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

Diff of /code/trunk/ChangeLog

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

revision 234 by ph10, Tue Sep 11 11:36:27 2007 UTC revision 1217 by ph10, Fri Nov 9 17:07:01 2012 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4  Version 7.4 10-Sep-07  Version 8.32 11-November-2012
5    -----------------------------
6    
7    1.  Improved JIT compiler optimizations for first character search and single
8        character iterators.
9    
10    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
11        Patch by Daniel Richard G.
12    
13    3.  Single character iterator optimizations in the JIT compiler.
14    
15    4.  Improved JIT compiler optimizations for character ranges.
16    
17    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
18        Reported by Giuseppe D'Angelo.
19    
20    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
21        of a line, was being set incorrectly in cases where .* appeared inside
22        atomic brackets at the start of a pattern, or where there was a subsequent
23        *PRUNE or *SKIP.
24    
25    7.  Improved instruction cache flush for POWER/PowerPC.
26        Patch by Daniel Richard G.
27    
28    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
29        grep:
30    
31        (a) There is now no limit to the number of patterns to be matched.
32    
33        (b) An error is given if a pattern is too long.
34    
35        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
36            are now supported.
37    
38        (d) --exclude-from and --include-from (multiple use) have been added.
39    
40        (e) Exclusions and inclusions now apply to all files and directories, not
41            just to those obtained from scanning a directory recursively.
42    
43        (f) Multiple uses of -f and --file-list are now supported.
44    
45        (g) In a Windows environment, the default for -d has been changed from
46            "read" (the GNU grep default) to "skip", because otherwise the presence
47            of a directory in the file list provokes an error.
48    
49        (h) The documentation has been revised and clarified in places.
50    
51    9.  Improve the matching speed of capturing brackets.
52    
53    10. Changed the meaning of \X so that it now matches a Unicode extended
54        grapheme cluster.
55    
56    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
57        out POSIX threads when JIT support is configured.
58    
59    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
60    
61    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
62        structure could go wrong in environments where size_t is not the same size
63        as int.
64    
65    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
66        set.
67    
68    15. The EBCDIC support had decayed; later updates to the code had included
69        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
70        general tidy up of EBCDIC-related issues, and the documentation was also
71        not quite right. There is now a test that can be run on ASCII systems to
72        check some of the EBCDIC-related things (but is it not a full test).
73    
74    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
75        in a small tidy to the code.
76    
77    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
78    
79    18. If the --only-matching (-o) option in pcregrep is specified multiple
80        times, each one causes appropriate output. For example, -o1 -o2 outputs the
81        substrings matched by the 1st and 2nd capturing parentheses. A separating
82        string can be specified by --om-separator (default empty).
83    
84    19. Improving the first n character searches.
85    
86    20. Turn case lists for horizontal and vertical white space into macros so that
87        they are defined only once.
88    
89    21. This set of changes together give more compatible Unicode case-folding
90        behaviour for characters that have more than one other case when UCP
91        support is available.
92    
93        (a) The Unicode property table now has offsets into a new table of sets of
94            three or more characters that are case-equivalent. The MultiStage2.py
95            script that generates these tables (the pcre_ucd.c file) now scans
96            CaseFolding.txt instead of UnicodeData.txt for character case
97            information.
98    
99        (b) The code for adding characters or ranges of characters to a character
100            class has been abstracted into a generalized function that also handles
101            case-independence. In UTF-mode with UCP support, this uses the new data
102            to handle characters with more than one other case.
103    
104        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
105            whose other case is greater than 256 are now correctly matched
106            caselessly. Previously, the high codepoint matched the low one, but not
107            vice versa.
108    
109        (d) The processing of \h, \H, \v, and \ in character classes now makes use
110            of the new class addition function, using character lists defined as
111            macros alongside the case definitions of 20 above.
112    
113        (e) Caseless back references now work with characters that have more than
114            one other case.
115    
116        (f) General caseless matching of characters with more than one other case
117            is supported.
118    
119    22. Unicode character properties were updated from Unicode 6.2.0
120    
121    23. Improved CMake support under Windows. Patch by Daniel Richard G.
122    
123    24. Add support for 32-bit character strings, and UTF-32
124    
125    25. Major JIT compiler update (code refactoring and bugfixing).
126        Experimental Sparc 32 support is added.
127    
128    26. Applied a modified version of Daniel Richard G's patch to create
129        pcre.h.generic and config.h.generic by "make" instead of in the
130        PrepareRelease script.
131    
132    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
133        pcre_compile.c when checking for a zero character.
134    
135    28. Introducing a native interface for JIT. Through this interface, the compiled
136        machine code can be directly executed. The purpose of this interface is to
137        provide fast pattern matching, so several sanity checks are not performed.
138        However, feature tests are still performed. The new interface provides
139        1.4x speedup compared to the old one.
140    
141    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
142        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
143        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
144    
145    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
146        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
147        the "old" RFC 2279). Instead, it ended up passing a negative length to
148        pcre_exec().
149    
150    31. Add support for GCC's visibility feature to hide internal functions.
151    
152    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
153        "unknown -C option" after outputting 0 or 1.
154    
155    33. There is now support for generating a code coverage report for the test
156        suite in environments where gcc is the compiler and lcov is installed. This
157        is mainly for the benefit of the developers.
158    
159    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
160        unaddressable using valgrind annotations, allowing valgrind to detect
161        invalid memory accesses. This is mainly for the benefit of the developers.
162    
163    
164    Version 8.31 06-July-2012
165    -------------------------
166    
167    1.  Fixing a wrong JIT test case and some compiler warnings.
168    
169    2.  Removed a bashism from the RunTest script.
170    
171    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
172        to unsigned type, result still unsigned" that was given by an MS compiler
173        on encountering the code "-sizeof(xxx)".
174    
175    4.  Partial matching support is added to the JIT compiler.
176    
177    5.  Fixed several bugs concerned with partial matching of items that consist
178        of more than one character:
179    
180        (a) /^(..)\1/ did not partially match "aba" because checking references was
181            done on an "all or nothing" basis. This also applied to repeated
182            references.
183    
184        (b) \R did not give a hard partial match if \r was found at the end of the
185            subject.
186    
187        (c) \X did not give a hard partial match after matching one or more
188            characters at the end of the subject.
189    
190        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
191            a partial match for the string "\r".
192    
193        (e) When newline was set to CRLF, the metacharacter "." did not recognize
194            a partial match for a CR character at the end of the subject string.
195    
196    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
197        running pcretest, the text "(JIT)" added to the output whenever JIT is
198        actually used to run the match.
199    
200    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
201        or /S+[+] with a digit between 1 and 7.
202    
203    8.  OP_NOT now supports any UTF character not just single-byte ones.
204    
205    9.  (*MARK) control verb is now supported by the JIT compiler.
206    
207    10. The command "./RunTest list" lists the available tests without actually
208        running any of them. (Because I keep forgetting what they all are.)
209    
210    11. Add PCRE_INFO_MAXLOOKBEHIND.
211    
212    12. Applied a (slightly modified) user-supplied patch that improves performance
213        when the heap is used for recursion (compiled with --disable-stack-for-
214        recursion). Instead of malloc and free for each heap frame each time a
215        logical recursion happens, frames are retained on a chain and re-used where
216        possible. This sometimes gives as much as 30% improvement.
217    
218    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
219        call.
220    
221    14. As documented, (*COMMIT) is now confined to within a positive assertion.
222    
223    15. It is now possible to link pcretest with libedit as an alternative to
224        libreadline.
225    
226    16. (*COMMIT) control verb is now supported by the JIT compiler.
227    
228    17. The Unicode data tables have been updated to Unicode 6.1.0.
229    
230    18. Added --file-list option to pcregrep.
231    
232    19. Added binary file support to pcregrep, including the -a, --binary-files,
233        -I, and --text options.
234    
235    20. The madvise function is renamed for posix_madvise for QNX compatibility
236        reasons. Fixed by Giuseppe D'Angelo.
237    
238    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
239    
240    22. Changed the option for creating symbolic links for 16-bit man pages from
241        -s to -sf so that re-installing does not cause issues.
242    
243    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
244    
245    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
246        matches in certain environments (the workspace was not being correctly
247        retained). Also added to pcre_dfa_exec() a simple plausibility check on
248        some of the workspace data at the beginning of a restart.
249    
250    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
251        was not doing so when it should - probably a typo introduced by SVN 528
252        (change 8.10/14).
253    
254    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
255        \w+ when the character tables indicated that \x{c4} was a word character.
256        There were several related cases, all because the tests for doing a table
257        lookup were testing for characters less than 127 instead of 255.
258    
259    27. If a pattern contains capturing parentheses that are not used in a match,
260        their slots in the ovector are set to -1. For those that are higher than
261        any matched groups, this happens at the end of processing. In the case when
262        there were back references that the ovector was too small to contain
263        (causing temporary malloc'd memory to be used during matching), and the
264        highest capturing number was not used, memory off the end of the ovector
265        was incorrectly being set to -1. (It was using the size of the temporary
266        memory instead of the true size.)
267    
268    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
269        ovector size, it uses memory at the end of the block that it has got.
270    
271    29. Check for an overlong MARK name and give an error at compile time. The
272        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
273    
274    30. JIT compiler update.
275    
276    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
277        Rill for the patch.
278    
279    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
280    
281    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
282    
283    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
284        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
285        was enabled.
286    
287    35. Improve JIT code generation for greedy plus quantifier.
288    
289    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
290        1 to "aa" instead of to an empty string. The bug affected repeated groups
291        that could potentially match an empty string.
292    
293    37. Optimizing single character iterators in JIT.
294    
295    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
296        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
297        codepoints that are too big for the mode are faulted, and in a UTF mode,
298        disallowed codepoints are also faulted.
299    
300    39. If PCRE was compiled with UTF support, in three places in the DFA
301        matcher there was code that should only have been obeyed in UTF mode, but
302        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
303        processing when bytes with values greater than 127 were present. In 16-bit
304        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
305        both cases the values are those that cannot be the first data item in a UTF
306        character. The three items that might have provoked this were recursions,
307        possessively repeated groups, and atomic groups.
308    
309    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
310        and pcregrep, because some OS require shared objects to be explicitly
311        passed to ld, causing the link step to fail if they are not.
312    
313    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
314        mode, patterns that started with \h* or \R* might be incorrectly matched.
315    
316    
317    Version 8.30 04-February-2012
318    -----------------------------
319    
320    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
321        name is defined in ctype.h.
322    
323    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
324        only in quite long subpatterns.
325    
326    3.  Removed the function pcre_info(), which has been obsolete and deprecated
327        since it was replaced by pcre_fullinfo() in February 2000.
328    
329    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
330        match a (*MARK), and the match failed at the start of the subject, a
331        reference to memory before the start of the subject could occur. This bug
332        was introduced by fix 17 of release 8.21.
333    
334    5.  A reference to an unset group with zero minimum repetition was giving
335        totally wrong answers (in non-JavaScript-compatibility mode). For example,
336        /(another)?(\1?)test/ matched against "hello world test". This bug was
337        introduced in release 8.13.
338    
339    6.  Add support for 16-bit character strings (a large amount of work involving
340        many changes and refactorings).
341    
342    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
343        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
344        from a file.
345    
346    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
347        rounding is not applied in this particular case).
348    
349    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
350        if they appear, or are escaped, in patterns.
351    
352    10. Get rid of a number of -Wunused-but-set-variable warnings.
353    
354    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
355        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
356        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
357        also returns the mark "x". This bug applied to capturing parentheses,
358        non-capturing parentheses, and atomic parentheses. It also applied to some
359        assertions.
360    
361    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
362        information out of configure.ac instead of relying on pcre.h.generic, which
363        is not stored in the repository.
364    
365    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
366        -lreadline.
367    
368    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
369    
370    15. Applied Graycode's patch to put the top-level frame on the stack rather
371        than the heap when not using the stack for recursion. This gives a
372        performance improvement in many cases when recursion is not deep.
373    
374    16. Experimental code added to "pcretest -C" to output the stack frame size.
375    
376    
377    Version 8.21 12-Dec-2011
378    ------------------------
379    
380    1.  Updating the JIT compiler.
381    
382    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
383        are added as well.
384    
385    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
386        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
387        calling _pcre_jit_exec. Some extra comments are added.
388    
389    4.  (*MARK) settings inside atomic groups that do not contain any capturing
390        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
391        was introduced by change 18 for 8.20.
392    
393    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
394        ECMA-262 standard.
395    
396    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
397        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
398        This bug was probably introduced by change 9 of 8.13.
399    
400    7.  While fixing 6 above, I noticed that a number of other items were being
401        incorrectly rejected as "not fixed length". This arose partly because newer
402        opcodes had not been added to the fixed-length checking code. I have (a)
403        corrected the bug and added tests for these items, and (b) arranged for an
404        error to occur if an unknown opcode is encountered while checking for fixed
405        length instead of just assuming "not fixed length". The items that were
406        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
407        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
408        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
409    
410    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
411        being incorrectly compiled and would have given unpredicatble results.
412    
413    9.  A possessively repeated subpattern with minimum repeat count greater than
414        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
415        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
416        the first (A) could occur when it should not.
417    
418    10. Add a cast and remove a redundant test from the code.
419    
420    11. JIT should use pcre_malloc/pcre_free for allocation.
421    
422    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
423        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
424        is anything other than /usr, -L is still shown).
425    
426    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
427    
428    14. Perl does not support \N without a following name in a [] class; PCRE now
429        also gives an error.
430    
431    15. If a forward reference was repeated with an upper limit of around 2000,
432        it caused the error "internal error: overran compiling workspace". The
433        maximum number of forward references (including repeats) was limited by the
434        internal workspace, and dependent on the LINK_SIZE. The code has been
435        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
436        the default depends on LINK_SIZE. There is a new upper limit (for safety)
437        of around 200,000 forward references. While doing this, I also speeded up
438        the filling in of repeated forward references.
439    
440    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
441        incorrectly expecting the subject to contain another "a" after the start.
442    
443    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
444        in the match, the SKIP should be ignored. This was not happening; instead
445        the SKIP was being treated as NOMATCH. For patterns such as
446        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
447        tested.
448    
449    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
450        now much more compatible with Perl, in particular in cases where the result
451        is a non-match for a non-anchored pattern. For example, if
452        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
453        "m", where previously it did not return a name. A side effect of this
454        change is that for partial matches, the last encountered mark name is
455        returned, as for non matches. A number of tests that were previously not
456        Perl-compatible have been moved into the Perl-compatible test files. The
457        refactoring has had the pleasing side effect of removing one argument from
458        the match() function, thus reducing its stack requirements.
459    
460    19. If the /S+ option was used in pcretest to study a pattern using JIT,
461        subsequent uses of /S (without +) incorrectly behaved like /S+.
462    
463    21. Retrieve executable code size support for the JIT compiler and fixing
464        some warnings.
465    
466    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
467        not work when the shorter character appeared right at the end of the
468        subject string.
469    
470    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
471        systems.
472    
473    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
474        output it when the /M option is used in pcretest.
475    
476    25. The CheckMan script was not being included in the distribution. Also, added
477        an explicit "perl" to run Perl scripts from the PrepareRelease script
478        because this is reportedly needed in Windows.
479    
480    26. If study data was being save in a file and studying had not found a set of
481        "starts with" bytes for the pattern, the data written to the file (though
482        never used) was taken from uninitialized memory and so caused valgrind to
483        complain.
484    
485    27. Updated RunTest.bat as provided by Sheri Pierce.
486    
487    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
488    
489    29. Computation of memory usage for the table of capturing group names was
490        giving an unnecessarily large value.
491    
492    
493    Version 8.20 21-Oct-2011
494    ------------------------
495    
496    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
497        a POSIX class. After further experiments with Perl, which convinced me that
498        Perl has bugs and confusions, a closing square bracket is no longer allowed
499        in a POSIX name. This bug also affected patterns with classes that started
500        with full stops.
501    
502    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
503        captured substring, but while checking the failing first alternative,
504        substring 1 is temporarily captured. If the output vector supplied to
505        pcre_exec() was not big enough for this capture, the yield of the function
506        was still zero ("insufficient space for captured substrings"). This cannot
507        be totally fixed without adding another stack variable, which seems a lot
508        of expense for a edge case. However, I have improved the situation in cases
509        such as /(a)(b)x|abc/ matched against "abc", where the return code
510        indicates that fewer than the maximum number of slots in the ovector have
511        been set.
512    
513    3.  Related to (2) above: when there are more back references in a pattern than
514        slots in the output vector, pcre_exec() uses temporary memory during
515        matching, and copies in the captures as far as possible afterwards. It was
516        using the entire output vector, but this conflicts with the specification
517        that only 2/3 is used for passing back captured substrings. Now it uses
518        only the first 2/3, for compatibility. This is, of course, another edge
519        case.
520    
521    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
522        main code base, and can be used by building with --enable-jit. When this is
523        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
524        runtime --no-jit option is given.
525    
526    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
527        ovector, the return from the function was zero, implying that there were
528        other matches that did not fit. The correct "exactly full" value is now
529        returned.
530    
531    6.  If a subpattern that was called recursively or as a subroutine contained
532        (*PRUNE) or any other control that caused it to give a non-standard return,
533        invalid errors such as "Error -26 (nested recursion at the same subject
534        position)" or even infinite loops could occur.
535    
536    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
537        computing the minimum length on reaching *ACCEPT, and so ended up with the
538        wrong value of 1 rather than 0. Further investigation indicates that
539        computing a minimum subject length in the presence of *ACCEPT is difficult
540        (think back references, subroutine calls), and so I have changed the code
541        so that no minimum is registered for a pattern that contains *ACCEPT.
542    
543    8.  If (*THEN) was present in the first (true) branch of a conditional group,
544        it was not handled as intended. [But see 16 below.]
545    
546    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
547        Sheri Pierce.
548    
549    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
550        the first byte in a match must be "a".
551    
552    11. Change 17 for 8.13 increased the recursion depth for patterns like
553        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
554        pattern contains any instances of (*THEN). If it does not, the old
555        optimizations are restored. It would be nice to do this on a per-group
556        basis, but at the moment that is not feasible.
557    
558    12. In some environments, the output of pcretest -C is CRLF terminated. This
559        broke RunTest's code that checks for the link size. A single white space
560        character after the value is now allowed for.
561    
562    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
563        For "fr", it uses the Windows-specific input and output files.
564    
565    14. If (*THEN) appeared in a group that was called recursively or as a
566        subroutine, it did not work as intended. [But see next item.]
567    
568    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
569        pattern fragments (but not containing any | characters). If A and B are
570        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
571        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
572        D. In other words, Perl considers parentheses that do not contain any |
573        characters to be part of a surrounding alternative, whereas PCRE was
574        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
575        differently. PCRE now behaves in the same way as Perl, except in the case
576        of subroutine/recursion calls such as (?1) which have in any case always
577        been different (but PCRE had them first :-).
578    
579    16. Related to 15 above: Perl does not treat the | in a conditional group as
580        creating alternatives. Such a group is treated in the same way as an
581        ordinary group without any | characters when processing (*THEN). PCRE has
582        been changed to match Perl's behaviour.
583    
584    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
585        RunGrepTest script failed.
586    
587    18. Change 22 for version 13 caused atomic groups to use more stack. This is
588        inevitable for groups that contain captures, but it can lead to a lot of
589        stack use in large patterns. The old behaviour has been restored for atomic
590        groups that do not contain any capturing parentheses.
591    
592    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
593        suppress the check for a minimum subject length at run time. (If it was
594        given to pcre_exec() or pcre_dfa_exec() it did work.)
595    
596    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
597        fail to work when decoding hex characters in data strings in EBCDIC
598        environments.
599    
600    21. It appears that in at least one Mac OS environment, the isxdigit() function
601        is implemented as a macro that evaluates to its argument more than once,
602        contravening the C 90 Standard (I haven't checked a later standard). There
603        was an instance in pcretest which caused it to go wrong when processing
604        \x{...} escapes in subject strings. The has been rewritten to avoid using
605        things like p++ in the argument of isxdigit().
606    
607    
608    Version 8.13 16-Aug-2011
609    ------------------------
610    
611    1.  The Unicode data tables have been updated to Unicode 6.0.0.
612    
613    2.  Two minor typos in pcre_internal.h have been fixed.
614    
615    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
616        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
617        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
618    
619    4.  There were a number of related bugs in the code for matching backrefences
620        caselessly in UTF-8 mode when codes for the characters concerned were
621        different numbers of bytes. For example, U+023A and U+2C65 are an upper
622        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
623        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
624        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
625        2-byte code at the end of the subject (it thought there wasn't enough data
626        left).
627    
628    5.  Comprehensive information about what went wrong is now returned by
629        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
630        as the output vector has at least 2 elements. The offset of the start of
631        the failing character and a reason code are placed in the vector.
632    
633    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
634        now returned is for the first byte of the failing character, instead of the
635        last byte inspected. This is an incompatible change, but I hope it is small
636        enough not to be a problem. It makes the returned offset consistent with
637        pcre_exec() and pcre_dfa_exec().
638    
639    7.  pcretest now gives a text phrase as well as the error number when
640        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
641        failure, the offset and reason code are output.
642    
643    8.  When \R was used with a maximizing quantifier it failed to skip backwards
644        over a \r\n pair if the subsequent match failed. Instead, it just skipped
645        back over a single character (\n). This seems wrong (because it treated the
646        two characters as a single entity when going forwards), conflicts with the
647        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
648        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
649        has been changed.
650    
651    9.  Some internal refactoring has changed the processing so that the handling
652        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
653        time (the PCRE_DOTALL option was changed this way some time ago: version
654        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
655        which was always a bit of a fudge. It also means that there is one less
656        argument for the match() function, which reduces its stack requirements
657        slightly. This change also fixes an incompatibility with Perl: the pattern
658        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
659    
660    10. More internal refactoring has drastically reduced the number of recursive
661        calls to match() for possessively repeated groups such as (abc)++ when
662        using pcre_exec().
663    
664    11. While implementing 10, a number of bugs in the handling of groups were
665        discovered and fixed:
666    
667        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
668        (a|)*(?1) gave a compile-time internal error.
669        ((a|)+)+  did not notice that the outer group could match an empty string.
670        (^a|^)+   was not marked as anchored.
671        (.*a|.*)+ was not marked as matching at start or after a newline.
672    
673    12. Yet more internal refactoring has removed another argument from the match()
674        function. Special calls to this function are now indicated by setting a
675        value in a variable in the "match data" data block.
676    
677    13. Be more explicit in pcre_study() instead of relying on "default" for
678        opcodes that mean there is no starting character; this means that when new
679        ones are added and accidentally left out of pcre_study(), testing should
680        pick them up.
681    
682    14. The -s option of pcretest has been documented for ages as being an old
683        synonym of -m (show memory usage). I have changed it to mean "force study
684        for every regex", that is, assume /S for every regex. This is similar to -i
685        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
686        using it. It makes it easier to run collections of tests with and without
687        study enabled, and thereby test pcre_study() more easily. All the standard
688        tests are now run with and without -s (but some patterns can be marked as
689        "never study" - see 20 below).
690    
691    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
692        restoration of the capturing data to the outer values was not happening
693        correctly.
694    
695    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
696        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
697        pattern had matched an empty string, and so incorrectly returned a no
698        match.
699    
700    17. There was optimizing code for the last branch of non-capturing parentheses,
701        and also for the obeyed branch of a conditional subexpression, which used
702        tail recursion to cut down on stack usage. Unfortunately, now that there is
703        the possibility of (*THEN) occurring in these branches, tail recursion is
704        no longer possible because the return has to be checked for (*THEN). These
705        two optimizations have therefore been removed. [But see 8.20/11 above.]
706    
707    18. If a pattern containing \R was studied, it was assumed that \R always
708        matched two bytes, thus causing the minimum subject length to be
709        incorrectly computed because \R can also match just one byte.
710    
711    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
712        was incorrectly computed.
713    
714    20. If /S is present twice on a test pattern in pcretest input, it now
715        *disables* studying, thereby overriding the use of -s on the command line
716        (see 14 above). This is necessary for one or two tests to keep the output
717        identical in both cases.
718    
719    21. When (*ACCEPT) was used in an assertion that matched an empty string and
720        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
721    
722    22. When an atomic group that contained a capturing parenthesis was
723        successfully matched, but the branch in which it appeared failed, the
724        capturing was not being forgotten if a higher numbered group was later
725        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
726        group 1 to "a", when in fact it should be unset. This applied to multi-
727        branched capturing and non-capturing groups, repeated or not, and also to
728        positive assertions (capturing in negative assertions does not happen
729        in PCRE) and also to nested atomic groups.
730    
731    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
732        subject after a captured substring, to make it easier to tell which of a
733        number of identical substrings has been captured.
734    
735    24. The way atomic groups are processed by pcre_exec() has been changed so that
736        if they are repeated, backtracking one repetition now resets captured
737        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
738        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
739        "aaa". Previously, it would have been "a". As part of this code
740        refactoring, the way recursive calls are handled has also been changed.
741    
742    25. If an assertion condition captured any substrings, they were not passed
743        back unless some other capturing happened later. For example, if
744        (?(?=(a))a) was matched against "a", no capturing was returned.
745    
746    26. When studying a pattern that contained subroutine calls or assertions,
747        the code for finding the minimum length of a possible match was handling
748        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
749        group 1 called group 2 while simultaneously a separate group 2 called group
750        1). A stack overflow occurred in this case. I have fixed this by limiting
751        the recursion depth to 10.
752    
753    27. Updated RunTest.bat in the distribution to the version supplied by Tom
754        Fortmann. This supports explicit test numbers on the command line, and has
755        argument validation and error reporting.
756    
757    28. An instance of \X with an unlimited repeat could fail if at any point the
758        first character it looked at was a mark character.
759    
760    29. Some minor code refactoring concerning Unicode properties and scripts
761        should reduce the stack requirement of match() slightly.
762    
763    30. Added the '=' option to pcretest to check the setting of unused capturing
764        slots at the end of the pattern, which are documented as being -1, but are
765        not included in the return count.
766    
767    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
768        compiled something random. Now it gives a compile-time error (as does
769        Perl).
770    
771    32. A *MARK encountered during the processing of a positive assertion is now
772        recorded and passed back (compatible with Perl).
773    
774    33. If --only-matching or --colour was set on a pcregrep call whose pattern
775        had alternative anchored branches, the search for a second match in a line
776        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
777        matched the line "0102" twice. The same bug affected patterns that started
778        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
779        twice.
780    
781    34. Previously, PCRE did not allow quantification of assertions. However, Perl
782        does, and because of capturing effects, quantifying parenthesized
783        assertions may at times be useful. Quantifiers are now allowed for
784        parenthesized assertions.
785    
786    35. A minor code tidy in pcre_compile() when checking options for \R usage.
787    
788    36. \g was being checked for fancy things in a character class, when it should
789        just be a literal "g".
790    
791    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
792        appearance of a nested POSIX class supersedes an apparent external class.
793        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
794        unescaped square brackets may also appear as part of class names. For
795        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
796        more like Perl. (But see 8.20/1 above.)
797    
798    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
799        was because it thought it was \N{name}, which is not supported).
800    
801    39. Add minix to OS list not supporting the -S option in pcretest.
802    
803    40. PCRE tries to detect cases of infinite recursion at compile time, but it
804        cannot analyze patterns in sufficient detail to catch mutual recursions
805        such as ((?1))((?2)). There is now a runtime test that gives an error if a
806        subgroup is called recursively as a subpattern for a second time at the
807        same position in the subject string. In previous releases this might have
808        been caught by the recursion limit, or it might have run out of stack.
809    
810    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
811        happen only once. PCRE was, however incorrectly giving a compile time error
812        "recursive call could loop indefinitely" because it cannot analyze the
813        pattern in sufficient detail. The compile time test no longer happens when
814        PCRE is compiling a conditional subpattern, but actual runaway loops are
815        now caught at runtime (see 40 above).
816    
817    42. It seems that Perl allows any characters other than a closing parenthesis
818        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
819        has been changed to be the same.
820    
821    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
822        as not to get warnings when autogen.sh is called. Also changed
823        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
824    
825    44. To help people who use pcregrep to scan files containing exceedingly long
826        lines, the following changes have been made:
827    
828        (a) The default value of the buffer size parameter has been increased from
829            8K to 20K. (The actual buffer used is three times this size.)
830    
831        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
832            PCRE is built.
833    
834        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
835            to be set at run time.
836    
837        (d) Numerical values in pcregrep options can be followed by K or M, for
838            example --buffer-size=50K.
839    
840        (e) If a line being scanned overflows pcregrep's buffer, an error is now
841            given and the return code is set to 2.
842    
843    45. Add a pointer to the latest mark to the callout data block.
844    
845    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
846        partial match of an empty string instead of no match. This was specific to
847        the use of ".".
848    
849    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
850        complete match instead of a partial match. This bug was dependent on both
851        the PCRE_UTF8 and PCRE_DOTALL options being set.
852    
853    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
854        starting byte set, because \b was not being ignored.
855    
856    
857    Version 8.12 15-Jan-2011
858    ------------------------
859    
860    1.  Fixed some typos in the markup of the man pages, and wrote a script that
861        checks for such things as part of the documentation building process.
862    
863    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
864        --match-limit and --recursion-limit options (added for 8.11). In
865        particular, this made one of the standard tests fail. (The integer value
866        went into the wrong half of a long int.)
867    
868    3.  If the --colour option was given to pcregrep with -v (invert match), it
869        did strange things, either producing crazy output, or crashing. It should,
870        of course, ignore a request for colour when reporting lines that do not
871        match.
872    
873    4.  Another pcregrep bug caused similar problems if --colour was specified with
874        -M (multiline) and the pattern match finished with a line ending.
875    
876    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
877        matched in multiline mode, the following line was shown as part of the
878        match. This seems wrong, so I have changed it.
879    
880    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
881        the check for further matches in the same line (so they could be coloured)
882        to overrun the end of the current line. If another match was found, it was
883        incorrectly shown (and then shown again when found in the next line).
884    
885    7.  If pcregrep was compiled under Windows, there was a reference to the
886        function pcregrep_exit() before it was defined. I am assuming this was
887        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
888        reported by a user. I've moved the definition above the reference.
889    
890    
891    Version 8.11 10-Dec-2010
892    ------------------------
893    
894    1.  (*THEN) was not working properly if there were untried alternatives prior
895        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
896        backtracked to try for "b" instead of moving to the next alternative branch
897        at the same level (in this case, to look for "c"). The Perl documentation
898        is clear that when (*THEN) is backtracked onto, it goes to the "next
899        alternative in the innermost enclosing group".
900    
901    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
902        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
903        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
904        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
905        (*THEN).
906    
907    3.  If \s appeared in a character class, it removed the VT character from
908        the class, even if it had been included by some previous item, for example
909        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
910        of \s, but is part of the POSIX "space" class.)
911    
912    4.  A partial match never returns an empty string (because you can always
913        match an empty string at the end of the subject); however the checking for
914        an empty string was starting at the "start of match" point. This has been
915        changed to the "earliest inspected character" point, because the returned
916        data for a partial match starts at this character. This means that, for
917        example, /(?<=abc)def/ gives a partial match for the subject "abc"
918        (previously it gave "no match").
919    
920    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
921        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
922        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
923        has an implication that the given string is incomplete (because a partial
924        match is preferred over a full match). For this reason, these items now
925        give a partial match in this situation. [Aside: previously, the one case
926        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
927        match rather than a full match, which was wrong by the old rules, but is
928        now correct.]
929    
930    6.  There was a bug in the handling of #-introduced comments, recognized when
931        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
932        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
933        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
934        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
935        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
936        places in pcre_compile().
937    
938    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
939        comments when looking ahead for named forward references to subpatterns,
940        the only newline sequence it recognized was NL. It now handles newlines
941        according to the set newline convention.
942    
943    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
944        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
945        cater for a lack of strerror(). These oversights have been fixed.
946    
947    9.  Added --match-limit and --recursion-limit to pcregrep.
948    
949    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
950    
951    11. When the -o option was used, pcregrep was setting a return code of 1, even
952        when matches were found, and --line-buffered was not being honoured.
953    
954    12. Added an optional parentheses number to the -o and --only-matching options
955        of pcregrep.
956    
957    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
958        can match an empty string. The code to do it in pcretest and pcredemo
959        needed fixing:
960    
961        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
962            only one byte after an empty string match just before CRLF (this case
963            just got forgotten; "any" and "anycrlf" were OK).
964    
965        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
966            mode when an empty string match preceded an ASCII character followed by
967            a non-ASCII character. (The code for advancing by one character rather
968            than one byte was nonsense.)
969    
970        (c) The pcredemo.c sample program did not have any code at all to handle
971            the cases when CRLF is a valid newline sequence.
972    
973    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
974        as a starting offset was within the subject string. There is now a new
975        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
976        negative or greater than the length of the string. In order to test this,
977        pcretest is extended to allow the setting of negative starting offsets.
978    
979    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
980        starting offset points to the beginning of a UTF-8 character was
981        unnecessarily clumsy. I tidied it up.
982    
983    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
984        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
985    
986    17. Nobody had reported that the --include_dir option, which was added in
987        release 7.7 should have been called --include-dir (hyphen, not underscore)
988        for compatibility with GNU grep. I have changed it to --include-dir, but
989        left --include_dir as an undocumented synonym, and the same for
990        --exclude-dir, though that is not available in GNU grep, at least as of
991        release 2.5.4.
992    
993    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
994        characters from a string of bytes) have been redefined so as not to use
995        loops, in order to improve performance in some environments. At the same
996        time, I abstracted some of the common code into auxiliary macros to save
997        repetition (this should not affect the compiled code).
998    
999    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1000        compile-time error is now given if \c is not followed by an ASCII
1001        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1002        different, and any byte value is allowed.)
1003    
1004    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1005        START_OPTIMIZE option, which is now allowed at compile time - but just
1006        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1007        to pcregrep and other applications that have no direct access to PCRE
1008        options. The new /Y option in pcretest sets this option when calling
1009        pcre_compile().
1010    
1011    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1012        back references. Groups containing recursive back references were forced to
1013        be atomic by that change, but in the case of named groups, the amount of
1014        memory required was incorrectly computed, leading to "Failed: internal
1015        error: code overflow". This has been fixed.
1016    
1017    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1018        pcretest.c, to avoid build problems in some Borland environments.
1019    
1020    
1021    Version 8.10 25-Jun-2010
1022    ------------------------
1023    
1024    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1025        THEN.
1026    
1027    2.  (*ACCEPT) was not working when inside an atomic group.
1028    
1029    3.  Inside a character class, \B is treated as a literal by default, but
1030        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1031        causes the error). The code is unchanged, but I tidied the documentation.
1032    
1033    4.  Inside a character class, PCRE always treated \R and \X as literals,
1034        whereas Perl faults them if its -w option is set. I have changed PCRE so
1035        that it faults them when PCRE_EXTRA is set.
1036    
1037    5.  Added support for \N, which always matches any character other than
1038        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1039    
1040    6.  When compiling pcregrep with newer versions of gcc which may have
1041        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1042        declared with attribute warn_unused_result" were given. Just casting the
1043        result to (void) does not stop the warnings; a more elaborate fudge is
1044        needed. I've used a macro to implement this.
1045    
1046    7.  Minor change to pcretest.c to avoid a compiler warning.
1047    
1048    8.  Added four artifical Unicode properties to help with an option to make
1049        \s etc use properties (see next item). The new properties are: Xan
1050        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1051    
1052    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1053        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1054        this option. Modified pcretest to add /W to test this facility. Added
1055        REG_UCP to make it available via the POSIX interface.
1056    
1057    10. Added --line-buffered to pcregrep.
1058    
1059    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1060        studied, and the match started with a letter with a code point greater than
1061        127 whose first byte was different to the first byte of the other case of
1062        the letter, the other case of this starting letter was not recognized
1063        (#976).
1064    
1065    12. If a pattern that was studied started with a repeated Unicode property
1066        test, for example, \p{Nd}+, there was the theoretical possibility of
1067        setting up an incorrect bitmap of starting bytes, but fortunately it could
1068        not have actually happened in practice until change 8 above was made (it
1069        added property types that matched character-matching opcodes).
1070    
1071    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1072        possible starting bytes for non-anchored patterns.
1073    
1074    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1075        \R, and also a number of cases that involve Unicode properties, both
1076        explicit and implicit when PCRE_UCP is set.
1077    
1078    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1079        input, it could crash or give wrong results if characters with values
1080        greater than 0xc0 were present in the subject string. (Detail: it assumed
1081        UTF-8 input when processing these items.)
1082    
1083    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1084        size_t is 64-bit (#991).
1085    
1086    17. Added a check for running out of memory when PCRE is compiled with
1087        --disable-stack-for-recursion (#990).
1088    
1089    18. If the last data line in a file for pcretest does not have a newline on
1090        the end, a newline was missing in the output.
1091    
1092    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1093        less than 128) in its various bitmaps. However, there is a facility for
1094        generating tables according to the current locale when PCRE is compiled. It
1095        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1096        characters, are recognized by isspace() and therefore were getting set in
1097        these tables, and indeed these tables seem to approximate to ISO 8859. This
1098        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1099        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1100        which of course cannot start UTF-8 characters. I have changed the code so
1101        that only real ASCII characters (less than 128) and the correct starting
1102        bytes for UTF-8 encodings are set for characters greater than 127 when in
1103        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1104        altogether.)
1105    
1106    20. Added the /T option to pcretest so as to be able to run tests with non-
1107        standard character tables, thus making it possible to include the tests
1108        used for 19 above in the standard set of tests.
1109    
1110    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1111        reference to a subpattern the other side of a comment that contains an
1112        opening parenthesis caused either an internal compiling error, or a
1113        reference to the wrong subpattern.
1114    
1115    
1116    Version 8.02 19-Mar-2010
1117    ------------------------
1118    
1119    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1120    
1121    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1122        configured.
1123    
1124    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1125        original author of that file, following a query about its status.
1126    
1127    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1128        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1129    
1130    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1131        quantifier applied to a forward-referencing subroutine call, could compile
1132        incorrect code or give the error "internal error: previously-checked
1133        referenced subpattern not found".
1134    
1135    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1136        variables to point to external functions. For these systems, therefore,
1137        pcre_malloc etc. are now initialized to local functions that call the
1138        relevant global functions.
1139    
1140    7.  There were two entries missing in the vectors called coptable and poptable
1141        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1142        I've fixed the data, and added a kludgy way of testing at compile time that
1143        the lengths are correct (equal to the number of opcodes).
1144    
1145    8.  Following on from 7, I added a similar kludge to check the length of the
1146        eint vector in pcreposix.c.
1147    
1148    9.  Error texts for pcre_compile() are held as one long string to avoid too
1149        much relocation at load time. To find a text, the string is searched,
1150        counting zeros. There was no check for running off the end of the string,
1151        which could happen if a new error number was added without updating the
1152        string.
1153    
1154    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1155    
1156    11. \K was not working if it appeared in an atomic group or in a group that
1157        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1158        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1159        the assertion is positive, but not if it is negative.
1160    
1161    12. Change 11 fortuitously reduced the size of the stack frame used in the
1162        "match()" function of pcre_exec.c by one pointer. Forthcoming
1163        implementation of support for (*MARK) will need an extra pointer on the
1164        stack; I have reserved it now, so that the stack frame size does not
1165        decrease.
1166    
1167    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1168        item in branch that calls a recursion is a subroutine call - as in the
1169        second branch in the above example - was incorrectly given the compile-
1170        time error "recursive call could loop indefinitely" because pcre_compile()
1171        was not correctly checking the subroutine for matching a non-empty string.
1172    
1173    14. The checks for overrunning compiling workspace could trigger after an
1174        overrun had occurred. This is a "should never occur" error, but it can be
1175        triggered by pathological patterns such as hundreds of nested parentheses.
1176        The checks now trigger 100 bytes before the end of the workspace.
1177    
1178    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1179    
1180    
1181    Version 8.01 19-Jan-2010
1182    ------------------------
1183    
1184    1.  If a pattern contained a conditional subpattern with only one branch (in
1185        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1186        computed the wrong minimum data length (which is of course zero for such
1187        subpatterns). This could cause incorrect "no match" results.
1188    
1189    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1190        the pattern is reset in the first branch, pcre_compile() failed with
1191        "internal error: code overflow at offset...". This happened only when
1192        the reset was to the original external option setting. (An optimization
1193        abstracts leading options settings into an external setting, which was the
1194        cause of this.)
1195    
1196    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1197        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1198        assertion pattern did not match (meaning that the assertion was true), it
1199        was incorrectly treated as false if the SKIP had been reached during the
1200        matching. This also applied to assertions used as conditions.
1201    
1202    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1203        assertion subpattern, including such a pattern used as a condition,
1204        unpredictable results occurred, instead of the error return
1205        PCRE_ERROR_DFA_UITEM.
1206    
1207    5.  The C++ GlobalReplace function was not working like Perl for the special
1208        situation when an empty string is matched. It now does the fancy magic
1209        stuff that is necessary.
1210    
1211    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1212        removed. (These were left over from very, very early versions of PCRE.)
1213    
1214    7.  Some cosmetic changes to the code to make life easier when compiling it
1215        as part of something else:
1216    
1217        (a) Change DEBUG to PCRE_DEBUG.
1218    
1219        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1220            called "current" as "current_branch", to prevent a collision with the
1221            Linux macro when compiled as a kernel module.
1222    
1223        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1224            prevent a collision with the Linux macro when compiled as a kernel
1225            module.
1226    
1227    8.  In pcre_compile() there are some checks for integer overflows that used to
1228        cast potentially large values to (double). This has been changed to that
1229        when building, a check for int64_t is made, and if it is found, it is used
1230        instead, thus avoiding the use of floating point arithmetic. (There is no
1231        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1232        double.
1233    
1234    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1235        2005 (difference between two addresses compared to an unsigned value).
1236    
1237    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1238        custom one, because of the following reported problem in Windows:
1239    
1240          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1241              under Win32.
1242          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1243              therefore missing the function definition.
1244          - The compiler thus generates a "C" signature for the test function.
1245          - The linker fails to find the "C" function.
1246          - PCRE fails to configure if asked to do so against libbz2.
1247    
1248    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1249        messages were output:
1250    
1251          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1252          rerunning libtoolize, to keep the correct libtool macros in-tree.
1253          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1254    
1255        I have done both of these things.
1256    
1257    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1258        most of the time, it *can* run out if it is given a pattern that contains a
1259        runaway infinite recursion. I updated the discussion in the pcrestack man
1260        page.
1261    
1262    13. Now that we have gone to the x.xx style of version numbers, the minor
1263        version may start with zero. Using 08 or 09 is a bad idea because users
1264        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1265        interpreted as invalid octal numbers. I've updated the previous comment in
1266        configure.ac, and also added a check that gives an error if 08 or 09 are
1267        used.
1268    
1269    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1270        causing partial matching to fail when the end of the subject matched \W
1271        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1272    
1273    15. There were some discrepancies between the declarations in pcre_internal.h
1274        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1275        their definitions. The declarations used "const uschar *" and the
1276        definitions used USPTR. Even though USPTR is normally defined as "const
1277        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1278        reported that: "This difference in casting confuses some C++ compilers, for
1279        example, SunCC recognizes above declarations as different functions and
1280        generates broken code for hbpcre." I have changed the declarations to use
1281        USPTR.
1282    
1283    16. GNU libtool is named differently on some systems. The autogen.sh script now
1284        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1285        (FreeBSD).
1286    
1287    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1288        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1289        comment: "Figure out how to create a longlong from a string: strtoll and
1290        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1291        instance, but it only takes 2 args instead of 3!"
1292    
1293    18. A subtle bug concerned with back references has been fixed by a change of
1294        specification, with a corresponding code fix. A pattern such as
1295        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1296        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1297        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1298        same bug. Such groups have to be quantified to be useful, or contained
1299        inside another quantified group. (If there's no repetition, the reference
1300        can never match.) The problem arises because, having left the group and
1301        moved on to the rest of the pattern, a later failure that backtracks into
1302        the group uses the captured value from the final iteration of the group
1303        rather than the correct earlier one. I have fixed this in PCRE by forcing
1304        any group that contains a reference to itself to be an atomic group; that
1305        is, there cannot be any backtracking into it once it has completed. This is
1306        similar to recursive and subroutine calls.
1307    
1308    
1309    Version 8.00 19-Oct-09
1310    ----------------------
1311    
1312    1.  The table for translating pcre_compile() error codes into POSIX error codes
1313        was out-of-date, and there was no check on the pcre_compile() error code
1314        being within the table. This could lead to an OK return being given in
1315        error.
1316    
1317    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1318        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1319        in a Windows environment.
1320    
1321    3.  The pcregrep --count option prints the count for each file even when it is
1322        zero, as does GNU grep. However, pcregrep was also printing all files when
1323        --files-with-matches was added. Now, when both options are given, it prints
1324        counts only for those files that have at least one match. (GNU grep just
1325        prints the file name in this circumstance, but including the count seems
1326        more useful - otherwise, why use --count?) Also ensured that the
1327        combination -clh just lists non-zero counts, with no names.
1328    
1329    4.  The long form of the pcregrep -F option was incorrectly implemented as
1330        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1331        but it seems right to fix it, and I didn't think it was worth preserving
1332        the old behaviour.
1333    
1334    5.  The command line items --regex=pattern and --regexp=pattern were not
1335        recognized by pcregrep, which required --regex pattern or --regexp pattern
1336        (with a space rather than an '='). The man page documented the '=' forms,
1337        which are compatible with GNU grep; these now work.
1338    
1339    6.  No libpcreposix.pc file was created for pkg-config; there was just
1340        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1341    
1342    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1343        when UCP support is not needed, by modifying the Python script that
1344        generates it from Unicode data files. This should not matter if the module
1345        is correctly used as a library, but I received one complaint about 50K of
1346        unwanted data. My guess is that the person linked everything into his
1347        program rather than using a library. Anyway, it does no harm.
1348    
1349    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1350        was a minimum greater than 1 for a wide character in a possessive
1351        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1352        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1353        character. Chaos in the form of incorrect output or a compiling loop could
1354        result.
1355    
1356    9.  The restrictions on what a pattern can contain when partial matching is
1357        requested for pcre_exec() have been removed. All patterns can now be
1358        partially matched by this function. In addition, if there are at least two
1359        slots in the offset vector, the offset of the earliest inspected character
1360        for the match and the offset of the end of the subject are set in them when
1361        PCRE_ERROR_PARTIAL is returned.
1362    
1363    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1364        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1365        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1366        and may be more useful for multi-segment matching.
1367    
1368    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1369        used to be given if ever the end of the subject was reached; now it is
1370        given only if matching could not proceed because another character was
1371        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1372        string "Z", which now yields "no match" instead of "partial match". In the
1373        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1374        final character ended with (*FAIL).
1375    
1376    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1377        if the pattern had a "must contain" character that was already found in the
1378        earlier partial match, unless partial matching was again requested. For
1379        example, with the pattern /dog.(body)?/, the "must contain" character is
1380        "g". If the first part-match was for the string "dog", restarting with
1381        "sbody" failed. This bug has been fixed.
1382    
1383    13. The string returned by pcre_dfa_exec() after a partial match has been
1384        changed so that it starts at the first inspected character rather than the
1385        first character of the match. This makes a difference only if the pattern
1386        starts with a lookbehind assertion or \b or \B (\K is not supported by
1387        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1388        matching functions compatible, and I think it's the right thing to do.
1389    
1390    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1391        so that the demonstration program is easily available in environments where
1392        PCRE has not been installed from source.
1393    
1394    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1395        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1396        library.
1397    
1398    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1399        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1400        is not the first non-POSIX option to be added. Clearly some people find
1401        these options useful.
1402    
1403    17. If a caller to the POSIX matching function regexec() passes a non-zero
1404        value for nmatch with a NULL value for pmatch, the value of
1405        nmatch is forced to zero.
1406    
1407    18. RunGrepTest did not have a test for the availability of the -u option of
1408        the diff command, as RunTest does. It now checks in the same way as
1409        RunTest, and also checks for the -b option.
1410    
1411    19. If an odd number of negated classes containing just a single character
1412        interposed, within parentheses, between a forward reference to a named
1413        subpattern and the definition of the subpattern, compilation crashed with
1414        an internal error, complaining that it could not find the referenced
1415        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1416        [The bug was that it was starting one character too far in when skipping
1417        over the character class, thus treating the ] as data rather than
1418        terminating the class. This meant it could skip too much.]
1419    
1420    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1421        /g option in pcretest when the pattern contains \K, which makes it possible
1422        to have an empty string match not at the start, even when the pattern is
1423        anchored. Updated pcretest and pcredemo to use this option.
1424    
1425    21. If the maximum number of capturing subpatterns in a recursion was greater
1426        than the maximum at the outer level, the higher number was returned, but
1427        with unset values at the outer level. The correct (outer level) value is
1428        now given.
1429    
1430    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1431        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1432        make it do so. The string so far is captured, making this feature
1433        compatible with Perl.
1434    
1435    23. The tests have been re-organized, adding tests 11 and 12, to make it
1436        possible to check the Perl 5.10 features against Perl 5.10.
1437    
1438    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1439        pattern matches a fixed length string. PCRE did not allow this; now it
1440        does. Neither allows recursion.
1441    
1442    25. I finally figured out how to implement a request to provide the minimum
1443        length of subject string that was needed in order to match a given pattern.
1444        (It was back references and recursion that I had previously got hung up
1445        on.) This code has now been added to pcre_study(); it finds a lower bound
1446        to the length of subject needed. It is not necessarily the greatest lower
1447        bound, but using it to avoid searching strings that are too short does give
1448        some useful speed-ups. The value is available to calling programs via
1449        pcre_fullinfo().
1450    
1451    26. While implementing 25, I discovered to my embarrassment that pcretest had
1452        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1453        study optimizations had never been tested with that matching function.
1454        Oops. What is worse, even when it was passed study data, there was a bug in
1455        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1456        were also very few tests of studied patterns with pcre_dfa_exec().
1457    
1458    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1459        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1460        on the other side of the coin, they are no longer allowed to have different
1461        names, because these cannot be distinguished in PCRE, and this has caused
1462        confusion. (This is a difference from Perl.)
1463    
1464    28. When duplicate subpattern names are present (necessarily with different
1465        numbers, as required by 27 above), and a test is made by name in a
1466        conditional pattern, either for a subpattern having been matched, or for
1467        recursion in such a pattern, all the associated numbered subpatterns are
1468        tested, and the overall condition is true if the condition is true for any
1469        one of them. This is the way Perl works, and is also more like the way
1470        testing by number works.
1471    
1472    
1473    Version 7.9 11-Apr-09
1474    ---------------------
1475    
1476    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1477        (pcretest), all targets were linked against these libraries. This included
1478        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1479        libraries. This caused unwanted dependencies to be created. This problem
1480        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1481        pcretest is linked with readline.
1482    
1483    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1484        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1485        moved outside it again, because FALSE and TRUE are already defined in AIX,
1486        but BOOL is not.
1487    
1488    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1489        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1490    
1491    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1492        hyphens) following file names and line numbers when outputting matching
1493        lines. This is not true; no spaces are inserted. I have also clarified the
1494        wording for the --colour (or --color) option.
1495    
1496    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1497        was not coloured; this is different to GNU grep, so I have changed it to be
1498        the same.
1499    
1500    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1501        each matching line was coloured. Now it goes on to look for further matches
1502        of any of the test patterns, which is the same behaviour as GNU grep.
1503    
1504    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1505        doesn't make sense to accept an empty string match in pcregrep, so I have
1506        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1507        seems to be how GNU grep behaves.
1508    
1509    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1510        start or after a newline", because the conditional assertion was not being
1511        correctly handled. The rule now is that both the assertion and what follows
1512        in the first alternative must satisfy the test.
1513    
1514    9.  If auto-callout was enabled in a pattern with a conditional group whose
1515        condition was an assertion, PCRE could crash during matching, both with
1516        pcre_exec() and pcre_dfa_exec().
1517    
1518    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1519        used for matching.
1520    
1521    11. Unicode property support in character classes was not working for
1522        characters (bytes) greater than 127 when not in UTF-8 mode.
1523    
1524    12. Added the -M command line option to pcretest.
1525    
1526    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1527    
1528    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1529    
1530    16. Added comments and documentation about mis-use of no_arg in the C++
1531        wrapper.
1532    
1533    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1534        from Martin Jerabek that uses macro names for all relevant character and
1535        string constants.
1536    
1537    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1538        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1539        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1540        these, but not everybody uses configure.
1541    
1542    19. A conditional group that had only one branch was not being correctly
1543        recognized as an item that could match an empty string. This meant that an
1544        enclosing group might also not be so recognized, causing infinite looping
1545        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1546        with the subject "ab", where knowledge that the repeated group can match
1547        nothing is needed in order to break the loop.
1548    
1549    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1550        exec(), but without supplying a callout function, matching went wrong.
1551    
1552    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1553        leak if the size of the offset vector was greater than 30. When the vector
1554        is smaller, the saved offsets during recursion go onto a local stack
1555        vector, but for larger vectors malloc() is used. It was failing to free
1556        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1557        error, in fact).
1558    
1559    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1560        heapframe that is used only when UTF-8 support is enabled. This caused no
1561        problem, but was untidy.
1562    
1563    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1564        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1565        included within another project.
1566    
1567    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1568        slightly modified by me:
1569    
1570          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1571              not building pcregrep.
1572    
1573          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1574              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1575    
1576    25. Forward references, both numeric and by name, in patterns that made use of
1577        duplicate group numbers, could behave incorrectly or give incorrect errors,
1578        because when scanning forward to find the reference group, PCRE was not
1579        taking into account the duplicate group numbers. A pattern such as
1580        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1581    
1582    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1583        the feature of a custom pointer more persuasive (as requested by a user).
1584    
1585    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1586        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1587        pre-defined.
1588    
1589    28. Added support for (*UTF8) at the start of a pattern.
1590    
1591    29. Arrange for flags added by the "release type" setting in CMake to be shown
1592        in the configuration summary.
1593    
1594    
1595    Version 7.8 05-Sep-08
1596    ---------------------
1597    
1598    1.  Replaced UCP searching code with optimized version as implemented for Ad
1599        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1600        stage table and inline lookup instead of a function, giving speed ups of 2
1601        to 5 times on some simple patterns that I tested. Permission was given to
1602        distribute the MultiStage2.py script that generates the tables (it's not in
1603        the tarball, but is in the Subversion repository).
1604    
1605    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1606        scripts.
1607    
1608    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1609        a group with a zero qualifier. The result of the study could be incorrect,
1610        or the function might crash, depending on the pattern.
1611    
1612    4.  Caseless matching was not working for non-ASCII characters in back
1613        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1614        It now works when Unicode Property Support is available.
1615    
1616    5.  In pcretest, an escape such as \x{de} in the data was always generating
1617        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1618        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1619        truncation.
1620    
1621    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1622    
1623    7.  Added two (int) casts to pcregrep when printing the difference of two
1624        pointers, in case they are 64-bit values.
1625    
1626    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1627        test 2 if it fails.
1628    
1629    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1630        and a #define of that name to empty if it is not externally set. This is to
1631        allow users of MSVC to set it if necessary.
1632    
1633    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1634        the convenience functions in the pcre_get.c source file.
1635    
1636    11. An option change at the start of a pattern that had top-level alternatives
1637        could cause overwriting and/or a crash. This command provoked a crash in
1638        some environments:
1639    
1640          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1641    
1642        This potential security problem was recorded as CVE-2008-2371.
1643    
1644    12. For a pattern where the match had to start at the beginning or immediately
1645        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1646        pcre_dfa_exec() could read past the end of the passed subject if there was
1647        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1648        pcretest so that it places the subject at the end of its malloc-ed buffer.
1649    
1650    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1651        exec() might read past the end of the data buffer in UTF-8 mode.
1652    
1653    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1654        the data contained the byte 0x85 as part of a UTF-8 character within its
1655        first line. This applied both to normal and DFA matching.
1656    
1657    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1658        /^[^d]*?$/8 failed to match "abc".
1659    
1660    16. Added a missing copyright notice to pcrecpp_internal.h.
1661    
1662    17. Make it more clear in the documentation that values returned from
1663        pcre_exec() in ovector are byte offsets, not character counts.
1664    
1665    18. Tidied a few places to stop certain compilers from issuing warnings.
1666    
1667    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1668        supplied by Stefan Weber. I made a further small update for 7.8 because
1669        there is a change of source arrangements: the pcre_searchfuncs.c module is
1670        replaced by pcre_ucd.c.
1671    
1672    
1673    Version 7.7 07-May-08
1674    ---------------------
1675    
1676    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1677        a string to a long long, pretend we don't even have a long long." This is
1678        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1679    
1680    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1681        pre-7.6 versions, which defined a global no_arg variable instead of putting
1682        it in the RE class. (See also #8 below.)
1683    
1684    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1685        Lopes.
1686    
1687    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1688    
1689        (1) The include/exclude patterns were being applied to the whole pathnames
1690            of files, instead of just to the final components.
1691    
1692        (2) If there was more than one level of directory, the subdirectories were
1693            skipped unless they satisfied the include/exclude conditions. This is
1694            inconsistent with GNU grep (and could even be seen as contrary to the
1695            pcregrep specification - which I improved to make it absolutely clear).
1696            The action now is always to scan all levels of directory, and just
1697            apply the include/exclude patterns to regular files.
1698    
1699    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1700        --exclude_dir in the tests to avoid scanning .svn directories.
1701    
1702    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1703        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1704        doesn't support NULs in patterns.
1705    
1706    7.  Added some missing "const"s to declarations of static tables in
1707        pcre_compile.c and pcre_dfa_exec.c.
1708    
1709    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1710        caused by fix #2  above. (Subsequently also a second patch to fix the
1711        first patch. And a third patch - this was a messy problem.)
1712    
1713    9.  Applied Craig's patch to remove the use of push_back().
1714    
1715    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1716        matching function regexec().
1717    
1718    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1719        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1720        references. PCRE supports relative numbers with this syntax (I don't think
1721        Oniguruma does).
1722    
1723    12. Previously, a group with a zero repeat such as (...){0} was completely
1724        omitted from the compiled regex. However, this means that if the group
1725        was called as a subroutine from elsewhere in the pattern, things went wrong
1726        (an internal error was given). Such groups are now left in the compiled
1727        pattern, with a new opcode that causes them to be skipped at execution
1728        time.
1729    
1730    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1731        to the way PCRE behaves:
1732    
1733        (a) A lone ] character is dis-allowed (Perl treats it as data).
1734    
1735        (b) A back reference to an unmatched subpattern matches an empty string
1736            (Perl fails the current match path).
1737    
1738        (c) A data ] in a character class must be notated as \] because if the
1739            first data character in a class is ], it defines an empty class. (In
1740            Perl it is not possible to have an empty class.) The empty class []
1741            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1742            The negative empty class [^] matches any one character, independently
1743            of the DOTALL setting.
1744    
1745    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1746        non-existent subpattern following a character class starting with ']' and
1747        containing () gave an internal compiling error instead of "reference to
1748        non-existent subpattern". Fortunately, when the pattern did exist, the
1749        compiled code was correct. (When scanning forwards to check for the
1750        existencd of the subpattern, it was treating the data ']' as terminating
1751        the class, so got the count wrong. When actually compiling, the reference
1752        was subsequently set up correctly.)
1753    
1754    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1755        it was being rejected as not supported by pcre_dfa_exec(), even though
1756        other assertions are supported. I have made pcre_dfa_exec() support
1757        (*FAIL).
1758    
1759    16. The implementation of 13c above involved the invention of a new opcode,
1760        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1761        cannot be changed at match time, I realized I could make a small
1762        improvement to matching performance by compiling OP_ALLANY instead of
1763        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1764        on the OP_ANY path.
1765    
1766    17. Compiling pcretest on Windows with readline support failed without the
1767        following two fixes: (1) Make the unistd.h include conditional on
1768        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1769    
1770    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1771        ncurses library to be included for pcretest when ReadLine support is
1772        requested, but also to allow for it to be overridden. This patch came from
1773        Daniel Bergström.
1774    
1775    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1776        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1777        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1778        spotting this.
1779    
1780    
1781    Version 7.6 28-Jan-08
1782    ---------------------
1783    
1784    1.  A character class containing a very large number of characters with
1785        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1786        overflow.
1787    
1788    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1789        HAVE_LONG_LONG is not defined.
1790    
1791    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1792        bring them up to date and include new features. This patch includes:
1793    
1794        - Fixed PH's badly added libz and libbz2 support.
1795        - Fixed a problem with static linking.
1796        - Added pcredemo. [But later removed - see 7 below.]
1797        - Fixed dftables problem and added an option.
1798        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1799            HAVE_LONG_LONG.
1800        - Added readline support for pcretest.
1801        - Added an listing of the option settings after cmake has run.
1802    
1803    4.  A user submitted a patch to Makefile that makes it easy to create
1804        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1805        Makefile.am that cause it to include this special target, without
1806        affecting anything else. Note that the same mingw target plus all
1807        the other distribution libraries and programs are now supported
1808        when configuring with CMake (see 6 below) instead of with
1809        Configure/Make.
1810    
1811    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1812        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1813        exported in the Windows port". It has not yet been confirmed that the patch
1814        solves the problem, but it does no harm.
1815    
1816    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1817        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1818        with CMake, and also correct the comment about stack recursion.
1819    
1820    7.  Remove the automatic building of pcredemo from the ./configure system and
1821        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1822        of a program that users should build themselves after PCRE is installed, so
1823        building it automatically is not really right. What is more, it gave
1824        trouble in some build environments.
1825    
1826    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1827    
1828    
1829    Version 7.5 10-Jan-08
1830    ---------------------
1831    
1832    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1833        values in parens when parsing an RE using the C++ wrapper."
1834    
1835    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1836        Characters greater than 255 were excluded from the class instead of being
1837        included.
1838    
1839    3.  The same bug as (2) above applied to negated POSIX classes such as
1840        [:^space:].
1841    
1842    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1843        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1844        I have changed it.
1845    
1846    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1847        first named subpattern) and a construct such as (?&a) would reference the
1848        first named subpattern whose name started with "a" (in other words, the
1849        length check was missing). Both these problems are fixed. "Subpattern name
1850        expected" is now given for (?&) (a zero-length name), and this patch also
1851        makes it give the same error for \k'' (previously it complained that that
1852        was a reference to a non-existent subpattern).
1853    
1854    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1855        this is right because (?- can be followed by option settings as well as by
1856        digits. I have, however, made the messages clearer.
1857    
1858    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1859        than the number used in the conditional) now cause a compile-time error.
1860        This is actually not compatible with Perl, which accepts such patterns, but
1861        treats the conditional as always being FALSE (as PCRE used to), but it
1862        seems to me that giving a diagnostic is better.
1863    
1864    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1865        and messages.
1866    
1867    9.  Fix two occurrences of "backslash" in comments that should have been
1868        "backspace".
1869    
1870    10. Remove two redundant lines of code that can never be obeyed (their function
1871        was moved elsewhere).
1872    
1873    11. The program that makes PCRE's Unicode character property table had a bug
1874        which caused it to generate incorrect table entries for sequences of
1875        characters that have the same character type, but are in different scripts.
1876        It amalgamated them into a single range, with the script of the first of
1877        them. In other words, some characters were in the wrong script. There were
1878        thirteen such cases, affecting characters in the following ranges:
1879    
1880          U+002b0 - U+002c1
1881          U+0060c - U+0060d
1882          U+0061e - U+00612
1883          U+0064b - U+0065e
1884          U+0074d - U+0076d
1885          U+01800 - U+01805
1886          U+01d00 - U+01d77
1887          U+01d9b - U+01dbf
1888          U+0200b - U+0200f
1889          U+030fc - U+030fe
1890          U+03260 - U+0327f
1891          U+0fb46 - U+0fbb1
1892          U+10450 - U+1049d
1893    
1894    12. The -o option (show only the matching part of a line) for pcregrep was not
1895        compatible with GNU grep in that, if there was more than one match in a
1896        line, it showed only the first of them. It now behaves in the same way as
1897        GNU grep.
1898    
1899    13. If the -o and -v options were combined for pcregrep, it printed a blank
1900        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1901        does the same. The return code can be used to tell if there were any
1902        non-matching lines.
1903    
1904    14. Added --file-offsets and --line-offsets to pcregrep.
1905    
1906    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1907        infinitely looping recursion. The bug was that positive lookaheads were not
1908        being skipped when checking for a possible empty match (negative lookaheads
1909        and both kinds of lookbehind were skipped).
1910    
1911    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1912        inclusion of <windows.h> to before rather than after the definition of
1913        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1914    
1915    17. Specifying a possessive quantifier with a specific limit for a Unicode
1916        character property caused pcre_compile() to compile bad code, which led at
1917        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1918        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1919        caused the error; without that there was no problem.
1920    
1921    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1922    
1923    19. Added --enable-pcretest-libreadline.
1924    
1925    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1926        RE::GlobalReplace(). As a result, the number of replacements returned was
1927        double what it should be. I removed one of the increments, but Craig sent a
1928        later patch that removed the other one (the right fix) and added unit tests
1929        that check the return values (which was not done before).
1930    
1931    21. Several CMake things:
1932    
1933        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1934            the names libpcre and libpcreposix, not just pcre and pcreposix.
1935    
1936        (2) The above change means that pcretest and pcregrep are now correctly
1937            linked with the newly-built libraries, not previously installed ones.
1938    
1939        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
1940    
1941    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1942        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1943        UTF-8 newline character). The key issue is that the pattern starts .*;
1944        this means that the match must be either at the beginning, or after a
1945        newline. The bug was in the code for advancing after a failed match and
1946        checking that the new position followed a newline. It was not taking
1947        account of UTF-8 characters correctly.
1948    
1949    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1950        character classes. PCRE was not treating the sequence [:...:] as a
1951        character class unless the ... were all letters. Perl, however, seems to
1952        allow any characters between [: and :], though of course it rejects as
1953        unknown any "names" that contain non-letters, because all the known class
1954        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1955        for example, whereas PCRE did not - it did not recognize a POSIX character
1956        class. This seemed a bit dangerous, so the code has been changed to be
1957        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1958        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1959        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1960        Perl does, and where it didn't before.
1961    
1962    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1963        Windows environments %n is disabled by default.
1964    
1965    
1966    Version 7.4 21-Sep-07
1967  ---------------------  ---------------------
1968    
1969  1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This  1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
# Line 25  Version 7.4 10-Sep-07 Line 1987  Version 7.4 10-Sep-07
1987    
1988  4.  Added options (at build time, compile time, exec time) to change \R from  4.  Added options (at build time, compile time, exec time) to change \R from
1989      matching any Unicode line ending sequence to just matching CR, LF, or CRLF.      matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
   
 5.  doc/pcresyntax.html was missing from the distribution.  
1990    
1991  6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward  5.  doc/pcresyntax.html was missing from the distribution.
1992    
1993    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1994      compatibility, even though it is no longer used.      compatibility, even though it is no longer used.
1995    
1996  7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and  7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1997      strtoull to pcrecpp.cc to select the available functions in WIN32 (where      strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1998      different names are used).      windows.h file is present (where different names are used). [This was
1999        reversed later after testing - see 16 below.]
2000    
2001    8.  Changed all #include <config.h> to #include "config.h". There were also
2002        some further <pcre.h> cases that I changed to "pcre.h".
2003    
2004    9.  When pcregrep was used with the --colour option, it missed the line ending
2005        sequence off the lines that it output.
2006    
2007    10. It was pointed out to me that arrays of string pointers cause lots of
2008        relocations when a shared library is dynamically loaded. A technique of
2009        using a single long string with a table of offsets can drastically reduce
2010        these. I have refactored PCRE in four places to do this. The result is
2011        dramatic:
2012    
2013          Originally:                          290
2014          After changing UCP table:            187
2015          After changing error message table:   43
2016          After changing table of "verbs"       36
2017          After changing table of Posix names   22
2018    
2019        Thanks to the folks working on Gregex for glib for this insight.
2020    
2021    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
2022        unicode-properties was also set.
2023    
2024    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
2025    
2026    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
2027        checked only for CRLF.
2028    
2029    14. Added casts to pcretest.c to avoid compiler warnings.
2030    
2031    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
2032    
2033    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
2034        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
2035        entirely. This removes changes made in 7 above.
2036    
2037    17. The CMake files have been updated, and there is now more information about
2038        building with CMake in the NON-UNIX-USE document.
2039    
2040    
2041  Version 7.3 28-Aug-07  Version 7.3 28-Aug-07

Legend:
Removed from v.234  
changed lines
  Added in v.1217

  ViewVC Help
Powered by ViewVC 1.1.5