/[pcre]/code/trunk/ChangeLog
ViewVC logotype

Diff of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 89 by nigel, Sat Feb 24 21:41:27 2007 UTC revision 213 by ph10, Wed Aug 15 11:34:14 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.3 09-Aug-07
5    ---------------------
6    
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:
12    
13          #include "pcre.h"
14    
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.
18    
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.
26    
27     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
28    
29     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
30        in debug output.
31    
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.
34    
35     6. ESC_K was left out of the EBCDIC table.
36    
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.
50    
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.
55    
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.
67    
68    10. Fixed a bug in the documentation for get/copy named substring when
69        duplicate names are permitted. If none of the named substrings are set, the
70        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
71        empty string.
72    
73    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
74        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
75        because the ] is interpreted as the first data character and the
76        terminating ] is not found. PCRE has been made compatible with Perl in this
77        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
78        cause memory overwriting.
79    
80    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
81        string has been matched (to stop an infinite loop). It was not recognizing
82        a conditional subpattern that could match an empty string if that
83        subpattern was within another subpattern. For example, it looped when
84        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
85        condition was not nested. This bug has been fixed.
86    
87    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
88        past the start of the subject in the presence of bytes with the top bit
89        set, for example "\x8aBCD".
90    
91    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
92        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
93    
94    14. Optimized (?!) to (*FAIL).
95    
96    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
97        This restricts code points to be within the range 0 to 0x10FFFF, excluding
98        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
99        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
100        does: it's just the validity check that is more restrictive.
101    
102    16. Inserted checks for integer overflows during escape sequence (backslash)
103        processing, and also fixed erroneous offset values for syntax errors during
104        backslash processing.
105    
106    
107    Version 7.2 19-Jun-07
108    ---------------------
109    
110     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
111        which is apparently normally available under Windows.
112    
113     2. Re-jig the pcregrep tests with different newline settings in an attempt
114        to make them independent of the local environment's newline setting.
115    
116     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
117    
118     4. Some of the "internals" tests were previously cut out when the link size
119        was not 2, because the output contained actual offsets. The recent new
120        "Z" feature of pcretest means that these can be cut out, making the tests
121        usable with all link sizes.
122    
123     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
124        stack recursion. This gives a massive performance boost under BSD, but just
125        a small improvement under Linux. However, it saves one field in the frame
126        in all cases.
127    
128     6. Added more features from the forthcoming Perl 5.10:
129    
130        (a) (?-n) (where n is a string of digits) is a relative subroutine or
131            recursion call. It refers to the nth most recently opened parentheses.
132    
133        (b) (?+n) is also a relative subroutine call; it refers to the nth next
134            to be opened parentheses.
135    
136        (c) Conditions that refer to capturing parentheses can be specified
137            relatively, for example, (?(-2)... or (?(+3)...
138    
139        (d) \K resets the start of the current match so that everything before
140            is not part of it.
141    
142        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
143    
144        (f) \g{name} is another synonym - part of Perl 5.10's unification of
145            reference syntax.
146    
147        (g) (?| introduces a group in which the numbering of parentheses in each
148            alternative starts with the same number.
149    
150        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
151    
152     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
153        PCRE_INFO_JCHANGED.
154    
155     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
156        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
157        for detecting groups that can match an empty string.
158    
159     9. A pattern with a very large number of alternatives (more than several
160        hundred) was running out of internal workspace during the pre-compile
161        phase, where pcre_compile() figures out how much memory will be needed. A
162        bit of new cunning has reduced the workspace needed for groups with
163        alternatives. The 1000-alternative test pattern now uses 12 bytes of
164        workspace instead of running out of the 4096 that are available.
165    
166    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
167    
168    11. Applied patch from Google to remove an optimization that didn't quite work.
169        The report of the bug said:
170    
171          pcrecpp::RE("a*").FullMatch("aaa") matches, while
172          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
173          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
174    
175    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
176        it matched the wrong number of bytes.
177    
178    
179    Version 7.1 24-Apr-07
180    ---------------------
181    
182     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
183        that is more "standard", making use of automake and other Autotools. There
184        is some re-arrangement of the files and adjustment of comments consequent
185        on this.
186    
187     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
188        for recursive directory scanning broke on some systems because the files
189        are not scanned in any specific order and on different systems the order
190        was different. A call to "sort" has been inserted into RunGrepTest for the
191        approprate test as a short-term fix. In the longer term there may be an
192        alternative.
193    
194     3. I had an email from Eric Raymond about problems translating some of PCRE's
195        man pages to HTML (despite the fact that I distribute HTML pages, some
196        people do their own conversions for various reasons). The problems
197        concerned the use of low-level troff macros .br and .in. I have therefore
198        removed all such uses from the man pages (some were redundant, some could
199        be replaced by .nf/.fi pairs). The 132html script that I use to generate
200        HTML has been updated to handle .nf/.fi and to complain if it encounters
201        .br or .in.
202    
203     4. Updated comments in configure.ac that get placed in config.h.in and also
204        arranged for config.h to be included in the distribution, with the name
205        config.h.generic, for the benefit of those who have to compile without
206        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
207    
208     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
209        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
210        makevp.bat for the current PCRE, using the additional files
211        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
212    
213     6. A Windows user reported a minor discrepancy with test 2, which turned out
214        to be caused by a trailing space on an input line that had got lost in his
215        copy. The trailing space was an accident, so I've just removed it.
216    
217     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
218        that is needed.
219    
220     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
221        as "const" (a) because they are and (b) because it helps the PHP
222        maintainers who have recently made a script to detect big data structures
223        in the php code that should be moved to the .rodata section. I remembered
224        to update Builducptable as well, so it won't revert if ucptable.h is ever
225        re-created.
226    
227     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
228        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
229        order to be able to cut out the UTF-8 tables in the latter when UTF-8
230        support is not required. This saves 1.5-2K of code, which is important in
231        some applications.
232    
233        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
234        so as not to refer to the tables, even though these functions will never be
235        called when UTF-8 support is disabled. Otherwise there are problems with a
236        shared library.
237    
238    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
239    
240        (a) It was defining its arguments as char * instead of void *.
241    
242        (b) It was assuming that all moves were upwards in memory; this was true
243            a long time ago when I wrote it, but is no longer the case.
244    
245        The emulated memove() is provided for those environments that have neither
246        memmove() nor bcopy(). I didn't think anyone used it these days, but that
247        is clearly not the case, as these two bugs were recently reported.
248    
249    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
250        and Detrail to create the HTML documentation, the .txt form of the man
251        pages, and it removes trailing spaces from listed files. It also creates
252        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
253        case, it wraps all the #defines with #ifndefs. This script should be run
254        before "make dist".
255    
256    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
257        with Unicode property support.
258    
259        (a) For a maximizing quantifier, if the two different cases of the
260            character were of different lengths in their UTF-8 codings (there are
261            some cases like this - I found 11), and the matching function had to
262            back up over a mixture of the two cases, it incorrectly assumed they
263            were both the same length.
264    
265        (b) When PCRE was configured to use the heap rather than the stack for
266            recursion during matching, it was not correctly preserving the data for
267            the other case of a UTF-8 character when checking ahead for a match
268            while processing a minimizing repeat. If the check also involved
269            matching a wide character, but failed, corruption could cause an
270            erroneous result when trying to check for a repeat of the original
271            character.
272    
273    13. Some tidying changes to the testing mechanism:
274    
275        (a) The RunTest script now detects the internal link size and whether there
276            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
277            values substituted by "configure". (The RunGrepTest script already did
278            this for UTF-8.) The configure.ac script no longer substitutes the
279            relevant variables.
280    
281        (b) The debugging options /B and /D in pcretest show the compiled bytecode
282            with length and offset values. This means that the output is different
283            for different internal link sizes. Test 2 is skipped for link sizes
284            other than 2 because of this, bypassing the problem. Unfortunately,
285            there was also a test in test 3 (the locale tests) that used /B and
286            failed for link sizes other than 2. Rather than cut the whole test out,
287            I have added a new /Z option to pcretest that replaces the length and
288            offset values with spaces. This is now used to make test 3 independent
289            of link size. (Test 2 will be tidied up later.)
290    
291    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
292        segmentation fault instead of returning the appropriate error message.
293    
294    15. In multiline mode when the newline sequence was set to "any", the pattern
295        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
296        This doesn't seem right; it now treats the CRLF combination as the line
297        ending, and so does not match in that case. It's only a pattern such as ^$
298        that would hit this one: something like ^ABC$ would have failed after \r
299        and then tried again after \r\n.
300    
301    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
302        in an attempt to make files that differ only in their line terminators
303        compare equal. This works on Linux.
304    
305    17. Under certain error circumstances pcregrep might try to free random memory
306        as it exited. This is now fixed, thanks to valgrind.
307    
308    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
309        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
310        was because its rules for how to advance for /g after matching an empty
311        string at the end of a line did not allow for this case. They now check for
312        it specially.
313    
314    20. pcretest is supposed to handle patterns and data of any length, by
315        extending its buffers when necessary. It was getting this wrong when the
316        buffer for a data line had to be extended.
317    
318    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
319        CRLF as a newline sequence.
320    
321    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
322        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
323        I have nevertheless tidied it up.
324    
325    23. Added some casts to kill warnings from HP-UX ia64 compiler.
326    
327    24. Added a man page for pcre-config.
328    
329    
330    Version 7.0 19-Dec-06
331    ---------------------
332    
333     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
334        moving to gcc 4.1.1.
335    
336     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
337        sys/time.h, which is documented as needed for this function. It doesn't
338        seem to matter on Linux, but it showed up on some releases of OS X.
339    
340     3. It seems that there are systems where bytes whose values are greater than
341        127 match isprint() in the "C" locale. The "C" locale should be the
342        default when a C program starts up. In most systems, only ASCII printing
343        characters match isprint(). This difference caused the output from pcretest
344        to vary, making some of the tests fail. I have changed pcretest so that:
345    
346        (a) When it is outputting text in the compiled version of a pattern, bytes
347            other than 32-126 are always shown as hex escapes.
348    
349        (b) When it is outputting text that is a matched part of a subject string,
350            it does the same, unless a different locale has been set for the match
351            (using the /L modifier). In this case, it uses isprint() to decide.
352    
353     4. Fixed a major bug that caused incorrect computation of the amount of memory
354        required for a compiled pattern when options that changed within the
355        pattern affected the logic of the preliminary scan that determines the
356        length. The relevant options are -x, and -i in UTF-8 mode. The result was
357        that the computed length was too small. The symptoms of this bug were
358        either the PCRE error "internal error: code overflow" from pcre_compile(),
359        or a glibc crash with a message such as "pcretest: free(): invalid next
360        size (fast)". Examples of patterns that provoked this bug (shown in
361        pcretest format) are:
362    
363          /(?-x: )/x
364          /(?x)(?-x: \s*#\s*)/
365          /((?i)[\x{c0}])/8
366          /(?i:[\x{c0}])/8
367    
368        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
369        is now done differently.
370    
371     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
372        wrapper classes; (b) implement a new function in the C++ scanner that is
373        more efficient than the old way of doing things because it avoids levels of
374        recursion in the regex matching; (c) add a paragraph to the documentation
375        for the FullMatch() function.
376    
377     6. The escape sequence \n was being treated as whatever was defined as
378        "newline". Not only was this contrary to the documentation, which states
379        that \n is character 10 (hex 0A), but it also went horribly wrong when
380        "newline" was defined as CRLF. This has been fixed.
381    
382     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
383        was being set to -1 for the "end of line" case (supposedly a value that no
384        character can have). Though this value is never used (the check for end of
385        line is "zero bytes in current character"), it caused compiler complaints.
386        I've changed it to 0xffffffff.
387    
388     8. In pcre_version.c, the version string was being built by a sequence of
389        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
390        string (as it is for production releases) called a macro with an empty
391        argument. The C standard says the result of this is undefined. The gcc
392        compiler treats it as an empty string (which was what was wanted) but it is
393        reported that Visual C gives an error. The source has been hacked around to
394        avoid this problem.
395    
396     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
397        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
398        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
399        of them did).
400    
401    10. Originally, pcretest opened its input and output without "b"; then I was
402        told that "b" was needed in some environments, so it was added for release
403        5.0 to both the input and output. (It makes no difference on Unix-like
404        systems.) Later I was told that it is wrong for the input on Windows. I've
405        now abstracted the modes into two macros, to make it easier to fiddle with
406        them, and removed "b" from the input mode under Windows.
407    
408    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
409    
410    12. Added -help and --help to pcretest as an official way of being reminded
411        of the options.
412    
413    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
414        and pcrecpp.cc because they annoy compilers at high warning levels.
415    
416    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
417    
418    15. Fixed an occurrence of == in configure.ac that should have been = (shell
419        scripts are not C programs :-) and which was not noticed because it works
420        on Linux.
421    
422    16. pcretest is supposed to handle any length of pattern and data line (as one
423        line or as a continued sequence of lines) by extending its input buffer if
424        necessary. This feature was broken for very long pattern lines, leading to
425        a string of junk being passed to pcre_compile() if the pattern was longer
426        than about 50K.
427    
428    17. I have done a major re-factoring of the way pcre_compile() computes the
429        amount of memory needed for a compiled pattern. Previously, there was code
430        that made a preliminary scan of the pattern in order to do this. That was
431        OK when PCRE was new, but as the facilities have expanded, it has become
432        harder and harder to keep it in step with the real compile phase, and there
433        have been a number of bugs (see for example, 4 above). I have now found a
434        cunning way of running the real compile function in a "fake" mode that
435        enables it to compute how much memory it would need, while actually only
436        ever using a few hundred bytes of working memory and without too many
437        tests of the mode. This should make future maintenance and development
438        easier. A side effect of this work is that the limit of 200 on the nesting
439        depth of parentheses has been removed (though this was never a serious
440        limitation, I suspect). However, there is a downside: pcre_compile() now
441        runs more slowly than before (30% or more, depending on the pattern). I
442        hope this isn't a big issue. There is no effect on runtime performance.
443    
444    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
445        newline (only possible for the last line of a file) and it was a
446        pattern that set a locale (followed by /Lsomething), pcretest crashed.
447    
448    19. Added additional timing features to pcretest. (1) The -tm option now times
449        matching only, not compiling. (2) Both -t and -tm can be followed, as a
450        separate command line item, by a number that specifies the number of
451        repeats to use when timing. The default is 50000; this gives better
452        precision, but takes uncomfortably long for very large patterns.
453    
454    20. Extended pcre_study() to be more clever in cases where a branch of a
455        subpattern has no definite first character. For example, (a*|b*)[cd] would
456        previously give no result from pcre_study(). Now it recognizes that the
457        first character must be a, b, c, or d.
458    
459    21. There was an incorrect error "recursive call could loop indefinitely" if
460        a subpattern (or the entire pattern) that was being tested for matching an
461        empty string contained only one non-empty item after a nested subpattern.
462        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
463        incorrectly, because the \d was being skipped in the check.
464    
465    22. The pcretest program now has a new pattern option /B and a command line
466        option -b, which is equivalent to adding /B to every pattern. This causes
467        it to show the compiled bytecode, without the additional information that
468        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
469        is the same as /B/I).
470    
471    23. A new optimization is now able automatically to treat some sequences such
472        as a*b as a*+b. More specifically, if something simple (such as a character
473        or a simple class like \d) has an unlimited quantifier, and is followed by
474        something that cannot possibly match the quantified thing, the quantifier
475        is automatically "possessified".
476    
477    24. A recursive reference to a subpattern whose number was greater than 39
478        went wrong under certain circumstances in UTF-8 mode. This bug could also
479        have affected the operation of pcre_study().
480    
481    25. Realized that a little bit of performance could be had by replacing
482        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
483    
484    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
485    
486    27. Possessive quantifiers such as a++ were previously implemented by turning
487        them into atomic groups such as ($>a+). Now they have their own opcodes,
488        which improves performance. This includes the automatically created ones
489        from 23 above.
490    
491    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
492        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
493        the first matched character to be a colon. This applied both to named and
494        numbered groups.
495    
496    29. The ucpinternal.h header file was missing its idempotency #ifdef.
497    
498    30. I was sent a "project" file called libpcre.a.dev which I understand makes
499        building PCRE on Windows easier, so I have included it in the distribution.
500    
501    31. There is now a check in pcretest against a ridiculously large number being
502        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
503        loop, the loop is abandoned.
504    
505    32. Forward references to subpatterns in conditions such as (?(2)...) where
506        subpattern 2 is defined later cause pcre_compile() to search forwards in
507        the pattern for the relevant set of parentheses. This search went wrong
508        when there were unescaped parentheses in a character class, parentheses
509        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
510    
511    33. "Subroutine" calls and backreferences were previously restricted to
512        referencing subpatterns earlier in the regex. This restriction has now
513        been removed.
514    
515    34. Added a number of extra features that are going to be in Perl 5.10. On the
516        whole, these are just syntactic alternatives for features that PCRE had
517        previously implemented using the Python syntax or my own invention. The
518        other formats are all retained for compatibility.
519    
520        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
521            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
522            also .NET compatible.
523    
524        (b) A recursion or subroutine call to a named group can now be defined as
525            (?&name) as well as (?P>name).
526    
527        (c) A backreference to a named group can now be defined as \k<name> or
528            \k'name' as well as (?P=name). The new forms, as well as being in Perl
529            5.10, are also .NET compatible.
530    
531        (d) A conditional reference to a named group can now use the syntax
532            (?(<name>) or (?('name') as well as (?(name).
533    
534        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
535            groups (named and numbered) that are never evaluated inline, but can be
536            called as "subroutines" from elsewhere. In effect, the DEFINE condition
537            is always false. There may be only one alternative in such a group.
538    
539        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
540            as the simple (?(R). The condition is true only if the most recent
541            recursion is that of the given number or name. It does not search out
542            through the entire recursion stack.
543    
544        (g) The escape \gN or \g{N} has been added, where N is a positive or
545            negative number, specifying an absolute or relative reference.
546    
547    35. Tidied to get rid of some further signed/unsigned compiler warnings and
548        some "unreachable code" warnings.
549    
550    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
551        things, this adds five new scripts.
552    
553    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
554        There were also incompatibilities regarding the handling of \Q..\E inside
555        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
556        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
557    
558    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
559        matches an empty string, and forcibly breaks the loop. There were bugs in
560        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
561        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
562        separate and independent bugs (that affected different cases) have been
563        fixed.
564    
565    39. Refactored the code to abolish the use of different opcodes for small
566        capturing bracket numbers. This is a tidy that I avoided doing when I
567        removed the limit on the number of capturing brackets for 3.5 back in 2001.
568        The new approach is not only tidier, it makes it possible to reduce the
569        memory needed to fix the previous bug (38).
570    
571    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
572        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
573        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
574        mode.
575    
576    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
577        report.
578    
579    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
580        copy construction and assignment in the C++ wrapper.
581    
582    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
583        couple of bugs that could have given wrong results in the "--newline=crlf"
584        case.
585    
586    44. Added a number of casts and did some reorganization of signed/unsigned int
587        variables following suggestions from Dair Grant. Also renamed the variable
588        "this" as "item" because it is a C++ keyword.
589    
590    45. Arranged for dftables to add
591    
592          #include "pcre_internal.h"
593    
594        to pcre_chartables.c because without it, gcc 4.x may remove the array
595        definition from the final binary if PCRE is built into a static library and
596        dead code stripping is activated.
597    
598    46. For an unanchored pattern, if a match attempt fails at the start of a
599        newline sequence, and the newline setting is CRLF or ANY, and the next two
600        characters are CRLF, advance by two characters instead of one.
601    
602    
603    Version 6.7 04-Jul-06
604    ---------------------
605    
606     1. In order to handle tests when input lines are enormously long, pcretest has
607        been re-factored so that it automatically extends its buffers when
608        necessary. The code is crude, but this _is_ just a test program. The
609        default size has been increased from 32K to 50K.
610    
611     2. The code in pcre_study() was using the value of the re argument before
612        testing it for NULL. (Of course, in any sensible call of the function, it
613        won't be NULL.)
614    
615     3. The memmove() emulation function in pcre_internal.h, which is used on
616        systems that lack both memmove() and bcopy() - that is, hardly ever -
617        was missing a "static" storage class specifier.
618    
619     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
620        containing an extended class (one that cannot be represented by a bitmap
621        because it contains high-valued characters or Unicode property items, e.g.
622        [\pZ]). Almost always one would set UTF-8 mode when processing such a
623        pattern, but PCRE should not loop if you do not (it no longer does).
624        [Detail: two cases were found: (a) a repeated subpattern containing an
625        extended class; (b) a recursive reference to a subpattern that followed a
626        previous extended class. It wasn't skipping over the extended class
627        correctly when UTF-8 mode was not set.]
628    
629     5. A negated single-character class was not being recognized as fixed-length
630        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
631        compile error "lookbehind assertion is not fixed length".
632    
633     6. The RunPerlTest auxiliary script was showing an unexpected difference
634        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
635        write a Perl script that can interpret lines of an input file either as
636        byte characters or as UTF-8, which is what "perltest" was being required to
637        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
638        can't do is switch easily at run time between having the "use utf8;" pragma
639        or not. In the end, I fudged it by using the RunPerlTest script to insert
640        "use utf8;" explicitly for the UTF-8 tests.
641    
642     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
643        the end of the subject string, contrary to the documentation and to what
644        Perl does. This was true of both matching functions. Now it matches only at
645        the start of the subject and immediately after *internal* newlines.
646    
647     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
648        a pointer to an int instead of a pointer to an unsigned long int. This
649        caused problems on 64-bit systems.
650    
651     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
652        instance of the 'standard' template library not being so standard".
653    
654    10. There was no check on the number of named subpatterns nor the maximum
655        length of a subpattern name. The product of these values is used to compute
656        the size of the memory block for a compiled pattern. By supplying a very
657        long subpattern name and a large number of named subpatterns, the size
658        computation could be caused to overflow. This is now prevented by limiting
659        the length of names to 32 characters, and the number of named subpatterns
660        to 10,000.
661    
662    11. Subpatterns that are repeated with specific counts have to be replicated in
663        the compiled pattern. The size of memory for this was computed from the
664        length of the subpattern and the repeat count. The latter is limited to
665        65535, but there was no limit on the former, meaning that integer overflow
666        could in principle occur. The compiled length of a repeated subpattern is
667        now limited to 30,000 bytes in order to prevent this.
668    
669    12. Added the optional facility to have named substrings with the same name.
670    
671    13. Added the ability to use a named substring as a condition, using the
672        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
673        are numbers (not recommended). Forward references are permitted.
674    
675    14. Added forward references in named backreferences (if you see what I mean).
676    
677    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
678        pattern could run off the end of the subject. For example, the pattern
679        "(?s)(.{1,5})"8 did this with the subject "ab".
680    
681    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
682        PCRE_CASELESS was set when matching characters that were quantified with ?
683        or *.
684    
685    17. A character class other than a single negated character that had a minimum
686        but no maximum quantifier - for example [ab]{6,} - was not handled
687        correctly by pce_dfa_exec(). It would match only one character.
688    
689    18. A valid (though odd) pattern that looked like a POSIX character
690        class but used an invalid character after [ (for example [[,abc,]]) caused
691        pcre_compile() to give the error "Failed: internal error: code overflow" or
692        in some cases to crash with a glibc free() error. This could even happen if
693        the pattern terminated after [[ but there just happened to be a sequence of
694        letters, a binary zero, and a closing ] in the memory that followed.
695    
696    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
697        over the years. Originally (before any Unicode support), just the bottom 8
698        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
699        output from "man perlunicode" includes this:
700    
701          The regular expression compiler produces polymorphic opcodes.  That
702          is, the pattern adapts to the data and automatically switches to
703          the Unicode character scheme when presented with Unicode data--or
704          instead uses a traditional byte scheme when presented with byte
705          data.
706    
707        Sadly, a wide octal escape does not cause a switch, and in a string with
708        no other multibyte characters, these octal escapes are treated as before.
709        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
710        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
711        Unicode string.
712    
713        I have not perpetrated such confusion in PCRE. Up till now, it took just
714        the bottom 8 bits, as in old Perl. I have now made octal escapes with
715        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
716        translate to the appropriate multibyte character.
717    
718    29. Applied some refactoring to reduce the number of warnings from Microsoft
719        and Borland compilers. This has included removing the fudge introduced
720        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
721        a warning about an unused variable.
722    
723    21. PCRE has not included VT (character 0x0b) in the set of whitespace
724        characters since release 4.0, because Perl (from release 5.004) does not.
725        [Or at least, is documented not to: some releases seem to be in conflict
726        with the documentation.] However, when a pattern was studied with
727        pcre_study() and all its branches started with \s, PCRE still included VT
728        as a possible starting character. Of course, this did no harm; it just
729        caused an unnecessary match attempt.
730    
731    22. Removed a now-redundant internal flag bit that recorded the fact that case
732        dependency changed within the pattern. This was once needed for "required
733        byte" processing, but is no longer used. This recovers a now-scarce options
734        bit. Also moved the least significant internal flag bit to the most-
735        significant bit of the word, which was not previously used (hangover from
736        the days when it was an int rather than a uint) to free up another bit for
737        the future.
738    
739    23. Added support for CRLF line endings as well as CR and LF. As well as the
740        default being selectable at build time, it can now be changed at runtime
741        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
742        specify that it is scanning data with non-default line endings.
743    
744    24. Changed the definition of CXXLINK to make it agree with the definition of
745        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
746    
747    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
748        recursions. This makes a big different to stack usage for some patterns.
749    
750    26. If a subpattern containing a named recursion or subroutine reference such
751        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
752        the space required for the compiled pattern went wrong and gave too small a
753        value. Depending on the environment, this could lead to "Failed: internal
754        error: code overflow at offset 49" or "glibc detected double free or
755        corruption" errors.
756    
757    27. Applied patches from Google (a) to support the new newline modes and (b) to
758        advance over multibyte UTF-8 characters in GlobalReplace.
759    
760    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
761        difference for some implementation of PCRE in some Windows version.
762    
763    29. Added some extra testing facilities to pcretest:
764    
765        \q<number>   in a data line sets the "match limit" value
766        \Q<number>   in a data line sets the "match recursion limt" value
767        -S <number>  sets the stack size, where <number> is in megabytes
768    
769        The -S option isn't available for Windows.
770    
771    
772  Version 6.6 06-Feb-06  Version 6.6 06-Feb-06
773  ---------------------  ---------------------
774    

Legend:
Removed from v.89  
changed lines
  Added in v.213

  ViewVC Help
Powered by ViewVC 1.1.5