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

Diff of /code/trunk/ChangeLog

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

revision 85 by nigel, Sat Feb 24 21:41:13 2007 UTC revision 150 by ph10, Tue Apr 17 08:22:40 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.1 12-Mar-07
5    ---------------------
6    
7     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
8        that is more "standard", making use of automake and other Autotools. There
9        is some re-arrangement of the files and adjustment of comments consequent
10        on this.
11    
12     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
13        for recursive directory scanning broke on some systems because the files
14        are not scanned in any specific order and on different systems the order
15        was different. A call to "sort" has been inserted into RunGrepTest for the
16        approprate test as a short-term fix. In the longer term there may be an
17        alternative.
18    
19     3. I had an email from Eric Raymond about problems translating some of PCRE's
20        man pages to HTML (despite the fact that I distribute HTML pages, some
21        people do their own conversions for various reasons). The problems
22        concerned the use of low-level troff macros .br and .in. I have therefore
23        removed all such uses from the man pages (some were redundant, some could
24        be replaced by .nf/.fi pairs). The 132html script that I use to generate
25        HTML has been updated to handle .nf/.fi and to complain if it encounters
26        .br or .in.
27    
28     4. Updated comments in configure.ac that get placed in config.h.in and also
29        arranged for config.h to be included in the distribution, with the name
30        config.h.generic, for the benefit of those who have to compile without
31        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
32    
33     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
34        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
35        makevp.bat for the current PCRE, using the additional files
36        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
37    
38     6. A Windows user reported a minor discrepancy with test 2, which turned out
39        to be caused by a trailing space on an input line that had got lost in his
40        copy. The trailing space was an accident, so I've just removed it.
41    
42     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
43        that is needed.
44    
45     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
46        as "const" (a) because they are and (b) because it helps the PHP
47        maintainers who have recently made a script to detect big data structures
48        in the php code that should be moved to the .rodata section. I remembered
49        to update Builducptable as well, so it won't revert if ucptable.h is ever
50        re-created.
51    
52     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
53        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
54        order to be able to cut out the UTF-8 tables in the latter when UTF-8
55        support is not required. This saves 1.5-2K of code, which is important in
56        some applications.
57    
58        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
59        so as not to refer to the tables, even though these functions will never be
60        called when UTF-8 support is disabled. Otherwise there are problems with a
61        shared library.
62    
63    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
64    
65        (a) It was defining its arguments as char * instead of void *.
66    
67        (b) It was assuming that all moves were upwards in memory; this was true
68            a long time ago when I wrote it, but is no longer the case.
69    
70        The emulated memove() is provided for those environments that have neither
71        memmove() nor bcopy(). I didn't think anyone used it these days, but that
72        is clearly not the case, as these two bugs were recently reported.
73    
74    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
75        and Detrail to create the HTML documentation, the .txt form of the man
76        pages, and it removes trailing spaces from listed files. It also creates
77        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
78        case, it wraps all the #defines with #ifndefs. This script should be run
79        before "make dist".
80    
81    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
82        with Unicode property support.
83    
84        (a) For a maximizing quantifier, if the two different cases of the
85            character were of different lengths in their UTF-8 codings (there are
86            some cases like this - I found 11), and the matching function had to
87            back up over a mixture of the two cases, it incorrectly assumed they
88            were both the same length.
89    
90        (b) When PCRE was configured to use the heap rather than the stack for
91            recursion during matching, it was not correctly preserving the data for
92            the other case of a UTF-8 character when checking ahead for a match
93            while processing a minimizing repeat. If the check also involved
94            matching a wide character, but failed, corruption could cause an
95            erroneous result when trying to check for a repeat of the original
96            character.
97    
98    13. Some tidying changes to the testing mechanism:
99    
100        (a) The RunTest script now detects the internal link size and whether there
101            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
102            values substituted by "configure". (The RunGrepTest script already did
103            this for UTF-8.) The configure.ac script no longer substitutes the
104            relevant variables.
105    
106        (b) The debugging options /B and /D in pcretest show the compiled bytecode
107            with length and offset values. This means that the output is different
108            for different internal link sizes. Test 2 is skipped for link sizes
109            other than 2 because of this, bypassing the problem. Unfortunately,
110            there was also a test in test 3 (the locale tests) that used /B and
111            failed for link sizes other than 2. Rather than cut the whole test out,
112            I have added a new /Z option to pcretest that replaces the length and
113            offset values with spaces. This is now used to make test 3 independent
114            of link size. (Test 2 will be tidied up later.)
115    
116    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
117        segmentation fault instead of returning the appropriate error message.
118    
119    15. In multiline mode when the newline sequence was set to "any", the pattern
120        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
121        This doesn't seem right; it now treats the CRLF combination as the line
122        ending, and so does not match in that case. It's only a pattern such as ^$
123        that would hit this one: something like ^ABC$ would have failed after \r
124        and then tried again after \r\n.
125    
126    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
127        in an attempt to make files that differ only in their line terminators
128        compare equal. This works on Linux.
129    
130    17. Under certain error circumstances pcregrep might try to free random memory
131        as it exited. This is now fixed, thanks to valgrind.
132    
133    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
134        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
135        was because its rules for how to advance for /g after matching an empty
136        string at the end of a line did not allow for this case. They now check for
137        it specially.
138    
139    20. pcretest is supposed to handle patterns and data of any length, by
140        extending its buffers when necessary. It was getting this wrong when the
141        buffer for a data line had to be extended.
142    
143    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
144        CRLF as a newline sequence.
145    
146    
147    Version 7.0 19-Dec-06
148    ---------------------
149    
150     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
151        moving to gcc 4.1.1.
152    
153     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
154        sys/time.h, which is documented as needed for this function. It doesn't
155        seem to matter on Linux, but it showed up on some releases of OS X.
156    
157     3. It seems that there are systems where bytes whose values are greater than
158        127 match isprint() in the "C" locale. The "C" locale should be the
159        default when a C program starts up. In most systems, only ASCII printing
160        characters match isprint(). This difference caused the output from pcretest
161        to vary, making some of the tests fail. I have changed pcretest so that:
162    
163        (a) When it is outputting text in the compiled version of a pattern, bytes
164            other than 32-126 are always shown as hex escapes.
165    
166        (b) When it is outputting text that is a matched part of a subject string,
167            it does the same, unless a different locale has been set for the match
168            (using the /L modifier). In this case, it uses isprint() to decide.
169    
170     4. Fixed a major bug that caused incorrect computation of the amount of memory
171        required for a compiled pattern when options that changed within the
172        pattern affected the logic of the preliminary scan that determines the
173        length. The relevant options are -x, and -i in UTF-8 mode. The result was
174        that the computed length was too small. The symptoms of this bug were
175        either the PCRE error "internal error: code overflow" from pcre_compile(),
176        or a glibc crash with a message such as "pcretest: free(): invalid next
177        size (fast)". Examples of patterns that provoked this bug (shown in
178        pcretest format) are:
179    
180          /(?-x: )/x
181          /(?x)(?-x: \s*#\s*)/
182          /((?i)[\x{c0}])/8
183          /(?i:[\x{c0}])/8
184    
185        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
186        is now done differently.
187    
188     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
189        wrapper classes; (b) implement a new function in the C++ scanner that is
190        more efficient than the old way of doing things because it avoids levels of
191        recursion in the regex matching; (c) add a paragraph to the documentation
192        for the FullMatch() function.
193    
194     6. The escape sequence \n was being treated as whatever was defined as
195        "newline". Not only was this contrary to the documentation, which states
196        that \n is character 10 (hex 0A), but it also went horribly wrong when
197        "newline" was defined as CRLF. This has been fixed.
198    
199     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
200        was being set to -1 for the "end of line" case (supposedly a value that no
201        character can have). Though this value is never used (the check for end of
202        line is "zero bytes in current character"), it caused compiler complaints.
203        I've changed it to 0xffffffff.
204    
205     8. In pcre_version.c, the version string was being built by a sequence of
206        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
207        string (as it is for production releases) called a macro with an empty
208        argument. The C standard says the result of this is undefined. The gcc
209        compiler treats it as an empty string (which was what was wanted) but it is
210        reported that Visual C gives an error. The source has been hacked around to
211        avoid this problem.
212    
213     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
214        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
215        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
216        of them did).
217    
218    10. Originally, pcretest opened its input and output without "b"; then I was
219        told that "b" was needed in some environments, so it was added for release
220        5.0 to both the input and output. (It makes no difference on Unix-like
221        systems.) Later I was told that it is wrong for the input on Windows. I've
222        now abstracted the modes into two macros, to make it easier to fiddle with
223        them, and removed "b" from the input mode under Windows.
224    
225    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
226    
227    12. Added -help and --help to pcretest as an official way of being reminded
228        of the options.
229    
230    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
231        and pcrecpp.cc because they annoy compilers at high warning levels.
232    
233    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
234    
235    15. Fixed an occurrence of == in configure.ac that should have been = (shell
236        scripts are not C programs :-) and which was not noticed because it works
237        on Linux.
238    
239    16. pcretest is supposed to handle any length of pattern and data line (as one
240        line or as a continued sequence of lines) by extending its input buffer if
241        necessary. This feature was broken for very long pattern lines, leading to
242        a string of junk being passed to pcre_compile() if the pattern was longer
243        than about 50K.
244    
245    17. I have done a major re-factoring of the way pcre_compile() computes the
246        amount of memory needed for a compiled pattern. Previously, there was code
247        that made a preliminary scan of the pattern in order to do this. That was
248        OK when PCRE was new, but as the facilities have expanded, it has become
249        harder and harder to keep it in step with the real compile phase, and there
250        have been a number of bugs (see for example, 4 above). I have now found a
251        cunning way of running the real compile function in a "fake" mode that
252        enables it to compute how much memory it would need, while actually only
253        ever using a few hundred bytes of working memory and without too many
254        tests of the mode. This should make future maintenance and development
255        easier. A side effect of this work is that the limit of 200 on the nesting
256        depth of parentheses has been removed (though this was never a serious
257        limitation, I suspect). However, there is a downside: pcre_compile() now
258        runs more slowly than before (30% or more, depending on the pattern). I
259        hope this isn't a big issue. There is no effect on runtime performance.
260    
261    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
262        newline (only possible for the last line of a file) and it was a
263        pattern that set a locale (followed by /Lsomething), pcretest crashed.
264    
265    19. Added additional timing features to pcretest. (1) The -tm option now times
266        matching only, not compiling. (2) Both -t and -tm can be followed, as a
267        separate command line item, by a number that specifies the number of
268        repeats to use when timing. The default is 50000; this gives better
269        precision, but takes uncomfortably long for very large patterns.
270    
271    20. Extended pcre_study() to be more clever in cases where a branch of a
272        subpattern has no definite first character. For example, (a*|b*)[cd] would
273        previously give no result from pcre_study(). Now it recognizes that the
274        first character must be a, b, c, or d.
275    
276    21. There was an incorrect error "recursive call could loop indefinitely" if
277        a subpattern (or the entire pattern) that was being tested for matching an
278        empty string contained only one non-empty item after a nested subpattern.
279        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
280        incorrectly, because the \d was being skipped in the check.
281    
282    22. The pcretest program now has a new pattern option /B and a command line
283        option -b, which is equivalent to adding /B to every pattern. This causes
284        it to show the compiled bytecode, without the additional information that
285        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
286        is the same as /B/I).
287    
288    23. A new optimization is now able automatically to treat some sequences such
289        as a*b as a*+b. More specifically, if something simple (such as a character
290        or a simple class like \d) has an unlimited quantifier, and is followed by
291        something that cannot possibly match the quantified thing, the quantifier
292        is automatically "possessified".
293    
294    24. A recursive reference to a subpattern whose number was greater than 39
295        went wrong under certain circumstances in UTF-8 mode. This bug could also
296        have affected the operation of pcre_study().
297    
298    25. Realized that a little bit of performance could be had by replacing
299        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
300    
301    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
302    
303    27. Possessive quantifiers such as a++ were previously implemented by turning
304        them into atomic groups such as ($>a+). Now they have their own opcodes,
305        which improves performance. This includes the automatically created ones
306        from 23 above.
307    
308    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
309        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
310        the first matched character to be a colon. This applied both to named and
311        numbered groups.
312    
313    29. The ucpinternal.h header file was missing its idempotency #ifdef.
314    
315    30. I was sent a "project" file called libpcre.a.dev which I understand makes
316        building PCRE on Windows easier, so I have included it in the distribution.
317    
318    31. There is now a check in pcretest against a ridiculously large number being
319        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
320        loop, the loop is abandoned.
321    
322    32. Forward references to subpatterns in conditions such as (?(2)...) where
323        subpattern 2 is defined later cause pcre_compile() to search forwards in
324        the pattern for the relevant set of parentheses. This search went wrong
325        when there were unescaped parentheses in a character class, parentheses
326        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
327    
328    33. "Subroutine" calls and backreferences were previously restricted to
329        referencing subpatterns earlier in the regex. This restriction has now
330        been removed.
331    
332    34. Added a number of extra features that are going to be in Perl 5.10. On the
333        whole, these are just syntactic alternatives for features that PCRE had
334        previously implemented using the Python syntax or my own invention. The
335        other formats are all retained for compatibility.
336    
337        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
338            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
339            also .NET compatible.
340    
341        (b) A recursion or subroutine call to a named group can now be defined as
342            (?&name) as well as (?P>name).
343    
344        (c) A backreference to a named group can now be defined as \k<name> or
345            \k'name' as well as (?P=name). The new forms, as well as being in Perl
346            5.10, are also .NET compatible.
347    
348        (d) A conditional reference to a named group can now use the syntax
349            (?(<name>) or (?('name') as well as (?(name).
350    
351        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
352            groups (named and numbered) that are never evaluated inline, but can be
353            called as "subroutines" from elsewhere. In effect, the DEFINE condition
354            is always false. There may be only one alternative in such a group.
355    
356        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
357            as the simple (?(R). The condition is true only if the most recent
358            recursion is that of the given number or name. It does not search out
359            through the entire recursion stack.
360    
361        (g) The escape \gN or \g{N} has been added, where N is a positive or
362            negative number, specifying an absolute or relative reference.
363    
364    35. Tidied to get rid of some further signed/unsigned compiler warnings and
365        some "unreachable code" warnings.
366    
367    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
368        things, this adds five new scripts.
369    
370    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
371        There were also incompatibilities regarding the handling of \Q..\E inside
372        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
373        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
374    
375    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
376        matches an empty string, and forcibly breaks the loop. There were bugs in
377        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
378        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
379        separate and independent bugs (that affected different cases) have been
380        fixed.
381    
382    39. Refactored the code to abolish the use of different opcodes for small
383        capturing bracket numbers. This is a tidy that I avoided doing when I
384        removed the limit on the number of capturing brackets for 3.5 back in 2001.
385        The new approach is not only tidier, it makes it possible to reduce the
386        memory needed to fix the previous bug (38).
387    
388    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
389        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
390        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
391        mode.
392    
393    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
394        report.
395    
396    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
397        copy construction and assignment in the C++ wrapper.
398    
399    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
400        couple of bugs that could have given wrong results in the "--newline=crlf"
401        case.
402    
403    44. Added a number of casts and did some reorganization of signed/unsigned int
404        variables following suggestions from Dair Grant. Also renamed the variable
405        "this" as "item" because it is a C++ keyword.
406    
407    45. Arranged for dftables to add
408    
409          #include "pcre_internal.h"
410    
411        to pcre_chartables.c because without it, gcc 4.x may remove the array
412        definition from the final binary if PCRE is built into a static library and
413        dead code stripping is activated.
414    
415    46. For an unanchored pattern, if a match attempt fails at the start of a
416        newline sequence, and the newline setting is CRLF or ANY, and the next two
417        characters are CRLF, advance by two characters instead of one.
418    
419    
420    Version 6.7 04-Jul-06
421    ---------------------
422    
423     1. In order to handle tests when input lines are enormously long, pcretest has
424        been re-factored so that it automatically extends its buffers when
425        necessary. The code is crude, but this _is_ just a test program. The
426        default size has been increased from 32K to 50K.
427    
428     2. The code in pcre_study() was using the value of the re argument before
429        testing it for NULL. (Of course, in any sensible call of the function, it
430        won't be NULL.)
431    
432     3. The memmove() emulation function in pcre_internal.h, which is used on
433        systems that lack both memmove() and bcopy() - that is, hardly ever -
434        was missing a "static" storage class specifier.
435    
436     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
437        containing an extended class (one that cannot be represented by a bitmap
438        because it contains high-valued characters or Unicode property items, e.g.
439        [\pZ]). Almost always one would set UTF-8 mode when processing such a
440        pattern, but PCRE should not loop if you do not (it no longer does).
441        [Detail: two cases were found: (a) a repeated subpattern containing an
442        extended class; (b) a recursive reference to a subpattern that followed a
443        previous extended class. It wasn't skipping over the extended class
444        correctly when UTF-8 mode was not set.]
445    
446     5. A negated single-character class was not being recognized as fixed-length
447        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
448        compile error "lookbehind assertion is not fixed length".
449    
450     6. The RunPerlTest auxiliary script was showing an unexpected difference
451        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
452        write a Perl script that can interpret lines of an input file either as
453        byte characters or as UTF-8, which is what "perltest" was being required to
454        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
455        can't do is switch easily at run time between having the "use utf8;" pragma
456        or not. In the end, I fudged it by using the RunPerlTest script to insert
457        "use utf8;" explicitly for the UTF-8 tests.
458    
459     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
460        the end of the subject string, contrary to the documentation and to what
461        Perl does. This was true of both matching functions. Now it matches only at
462        the start of the subject and immediately after *internal* newlines.
463    
464     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
465        a pointer to an int instead of a pointer to an unsigned long int. This
466        caused problems on 64-bit systems.
467    
468     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
469        instance of the 'standard' template library not being so standard".
470    
471    10. There was no check on the number of named subpatterns nor the maximum
472        length of a subpattern name. The product of these values is used to compute
473        the size of the memory block for a compiled pattern. By supplying a very
474        long subpattern name and a large number of named subpatterns, the size
475        computation could be caused to overflow. This is now prevented by limiting
476        the length of names to 32 characters, and the number of named subpatterns
477        to 10,000.
478    
479    11. Subpatterns that are repeated with specific counts have to be replicated in
480        the compiled pattern. The size of memory for this was computed from the
481        length of the subpattern and the repeat count. The latter is limited to
482        65535, but there was no limit on the former, meaning that integer overflow
483        could in principle occur. The compiled length of a repeated subpattern is
484        now limited to 30,000 bytes in order to prevent this.
485    
486    12. Added the optional facility to have named substrings with the same name.
487    
488    13. Added the ability to use a named substring as a condition, using the
489        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
490        are numbers (not recommended). Forward references are permitted.
491    
492    14. Added forward references in named backreferences (if you see what I mean).
493    
494    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
495        pattern could run off the end of the subject. For example, the pattern
496        "(?s)(.{1,5})"8 did this with the subject "ab".
497    
498    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
499        PCRE_CASELESS was set when matching characters that were quantified with ?
500        or *.
501    
502    17. A character class other than a single negated character that had a minimum
503        but no maximum quantifier - for example [ab]{6,} - was not handled
504        correctly by pce_dfa_exec(). It would match only one character.
505    
506    18. A valid (though odd) pattern that looked like a POSIX character
507        class but used an invalid character after [ (for example [[,abc,]]) caused
508        pcre_compile() to give the error "Failed: internal error: code overflow" or
509        in some cases to crash with a glibc free() error. This could even happen if
510        the pattern terminated after [[ but there just happened to be a sequence of
511        letters, a binary zero, and a closing ] in the memory that followed.
512    
513    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
514        over the years. Originally (before any Unicode support), just the bottom 8
515        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
516        output from "man perlunicode" includes this:
517    
518          The regular expression compiler produces polymorphic opcodes.  That
519          is, the pattern adapts to the data and automatically switches to
520          the Unicode character scheme when presented with Unicode data--or
521          instead uses a traditional byte scheme when presented with byte
522          data.
523    
524        Sadly, a wide octal escape does not cause a switch, and in a string with
525        no other multibyte characters, these octal escapes are treated as before.
526        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
527        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
528        Unicode string.
529    
530        I have not perpetrated such confusion in PCRE. Up till now, it took just
531        the bottom 8 bits, as in old Perl. I have now made octal escapes with
532        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
533        translate to the appropriate multibyte character.
534    
535    29. Applied some refactoring to reduce the number of warnings from Microsoft
536        and Borland compilers. This has included removing the fudge introduced
537        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
538        a warning about an unused variable.
539    
540    21. PCRE has not included VT (character 0x0b) in the set of whitespace
541        characters since release 4.0, because Perl (from release 5.004) does not.
542        [Or at least, is documented not to: some releases seem to be in conflict
543        with the documentation.] However, when a pattern was studied with
544        pcre_study() and all its branches started with \s, PCRE still included VT
545        as a possible starting character. Of course, this did no harm; it just
546        caused an unnecessary match attempt.
547    
548    22. Removed a now-redundant internal flag bit that recorded the fact that case
549        dependency changed within the pattern. This was once needed for "required
550        byte" processing, but is no longer used. This recovers a now-scarce options
551        bit. Also moved the least significant internal flag bit to the most-
552        significant bit of the word, which was not previously used (hangover from
553        the days when it was an int rather than a uint) to free up another bit for
554        the future.
555    
556    23. Added support for CRLF line endings as well as CR and LF. As well as the
557        default being selectable at build time, it can now be changed at runtime
558        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
559        specify that it is scanning data with non-default line endings.
560    
561    24. Changed the definition of CXXLINK to make it agree with the definition of
562        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
563    
564    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
565        recursions. This makes a big different to stack usage for some patterns.
566    
567    26. If a subpattern containing a named recursion or subroutine reference such
568        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
569        the space required for the compiled pattern went wrong and gave too small a
570        value. Depending on the environment, this could lead to "Failed: internal
571        error: code overflow at offset 49" or "glibc detected double free or
572        corruption" errors.
573    
574    27. Applied patches from Google (a) to support the new newline modes and (b) to
575        advance over multibyte UTF-8 characters in GlobalReplace.
576    
577    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
578        difference for some implementation of PCRE in some Windows version.
579    
580    29. Added some extra testing facilities to pcretest:
581    
582        \q<number>   in a data line sets the "match limit" value
583        \Q<number>   in a data line sets the "match recursion limt" value
584        -S <number>  sets the stack size, where <number> is in megabytes
585    
586        The -S option isn't available for Windows.
587    
588    
589    Version 6.6 06-Feb-06
590    ---------------------
591    
592     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
593        in pcreposix.h. I have copied the definition from pcre.h.
594    
595     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
596        because pcre.h is no longer a built file.
597    
598     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
599        not normally included in the compiled code.
600    
601    
602    Version 6.5 01-Feb-06
603    ---------------------
604    
605     1. When using the partial match feature with pcre_dfa_exec(), it was not
606        anchoring the second and subsequent partial matches at the new starting
607        point. This could lead to incorrect results. For example, with the pattern
608        /1234/, partially matching against "123" and then "a4" gave a match.
609    
610     2. Changes to pcregrep:
611    
612        (a) All non-match returns from pcre_exec() were being treated as failures
613            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
614            error message is output. Some extra information is given for the
615            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
616            probably the only errors that are likely to be caused by users (by
617            specifying a regex that has nested indefinite repeats, for instance).
618            If there are more than 20 of these errors, pcregrep is abandoned.
619    
620        (b) A binary zero was treated as data while matching, but terminated the
621            output line if it was written out. This has been fixed: binary zeroes
622            are now no different to any other data bytes.
623    
624        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
625            used to set a locale for matching. The --locale=xxxx long option has
626            been added (no short equivalent) to specify a locale explicitly on the
627            pcregrep command, overriding the environment variables.
628    
629        (d) When -B was used with -n, some line numbers in the output were one less
630            than they should have been.
631    
632        (e) Added the -o (--only-matching) option.
633    
634        (f) If -A or -C was used with -c (count only), some lines of context were
635            accidentally printed for the final match.
636    
637        (g) Added the -H (--with-filename) option.
638    
639        (h) The combination of options -rh failed to suppress file names for files
640            that were found from directory arguments.
641    
642        (i) Added the -D (--devices) and -d (--directories) options.
643    
644        (j) Added the -F (--fixed-strings) option.
645    
646        (k) Allow "-" to be used as a file name for -f as well as for a data file.
647    
648        (l) Added the --colo(u)r option.
649    
650        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
651            is not present by default.
652    
653     3. A nasty bug was discovered in the handling of recursive patterns, that is,
654        items such as (?R) or (?1), when the recursion could match a number of
655        alternatives. If it matched one of the alternatives, but subsequently,
656        outside the recursion, there was a failure, the code tried to back up into
657        the recursion. However, because of the way PCRE is implemented, this is not
658        possible, and the result was an incorrect result from the match.
659    
660        In order to prevent this happening, the specification of recursion has
661        been changed so that all such subpatterns are automatically treated as
662        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
663    
664     4. I had overlooked the fact that, in some locales, there are characters for
665        which isalpha() is true but neither isupper() nor islower() are true. In
666        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
667        and ordfeminine) are like this. This affected the treatment of \w and \W
668        when they appeared in character classes, but not when they appeared outside
669        a character class. The bit map for "word" characters is now created
670        separately from the results of isalnum() instead of just taking it from the
671        upper, lower, and digit maps. (Plus the underscore character, of course.)
672    
673     5. The above bug also affected the handling of POSIX character classes such as
674        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
675        permanent tables. Instead, the bit maps for such a class were previously
676        created as the appropriate unions of the upper, lower, and digit bitmaps.
677        Now they are created by subtraction from the [[:word:]] class, which has
678        its own bitmap.
679    
680     6. The [[:blank:]] character class matches horizontal, but not vertical space.
681        It is created by subtracting the vertical space characters (\x09, \x0a,
682        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
683        subtraction was done in the overall bitmap for a character class, meaning
684        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
685        be recognized. This bug has been fixed.
686    
687     7. Patches from the folks at Google:
688    
689          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
690          real life, but is still worth protecting against".
691    
692          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
693          regular expressions".
694    
695          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
696          have it.
697    
698          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
699          "configure" and the latter not, in order to fix a problem somebody had
700          with compiling the Arg class on HP-UX.
701    
702          (e) Improve the error-handling of the C++ wrapper a little bit.
703    
704          (f) New tests for checking recursion limiting.
705    
706     8. The pcre_memmove() function, which is used only if the environment does not
707        have a standard memmove() function (and is therefore rarely compiled),
708        contained two bugs: (a) use of int instead of size_t, and (b) it was not
709        returning a result (though PCRE never actually uses the result).
710    
711     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
712        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
713        returned instead of calling malloc() with an overflowing number that would
714        most likely cause subsequent chaos.
715    
716    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
717    
718    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
719        with this option is matched, the nmatch and pmatch options of regexec() are
720        ignored.
721    
722    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
723        provided in case anyone wants to the the POSIX interface with UTF-8
724        strings.
725    
726    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
727        C++ linking (needed for some HP-UX environments).
728    
729    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
730        (unused parameter) and in the pcre_printint() function (omitted "default"
731        switch label when the default is to do nothing).
732    
733    15. Added some code to make it possible, when PCRE is compiled as a C++
734        library, to replace subject pointers for pcre_exec() with a smart pointer
735        class, thus making it possible to process discontinuous strings.
736    
737    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
738        much the same function. They were added by different people who were trying
739        to make PCRE easy to compile on non-Unix systems. It has been suggested
740        that PCRE_EXPORT be abolished now that there is more automatic apparatus
741        for compiling on Windows systems. I have therefore replaced it with
742        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
743        defaults to "extern" for C or "extern C" for C++, which works fine on
744        Unix-like systems. It is now possible to override the value of PCRE_DATA_
745        SCOPE with something explicit in config.h. In addition:
746    
747        (a) pcreposix.h still had just "extern" instead of either of these macros;
748            I have replaced it with PCRE_DATA_SCOPE.
749    
750        (b) Functions such as _pcre_xclass(), which are internal to the library,
751            but external in the C sense, all had PCRE_EXPORT in their definitions.
752            This is apparently wrong for the Windows case, so I have removed it.
753            (It makes no difference on Unix-like systems.)
754    
755    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
756        of recursive calls to match(). This is different to MATCH_LIMIT because
757        that limits the total number of calls to match(), not all of which increase
758        the depth of recursion. Limiting the recursion depth limits the amount of
759        stack (or heap if NO_RECURSE is set) that is used. The default can be set
760        when PCRE is compiled, and changed at run time. A patch from Google adds
761        this functionality to the C++ interface.
762    
763    18. Changes to the handling of Unicode character properties:
764    
765        (a) Updated the table to Unicode 4.1.0.
766    
767        (b) Recognize characters that are not in the table as "Cn" (undefined).
768    
769        (c) I revised the way the table is implemented to a much improved format
770            which includes recognition of ranges. It now supports the ranges that
771            are defined in UnicodeData.txt, and it also amalgamates other
772            characters into ranges. This has reduced the number of entries in the
773            table from around 16,000 to around 3,000, thus reducing its size
774            considerably. I realized I did not need to use a tree structure after
775            all - a binary chop search is just as efficient. Having reduced the
776            number of entries, I extended their size from 6 bytes to 8 bytes to
777            allow for more data.
778    
779        (d) Added support for Unicode script names via properties such as \p{Han}.
780    
781    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
782        matching that character.
783    
784    20. When matching a repeated Unicode property with a minimum greater than zero,
785        (for example \pL{2,}), PCRE could look past the end of the subject if it
786        reached it while seeking the minimum number of characters. This could
787        happen only if some of the characters were more than one byte long, because
788        there is a check for at least the minimum number of bytes.
789    
790    21. Refactored the implementation of \p and \P so as to be more general, to
791        allow for more different types of property in future. This has changed the
792        compiled form incompatibly. Anybody with saved compiled patterns that use
793        \p or \P will have to recompile them.
794    
795    22. Added "Any" and "L&" to the supported property types.
796    
797    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
798        but give a compile time error if the value is greater than 0xff.
799    
800    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
801        accidentally not being installed or uninstalled.
802    
803    25. The pcre.h file was built from pcre.h.in, but the only changes that were
804        made were to insert the current release number. This seemed silly, because
805        it made things harder for people building PCRE on systems that don't run
806        "configure". I have turned pcre.h into a distributed file, no longer built
807        by "configure", with the version identification directly included. There is
808        no longer a pcre.h.in file.
809    
810        However, this change necessitated a change to the pcre-config script as
811        well. It is built from pcre-config.in, and one of the substitutions was the
812        release number. I have updated configure.ac so that ./configure now finds
813        the release number by grepping pcre.h.
814    
815    26. Added the ability to run the tests under valgrind.
816    
817    
818  Version 6.4 05-Sep-05  Version 6.4 05-Sep-05
819  ---------------------  ---------------------
820    

Legend:
Removed from v.85  
changed lines
  Added in v.150

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12