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

Diff of /code/trunk/ChangeLog

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

revision 69 by nigel, Sat Feb 24 21:40:18 2007 UTC revision 192 by ph10, Mon Jul 30 11:23:21 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.3 05-Jul-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    
33    
34    Version 7.2 19-Jun-07
35    ---------------------
36    
37     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
38        which is apparently normally available under Windows.
39    
40     2. Re-jig the pcregrep tests with different newline settings in an attempt
41        to make them independent of the local environment's newline setting.
42    
43     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
44    
45     4. Some of the "internals" tests were previously cut out when the link size
46        was not 2, because the output contained actual offsets. The recent new
47        "Z" feature of pcretest means that these can be cut out, making the tests
48        usable with all link sizes.
49    
50     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
51        stack recursion. This gives a massive performance boost under BSD, but just
52        a small improvement under Linux. However, it saves one field in the frame
53        in all cases.
54    
55     6. Added more features from the forthcoming Perl 5.10:
56    
57        (a) (?-n) (where n is a string of digits) is a relative subroutine or
58            recursion call. It refers to the nth most recently opened parentheses.
59    
60        (b) (?+n) is also a relative subroutine call; it refers to the nth next
61            to be opened parentheses.
62    
63        (c) Conditions that refer to capturing parentheses can be specified
64            relatively, for example, (?(-2)... or (?(+3)...
65    
66        (d) \K resets the start of the current match so that everything before
67            is not part of it.
68    
69        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
70    
71        (f) \g{name} is another synonym - part of Perl 5.10's unification of
72            reference syntax.
73    
74        (g) (?| introduces a group in which the numbering of parentheses in each
75            alternative starts with the same number.
76    
77        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
78    
79     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
80        PCRE_INFO_JCHANGED.
81    
82     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
83        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
84        for detecting groups that can match an empty string.
85    
86     9. A pattern with a very large number of alternatives (more than several
87        hundred) was running out of internal workspace during the pre-compile
88        phase, where pcre_compile() figures out how much memory will be needed. A
89        bit of new cunning has reduced the workspace needed for groups with
90        alternatives. The 1000-alternative test pattern now uses 12 bytes of
91        workspace instead of running out of the 4096 that are available.
92    
93    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
94    
95    11. Applied patch from Google to remove an optimization that didn't quite work.
96        The report of the bug said:
97    
98          pcrecpp::RE("a*").FullMatch("aaa") matches, while
99          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
100          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
101    
102    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
103        it matched the wrong number of bytes.
104    
105    
106    Version 7.1 24-Apr-07
107    ---------------------
108    
109     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
110        that is more "standard", making use of automake and other Autotools. There
111        is some re-arrangement of the files and adjustment of comments consequent
112        on this.
113    
114     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
115        for recursive directory scanning broke on some systems because the files
116        are not scanned in any specific order and on different systems the order
117        was different. A call to "sort" has been inserted into RunGrepTest for the
118        approprate test as a short-term fix. In the longer term there may be an
119        alternative.
120    
121     3. I had an email from Eric Raymond about problems translating some of PCRE's
122        man pages to HTML (despite the fact that I distribute HTML pages, some
123        people do their own conversions for various reasons). The problems
124        concerned the use of low-level troff macros .br and .in. I have therefore
125        removed all such uses from the man pages (some were redundant, some could
126        be replaced by .nf/.fi pairs). The 132html script that I use to generate
127        HTML has been updated to handle .nf/.fi and to complain if it encounters
128        .br or .in.
129    
130     4. Updated comments in configure.ac that get placed in config.h.in and also
131        arranged for config.h to be included in the distribution, with the name
132        config.h.generic, for the benefit of those who have to compile without
133        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
134    
135     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
136        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
137        makevp.bat for the current PCRE, using the additional files
138        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
139    
140     6. A Windows user reported a minor discrepancy with test 2, which turned out
141        to be caused by a trailing space on an input line that had got lost in his
142        copy. The trailing space was an accident, so I've just removed it.
143    
144     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
145        that is needed.
146    
147     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
148        as "const" (a) because they are and (b) because it helps the PHP
149        maintainers who have recently made a script to detect big data structures
150        in the php code that should be moved to the .rodata section. I remembered
151        to update Builducptable as well, so it won't revert if ucptable.h is ever
152        re-created.
153    
154     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
155        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
156        order to be able to cut out the UTF-8 tables in the latter when UTF-8
157        support is not required. This saves 1.5-2K of code, which is important in
158        some applications.
159    
160        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
161        so as not to refer to the tables, even though these functions will never be
162        called when UTF-8 support is disabled. Otherwise there are problems with a
163        shared library.
164    
165    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
166    
167        (a) It was defining its arguments as char * instead of void *.
168    
169        (b) It was assuming that all moves were upwards in memory; this was true
170            a long time ago when I wrote it, but is no longer the case.
171    
172        The emulated memove() is provided for those environments that have neither
173        memmove() nor bcopy(). I didn't think anyone used it these days, but that
174        is clearly not the case, as these two bugs were recently reported.
175    
176    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
177        and Detrail to create the HTML documentation, the .txt form of the man
178        pages, and it removes trailing spaces from listed files. It also creates
179        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
180        case, it wraps all the #defines with #ifndefs. This script should be run
181        before "make dist".
182    
183    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
184        with Unicode property support.
185    
186        (a) For a maximizing quantifier, if the two different cases of the
187            character were of different lengths in their UTF-8 codings (there are
188            some cases like this - I found 11), and the matching function had to
189            back up over a mixture of the two cases, it incorrectly assumed they
190            were both the same length.
191    
192        (b) When PCRE was configured to use the heap rather than the stack for
193            recursion during matching, it was not correctly preserving the data for
194            the other case of a UTF-8 character when checking ahead for a match
195            while processing a minimizing repeat. If the check also involved
196            matching a wide character, but failed, corruption could cause an
197            erroneous result when trying to check for a repeat of the original
198            character.
199    
200    13. Some tidying changes to the testing mechanism:
201    
202        (a) The RunTest script now detects the internal link size and whether there
203            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
204            values substituted by "configure". (The RunGrepTest script already did
205            this for UTF-8.) The configure.ac script no longer substitutes the
206            relevant variables.
207    
208        (b) The debugging options /B and /D in pcretest show the compiled bytecode
209            with length and offset values. This means that the output is different
210            for different internal link sizes. Test 2 is skipped for link sizes
211            other than 2 because of this, bypassing the problem. Unfortunately,
212            there was also a test in test 3 (the locale tests) that used /B and
213            failed for link sizes other than 2. Rather than cut the whole test out,
214            I have added a new /Z option to pcretest that replaces the length and
215            offset values with spaces. This is now used to make test 3 independent
216            of link size. (Test 2 will be tidied up later.)
217    
218    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
219        segmentation fault instead of returning the appropriate error message.
220    
221    15. In multiline mode when the newline sequence was set to "any", the pattern
222        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
223        This doesn't seem right; it now treats the CRLF combination as the line
224        ending, and so does not match in that case. It's only a pattern such as ^$
225        that would hit this one: something like ^ABC$ would have failed after \r
226        and then tried again after \r\n.
227    
228    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
229        in an attempt to make files that differ only in their line terminators
230        compare equal. This works on Linux.
231    
232    17. Under certain error circumstances pcregrep might try to free random memory
233        as it exited. This is now fixed, thanks to valgrind.
234    
235    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
236        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
237        was because its rules for how to advance for /g after matching an empty
238        string at the end of a line did not allow for this case. They now check for
239        it specially.
240    
241    20. pcretest is supposed to handle patterns and data of any length, by
242        extending its buffers when necessary. It was getting this wrong when the
243        buffer for a data line had to be extended.
244    
245    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
246        CRLF as a newline sequence.
247    
248    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
249        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
250        I have nevertheless tidied it up.
251    
252    23. Added some casts to kill warnings from HP-UX ia64 compiler.
253    
254    24. Added a man page for pcre-config.
255    
256    
257    Version 7.0 19-Dec-06
258    ---------------------
259    
260     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
261        moving to gcc 4.1.1.
262    
263     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
264        sys/time.h, which is documented as needed for this function. It doesn't
265        seem to matter on Linux, but it showed up on some releases of OS X.
266    
267     3. It seems that there are systems where bytes whose values are greater than
268        127 match isprint() in the "C" locale. The "C" locale should be the
269        default when a C program starts up. In most systems, only ASCII printing
270        characters match isprint(). This difference caused the output from pcretest
271        to vary, making some of the tests fail. I have changed pcretest so that:
272    
273        (a) When it is outputting text in the compiled version of a pattern, bytes
274            other than 32-126 are always shown as hex escapes.
275    
276        (b) When it is outputting text that is a matched part of a subject string,
277            it does the same, unless a different locale has been set for the match
278            (using the /L modifier). In this case, it uses isprint() to decide.
279    
280     4. Fixed a major bug that caused incorrect computation of the amount of memory
281        required for a compiled pattern when options that changed within the
282        pattern affected the logic of the preliminary scan that determines the
283        length. The relevant options are -x, and -i in UTF-8 mode. The result was
284        that the computed length was too small. The symptoms of this bug were
285        either the PCRE error "internal error: code overflow" from pcre_compile(),
286        or a glibc crash with a message such as "pcretest: free(): invalid next
287        size (fast)". Examples of patterns that provoked this bug (shown in
288        pcretest format) are:
289    
290          /(?-x: )/x
291          /(?x)(?-x: \s*#\s*)/
292          /((?i)[\x{c0}])/8
293          /(?i:[\x{c0}])/8
294    
295        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
296        is now done differently.
297    
298     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
299        wrapper classes; (b) implement a new function in the C++ scanner that is
300        more efficient than the old way of doing things because it avoids levels of
301        recursion in the regex matching; (c) add a paragraph to the documentation
302        for the FullMatch() function.
303    
304     6. The escape sequence \n was being treated as whatever was defined as
305        "newline". Not only was this contrary to the documentation, which states
306        that \n is character 10 (hex 0A), but it also went horribly wrong when
307        "newline" was defined as CRLF. This has been fixed.
308    
309     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
310        was being set to -1 for the "end of line" case (supposedly a value that no
311        character can have). Though this value is never used (the check for end of
312        line is "zero bytes in current character"), it caused compiler complaints.
313        I've changed it to 0xffffffff.
314    
315     8. In pcre_version.c, the version string was being built by a sequence of
316        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
317        string (as it is for production releases) called a macro with an empty
318        argument. The C standard says the result of this is undefined. The gcc
319        compiler treats it as an empty string (which was what was wanted) but it is
320        reported that Visual C gives an error. The source has been hacked around to
321        avoid this problem.
322    
323     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
324        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
325        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
326        of them did).
327    
328    10. Originally, pcretest opened its input and output without "b"; then I was
329        told that "b" was needed in some environments, so it was added for release
330        5.0 to both the input and output. (It makes no difference on Unix-like
331        systems.) Later I was told that it is wrong for the input on Windows. I've
332        now abstracted the modes into two macros, to make it easier to fiddle with
333        them, and removed "b" from the input mode under Windows.
334    
335    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
336    
337    12. Added -help and --help to pcretest as an official way of being reminded
338        of the options.
339    
340    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
341        and pcrecpp.cc because they annoy compilers at high warning levels.
342    
343    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
344    
345    15. Fixed an occurrence of == in configure.ac that should have been = (shell
346        scripts are not C programs :-) and which was not noticed because it works
347        on Linux.
348    
349    16. pcretest is supposed to handle any length of pattern and data line (as one
350        line or as a continued sequence of lines) by extending its input buffer if
351        necessary. This feature was broken for very long pattern lines, leading to
352        a string of junk being passed to pcre_compile() if the pattern was longer
353        than about 50K.
354    
355    17. I have done a major re-factoring of the way pcre_compile() computes the
356        amount of memory needed for a compiled pattern. Previously, there was code
357        that made a preliminary scan of the pattern in order to do this. That was
358        OK when PCRE was new, but as the facilities have expanded, it has become
359        harder and harder to keep it in step with the real compile phase, and there
360        have been a number of bugs (see for example, 4 above). I have now found a
361        cunning way of running the real compile function in a "fake" mode that
362        enables it to compute how much memory it would need, while actually only
363        ever using a few hundred bytes of working memory and without too many
364        tests of the mode. This should make future maintenance and development
365        easier. A side effect of this work is that the limit of 200 on the nesting
366        depth of parentheses has been removed (though this was never a serious
367        limitation, I suspect). However, there is a downside: pcre_compile() now
368        runs more slowly than before (30% or more, depending on the pattern). I
369        hope this isn't a big issue. There is no effect on runtime performance.
370    
371    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
372        newline (only possible for the last line of a file) and it was a
373        pattern that set a locale (followed by /Lsomething), pcretest crashed.
374    
375    19. Added additional timing features to pcretest. (1) The -tm option now times
376        matching only, not compiling. (2) Both -t and -tm can be followed, as a
377        separate command line item, by a number that specifies the number of
378        repeats to use when timing. The default is 50000; this gives better
379        precision, but takes uncomfortably long for very large patterns.
380    
381    20. Extended pcre_study() to be more clever in cases where a branch of a
382        subpattern has no definite first character. For example, (a*|b*)[cd] would
383        previously give no result from pcre_study(). Now it recognizes that the
384        first character must be a, b, c, or d.
385    
386    21. There was an incorrect error "recursive call could loop indefinitely" if
387        a subpattern (or the entire pattern) that was being tested for matching an
388        empty string contained only one non-empty item after a nested subpattern.
389        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
390        incorrectly, because the \d was being skipped in the check.
391    
392    22. The pcretest program now has a new pattern option /B and a command line
393        option -b, which is equivalent to adding /B to every pattern. This causes
394        it to show the compiled bytecode, without the additional information that
395        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
396        is the same as /B/I).
397    
398    23. A new optimization is now able automatically to treat some sequences such
399        as a*b as a*+b. More specifically, if something simple (such as a character
400        or a simple class like \d) has an unlimited quantifier, and is followed by
401        something that cannot possibly match the quantified thing, the quantifier
402        is automatically "possessified".
403    
404    24. A recursive reference to a subpattern whose number was greater than 39
405        went wrong under certain circumstances in UTF-8 mode. This bug could also
406        have affected the operation of pcre_study().
407    
408    25. Realized that a little bit of performance could be had by replacing
409        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
410    
411    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
412    
413    27. Possessive quantifiers such as a++ were previously implemented by turning
414        them into atomic groups such as ($>a+). Now they have their own opcodes,
415        which improves performance. This includes the automatically created ones
416        from 23 above.
417    
418    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
419        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
420        the first matched character to be a colon. This applied both to named and
421        numbered groups.
422    
423    29. The ucpinternal.h header file was missing its idempotency #ifdef.
424    
425    30. I was sent a "project" file called libpcre.a.dev which I understand makes
426        building PCRE on Windows easier, so I have included it in the distribution.
427    
428    31. There is now a check in pcretest against a ridiculously large number being
429        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
430        loop, the loop is abandoned.
431    
432    32. Forward references to subpatterns in conditions such as (?(2)...) where
433        subpattern 2 is defined later cause pcre_compile() to search forwards in
434        the pattern for the relevant set of parentheses. This search went wrong
435        when there were unescaped parentheses in a character class, parentheses
436        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
437    
438    33. "Subroutine" calls and backreferences were previously restricted to
439        referencing subpatterns earlier in the regex. This restriction has now
440        been removed.
441    
442    34. Added a number of extra features that are going to be in Perl 5.10. On the
443        whole, these are just syntactic alternatives for features that PCRE had
444        previously implemented using the Python syntax or my own invention. The
445        other formats are all retained for compatibility.
446    
447        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
448            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
449            also .NET compatible.
450    
451        (b) A recursion or subroutine call to a named group can now be defined as
452            (?&name) as well as (?P>name).
453    
454        (c) A backreference to a named group can now be defined as \k<name> or
455            \k'name' as well as (?P=name). The new forms, as well as being in Perl
456            5.10, are also .NET compatible.
457    
458        (d) A conditional reference to a named group can now use the syntax
459            (?(<name>) or (?('name') as well as (?(name).
460    
461        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
462            groups (named and numbered) that are never evaluated inline, but can be
463            called as "subroutines" from elsewhere. In effect, the DEFINE condition
464            is always false. There may be only one alternative in such a group.
465    
466        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
467            as the simple (?(R). The condition is true only if the most recent
468            recursion is that of the given number or name. It does not search out
469            through the entire recursion stack.
470    
471        (g) The escape \gN or \g{N} has been added, where N is a positive or
472            negative number, specifying an absolute or relative reference.
473    
474    35. Tidied to get rid of some further signed/unsigned compiler warnings and
475        some "unreachable code" warnings.
476    
477    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
478        things, this adds five new scripts.
479    
480    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
481        There were also incompatibilities regarding the handling of \Q..\E inside
482        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
483        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
484    
485    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
486        matches an empty string, and forcibly breaks the loop. There were bugs in
487        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
488        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
489        separate and independent bugs (that affected different cases) have been
490        fixed.
491    
492    39. Refactored the code to abolish the use of different opcodes for small
493        capturing bracket numbers. This is a tidy that I avoided doing when I
494        removed the limit on the number of capturing brackets for 3.5 back in 2001.
495        The new approach is not only tidier, it makes it possible to reduce the
496        memory needed to fix the previous bug (38).
497    
498    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
499        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
500        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
501        mode.
502    
503    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
504        report.
505    
506    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
507        copy construction and assignment in the C++ wrapper.
508    
509    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
510        couple of bugs that could have given wrong results in the "--newline=crlf"
511        case.
512    
513    44. Added a number of casts and did some reorganization of signed/unsigned int
514        variables following suggestions from Dair Grant. Also renamed the variable
515        "this" as "item" because it is a C++ keyword.
516    
517    45. Arranged for dftables to add
518    
519          #include "pcre_internal.h"
520    
521        to pcre_chartables.c because without it, gcc 4.x may remove the array
522        definition from the final binary if PCRE is built into a static library and
523        dead code stripping is activated.
524    
525    46. For an unanchored pattern, if a match attempt fails at the start of a
526        newline sequence, and the newline setting is CRLF or ANY, and the next two
527        characters are CRLF, advance by two characters instead of one.
528    
529    
530    Version 6.7 04-Jul-06
531    ---------------------
532    
533     1. In order to handle tests when input lines are enormously long, pcretest has
534        been re-factored so that it automatically extends its buffers when
535        necessary. The code is crude, but this _is_ just a test program. The
536        default size has been increased from 32K to 50K.
537    
538     2. The code in pcre_study() was using the value of the re argument before
539        testing it for NULL. (Of course, in any sensible call of the function, it
540        won't be NULL.)
541    
542     3. The memmove() emulation function in pcre_internal.h, which is used on
543        systems that lack both memmove() and bcopy() - that is, hardly ever -
544        was missing a "static" storage class specifier.
545    
546     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
547        containing an extended class (one that cannot be represented by a bitmap
548        because it contains high-valued characters or Unicode property items, e.g.
549        [\pZ]). Almost always one would set UTF-8 mode when processing such a
550        pattern, but PCRE should not loop if you do not (it no longer does).
551        [Detail: two cases were found: (a) a repeated subpattern containing an
552        extended class; (b) a recursive reference to a subpattern that followed a
553        previous extended class. It wasn't skipping over the extended class
554        correctly when UTF-8 mode was not set.]
555    
556     5. A negated single-character class was not being recognized as fixed-length
557        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
558        compile error "lookbehind assertion is not fixed length".
559    
560     6. The RunPerlTest auxiliary script was showing an unexpected difference
561        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
562        write a Perl script that can interpret lines of an input file either as
563        byte characters or as UTF-8, which is what "perltest" was being required to
564        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
565        can't do is switch easily at run time between having the "use utf8;" pragma
566        or not. In the end, I fudged it by using the RunPerlTest script to insert
567        "use utf8;" explicitly for the UTF-8 tests.
568    
569     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
570        the end of the subject string, contrary to the documentation and to what
571        Perl does. This was true of both matching functions. Now it matches only at
572        the start of the subject and immediately after *internal* newlines.
573    
574     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
575        a pointer to an int instead of a pointer to an unsigned long int. This
576        caused problems on 64-bit systems.
577    
578     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
579        instance of the 'standard' template library not being so standard".
580    
581    10. There was no check on the number of named subpatterns nor the maximum
582        length of a subpattern name. The product of these values is used to compute
583        the size of the memory block for a compiled pattern. By supplying a very
584        long subpattern name and a large number of named subpatterns, the size
585        computation could be caused to overflow. This is now prevented by limiting
586        the length of names to 32 characters, and the number of named subpatterns
587        to 10,000.
588    
589    11. Subpatterns that are repeated with specific counts have to be replicated in
590        the compiled pattern. The size of memory for this was computed from the
591        length of the subpattern and the repeat count. The latter is limited to
592        65535, but there was no limit on the former, meaning that integer overflow
593        could in principle occur. The compiled length of a repeated subpattern is
594        now limited to 30,000 bytes in order to prevent this.
595    
596    12. Added the optional facility to have named substrings with the same name.
597    
598    13. Added the ability to use a named substring as a condition, using the
599        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
600        are numbers (not recommended). Forward references are permitted.
601    
602    14. Added forward references in named backreferences (if you see what I mean).
603    
604    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
605        pattern could run off the end of the subject. For example, the pattern
606        "(?s)(.{1,5})"8 did this with the subject "ab".
607    
608    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
609        PCRE_CASELESS was set when matching characters that were quantified with ?
610        or *.
611    
612    17. A character class other than a single negated character that had a minimum
613        but no maximum quantifier - for example [ab]{6,} - was not handled
614        correctly by pce_dfa_exec(). It would match only one character.
615    
616    18. A valid (though odd) pattern that looked like a POSIX character
617        class but used an invalid character after [ (for example [[,abc,]]) caused
618        pcre_compile() to give the error "Failed: internal error: code overflow" or
619        in some cases to crash with a glibc free() error. This could even happen if
620        the pattern terminated after [[ but there just happened to be a sequence of
621        letters, a binary zero, and a closing ] in the memory that followed.
622    
623    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
624        over the years. Originally (before any Unicode support), just the bottom 8
625        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
626        output from "man perlunicode" includes this:
627    
628          The regular expression compiler produces polymorphic opcodes.  That
629          is, the pattern adapts to the data and automatically switches to
630          the Unicode character scheme when presented with Unicode data--or
631          instead uses a traditional byte scheme when presented with byte
632          data.
633    
634        Sadly, a wide octal escape does not cause a switch, and in a string with
635        no other multibyte characters, these octal escapes are treated as before.
636        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
637        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
638        Unicode string.
639    
640        I have not perpetrated such confusion in PCRE. Up till now, it took just
641        the bottom 8 bits, as in old Perl. I have now made octal escapes with
642        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
643        translate to the appropriate multibyte character.
644    
645    29. Applied some refactoring to reduce the number of warnings from Microsoft
646        and Borland compilers. This has included removing the fudge introduced
647        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
648        a warning about an unused variable.
649    
650    21. PCRE has not included VT (character 0x0b) in the set of whitespace
651        characters since release 4.0, because Perl (from release 5.004) does not.
652        [Or at least, is documented not to: some releases seem to be in conflict
653        with the documentation.] However, when a pattern was studied with
654        pcre_study() and all its branches started with \s, PCRE still included VT
655        as a possible starting character. Of course, this did no harm; it just
656        caused an unnecessary match attempt.
657    
658    22. Removed a now-redundant internal flag bit that recorded the fact that case
659        dependency changed within the pattern. This was once needed for "required
660        byte" processing, but is no longer used. This recovers a now-scarce options
661        bit. Also moved the least significant internal flag bit to the most-
662        significant bit of the word, which was not previously used (hangover from
663        the days when it was an int rather than a uint) to free up another bit for
664        the future.
665    
666    23. Added support for CRLF line endings as well as CR and LF. As well as the
667        default being selectable at build time, it can now be changed at runtime
668        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
669        specify that it is scanning data with non-default line endings.
670    
671    24. Changed the definition of CXXLINK to make it agree with the definition of
672        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
673    
674    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
675        recursions. This makes a big different to stack usage for some patterns.
676    
677    26. If a subpattern containing a named recursion or subroutine reference such
678        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
679        the space required for the compiled pattern went wrong and gave too small a
680        value. Depending on the environment, this could lead to "Failed: internal
681        error: code overflow at offset 49" or "glibc detected double free or
682        corruption" errors.
683    
684    27. Applied patches from Google (a) to support the new newline modes and (b) to
685        advance over multibyte UTF-8 characters in GlobalReplace.
686    
687    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
688        difference for some implementation of PCRE in some Windows version.
689    
690    29. Added some extra testing facilities to pcretest:
691    
692        \q<number>   in a data line sets the "match limit" value
693        \Q<number>   in a data line sets the "match recursion limt" value
694        -S <number>  sets the stack size, where <number> is in megabytes
695    
696        The -S option isn't available for Windows.
697    
698    
699    Version 6.6 06-Feb-06
700    ---------------------
701    
702     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
703        in pcreposix.h. I have copied the definition from pcre.h.
704    
705     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
706        because pcre.h is no longer a built file.
707    
708     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
709        not normally included in the compiled code.
710    
711    
712    Version 6.5 01-Feb-06
713    ---------------------
714    
715     1. When using the partial match feature with pcre_dfa_exec(), it was not
716        anchoring the second and subsequent partial matches at the new starting
717        point. This could lead to incorrect results. For example, with the pattern
718        /1234/, partially matching against "123" and then "a4" gave a match.
719    
720     2. Changes to pcregrep:
721    
722        (a) All non-match returns from pcre_exec() were being treated as failures
723            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
724            error message is output. Some extra information is given for the
725            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
726            probably the only errors that are likely to be caused by users (by
727            specifying a regex that has nested indefinite repeats, for instance).
728            If there are more than 20 of these errors, pcregrep is abandoned.
729    
730        (b) A binary zero was treated as data while matching, but terminated the
731            output line if it was written out. This has been fixed: binary zeroes
732            are now no different to any other data bytes.
733    
734        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
735            used to set a locale for matching. The --locale=xxxx long option has
736            been added (no short equivalent) to specify a locale explicitly on the
737            pcregrep command, overriding the environment variables.
738    
739        (d) When -B was used with -n, some line numbers in the output were one less
740            than they should have been.
741    
742        (e) Added the -o (--only-matching) option.
743    
744        (f) If -A or -C was used with -c (count only), some lines of context were
745            accidentally printed for the final match.
746    
747        (g) Added the -H (--with-filename) option.
748    
749        (h) The combination of options -rh failed to suppress file names for files
750            that were found from directory arguments.
751    
752        (i) Added the -D (--devices) and -d (--directories) options.
753    
754        (j) Added the -F (--fixed-strings) option.
755    
756        (k) Allow "-" to be used as a file name for -f as well as for a data file.
757    
758        (l) Added the --colo(u)r option.
759    
760        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
761            is not present by default.
762    
763     3. A nasty bug was discovered in the handling of recursive patterns, that is,
764        items such as (?R) or (?1), when the recursion could match a number of
765        alternatives. If it matched one of the alternatives, but subsequently,
766        outside the recursion, there was a failure, the code tried to back up into
767        the recursion. However, because of the way PCRE is implemented, this is not
768        possible, and the result was an incorrect result from the match.
769    
770        In order to prevent this happening, the specification of recursion has
771        been changed so that all such subpatterns are automatically treated as
772        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
773    
774     4. I had overlooked the fact that, in some locales, there are characters for
775        which isalpha() is true but neither isupper() nor islower() are true. In
776        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
777        and ordfeminine) are like this. This affected the treatment of \w and \W
778        when they appeared in character classes, but not when they appeared outside
779        a character class. The bit map for "word" characters is now created
780        separately from the results of isalnum() instead of just taking it from the
781        upper, lower, and digit maps. (Plus the underscore character, of course.)
782    
783     5. The above bug also affected the handling of POSIX character classes such as
784        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
785        permanent tables. Instead, the bit maps for such a class were previously
786        created as the appropriate unions of the upper, lower, and digit bitmaps.
787        Now they are created by subtraction from the [[:word:]] class, which has
788        its own bitmap.
789    
790     6. The [[:blank:]] character class matches horizontal, but not vertical space.
791        It is created by subtracting the vertical space characters (\x09, \x0a,
792        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
793        subtraction was done in the overall bitmap for a character class, meaning
794        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
795        be recognized. This bug has been fixed.
796    
797     7. Patches from the folks at Google:
798    
799          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
800          real life, but is still worth protecting against".
801    
802          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
803          regular expressions".
804    
805          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
806          have it.
807    
808          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
809          "configure" and the latter not, in order to fix a problem somebody had
810          with compiling the Arg class on HP-UX.
811    
812          (e) Improve the error-handling of the C++ wrapper a little bit.
813    
814          (f) New tests for checking recursion limiting.
815    
816     8. The pcre_memmove() function, which is used only if the environment does not
817        have a standard memmove() function (and is therefore rarely compiled),
818        contained two bugs: (a) use of int instead of size_t, and (b) it was not
819        returning a result (though PCRE never actually uses the result).
820    
821     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
822        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
823        returned instead of calling malloc() with an overflowing number that would
824        most likely cause subsequent chaos.
825    
826    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
827    
828    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
829        with this option is matched, the nmatch and pmatch options of regexec() are
830        ignored.
831    
832    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
833        provided in case anyone wants to the the POSIX interface with UTF-8
834        strings.
835    
836    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
837        C++ linking (needed for some HP-UX environments).
838    
839    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
840        (unused parameter) and in the pcre_printint() function (omitted "default"
841        switch label when the default is to do nothing).
842    
843    15. Added some code to make it possible, when PCRE is compiled as a C++
844        library, to replace subject pointers for pcre_exec() with a smart pointer
845        class, thus making it possible to process discontinuous strings.
846    
847    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
848        much the same function. They were added by different people who were trying
849        to make PCRE easy to compile on non-Unix systems. It has been suggested
850        that PCRE_EXPORT be abolished now that there is more automatic apparatus
851        for compiling on Windows systems. I have therefore replaced it with
852        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
853        defaults to "extern" for C or "extern C" for C++, which works fine on
854        Unix-like systems. It is now possible to override the value of PCRE_DATA_
855        SCOPE with something explicit in config.h. In addition:
856    
857        (a) pcreposix.h still had just "extern" instead of either of these macros;
858            I have replaced it with PCRE_DATA_SCOPE.
859    
860        (b) Functions such as _pcre_xclass(), which are internal to the library,
861            but external in the C sense, all had PCRE_EXPORT in their definitions.
862            This is apparently wrong for the Windows case, so I have removed it.
863            (It makes no difference on Unix-like systems.)
864    
865    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
866        of recursive calls to match(). This is different to MATCH_LIMIT because
867        that limits the total number of calls to match(), not all of which increase
868        the depth of recursion. Limiting the recursion depth limits the amount of
869        stack (or heap if NO_RECURSE is set) that is used. The default can be set
870        when PCRE is compiled, and changed at run time. A patch from Google adds
871        this functionality to the C++ interface.
872    
873    18. Changes to the handling of Unicode character properties:
874    
875        (a) Updated the table to Unicode 4.1.0.
876    
877        (b) Recognize characters that are not in the table as "Cn" (undefined).
878    
879        (c) I revised the way the table is implemented to a much improved format
880            which includes recognition of ranges. It now supports the ranges that
881            are defined in UnicodeData.txt, and it also amalgamates other
882            characters into ranges. This has reduced the number of entries in the
883            table from around 16,000 to around 3,000, thus reducing its size
884            considerably. I realized I did not need to use a tree structure after
885            all - a binary chop search is just as efficient. Having reduced the
886            number of entries, I extended their size from 6 bytes to 8 bytes to
887            allow for more data.
888    
889        (d) Added support for Unicode script names via properties such as \p{Han}.
890    
891    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
892        matching that character.
893    
894    20. When matching a repeated Unicode property with a minimum greater than zero,
895        (for example \pL{2,}), PCRE could look past the end of the subject if it
896        reached it while seeking the minimum number of characters. This could
897        happen only if some of the characters were more than one byte long, because
898        there is a check for at least the minimum number of bytes.
899    
900    21. Refactored the implementation of \p and \P so as to be more general, to
901        allow for more different types of property in future. This has changed the
902        compiled form incompatibly. Anybody with saved compiled patterns that use
903        \p or \P will have to recompile them.
904    
905    22. Added "Any" and "L&" to the supported property types.
906    
907    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
908        but give a compile time error if the value is greater than 0xff.
909    
910    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
911        accidentally not being installed or uninstalled.
912    
913    25. The pcre.h file was built from pcre.h.in, but the only changes that were
914        made were to insert the current release number. This seemed silly, because
915        it made things harder for people building PCRE on systems that don't run
916        "configure". I have turned pcre.h into a distributed file, no longer built
917        by "configure", with the version identification directly included. There is
918        no longer a pcre.h.in file.
919    
920        However, this change necessitated a change to the pcre-config script as
921        well. It is built from pcre-config.in, and one of the substitutions was the
922        release number. I have updated configure.ac so that ./configure now finds
923        the release number by grepping pcre.h.
924    
925    26. Added the ability to run the tests under valgrind.
926    
927    
928    Version 6.4 05-Sep-05
929    ---------------------
930    
931     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
932        "--" to be printed when multiple files were scanned, even when none of the
933        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
934        consider it to be a bug, and have restored the previous behaviour.
935    
936     2. A couple of code tidies to get rid of compiler warnings.
937    
938     3. The pcretest program used to cheat by referring to symbols in the library
939        whose names begin with _pcre_. These are internal symbols that are not
940        really supposed to be visible externally, and in some environments it is
941        possible to suppress them. The cheating is now confined to including
942        certain files from the library's source, which is a bit cleaner.
943    
944     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
945        file's purpose clearer.
946    
947     5. Reorganized pcre_ucp_findchar().
948    
949    
950    Version 6.3 15-Aug-05
951    ---------------------
952    
953     1. The file libpcre.pc.in did not have general read permission in the tarball.
954    
955     2. There were some problems when building without C++ support:
956    
957        (a) If C++ support was not built, "make install" and "make test" still
958            tried to test it.
959    
960        (b) There were problems when the value of CXX was explicitly set. Some
961            changes have been made to try to fix these, and ...
962    
963        (c) --disable-cpp can now be used to explicitly disable C++ support.
964    
965        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
966            backslash in a target when C++ was disabled. This confuses some
967            versions of "make", apparently. Using an intermediate variable solves
968            this. (Same for CPP_LOBJ.)
969    
970     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
971        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
972        necessary on certain architectures.
973    
974     4. Added a setting of -export-symbols-regex to the link command to remove
975        those symbols that are exported in the C sense, but actually are local
976        within the library, and not documented. Their names all begin with
977        "_pcre_". This is not a perfect job, because (a) we have to except some
978        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
979        available (and never for static libraries). I have made a note to try to
980        find a way round (a) in the future.
981    
982    
983    Version 6.2 01-Aug-05
984    ---------------------
985    
986     1. There was no test for integer overflow of quantifier values. A construction
987        such as {1111111111111111} would give undefined results. What is worse, if
988        a minimum quantifier for a parenthesized subpattern overflowed and became
989        negative, the calculation of the memory size went wrong. This could have
990        led to memory overwriting.
991    
992     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
993    
994     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
995        operating environments where this matters.
996    
997     4. Applied Giuseppe Maxia's patch to add additional features for controlling
998        PCRE options from within the C++ wrapper.
999    
1000     5. Named capturing subpatterns were not being correctly counted when a pattern
1001        was compiled. This caused two problems: (a) If there were more than 100
1002        such subpatterns, the calculation of the memory needed for the whole
1003        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1004        back references of the form \12, where the number was greater than 9, were
1005        not recognized as back references, even though there were sufficient
1006        previous subpatterns.
1007    
1008     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1009        versions of gcc, e.g. 2.95.4.
1010    
1011    
1012    Version 6.1 21-Jun-05
1013    ---------------------
1014    
1015     1. There was one reference to the variable "posix" in pcretest.c that was not
1016        surrounded by "#if !defined NOPOSIX".
1017    
1018     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1019        the cross-check on the old pcre_info() function, for the benefit of the
1020        cut-down version of PCRE that is currently imported into Exim.
1021    
1022     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1023        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1024        patterns that start (?i) but I don't think that matters. The use of (?i) is
1025        just an example; this all applies to the other options as well.
1026    
1027     4. Since libtool seems to echo the compile commands it is issuing, the output
1028        from "make" can be reduced a bit by putting "@" in front of each libtool
1029        compile command.
1030    
1031     5. Patch from the folks at Google for configure.in to be a bit more thorough
1032        in checking for a suitable C++ installation before trying to compile the
1033        C++ stuff. This should fix a reported problem when a compiler was present,
1034        but no suitable headers.
1035    
1036     6. The man pages all had just "PCRE" as their title. I have changed them to
1037        be the relevant file name. I have also arranged that these names are
1038        retained in the file doc/pcre.txt, which is a concatenation in text format
1039        of all the man pages except the little individual ones for each function.
1040    
1041     7. The NON-UNIX-USE file had not been updated for the different set of source
1042        files that come with release 6. I also added a few comments about the C++
1043        wrapper.
1044    
1045    
1046    Version 6.0 07-Jun-05
1047    ---------------------
1048    
1049     1. Some minor internal re-organization to help with my DFA experiments.
1050    
1051     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1052        didn't matter for the library itself when fully configured, but did matter
1053        when compiling without UCP support, or within Exim, where the ucp files are
1054        not imported.
1055    
1056     3. Refactoring of the library code to split up the various functions into
1057        different source modules. The addition of the new DFA matching code (see
1058        below) to a single monolithic source would have made it really too
1059        unwieldy, quite apart from causing all the code to be include in a
1060        statically linked application, when only some functions are used. This is
1061        relevant even without the DFA addition now that patterns can be compiled in
1062        one application and matched in another.
1063    
1064        The downside of splitting up is that there have to be some external
1065        functions and data tables that are used internally in different modules of
1066        the library but which are not part of the API. These have all had their
1067        names changed to start with "_pcre_" so that they are unlikely to clash
1068        with other external names.
1069    
1070     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1071        a different (DFA) algorithm. Although it is slower than the original
1072        function, it does have some advantages for certain types of matching
1073        problem.
1074    
1075     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1076        including restarting after a partial match.
1077    
1078     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1079        defined when compiling for Windows was sent to me. I have put it into the
1080        code, though I have no means of testing or verifying it.
1081    
1082     7. Added the pcre_refcount() auxiliary function.
1083    
1084     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1085        match before or at the first newline in the subject string. In pcretest,
1086        the /f option on a pattern can be used to set this.
1087    
1088     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1089        would behave wrongly. This has been present in PCRE since release 4.0.
1090    
1091    10. A number of changes to the pcregrep command:
1092    
1093        (a) Refactored how -x works; insert ^(...)$ instead of setting
1094            PCRE_ANCHORED and checking the length, in preparation for adding
1095            something similar for -w.
1096    
1097        (b) Added the -w (match as a word) option.
1098    
1099        (c) Refactored the way lines are read and buffered so as to have more
1100            than one at a time available.
1101    
1102        (d) Implemented a pcregrep test script.
1103    
1104        (e) Added the -M (multiline match) option. This allows patterns to match
1105            over several lines of the subject. The buffering ensures that at least
1106            8K, or the rest of the document (whichever is the shorter) is available
1107            for matching (and similarly the previous 8K for lookbehind assertions).
1108    
1109        (f) Changed the --help output so that it now says
1110    
1111              -w, --word-regex(p)
1112    
1113            instead of two lines, one with "regex" and the other with "regexp"
1114            because that confused at least one person since the short forms are the
1115            same. (This required a bit of code, as the output is generated
1116            automatically from a table. It wasn't just a text change.)
1117    
1118        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1119            option but starts with a hyphen. Could be a pattern or a path name
1120            starting with a hyphen, for instance.
1121    
1122        (h) "-" can be given as a file name to represent stdin.
1123    
1124        (i) When file names are being printed, "(standard input)" is used for
1125            the standard input, for compatibility with GNU grep. Previously
1126            "<stdin>" was used.
1127    
1128        (j) The option --label=xxx can be used to supply a name to be used for
1129            stdin when file names are being printed. There is no short form.
1130    
1131        (k) Re-factored the options decoding logic because we are going to add
1132            two more options that take data. Such options can now be given in four
1133            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1134    
1135        (l) Added the -A, -B, and -C options for requesting that lines of context
1136            around matches be printed.
1137    
1138        (m) Added the -L option to print the names of files that do not contain
1139            any matching lines, that is, the complement of -l.
1140    
1141        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1142            continue to scan other files.
1143    
1144        (o) The -s option was incorrectly implemented. For compatibility with other
1145            greps, it now suppresses the error message for a non-existent or non-
1146            accessible file (but not the return code). There is a new option called
1147            -q that suppresses the output of matching lines, which was what -s was
1148            previously doing.
1149    
1150        (p) Added --include and --exclude options to specify files for inclusion
1151            and exclusion when recursing.
1152    
1153    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1154        Hopefully, it now does.
1155    
1156    12. Missing cast in pcre_study().
1157    
1158    13. Added an "uninstall" target to the makefile.
1159    
1160    14. Replaced "extern" in the function prototypes in Makefile.in with
1161        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1162        world, but is set differently for Windows.
1163    
1164    15. Added a second compiling function called pcre_compile2(). The only
1165        difference is that it has an extra argument, which is a pointer to an
1166        integer error code. When there is a compile-time failure, this is set
1167        non-zero, in addition to the error test pointer being set to point to an
1168        error message. The new argument may be NULL if no error number is required
1169        (but then you may as well call pcre_compile(), which is now just a
1170        wrapper). This facility is provided because some applications need a
1171        numeric error indication, but it has also enabled me to tidy up the way
1172        compile-time errors are handled in the POSIX wrapper.
1173    
1174    16. Added VPATH=.libs to the makefile; this should help when building with one
1175        prefix path and installing with another. (Or so I'm told by someone who
1176        knows more about this stuff than I do.)
1177    
1178    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1179        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1180        match everything, including newlines. This is not POSIX-compatible, but
1181        somebody wanted the feature. From pcretest it can be activated by using
1182        both the P and the s flags.
1183    
1184    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1185    
1186    19. libpcre.pc was being incorrectly installed as executable.
1187    
1188    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1189        it now also looks for '\r' so that it will work unmodified on Windows.
1190    
1191    21. Added Google's contributed C++ wrapper to the distribution.
1192    
1193    22. Added some untidy missing memory free() calls in pcretest, to keep
1194        Electric Fence happy when testing.
1195    
1196    
1197    
1198    Version 5.0 13-Sep-04
1199    ---------------------
1200    
1201     1. Internal change: literal characters are no longer packed up into items
1202        containing multiple characters in a single byte-string. Each character
1203        is now matched using a separate opcode. However, there may be more than one
1204        byte in the character in UTF-8 mode.
1205    
1206     2. The pcre_callout_block structure has two new fields: pattern_position and
1207        next_item_length. These contain the offset in the pattern to the next match
1208        item, and its length, respectively.
1209    
1210     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1211        insertion of callouts before each pattern item. Added the /C option to
1212        pcretest to make use of this.
1213    
1214     4. On the advice of a Windows user, the lines
1215    
1216          #if defined(_WIN32) || defined(WIN32)
1217          _setmode( _fileno( stdout ), 0x8000 );
1218          #endif  /* defined(_WIN32) || defined(WIN32) */
1219    
1220        have been added to the source of pcretest. This apparently does useful
1221        magic in relation to line terminators.
1222    
1223     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1224        for the benefit of those environments where the "b" makes a difference.
1225    
1226     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1227        to know about it. I have put a hack into configure.in that adds in code
1228        to set GCC=yes if CC=icc. This seems to end up at a point in the
1229        generated configure script that is early enough to affect the setting of
1230        compiler options, which is what is needed, but I have no means of testing
1231        whether it really works. (The user who reported this had patched the
1232        generated configure script, which of course I cannot do.)
1233    
1234        LATER: After change 22 below (new libtool files), the configure script
1235        seems to know about icc (and also ecc). Therefore, I have commented out
1236        this hack in configure.in.
1237    
1238     7. Added support for pkg-config (2 patches were sent in).
1239    
1240     8. Negated POSIX character classes that used a combination of internal tables
1241        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1242        [[:^ascii]]. Typically, they would match almost any characters. The other
1243        POSIX classes were not broken in this way.
1244    
1245     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1246        to find the match, as PCRE was deluded into thinking that the match had to
1247        start at the start point or following a newline. The same bug applied to
1248        patterns with negative forward assertions or any backward assertions
1249        preceding ".*" at the start, unless the pattern required a fixed first
1250        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1251    
1252    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1253        starting at the last subject character, bytes beyond the end of the subject
1254        string were read.
1255    
1256    11. Renamed the variable "class" as "classbits" to make life easier for C++
1257        users. (Previously there was a macro definition, but it apparently wasn't
1258        enough.)
1259    
1260    12. Added the new field "tables" to the extra data so that tables can be passed
1261        in at exec time, or the internal tables can be re-selected. This allows
1262        a compiled regex to be saved and re-used at a later time by a different
1263        program that might have everything at different addresses.
1264    
1265    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1266        -R library as well as a -L library.
1267    
1268    14. The debugging options of pcretest (-d on the command line or D on a
1269        pattern) showed incorrect output for anything following an extended class
1270        that contained multibyte characters and which was followed by a quantifier.
1271    
1272    15. Added optional support for general category Unicode character properties
1273        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1274        support. It adds about 90K to the size of the library. The meanings of the
1275        inbuilt class escapes such as \d and \s have NOT been changed.
1276    
1277    16. Updated pcredemo.c to include calls to free() to release the memory for the
1278        compiled pattern.
1279    
1280    17. The generated file chartables.c was being created in the source directory
1281        instead of in the building directory. This caused the build to fail if the
1282        source directory was different from the building directory, and was
1283        read-only.
1284    
1285    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1286        file. No doubt somebody will tell me if they don't make sense... Also added
1287        Dan Mooney's comments about building on OpenVMS.
1288    
1289    19. Added support for partial matching via the PCRE_PARTIAL option for
1290        pcre_exec() and the \P data escape in pcretest.
1291    
1292    20. Extended pcretest with 3 new pattern features:
1293    
1294        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1295              write the compiled pattern to the file whose name is "rest-of-line".
1296              This is a straight binary dump of the data, with the saved pointer to
1297              the character tables forced to be NULL. The study data, if any, is
1298              written too. After writing, pcretest reads a new pattern.
1299    
1300        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1301              compiled pattern from the given file. There must not be any
1302              occurrences of "<" in the file name (pretty unlikely); if there are,
1303              pcretest will instead treat the initial "<" as a pattern delimiter.
1304              After reading in the pattern, pcretest goes on to read data lines as
1305              usual.
1306    
1307        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1308              and 16-bit fields in a compiled pattern, to simulate a pattern that
1309              was compiled on a host of opposite endianness.
1310    
1311    21. The pcre-exec() function can now cope with patterns that were compiled on
1312        hosts of opposite endianness, with this restriction:
1313    
1314          As for any compiled expression that is saved and used later, the tables
1315          pointer field cannot be preserved; the extra_data field in the arguments
1316          to pcre_exec() should be used to pass in a tables address if a value
1317          other than the default internal tables were used at compile time.
1318    
1319    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1320        now diagnosed as an error. Previously, most of the time, a negative number
1321        would have been treated as zero, but if in addition "ovector" was passed as
1322        NULL, a crash could occur.
1323    
1324    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1325        new versions from the libtool 1.5 distribution (the last one is a copy of
1326        a file called libtool.m4). This seems to have fixed the need to patch
1327        "configure" to support Darwin 1.3 (which I used to do). However, I still
1328        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1329        workstation).
1330    
1331    24. Changed the PCRE licence to be the more standard "BSD" licence.
1332    
1333    
1334    Version 4.5 01-Dec-03
1335    ---------------------
1336    
1337     1. There has been some re-arrangement of the code for the match() function so
1338        that it can be compiled in a version that does not call itself recursively.
1339        Instead, it keeps those local variables that need separate instances for
1340        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1341        needs to "recurse". Keeping track of where control must go is done by means
1342        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1343        hide most of the details from the main code, and operates only if
1344        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1345        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1346        operating.
1347    
1348        To make it easier for callers to provide specially tailored get/free
1349        functions for this usage, two new functions, pcre_stack_malloc, and
1350        pcre_stack_free, are used. They are always called in strict stacking order,
1351        and the size of block requested is always the same.
1352    
1353        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1354        PCRE has been compiled to use the stack or the heap for recursion. The
1355        -C option of pcretest uses this to show which version is compiled.
1356    
1357        A new data escape \S, is added to pcretest; it causes the amounts of store
1358        obtained and freed by both kinds of malloc/free at match time to be added
1359        to the output.
1360    
1361     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1362        what's available on my current Linux desktop machine.
1363    
1364     3. When matching a UTF-8 string, the test for a valid string at the start has
1365        been extended. If start_offset is not zero, PCRE now checks that it points
1366        to a byte that is the start of a UTF-8 character. If not, it returns
1367        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1368        this is necessary because there may be backward assertions in the pattern.
1369        When matching the same subject several times, it may save resources to use
1370        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1371    
1372     4. The code for checking the validity of UTF-8 strings has been tightened so
1373        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1374        containing "overlong sequences".
1375    
1376     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1377        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1378        should have been "&", but it just so happened that all the cases this let
1379        through by mistake were picked up later in the function.
1380    
1381     6. I had used a variable called "isblank" - this is a C99 function, causing
1382        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1383    
1384     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1385        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1386        so the version is shown; (c) stop "make test" echoing "./RunTest".
1387    
1388     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1389    
1390     9. The prototype for memmove() for systems that don't have it was using
1391        size_t, but the inclusion of the header that defines size_t was later. I've
1392        moved the #includes for the C headers earlier to avoid this.
1393    
1394    10. Added some adjustments to the code to make it easier to compiler on certain
1395        special systems:
1396    
1397          (a) Some "const" qualifiers were missing.
1398          (b) Added the macro EXPORT before all exported functions; by default this
1399              is defined to be empty.
1400          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1401              that it reads its output file name as an argument instead of writing
1402              to the standard output and assuming this can be redirected.
1403    
1404    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1405        class containing characters with values greater than 255, PCRE compilation
1406        went into a loop.
1407    
1408    12. A recursive reference to a subpattern that was within another subpattern
1409        that had a minimum quantifier of zero caused PCRE to crash. For example,
1410        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1411        recursion. If the recursively-called subpattern itself had a zero repeat,
1412        that was OK.
1413    
1414    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1415        buffer into which it was copied (for escape processing) was still set at
1416        1024, so long lines caused crashes.
1417    
1418    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1419        "internal error: code overflow...". This applied to any character class
1420        that was followed by a possessive quantifier.
1421    
1422    15. Modified the Makefile to add libpcre.la as a prerequisite for
1423        libpcreposix.la because I was told this is needed for a parallel build to
1424        work.
1425    
1426    16. If a pattern that contained .* following optional items at the start was
1427        studied, the wrong optimizing data was generated, leading to matching
1428        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1429        matching string must start with a or b or c. The correct conclusion for
1430        this pattern is that a match can start with any character.
1431    
1432    
1433    Version 4.4 13-Aug-03
1434    ---------------------
1435    
1436     1. In UTF-8 mode, a character class containing characters with values between
1437        127 and 255 was not handled correctly if the compiled pattern was studied.
1438        In fixing this, I have also improved the studying algorithm for such
1439        classes (slightly).
1440    
1441     2. Three internal functions had redundant arguments passed to them. Removal
1442        might give a very teeny performance improvement.
1443    
1444     3. Documentation bug: the value of the capture_top field in a callout is *one
1445        more than* the number of the hightest numbered captured substring.
1446    
1447     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1448        in incorrectly linking with a previously installed version. They now link
1449        explicitly with libpcre.la.
1450    
1451     5. configure.in no longer needs to recognize Cygwin specially.
1452    
1453     6. A problem in pcre.in for Windows platforms is fixed.
1454    
1455     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1456        pcretest, it used to include the size of the study block as part of its
1457        output. Unfortunately, the structure contains a field that has a different
1458        size on different hardware architectures. This meant that the tests that
1459        showed this size failed. As the block is currently always of a fixed size,
1460        this information isn't actually particularly useful in pcretest output, so
1461        I have just removed it.
1462    
1463     8. Three pre-processor statements accidentally did not start in column 1.
1464        Sadly, there are *still* compilers around that complain, even though
1465        standard C has not required this for well over a decade. Sigh.
1466    
1467     9. In pcretest, the code for checking callouts passed small integers in the
1468        callout_data field, which is a void * field. However, some picky compilers
1469        complained about the casts involved for this on 64-bit systems. Now
1470        pcretest passes the address of the small integer instead, which should get
1471        rid of the warnings.
1472    
1473    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1474        both compile and run time, and gives an error if an invalid UTF-8 sequence
1475        is found. There is a option for disabling this check in cases where the
1476        string is known to be correct and/or the maximum performance is wanted.
1477    
1478    11. In response to a bug report, I changed one line in Makefile.in from
1479    
1480            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1481        to
1482            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1483    
1484        to look similar to other lines, but I have no way of telling whether this
1485        is the right thing to do, as I do not use Windows. No doubt I'll get told
1486        if it's wrong...
1487    
1488    
1489  Version 4.3 21-May-03  Version 4.3 21-May-03
1490  ---------------------  ---------------------
1491    

Legend:
Removed from v.69  
changed lines
  Added in v.192

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12