than standard compilers are
1631 required to support.
1633 2. Compiling with Sun's compiler found a few more places where the code could
1634 be tidied up in order to avoid warnings.
1636 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1637 first of these names is deprecated in the latest Autoconf in favour of the name
1638 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1639 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1640 analogy I have changed it to CFLAGS_FOR_BUILD.
1642 4. Added -no-undefined to the linking command in the Makefile, because this is
1643 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1644 linking step for the pcreposix library.
1646 5. PCRE was failing to diagnose the case of two named groups with the same
1647 name.
1649 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1650 literal character that is needed in the subject for a match, and scans along to
1651 ensure that it is present before embarking on the full matching process. This
1652 saves time in cases of nested unlimited repeats that are never going to match.
1653 Problem: the scan can take a lot of time if the subject is very long (e.g.
1654 megabytes), thus penalizing straightforward matches. It is now done only if the
1655 amount of subject to be scanned is less than 1000 bytes.
1657 7. A lesser problem with the same optimization is that it was recording the
1658 first character of an anchored pattern as "needed", thus provoking a search
1659 right along the subject, even when the first match of the pattern was going to
1660 fail. The "needed" character is now not set for anchored patterns, unless it
1661 follows something in the pattern that is of non-fixed length. Thus, it still
1662 fulfils its original purpose of finding quick non-matches in cases of nested
1663 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1666 Version 4.0 17-Feb-03
1667 ---------------------
1669 1. If a comment in an extended regex that started immediately after a meta-item
1670 extended to the end of string, PCRE compiled incorrect data. This could lead to
1671 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1673 2. Moved to autoconf 2.53 and libtool 1.4.2.
1675 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1676 the special perltest8 script is no longer needed - all the tests can be run
1677 from a single perltest script.
1679 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1680 by \s. It has now been removed in PCRE. This means it isn't recognized as
1681 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1682 class [:space:] *does* include VT, thereby creating a mess.
1684 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1685 space and tab.
1687 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1688 its new features into the main test script, reducing the number of scripts.
1690 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1691 were backward compatible, and made the (?i) apply to the whole pattern, as if
1692 /i were given. Now it behaves more logically, and applies the option setting
1693 only to what follows. PCRE has been changed to follow suit. However, if it
1694 finds options settings right at the start of the pattern, it extracts them into
1695 the global options, as before. Thus, they show up in the info data.
1697 8. Added support for the \Q...\E escape sequence. Characters in between are
1698 treated as literals. This is slightly different from Perl in that $ and @ are
1699 also handled as literals inside the quotes. In Perl, they will cause variable
1700 interpolation. Note the following examples:
1702 Pattern PCRE matches Perl matches
1704 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
1705 \Qabc\$xyz\E abc\$xyz abc\$xyz
1706 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
1708 For compatibility with Perl, \Q...\E sequences are recognized inside character
1709 classes as well as outside them.
1711 9. Re-organized 3 code statements in pcretest to avoid "overflow in
1712 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1713 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1714 signed/unsigned warnings.
1716 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1717 option for pcretest, so I've replaced it by a simple function that does just
1718 that job.
1720 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1721 "pcregrep -".
1723 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1724 Java package. This provides some syntactic sugar for simple cases of what my
1725 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1726 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1727 item, you can use this simplified notation. Note that only makes sense with
1728 greedy quantifiers. Consequently, the use of the possessive quantifier forces
1729 greediness, whatever the setting of the PCRE_UNGREEDY option.
1731 13. A change of greediness default within a pattern was not taking effect at
1732 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1733 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1734 was abstracted outside.
1736 14. PCRE now supports the \G assertion. It is true when the current matching
1737 position is at the start point of the match. This differs from \A when the
1738 starting offset is non-zero. Used with the /g option of pcretest (or similar
1739 code), it works in the same way as it does for Perl's /g option. If all
1740 alternatives of a regex begin with \G, the expression is anchored to the start
1741 match position, and the "anchored" flag is set in the compiled expression.
1743 15. Some bugs concerning the handling of certain option changes within patterns
1744 have been fixed. These applied to options other than (?ims). For example,
1745 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1746 been the other way round. Some of this was related to change 7 above.
1748 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1749 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1750 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1751 POSIX classes only within a class (e.g. /[[:alpha:]]/).
1753 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1754 mode. Unlike ".", it always matches newline, whatever the setting of
1755 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1756 assertions. Perl allows it, but it doesn't (in general) work because it can't
1757 calculate the length of the lookbehind. At least, that's the case for Perl
1758 5.8.0 - I've been told they are going to document that it doesn't work in
1759 future.
1761 18. Added an error diagnosis for escapes that PCRE does not support: these are
1762 \L, \l, \N, \P, \p, \U, \u, and \X.
1764 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1765 reading past the end of the pattern in cases such as /[abcd/.
1767 20. PCRE was getting more memory than necessary for patterns with classes that
1768 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1770 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1771 compiling PCRE for use with Virtual Pascal.
1773 22. Small fix to the Makefile to make it work properly if the build is done
1774 outside the source tree.
1776 23. Added a new extension: a condition to go with recursion. If a conditional
1777 subpattern starts with (?(R) the "true" branch is used if recursion has
1778 happened, whereas the "false" branch is used only at the top level.
1780 24. When there was a very long string of literal characters (over 255 bytes
1781 without UTF support, over 250 bytes with UTF support), the computation of how
1782 much memory was required could be incorrect, leading to segfaults or other
1783 strange effects.
1785 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1786 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1787 there was a subsequent back reference to those brackets. This meant that, for
1788 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1789 possible to check for precisely this case. All we can do is abandon the
1790 optimization if .* occurs inside capturing brackets when there are any back
1791 references whatsoever. (See below for a better fix that came later.)
1793 26. The handling of the optimization for finding the first character of a
1794 non-anchored pattern, and for finding a character that is required later in the
1795 match were failing in some cases. This didn't break the matching; it just
1796 failed to optimize when it could. The way this is done has been re-implemented.
1798 27. Fixed typo in error message for invalid (?R item (it said "(?p").
1800 28. Added a new feature that provides some of the functionality that Perl
1801 provides with (?{...}). The facility is termed a "callout". The way it is done
1802 in PCRE is for the caller to provide an optional function, by setting
1803 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1804 global variable. By default it is unset, which disables all calling out. To get
1805 the function called, the regex must include (?C) at appropriate points. This
1806 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1807 This provides a means of identifying different callout points. When PCRE
1808 reaches such a point in the regex, if pcre_callout has been set, the external
1809 function is called. It is provided with data in a structure called
1810 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1811 matching continues; if it returns a non-zero value, the match at the current
1812 point fails. However, backtracking will occur if possible. [This was changed
1813 later and other features added - see item 49 below.]
1815 29. pcretest is upgraded to test the callout functionality. It provides a
1816 callout function that displays information. By default, it shows the start of
1817 the match and the current position in the text. There are some new data escapes
1818 to vary what happens:
1820 \C+ in addition, show current contents of captured substrings
1821 \C- do not supply a callout function
1822 \C!n return 1 when callout number n is reached
1823 \C!n!m return 1 when callout number n is reached for the mth time
1825 30. If pcregrep was called with the -l option and just a single file name, it
1826 output "<stdin>" if a match was found, instead of the file name.
1828 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1829 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1830 pcre_exec(). This saves a malloc/free per call. The default value of
1831 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1832 when configuring.
1834 32. The default maximum size of a compiled pattern is 64K. There have been a
1835 few cases of people hitting this limit. The code now uses macros to handle the
1836 storing of links as offsets within the compiled pattern. It defaults to 2-byte
1837 links, but this can be changed to 3 or 4 bytes by --with-link-size when
1838 configuring. Tests 2 and 5 work only with 2-byte links because they output
1839 debugging information about compiled patterns.
1841 33. Internal code re-arrangements:
1843 (a) Moved the debugging function for printing out a compiled regex into
1844 its own source file (printint.c) and used #include to pull it into
1845 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1846 separate copies.
1848 (b) Defined the list of op-code names for debugging as a macro in
1849 internal.h so that it is next to the definition of the opcodes.
1851 (c) Defined a table of op-code lengths for simpler skipping along compiled
1852 code. This is again a macro in internal.h so that it is next to the
1853 definition of the opcodes.
1855 34. Added support for recursive calls to individual subpatterns, along the
1856 lines of Robin Houston's patch (but implemented somewhat differently).
1858 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1859 allow it to read and process whole directories in Win32. This code was
1860 contributed by Lionel Fourquaux; it has not been tested by me.
1862 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1863 used to name a group. Names consist of alphanumerics and underscores, and must
1864 be unique. Back references use the syntax (?P=name) and recursive calls use
1865 (?P>name) which is a PCRE extension to the Python extension. Groups still have
1866 numbers. The function pcre_fullinfo() can be used after compilation to extract
1867 a name/number map. There are three relevant calls:
1869 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
1870 PCRE_INFO_NAMECOUNT yields the number of entries
1871 PCRE_INFO_NAMETABLE yields a pointer to the map.
1873 The map is a vector of fixed-size entries. The size of each entry depends on
1874 the length of the longest name used. The first two bytes of each entry are the
1875 group number, most significant byte first. There follows the corresponding
1876 name, zero terminated. The names are in alphabetical order.
1878 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1879 case instead of 255. Making it the same both with and without UTF-8 support
1880 means that the same test output works with both.
1882 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1883 calling malloc() with a zero argument.
1885 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1886 optimization. I've improved things by keeping a bitmap of backreferences with
1887 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1888 fact referenced, the optimization can be applied. It is unlikely that a
1889 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1890 the match to follow \n) will appear inside brackets with a number greater than
1891 31, but if it does, any back reference > 31 suppresses the optimization.
1893 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1894 of disabling numbered capturing parentheses. Any opening parenthesis that is
1895 not followed by ? behaves as if it were followed by ?: but named parentheses
1896 can still be used for capturing (and they will acquire numbers in the usual
1897 way).
1899 41. Redesigned the return codes from the match() function into yes/no/error so
1900 that errors can be passed back from deep inside the nested calls. A malloc
1901 failure while inside a recursive subpattern call now causes the
1902 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
1904 42. It is now possible to set a limit on the number of times the match()
1905 function is called in a call to pcre_exec(). This facility makes it possible to
1906 limit the amount of recursion and backtracking, though not in a directly
1907 obvious way, because the match() function is used in a number of different
1908 circumstances. The count starts from zero for each position in the subject
1909 string (for non-anchored patterns). The default limit is, for compatibility, a
1910 large number, namely 10 000 000. You can change this in two ways:
1912 (a) When configuring PCRE before making, you can use --with-match-limit=n
1913 to set a default value for the compiled library.
1915 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
1916 a different value is set. See 45 below.
1918 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1920 43. Added a new function pcre_config(int, void *) to enable run-time extraction
1921 of things that can be changed at compile time. The first argument specifies
1922 what is wanted and the second points to where the information is to be placed.
1923 The current list of available information is:
1927 The output is an integer that is set to one if UTF-8 support is available;
1928 otherwise it is set to zero.
1932 The output is an integer that it set to the value of the code that is used for
1933 newline. It is either LF (10) or CR (13).
1937 The output is an integer that contains the number of bytes used for internal
1938 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
1942 The output is an integer that contains the threshold above which the POSIX
1943 interface uses malloc() for output vectors. See item 31 above.
1947 The output is an unsigned integer that contains the default limit of the number
1948 of match() calls in a pcre_exec() execution. See 42 above.
1950 44. pcretest has been upgraded by the addition of the -C option. This causes it
1951 to extract all the available output from the new pcre_config() function, and to
1952 output it. The program then exits immediately.
1954 45. A need has arisen to pass over additional data with calls to pcre_exec() in
1955 order to support additional features. One way would have been to define
1956 pcre_exec2() (for example) with extra arguments, but this would not have been
1957 extensible, and would also have required all calls to the original function to
1958 be mapped to the new one. Instead, I have chosen to extend the mechanism that
1959 is used for passing in "extra" data from pcre_study().
1961 The pcre_extra structure is now exposed and defined in pcre.h. It currently
1962 contains the following fields:
1964 flags a bitmap indicating which of the following fields are set
1965 study_data opaque data from pcre_study()
1966 match_limit a way of specifying a limit on match() calls for a specific
1967 call to pcre_exec()
1968 callout_data data for callouts (see 49 below)
1970 The flag bits are also defined in pcre.h, and are
1976 The pcre_study() function now returns one of these new pcre_extra blocks, with
1977 the actual study data pointed to by the study_data field, and the
1978 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
1979 before. That is, this change is entirely upwards-compatible and requires no
1980 change to existing code.
1982 If you want to pass in additional data to pcre_exec(), you can either place it
1983 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
1984 block.
1986 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
1987 data string contains the escape sequence \M, pcretest calls pcre_exec() several
1988 times with different match limits, until it finds the minimum value needed for
1989 pcre_exec() to complete. The value is then output. This can be instructive; for
1990 most simple matches the number is quite small, but for pathological cases it
1991 gets very large very quickly.
1993 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
1994 returns the size of the data block pointed to by the study_data field in a
1995 pcre_extra block, that is, the value that was passed as the argument to
1996 pcre_malloc() when PCRE was getting memory in which to place the information
1997 created by pcre_study(). The fourth argument should point to a size_t variable.
1998 pcretest has been extended so that this information is shown after a successful
1999 pcre_study() call when information about the compiled regex is being displayed.
2001 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2002 because what follows is always an absolute path. (Later: it turns out that this
2003 is more than cosmetic for MinGW, because it doesn't like empty path
2004 components.)
2006 49. Some changes have been made to the callout feature (see 28 above):
2008 (i) A callout function now has three choices for what it returns:
2010 0 => success, carry on matching
2011 > 0 => failure at this point, but backtrack if possible
2012 < 0 => serious error, return this value from pcre_exec()
2014 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2015 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2016 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2017 use by callout functions. It will never be used by PCRE itself.
2019 (ii) The pcre_extra structure (see 45 above) has a void * field called
2020 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2021 pcre_callout_block structure has a field of the same name. The contents of
2022 the field passed in the pcre_extra structure are passed to the callout
2023 function in the corresponding field in the callout block. This makes it
2024 easier to use the same callout-containing regex from multiple threads. For
2025 testing, the pcretest program has a new data escape
2027 \C*n pass the number n (may be negative) as callout_data
2029 If the callout function in pcretest receives a non-zero value as
2030 callout_data, it returns that value.
2032 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2033 there were some redundant $(CFLAGS) in commands that are now specified as
2034 $(LINK), which already includes $(CFLAGS).
2036 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2037 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2038 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2039 one-byte characters throughout. Note that case-insensitive matching applies
2040 only to characters whose values are less than 256. PCRE doesn't support the
2041 notion of cases for higher-valued characters.
2043 (i) A character class whose characters are all within 0-255 is handled as
2044 a bit map, and the map is inverted for negative classes. Previously, a
2045 character > 255 always failed to match such a class; however it should
2046 match if the class was a negative one (e.g. [^ab]). This has been fixed.
2048 (ii) A negated character class with a single character < 255 is coded as
2049 "not this character" (OP_NOT). This wasn't working properly when the test
2050 character was multibyte, either singly or repeated.
2052 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2053 mode, for example: \x{100}{2,3}.
2055 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2056 singly or repeated) now correctly test multibyte characters. However,
2057 PCRE doesn't recognize any characters with values greater than 255 as
2058 digits, spaces, or word characters. Such characters always match \D, \S,
2059 and \W, and never match \d, \s, or \w.
2061 (v) Classes may now contain characters and character ranges with values
2062 greater than 255. For example: [ab\x{100}-\x{400}].
2064 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
2065 PCRE in UTF-8 mode.
2067 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2068 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2069 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2070 value.)
2072 53. The single man page has become too large. I have therefore split it up into
2073 a number of separate man pages. These also give rise to individual HTML pages;
2074 these are now put in a separate directory, and there is an index.html page that
2075 lists them all. Some hyperlinking between the pages has been installed.
2077 54. Added convenience functions for handling named capturing parentheses.
2079 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2080 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2081 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2082 are faulted.
2084 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2085 calling configure. These values are used when compiling the dftables.c program
2086 which is run to generate the source of the default character tables. They
2087 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2088 you will need to set these values.
2090 57. Updated the building process for Windows DLL, as provided by Fred Cox.
2093 Version 3.9 02-Jan-02
2094 ---------------------
2096 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2098 2. If --disable-static was given, the building process failed when trying to
2099 build pcretest and pcregrep. (For some reason it was using libtool to compile
2100 them, which is not right, as they aren't part of the library.)
2103 Version 3.8 18-Dec-01
2104 ---------------------
2106 1. The experimental UTF-8 code was completely screwed up. It was packing the
2107 bytes in the wrong order. How dumb can you get?
2110 Version 3.7 29-Oct-01
2111 ---------------------
2113 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2114 This caused pcretest, when used on the test data, to segfault. Unfortunately,
2115 this didn't happen under Solaris 8, where I normally test things.
2117 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2118 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2119 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2120 this fix an hour or so after the initial 3.7 release.)
2123 Version 3.6 23-Oct-01
2124 ---------------------
2126 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2127 offsets passed as NULL with zero offset count.
2129 2. The config.guess and config.sub files had not been updated when I moved to
2130 the latest autoconf.
2133 Version 3.5 15-Aug-01
2134 ---------------------
2136 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2137 had been forgotten.
2139 2. By using declared but undefined structures, we can avoid using "void"
2140 definitions in pcre.h while keeping the internal definitions of the structures
2141 private.
2143 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2144 user point of view, this means that both static and shared libraries are built
2145 by default, but this can be individually controlled. More of the work of
2146 handling this static/shared cases is now inside libtool instead of PCRE's make
2147 file.
2149 4. The pcretest utility is now installed along with pcregrep because it is
2150 useful for users (to test regexs) and by doing this, it automatically gets
2151 relinked by libtool. The documentation has been turned into a man page, so
2152 there are now .1, .txt, and .html versions in /doc.
2154 5. Upgrades to pcregrep:
2155 (i) Added long-form option names like gnu grep.
2156 (ii) Added --help to list all options with an explanatory phrase.
2157 (iii) Added -r, --recursive to recurse into sub-directories.
2158 (iv) Added -f, --file to read patterns from a file.
2160 6. pcre_exec() was referring to its "code" argument before testing that
2161 argument for NULL (and giving an error if it was NULL).
2163 7. Upgraded Makefile.in to allow for compiling in a different directory from
2164 the source directory.
2166 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2167 options bits, the pointer it was passed was to an int instead of to an unsigned
2168 long int. This mattered only on 64-bit systems.
2170 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2171 generated) instead of pcre.in, which it its source. Also made the same change
2172 in several of the .c files.
2174 10. A new release of gcc defines printf() as a macro, which broke pcretest
2175 because it had an ifdef in the middle of a string argument for printf(). Fixed
2176 by using separate calls to printf().
2178 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2179 script, to force use of CR or LF instead of \n in the source. On non-Unix
2180 systems, the value can be set in config.h.
2182 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2183 absolute limit. Changed the text of the error message to make this clear, and
2184 likewise updated the man page.
2186 13. The limit of 99 on the number of capturing subpatterns has been removed.
2187 The new limit is 65535, which I hope will not be a "real" limit.
2190 Version 3.4 22-Aug-00
2191 ---------------------
2193 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2195 2. Diagnose condition (?(0) as an error instead of crashing on matching.
2198 Version 3.3 01-Aug-00
2199 ---------------------
2201 1. If an octal character was given, but the value was greater than \377, it
2202 was not getting masked to the least significant bits, as documented. This could
2203 lead to crashes in some systems.
2205 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2206 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2208 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2209 These just pass their arguments on to (pcre_free)(), but they are provided
2210 because some uses of PCRE bind it to non-C systems that can call its functions,
2211 but cannot call free() or pcre_free() directly.
2213 4. Add "make test" as a synonym for "make check". Corrected some comments in
2214 the Makefile.
2216 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2217 Makefile.
2219 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2220 command called pgrep for grepping around the active processes.
2222 7. Added the beginnings of support for UTF-8 character strings.
2224 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2225 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2226 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2227 out for the ar command.)
2230 Version 3.2 12-May-00
2231 ---------------------
2233 This is purely a bug fixing release.
2235 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2236 of ZA. This was just one example of several cases that could provoke this bug,
2237 which was introduced by change 9 of version 2.00. The code for breaking
2238 infinite loops after an iteration that matches an empty string was't working
2239 correctly.
2241 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2242 when matched against abbab (for example). After matching an empty string, it
2243 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2244 caused it to match further down the string than it should.
2246 3. The code contained an inclusion of sys/types.h. It isn't clear why this
2247 was there because it doesn't seem to be needed, and it causes trouble on some
2248 systems, as it is not a Standard C header. It has been removed.
2250 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2251 were reported on the Macintosh. The changes were from
2253 while ((c = *(++ptr)) != 0 && c != '\n');
2254 to
2255 while ((c = *(++ptr)) != 0 && c != '\n') ;
2257 Totally extraordinary, but if that's what it takes...
2259 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2260 available. Added HAVE_BCOPY and an autoconf test for it; if neither
2261 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2262 assumes the way PCRE uses memmove() (always moving upwards).
2264 6. PCRE is being used in one environment where strchr() is not available. There
2265 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2266 faster code anyway.
2269 Version 3.1 09-Feb-00
2270 ---------------------
2272 The only change in this release is the fixing of some bugs in Makefile.in for
2273 the "install" target:
2275 (1) It was failing to install pcreposix.h.
2277 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2280 Version 3.0 01-Feb-00
2281 ---------------------
2283 1. Add support for the /+ modifier to perltest (to output $` like it does in
2284 pcretest).
2286 2. Add support for the /g modifier to perltest.
2288 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2289 matches null strings.
2291 4. Fix perltest so that it doesn't do unwanted things when fed an empty
2292 pattern. Perl treats empty patterns specially - it reuses the most recent
2293 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2294 effect.
2296 5. The POSIX interface was broken in that it was just handing over the POSIX
2297 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2298 required a bigger vector, with some working space on the end. This means that
2299 the POSIX wrapper now has to get and free some memory, and copy the results.
2301 6. Added some simple autoconf support, placing the test data and the
2302 documentation in separate directories, re-organizing some of the
2303 information files, and making it build pcre-config (a GNU standard). Also added
2304 libtool support for building PCRE as a shared library, which is now the
2305 default.
2307 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2308 09 are not valid octal constants. Single digits will be used for minor values
2309 less than 10.
2311 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2312 existing programs that set these in the POSIX interface can use PCRE without
2313 modification.
2315 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2316 return all that pcre_info() returns, plus additional data. The pcre_info()
2317 function is retained for compatibility, but is considered to be obsolete.
2319 10. Added experimental recursion feature (?R) to handle one common case that
2320 Perl 5.6 will be able to do with (?p{...}).
2322 11. Added support for POSIX character classes like [:alpha:], which Perl is
2323 adopting.
2326 Version 2.08 31-Aug-99
2327 ----------------------
2329 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2330 trying to match at the startoffset position, but instead was moving forward to
2331 the next newline as if a previous match had failed.
2333 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2334 and could get into a loop if a null string was matched other than at the start
2335 of the subject.
2337 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2338 be distinguished at compile time, and for completeness also added PCRE_DATE.
2340 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2341 in GnuWin32 environments.
2344 Version 2.07 29-Jul-99
2345 ----------------------
2347 1. The documentation is now supplied in plain text form and HTML as well as in
2348 the form of man page sources.
2350 2. C++ compilers don't like assigning (void *) values to other pointer types.
2351 In particular this affects malloc(). Although there is no problem in Standard
2352 C, I've put in casts to keep C++ compilers happy.
2354 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2355 should be (const char *).
2357 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2358 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2359 However, I haven't made this a standard facility. The documentation doesn't
2360 mention it, and the Makefile doesn't support it.
2362 5. The Makefile now contains an "install" target, with editable destinations at
2363 the top of the file. The pcretest program is not installed.
2365 6. pgrep -V now gives the PCRE version number and date.
2367 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2368 causing the entire string to be ignored, instead of just the last character.
2370 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2371 non-matching string, it can take a very, very long time, even for strings of
2372 quite modest length, because of the nested recursion. PCRE now does better in
2373 some of these cases. It does this by remembering the last required literal
2374 character in the pattern, and pre-searching the subject to ensure it is present
2375 before running the real match. In other words, it applies a heuristic to detect
2376 some types of certain failure quickly, and in the above example, if presented
2377 with a string that has no trailing " it gives "no match" very quickly.
2379 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2380 other alternatives are tried instead.
2383 Version 2.06 09-Jun-99
2384 ----------------------
2386 1. Change pcretest's output for amount of store used to show just the code
2387 space, because the remainder (the data block) varies in size between 32-bit and
2388 64-bit systems.
2390 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2391 start matching at. This allows lookbehinds to work when searching for multiple
2392 occurrences in a string.
2394 3. Added additional options to pcretest for testing multiple occurrences:
2396 /+ outputs the rest of the string that follows a match
2397 /g loops for multiple occurrences, using the new startoffset argument
2398 /G loops for multiple occurrences by passing an incremented pointer
2400 4. PCRE wasn't doing the "first character" optimization for patterns starting
2401 with \b or \B, though it was doing it for other lookbehind assertions. That is,
2402 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2403 the letter 'x'. On long subject strings, this gives a significant speed-up.
2406 Version 2.05 21-Apr-99
2407 ----------------------
2409 1. Changed the type of magic_number from int to long int so that it works
2410 properly on 16-bit systems.
2412 2. Fixed a bug which caused patterns starting with .* not to work correctly
2413 when the subject string contained newline characters. PCRE was assuming
2414 anchoring for such patterns in all cases, which is not correct because .* will
2415 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2416 DOTALL is set at top level; otherwise it knows that patterns starting with .*
2417 must be retried after every newline in the subject.
2420 Version 2.04 18-Feb-99
2421 ----------------------
2423 1. For parenthesized subpatterns with repeats whose minimum was zero, the
2424 computation of the store needed to hold the pattern was incorrect (too large).
2425 If such patterns were nested a few deep, this could multiply and become a real
2426 problem.
2428 2. Added /M option to pcretest to show the memory requirement of a specific
2429 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2431 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2432 compiled in such a way that the backtracking after subsequent failure was
2433 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2434 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2437 Version 2.03 02-Feb-99
2438 ----------------------
2440 1. Fixed typo and small mistake in man page.
2442 2. Added 4th condition (GPL supersedes if conflict) and created separate
2443 LICENCE file containing the conditions.
2445 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2446 Perl, that is the internal \ allows the delimiter to be included in the
2447 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2448 the final delimiter, add \ to the end of the pattern (to test the error).
2450 4. Added the convenience functions for extracting substrings after a successful
2451 match. Updated pcretest to make it able to test these functions.
2454 Version 2.02 14-Jan-99
2455 ----------------------
2457 1. Initialized the working variables associated with each extraction so that
2458 their saving and restoring doesn't refer to uninitialized store.
2460 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2461 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2462 fix the problem.
2464 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2465 calls, and wasn't printing the correct value for compiling calls. Increased the
2466 default value of LOOPREPEAT, and the number of significant figures in the
2467 times.
2469 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2471 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2472 a building problem on Windows NT with a FAT file system.
2475 Version 2.01 21-Oct-98
2476 ----------------------
2478 1. Changed the API for pcre_compile() to allow for the provision of a pointer
2479 to character tables built by pcre_maketables() in the current locale. If NULL
2480 is passed, the default tables are used.
2483 Version 2.00 24-Sep-98
2484 ----------------------
2486 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2487 it any more.
2489 2. Allow quantification of (?>) groups, and make it work correctly.
2491 3. The first character computation wasn't working for (?>) groups.
2493 4. Correct the implementation of \Z (it is permitted to match on the \n at the
2494 end of the subject) and add 5.005's \z, which really does match only at the
2495 very end of the subject.
2497 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2499 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2500 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2501 localized options. All options to pcre_study() were also removed.
2503 7. Add other new features from 5.005:
2505 $(?<= positive lookbehind
2506 $(?<! negative lookbehind
2507 (?imsx-imsx) added the unsetting capability
2508 such a setting is global if at outer level; local otherwise
2509 (?imsx-imsx:) non-capturing groups with option setting
2510 (?(cond)re|re) conditional pattern matching
2512 A backreference to itself in a repeated group matches the previous
2513 captured string.
2515 8. General tidying up of studying (both automatic and via "study")
2516 consequential on the addition of new assertions.
2518 9. As in 5.005, unlimited repeated groups that could match an empty substring
2519 are no longer faulted at compile time. Instead, the loop is forcibly broken at
2520 runtime if any iteration does actually match an empty substring.
2522 10. Include the RunTest script in the distribution.
2524 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2525 discrepancies, some of which were old and were also with respect to 5.004. They
2526 have now been fixed.
2529 Version 1.09 28-Apr-98
2530 ----------------------
2532 1. A negated single character class followed by a quantifier with a minimum
2533 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
2534 program crashes, or just wrong answers. This did not apply to negated classes
2535 containing more than one character, or to minima other than one.
2538 Version 1.08 27-Mar-98
2539 ----------------------
2541 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
2543 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
2544 latter must appear before anything that relies on it in the pattern.
2547 Version 1.07 16-Feb-98
2548 ----------------------
2550 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
2551 repeat of a potentially empty string).
2554 Version 1.06 23-Jan-98
2555 ----------------------
2557 1. Added Markus Oberhumer's little patches for C++.
2559 2. Literal strings longer than 255 characters were broken.
2562 Version 1.05 23-Dec-97
2563 ----------------------
2565 1. Negated character classes containing more than one character were failing if
2566 PCRE_CASELESS was set at run time.
2569 Version 1.04 19-Dec-97
2570 ----------------------
2572 1. Corrected the man page, where some "const" qualifiers had been omitted.
2574 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
2575 input syntax.
2577 3. Fixed memory leak which occurred when a regex with back references was
2578 matched with an offsets vector that wasn't big enough. The temporary memory
2579 that is used in this case wasn't being freed if the match failed.
2581 4. Tidied pcretest to ensure it frees memory that it gets.
2583 5. Temporary memory was being obtained in the case where the passed offsets
2584 vector was exactly big enough.
2586 6. Corrected definition of offsetof() from change 5 below.
2588 7. I had screwed up change 6 below and broken the rules for the use of
2589 setjmp(). Now fixed.
2592 Version 1.03 18-Dec-97
2593 ----------------------
2595 1. A erroneous regex with a missing opening parenthesis was correctly
2596 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
2597 on some systems.
2599 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
2600 it was reported that one broken compiler failed on the former because "code" is
2601 also an independent variable.
2603 3. The erroneous regex a[]b caused an array overrun reference.
2605 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
2606 fail on data ending with that character. (It was going on too far, and checking
2607 the next character, typically a binary zero.) This was specific to the
2608 optimized code for single-character negative classes.
2610 5. Added a contributed patch from the TIN world which does the following:
2612 + Add an undef for memmove, in case the the system defines a macro for it.
2614 + Add a definition of offsetof(), in case there isn't one. (I don't know
2615 the reason behind this - offsetof() is part of the ANSI standard - but
2616 it does no harm).
2618 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
2619 most of the places where whitespace preceded '#'. I have given up and
2620 allowed the remaining 2 cases to be at the margin.
2622 + Rename some variables in pcre to eliminate shadowing. This seems very
2623 pedantic, but does no harm, of course.
2625 6. Moved the call to setjmp() into its own function, to get rid of warnings
2626 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
2628 7. Constructs such as \d{8,} were compiling into the equivalent of
2629 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
2630 outcome, but in this particular case used more store than had been allocated,
2631 which caused the bug to be discovered because it threw up an internal error.
2633 8. The debugging code in both pcre and pcretest for outputting the compiled
2634 form of a regex was going wrong in the case of back references followed by
2635 curly-bracketed repeats.
2638 Version 1.02 12-Dec-97
2639 ----------------------
2641 1. Typos in pcre.3 and comments in the source fixed.
2643 2. Applied a contributed patch to get rid of places where it used to remove
2644 'const' from variables, and fixed some signed/unsigned and uninitialized
2645 variable warnings.
2647 3. Added the "runtest" target to Makefile.
2649 4. Set default compiler flag to -O2 rather than just -O.
2652 Version 1.01 19-Nov-97
2653 ----------------------
2655 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
2656 like /([ab]*)*/, that is, for classes with more than one character in them.
2658 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
2659 as /((?>a*))*/ (a PCRE_EXTRA facility).
2662 Version 1.00 18-Nov-97
2663 ----------------------
2665 1. Added compile-time macros to support systems such as SunOS4 which don't have
2666 memmove() or strerror() but have other things that can be used instead.
2668 2. Arranged that "make clean" removes the executables.
2671 Version 0.99 27-Oct-97
2672 ----------------------
2674 1. Fixed bug in code for optimizing classes with only one character. It was
2675 initializing a 32-byte map regardless, which could cause it to run off the end
2676 of the memory it had got.
2678 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
2681 Version 0.98 22-Oct-97
2682 ----------------------
2684 1. Fixed bug in code for handling temporary memory usage when there are more
2685 back references than supplied space in the ovector. This could cause segfaults.
2688 Version 0.97 21-Oct-97
2689 ----------------------
2691 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
2693 2. Optimized negated single characters not to use a bit map.
2695 3. Brought error texts together as macro definitions; clarified some of them;
2696 fixed one that was wrong - it said "range out of order" when it meant "invalid
2697 escape sequence".
2699 4. Changed some char * arguments to const char *.
2701 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
2703 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
2704 pcretest.
2707 Version 0.96 16-Oct-97
2708 ----------------------
2710 1. Added a simple "pgrep" utility to the distribution.
2712 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
2713 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
2714 where "ddd" means "one or more decimal digits".
2716 3. Fixed serious bug. If a pattern had a back reference, but the call to
2717 pcre_exec() didn't supply a large enough ovector to record the related
2718 identifying subpattern, the match always failed. PCRE now remembers the number
2719 of the largest back reference, and gets some temporary memory in which to save
2720 the offsets during matching if necessary, in order to ensure that
2721 backreferences always work.
2723 4. Increased the compatibility with Perl in a number of ways:
2725 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
2726 to request this handling. The option can be set at compile or exec time.
2728 (b) $ matches before a terminating newline by default; an option
2729 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
2730 mode). The option can be set at compile or exec time.
2732 (c) The handling of \ followed by a digit other than 0 is now supposed to be
2733 the same as Perl's. If the decimal number it represents is less than 10
2734 or there aren't that many previous left capturing parentheses, an octal
2735 escape is read. Inside a character class, it's always an octal escape,
2736 even if it is a single digit.
2738 (d) An escaped but undefined alphabetic character is taken as a literal,
2739 unless PCRE_EXTRA is set. Currently this just reserves the remaining
2740 escapes.
2742 (e) {0} is now permitted. (The previous item is removed from the compiled
2743 pattern).
2745 5. Changed all the names of code files so that the basic parts are no longer
2746 than 10 characters, and abolished the teeny "globals.c" file.
2748 6. Changed the handling of character classes; they are now done with a 32-byte
2749 bit map always.
2751 7. Added the -d and /D options to pcretest to make it possible to look at the
2752 internals of compilation without having to recompile pcre.
2755 Version 0.95 23-Sep-97
2756 ----------------------
2758 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
2759 \x20 at the start of a run of normal characters. These were being treated as
2760 real characters, instead of the source characters being re-checked.
2763 Version 0.94 18-Sep-97
2764 ----------------------
2766 1. The functions are now thread-safe, with the caveat that the global variables
2767 containing pointers to malloc() and free() or alternative functions are the
2768 same for all threads.
2770 2. Get pcre_study() to generate a bitmap of initial characters for non-
2771 anchored patterns when this is possible, and use it if passed to pcre_exec().
2774 Version 0.93 15-Sep-97
2775 ----------------------
2777 1. /(b)|(:+)/ was computing an incorrect first character.
2779 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
2780 but not actually doing anything yet.
2782 3. Treat "-" characters in classes that cannot be part of ranges as literals,
2783 as Perl does (e.g. [-az] or [az-]).
2785 4. Set the anchored flag if a branch starts with .* or .*? because that tests
2786 all possible positions.
2788 5. Split up into different modules to avoid including unneeded functions in a
2789 compiled binary. However, compile and exec are still in one module. The "study"
2790 function is split off.
2792 6. The character tables are now in a separate module whose source is generated
2793 by an auxiliary program - but can then be edited by hand if required. There are
2794 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
2795 toupper() in the code.
2797 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
2798 make them global. Abolish the function for setting them, as the caller can now
2799 set them directly.
2802 Version 0.92 11-Sep-97
2803 ----------------------
2805 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
2806 (e.g. /a{1,3}/) was broken (I mis-optimized it).
2808 2. Caseless matching was not working in character classes if the characters in
2809 the pattern were in upper case.
2811 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
2813 4. Make PCRE_ANCHORED public and accept as a compile option.
2815 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
2816 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
2817 pass them.
2819 6. Give an error if bad option bits passed at compile or run time.
2821 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
2822 pcretest to cause it to pass that flag.
2824 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
2825 options, and the first character, if set.
2827 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
2830 Version 0.91 10-Sep-97
2831 ----------------------
2833 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
2834 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
2836 2. PCRE was looping on encountering an indefinitely repeated back reference to
2837 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
2838 Perl does - treats the match as successful.
2840 ****


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

  ViewVC Help
Powered by ViewVC 1.1.5