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

Diff of /code/trunk/ChangeLog

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

revision 158 by ph10, Tue May 1 10:30:14 2007 UTC revision 1226 by ph10, Fri Nov 30 09:14:29 2012 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4  Version 7.2 01-May-07  Version 8.32 30-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    25. (*UTF) can now be used to start a pattern in any of the three libraries.
164    
165    26. Give configure error if --enable-cpp but no C++ compiler found.
166    
167    
168    Version 8.31 06-July-2012
169    -------------------------
170    
171    1.  Fixing a wrong JIT test case and some compiler warnings.
172    
173    2.  Removed a bashism from the RunTest script.
174    
175    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
176        to unsigned type, result still unsigned" that was given by an MS compiler
177        on encountering the code "-sizeof(xxx)".
178    
179    4.  Partial matching support is added to the JIT compiler.
180    
181    5.  Fixed several bugs concerned with partial matching of items that consist
182        of more than one character:
183    
184        (a) /^(..)\1/ did not partially match "aba" because checking references was
185            done on an "all or nothing" basis. This also applied to repeated
186            references.
187    
188        (b) \R did not give a hard partial match if \r was found at the end of the
189            subject.
190    
191        (c) \X did not give a hard partial match after matching one or more
192            characters at the end of the subject.
193    
194        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
195            a partial match for the string "\r".
196    
197        (e) When newline was set to CRLF, the metacharacter "." did not recognize
198            a partial match for a CR character at the end of the subject string.
199    
200    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
201        running pcretest, the text "(JIT)" added to the output whenever JIT is
202        actually used to run the match.
203    
204    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
205        or /S+[+] with a digit between 1 and 7.
206    
207    8.  OP_NOT now supports any UTF character not just single-byte ones.
208    
209    9.  (*MARK) control verb is now supported by the JIT compiler.
210    
211    10. The command "./RunTest list" lists the available tests without actually
212        running any of them. (Because I keep forgetting what they all are.)
213    
214    11. Add PCRE_INFO_MAXLOOKBEHIND.
215    
216    12. Applied a (slightly modified) user-supplied patch that improves performance
217        when the heap is used for recursion (compiled with --disable-stack-for-
218        recursion). Instead of malloc and free for each heap frame each time a
219        logical recursion happens, frames are retained on a chain and re-used where
220        possible. This sometimes gives as much as 30% improvement.
221    
222    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
223        call.
224    
225    14. As documented, (*COMMIT) is now confined to within a positive assertion.
226    
227    15. It is now possible to link pcretest with libedit as an alternative to
228        libreadline.
229    
230    16. (*COMMIT) control verb is now supported by the JIT compiler.
231    
232    17. The Unicode data tables have been updated to Unicode 6.1.0.
233    
234    18. Added --file-list option to pcregrep.
235    
236    19. Added binary file support to pcregrep, including the -a, --binary-files,
237        -I, and --text options.
238    
239    20. The madvise function is renamed for posix_madvise for QNX compatibility
240        reasons. Fixed by Giuseppe D'Angelo.
241    
242    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
243    
244    22. Changed the option for creating symbolic links for 16-bit man pages from
245        -s to -sf so that re-installing does not cause issues.
246    
247    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
248    
249    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
250        matches in certain environments (the workspace was not being correctly
251        retained). Also added to pcre_dfa_exec() a simple plausibility check on
252        some of the workspace data at the beginning of a restart.
253    
254    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
255        was not doing so when it should - probably a typo introduced by SVN 528
256        (change 8.10/14).
257    
258    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
259        \w+ when the character tables indicated that \x{c4} was a word character.
260        There were several related cases, all because the tests for doing a table
261        lookup were testing for characters less than 127 instead of 255.
262    
263    27. If a pattern contains capturing parentheses that are not used in a match,
264        their slots in the ovector are set to -1. For those that are higher than
265        any matched groups, this happens at the end of processing. In the case when
266        there were back references that the ovector was too small to contain
267        (causing temporary malloc'd memory to be used during matching), and the
268        highest capturing number was not used, memory off the end of the ovector
269        was incorrectly being set to -1. (It was using the size of the temporary
270        memory instead of the true size.)
271    
272    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
273        ovector size, it uses memory at the end of the block that it has got.
274    
275    29. Check for an overlong MARK name and give an error at compile time. The
276        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
277    
278    30. JIT compiler update.
279    
280    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
281        Rill for the patch.
282    
283    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
284    
285    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
286    
287    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
288        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
289        was enabled.
290    
291    35. Improve JIT code generation for greedy plus quantifier.
292    
293    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
294        1 to "aa" instead of to an empty string. The bug affected repeated groups
295        that could potentially match an empty string.
296    
297    37. Optimizing single character iterators in JIT.
298    
299    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
300        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
301        codepoints that are too big for the mode are faulted, and in a UTF mode,
302        disallowed codepoints are also faulted.
303    
304    39. If PCRE was compiled with UTF support, in three places in the DFA
305        matcher there was code that should only have been obeyed in UTF mode, but
306        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
307        processing when bytes with values greater than 127 were present. In 16-bit
308        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
309        both cases the values are those that cannot be the first data item in a UTF
310        character. The three items that might have provoked this were recursions,
311        possessively repeated groups, and atomic groups.
312    
313    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
314        and pcregrep, because some OS require shared objects to be explicitly
315        passed to ld, causing the link step to fail if they are not.
316    
317    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
318        mode, patterns that started with \h* or \R* might be incorrectly matched.
319    
320    
321    Version 8.30 04-February-2012
322    -----------------------------
323    
324    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
325        name is defined in ctype.h.
326    
327    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
328        only in quite long subpatterns.
329    
330    3.  Removed the function pcre_info(), which has been obsolete and deprecated
331        since it was replaced by pcre_fullinfo() in February 2000.
332    
333    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
334        match a (*MARK), and the match failed at the start of the subject, a
335        reference to memory before the start of the subject could occur. This bug
336        was introduced by fix 17 of release 8.21.
337    
338    5.  A reference to an unset group with zero minimum repetition was giving
339        totally wrong answers (in non-JavaScript-compatibility mode). For example,
340        /(another)?(\1?)test/ matched against "hello world test". This bug was
341        introduced in release 8.13.
342    
343    6.  Add support for 16-bit character strings (a large amount of work involving
344        many changes and refactorings).
345    
346    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
347        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
348        from a file.
349    
350    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
351        rounding is not applied in this particular case).
352    
353    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
354        if they appear, or are escaped, in patterns.
355    
356    10. Get rid of a number of -Wunused-but-set-variable warnings.
357    
358    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
359        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
360        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
361        also returns the mark "x". This bug applied to capturing parentheses,
362        non-capturing parentheses, and atomic parentheses. It also applied to some
363        assertions.
364    
365    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
366        information out of configure.ac instead of relying on pcre.h.generic, which
367        is not stored in the repository.
368    
369    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
370        -lreadline.
371    
372    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
373    
374    15. Applied Graycode's patch to put the top-level frame on the stack rather
375        than the heap when not using the stack for recursion. This gives a
376        performance improvement in many cases when recursion is not deep.
377    
378    16. Experimental code added to "pcretest -C" to output the stack frame size.
379    
380    
381    Version 8.21 12-Dec-2011
382    ------------------------
383    
384    1.  Updating the JIT compiler.
385    
386    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
387        are added as well.
388    
389    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
390        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
391        calling _pcre_jit_exec. Some extra comments are added.
392    
393    4.  (*MARK) settings inside atomic groups that do not contain any capturing
394        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
395        was introduced by change 18 for 8.20.
396    
397    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
398        ECMA-262 standard.
399    
400    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
401        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
402        This bug was probably introduced by change 9 of 8.13.
403    
404    7.  While fixing 6 above, I noticed that a number of other items were being
405        incorrectly rejected as "not fixed length". This arose partly because newer
406        opcodes had not been added to the fixed-length checking code. I have (a)
407        corrected the bug and added tests for these items, and (b) arranged for an
408        error to occur if an unknown opcode is encountered while checking for fixed
409        length instead of just assuming "not fixed length". The items that were
410        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
411        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
412        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
413    
414    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
415        being incorrectly compiled and would have given unpredicatble results.
416    
417    9.  A possessively repeated subpattern with minimum repeat count greater than
418        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
419        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
420        the first (A) could occur when it should not.
421    
422    10. Add a cast and remove a redundant test from the code.
423    
424    11. JIT should use pcre_malloc/pcre_free for allocation.
425    
426    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
427        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
428        is anything other than /usr, -L is still shown).
429    
430    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
431    
432    14. Perl does not support \N without a following name in a [] class; PCRE now
433        also gives an error.
434    
435    15. If a forward reference was repeated with an upper limit of around 2000,
436        it caused the error "internal error: overran compiling workspace". The
437        maximum number of forward references (including repeats) was limited by the
438        internal workspace, and dependent on the LINK_SIZE. The code has been
439        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
440        the default depends on LINK_SIZE. There is a new upper limit (for safety)
441        of around 200,000 forward references. While doing this, I also speeded up
442        the filling in of repeated forward references.
443    
444    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
445        incorrectly expecting the subject to contain another "a" after the start.
446    
447    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
448        in the match, the SKIP should be ignored. This was not happening; instead
449        the SKIP was being treated as NOMATCH. For patterns such as
450        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
451        tested.
452    
453    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
454        now much more compatible with Perl, in particular in cases where the result
455        is a non-match for a non-anchored pattern. For example, if
456        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
457        "m", where previously it did not return a name. A side effect of this
458        change is that for partial matches, the last encountered mark name is
459        returned, as for non matches. A number of tests that were previously not
460        Perl-compatible have been moved into the Perl-compatible test files. The
461        refactoring has had the pleasing side effect of removing one argument from
462        the match() function, thus reducing its stack requirements.
463    
464    19. If the /S+ option was used in pcretest to study a pattern using JIT,
465        subsequent uses of /S (without +) incorrectly behaved like /S+.
466    
467    21. Retrieve executable code size support for the JIT compiler and fixing
468        some warnings.
469    
470    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
471        not work when the shorter character appeared right at the end of the
472        subject string.
473    
474    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
475        systems.
476    
477    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
478        output it when the /M option is used in pcretest.
479    
480    25. The CheckMan script was not being included in the distribution. Also, added
481        an explicit "perl" to run Perl scripts from the PrepareRelease script
482        because this is reportedly needed in Windows.
483    
484    26. If study data was being save in a file and studying had not found a set of
485        "starts with" bytes for the pattern, the data written to the file (though
486        never used) was taken from uninitialized memory and so caused valgrind to
487        complain.
488    
489    27. Updated RunTest.bat as provided by Sheri Pierce.
490    
491    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
492    
493    29. Computation of memory usage for the table of capturing group names was
494        giving an unnecessarily large value.
495    
496    
497    Version 8.20 21-Oct-2011
498    ------------------------
499    
500    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
501        a POSIX class. After further experiments with Perl, which convinced me that
502        Perl has bugs and confusions, a closing square bracket is no longer allowed
503        in a POSIX name. This bug also affected patterns with classes that started
504        with full stops.
505    
506    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
507        captured substring, but while checking the failing first alternative,
508        substring 1 is temporarily captured. If the output vector supplied to
509        pcre_exec() was not big enough for this capture, the yield of the function
510        was still zero ("insufficient space for captured substrings"). This cannot
511        be totally fixed without adding another stack variable, which seems a lot
512        of expense for a edge case. However, I have improved the situation in cases
513        such as /(a)(b)x|abc/ matched against "abc", where the return code
514        indicates that fewer than the maximum number of slots in the ovector have
515        been set.
516    
517    3.  Related to (2) above: when there are more back references in a pattern than
518        slots in the output vector, pcre_exec() uses temporary memory during
519        matching, and copies in the captures as far as possible afterwards. It was
520        using the entire output vector, but this conflicts with the specification
521        that only 2/3 is used for passing back captured substrings. Now it uses
522        only the first 2/3, for compatibility. This is, of course, another edge
523        case.
524    
525    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
526        main code base, and can be used by building with --enable-jit. When this is
527        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
528        runtime --no-jit option is given.
529    
530    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
531        ovector, the return from the function was zero, implying that there were
532        other matches that did not fit. The correct "exactly full" value is now
533        returned.
534    
535    6.  If a subpattern that was called recursively or as a subroutine contained
536        (*PRUNE) or any other control that caused it to give a non-standard return,
537        invalid errors such as "Error -26 (nested recursion at the same subject
538        position)" or even infinite loops could occur.
539    
540    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
541        computing the minimum length on reaching *ACCEPT, and so ended up with the
542        wrong value of 1 rather than 0. Further investigation indicates that
543        computing a minimum subject length in the presence of *ACCEPT is difficult
544        (think back references, subroutine calls), and so I have changed the code
545        so that no minimum is registered for a pattern that contains *ACCEPT.
546    
547    8.  If (*THEN) was present in the first (true) branch of a conditional group,
548        it was not handled as intended. [But see 16 below.]
549    
550    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
551        Sheri Pierce.
552    
553    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
554        the first byte in a match must be "a".
555    
556    11. Change 17 for 8.13 increased the recursion depth for patterns like
557        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
558        pattern contains any instances of (*THEN). If it does not, the old
559        optimizations are restored. It would be nice to do this on a per-group
560        basis, but at the moment that is not feasible.
561    
562    12. In some environments, the output of pcretest -C is CRLF terminated. This
563        broke RunTest's code that checks for the link size. A single white space
564        character after the value is now allowed for.
565    
566    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
567        For "fr", it uses the Windows-specific input and output files.
568    
569    14. If (*THEN) appeared in a group that was called recursively or as a
570        subroutine, it did not work as intended. [But see next item.]
571    
572    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
573        pattern fragments (but not containing any | characters). If A and B are
574        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
575        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
576        D. In other words, Perl considers parentheses that do not contain any |
577        characters to be part of a surrounding alternative, whereas PCRE was
578        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
579        differently. PCRE now behaves in the same way as Perl, except in the case
580        of subroutine/recursion calls such as (?1) which have in any case always
581        been different (but PCRE had them first :-).
582    
583    16. Related to 15 above: Perl does not treat the | in a conditional group as
584        creating alternatives. Such a group is treated in the same way as an
585        ordinary group without any | characters when processing (*THEN). PCRE has
586        been changed to match Perl's behaviour.
587    
588    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
589        RunGrepTest script failed.
590    
591    18. Change 22 for version 13 caused atomic groups to use more stack. This is
592        inevitable for groups that contain captures, but it can lead to a lot of
593        stack use in large patterns. The old behaviour has been restored for atomic
594        groups that do not contain any capturing parentheses.
595    
596    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
597        suppress the check for a minimum subject length at run time. (If it was
598        given to pcre_exec() or pcre_dfa_exec() it did work.)
599    
600    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
601        fail to work when decoding hex characters in data strings in EBCDIC
602        environments.
603    
604    21. It appears that in at least one Mac OS environment, the isxdigit() function
605        is implemented as a macro that evaluates to its argument more than once,
606        contravening the C 90 Standard (I haven't checked a later standard). There
607        was an instance in pcretest which caused it to go wrong when processing
608        \x{...} escapes in subject strings. The has been rewritten to avoid using
609        things like p++ in the argument of isxdigit().
610    
611    
612    Version 8.13 16-Aug-2011
613    ------------------------
614    
615    1.  The Unicode data tables have been updated to Unicode 6.0.0.
616    
617    2.  Two minor typos in pcre_internal.h have been fixed.
618    
619    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
620        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
621        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
622    
623    4.  There were a number of related bugs in the code for matching backrefences
624        caselessly in UTF-8 mode when codes for the characters concerned were
625        different numbers of bytes. For example, U+023A and U+2C65 are an upper
626        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
627        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
628        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
629        2-byte code at the end of the subject (it thought there wasn't enough data
630        left).
631    
632    5.  Comprehensive information about what went wrong is now returned by
633        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
634        as the output vector has at least 2 elements. The offset of the start of
635        the failing character and a reason code are placed in the vector.
636    
637    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
638        now returned is for the first byte of the failing character, instead of the
639        last byte inspected. This is an incompatible change, but I hope it is small
640        enough not to be a problem. It makes the returned offset consistent with
641        pcre_exec() and pcre_dfa_exec().
642    
643    7.  pcretest now gives a text phrase as well as the error number when
644        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
645        failure, the offset and reason code are output.
646    
647    8.  When \R was used with a maximizing quantifier it failed to skip backwards
648        over a \r\n pair if the subsequent match failed. Instead, it just skipped
649        back over a single character (\n). This seems wrong (because it treated the
650        two characters as a single entity when going forwards), conflicts with the
651        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
652        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
653        has been changed.
654    
655    9.  Some internal refactoring has changed the processing so that the handling
656        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
657        time (the PCRE_DOTALL option was changed this way some time ago: version
658        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
659        which was always a bit of a fudge. It also means that there is one less
660        argument for the match() function, which reduces its stack requirements
661        slightly. This change also fixes an incompatibility with Perl: the pattern
662        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
663    
664    10. More internal refactoring has drastically reduced the number of recursive
665        calls to match() for possessively repeated groups such as (abc)++ when
666        using pcre_exec().
667    
668    11. While implementing 10, a number of bugs in the handling of groups were
669        discovered and fixed:
670    
671        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
672        (a|)*(?1) gave a compile-time internal error.
673        ((a|)+)+  did not notice that the outer group could match an empty string.
674        (^a|^)+   was not marked as anchored.
675        (.*a|.*)+ was not marked as matching at start or after a newline.
676    
677    12. Yet more internal refactoring has removed another argument from the match()
678        function. Special calls to this function are now indicated by setting a
679        value in a variable in the "match data" data block.
680    
681    13. Be more explicit in pcre_study() instead of relying on "default" for
682        opcodes that mean there is no starting character; this means that when new
683        ones are added and accidentally left out of pcre_study(), testing should
684        pick them up.
685    
686    14. The -s option of pcretest has been documented for ages as being an old
687        synonym of -m (show memory usage). I have changed it to mean "force study
688        for every regex", that is, assume /S for every regex. This is similar to -i
689        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
690        using it. It makes it easier to run collections of tests with and without
691        study enabled, and thereby test pcre_study() more easily. All the standard
692        tests are now run with and without -s (but some patterns can be marked as
693        "never study" - see 20 below).
694    
695    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
696        restoration of the capturing data to the outer values was not happening
697        correctly.
698    
699    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
700        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
701        pattern had matched an empty string, and so incorrectly returned a no
702        match.
703    
704    17. There was optimizing code for the last branch of non-capturing parentheses,
705        and also for the obeyed branch of a conditional subexpression, which used
706        tail recursion to cut down on stack usage. Unfortunately, now that there is
707        the possibility of (*THEN) occurring in these branches, tail recursion is
708        no longer possible because the return has to be checked for (*THEN). These
709        two optimizations have therefore been removed. [But see 8.20/11 above.]
710    
711    18. If a pattern containing \R was studied, it was assumed that \R always
712        matched two bytes, thus causing the minimum subject length to be
713        incorrectly computed because \R can also match just one byte.
714    
715    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
716        was incorrectly computed.
717    
718    20. If /S is present twice on a test pattern in pcretest input, it now
719        *disables* studying, thereby overriding the use of -s on the command line
720        (see 14 above). This is necessary for one or two tests to keep the output
721        identical in both cases.
722    
723    21. When (*ACCEPT) was used in an assertion that matched an empty string and
724        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
725    
726    22. When an atomic group that contained a capturing parenthesis was
727        successfully matched, but the branch in which it appeared failed, the
728        capturing was not being forgotten if a higher numbered group was later
729        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
730        group 1 to "a", when in fact it should be unset. This applied to multi-
731        branched capturing and non-capturing groups, repeated or not, and also to
732        positive assertions (capturing in negative assertions does not happen
733        in PCRE) and also to nested atomic groups.
734    
735    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
736        subject after a captured substring, to make it easier to tell which of a
737        number of identical substrings has been captured.
738    
739    24. The way atomic groups are processed by pcre_exec() has been changed so that
740        if they are repeated, backtracking one repetition now resets captured
741        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
742        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
743        "aaa". Previously, it would have been "a". As part of this code
744        refactoring, the way recursive calls are handled has also been changed.
745    
746    25. If an assertion condition captured any substrings, they were not passed
747        back unless some other capturing happened later. For example, if
748        (?(?=(a))a) was matched against "a", no capturing was returned.
749    
750    26. When studying a pattern that contained subroutine calls or assertions,
751        the code for finding the minimum length of a possible match was handling
752        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
753        group 1 called group 2 while simultaneously a separate group 2 called group
754        1). A stack overflow occurred in this case. I have fixed this by limiting
755        the recursion depth to 10.
756    
757    27. Updated RunTest.bat in the distribution to the version supplied by Tom
758        Fortmann. This supports explicit test numbers on the command line, and has
759        argument validation and error reporting.
760    
761    28. An instance of \X with an unlimited repeat could fail if at any point the
762        first character it looked at was a mark character.
763    
764    29. Some minor code refactoring concerning Unicode properties and scripts
765        should reduce the stack requirement of match() slightly.
766    
767    30. Added the '=' option to pcretest to check the setting of unused capturing
768        slots at the end of the pattern, which are documented as being -1, but are
769        not included in the return count.
770    
771    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
772        compiled something random. Now it gives a compile-time error (as does
773        Perl).
774    
775    32. A *MARK encountered during the processing of a positive assertion is now
776        recorded and passed back (compatible with Perl).
777    
778    33. If --only-matching or --colour was set on a pcregrep call whose pattern
779        had alternative anchored branches, the search for a second match in a line
780        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
781        matched the line "0102" twice. The same bug affected patterns that started
782        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
783        twice.
784    
785    34. Previously, PCRE did not allow quantification of assertions. However, Perl
786        does, and because of capturing effects, quantifying parenthesized
787        assertions may at times be useful. Quantifiers are now allowed for
788        parenthesized assertions.
789    
790    35. A minor code tidy in pcre_compile() when checking options for \R usage.
791    
792    36. \g was being checked for fancy things in a character class, when it should
793        just be a literal "g".
794    
795    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
796        appearance of a nested POSIX class supersedes an apparent external class.
797        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
798        unescaped square brackets may also appear as part of class names. For
799        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
800        more like Perl. (But see 8.20/1 above.)
801    
802    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
803        was because it thought it was \N{name}, which is not supported).
804    
805    39. Add minix to OS list not supporting the -S option in pcretest.
806    
807    40. PCRE tries to detect cases of infinite recursion at compile time, but it
808        cannot analyze patterns in sufficient detail to catch mutual recursions
809        such as ((?1))((?2)). There is now a runtime test that gives an error if a
810        subgroup is called recursively as a subpattern for a second time at the
811        same position in the subject string. In previous releases this might have
812        been caught by the recursion limit, or it might have run out of stack.
813    
814    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
815        happen only once. PCRE was, however incorrectly giving a compile time error
816        "recursive call could loop indefinitely" because it cannot analyze the
817        pattern in sufficient detail. The compile time test no longer happens when
818        PCRE is compiling a conditional subpattern, but actual runaway loops are
819        now caught at runtime (see 40 above).
820    
821    42. It seems that Perl allows any characters other than a closing parenthesis
822        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
823        has been changed to be the same.
824    
825    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
826        as not to get warnings when autogen.sh is called. Also changed
827        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
828    
829    44. To help people who use pcregrep to scan files containing exceedingly long
830        lines, the following changes have been made:
831    
832        (a) The default value of the buffer size parameter has been increased from
833            8K to 20K. (The actual buffer used is three times this size.)
834    
835        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
836            PCRE is built.
837    
838        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
839            to be set at run time.
840    
841        (d) Numerical values in pcregrep options can be followed by K or M, for
842            example --buffer-size=50K.
843    
844        (e) If a line being scanned overflows pcregrep's buffer, an error is now
845            given and the return code is set to 2.
846    
847    45. Add a pointer to the latest mark to the callout data block.
848    
849    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
850        partial match of an empty string instead of no match. This was specific to
851        the use of ".".
852    
853    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
854        complete match instead of a partial match. This bug was dependent on both
855        the PCRE_UTF8 and PCRE_DOTALL options being set.
856    
857    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
858        starting byte set, because \b was not being ignored.
859    
860    
861    Version 8.12 15-Jan-2011
862    ------------------------
863    
864    1.  Fixed some typos in the markup of the man pages, and wrote a script that
865        checks for such things as part of the documentation building process.
866    
867    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
868        --match-limit and --recursion-limit options (added for 8.11). In
869        particular, this made one of the standard tests fail. (The integer value
870        went into the wrong half of a long int.)
871    
872    3.  If the --colour option was given to pcregrep with -v (invert match), it
873        did strange things, either producing crazy output, or crashing. It should,
874        of course, ignore a request for colour when reporting lines that do not
875        match.
876    
877    4.  Another pcregrep bug caused similar problems if --colour was specified with
878        -M (multiline) and the pattern match finished with a line ending.
879    
880    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
881        matched in multiline mode, the following line was shown as part of the
882        match. This seems wrong, so I have changed it.
883    
884    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
885        the check for further matches in the same line (so they could be coloured)
886        to overrun the end of the current line. If another match was found, it was
887        incorrectly shown (and then shown again when found in the next line).
888    
889    7.  If pcregrep was compiled under Windows, there was a reference to the
890        function pcregrep_exit() before it was defined. I am assuming this was
891        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
892        reported by a user. I've moved the definition above the reference.
893    
894    
895    Version 8.11 10-Dec-2010
896    ------------------------
897    
898    1.  (*THEN) was not working properly if there were untried alternatives prior
899        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
900        backtracked to try for "b" instead of moving to the next alternative branch
901        at the same level (in this case, to look for "c"). The Perl documentation
902        is clear that when (*THEN) is backtracked onto, it goes to the "next
903        alternative in the innermost enclosing group".
904    
905    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
906        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
907        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
908        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
909        (*THEN).
910    
911    3.  If \s appeared in a character class, it removed the VT character from
912        the class, even if it had been included by some previous item, for example
913        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
914        of \s, but is part of the POSIX "space" class.)
915    
916    4.  A partial match never returns an empty string (because you can always
917        match an empty string at the end of the subject); however the checking for
918        an empty string was starting at the "start of match" point. This has been
919        changed to the "earliest inspected character" point, because the returned
920        data for a partial match starts at this character. This means that, for
921        example, /(?<=abc)def/ gives a partial match for the subject "abc"
922        (previously it gave "no match").
923    
924    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
925        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
926        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
927        has an implication that the given string is incomplete (because a partial
928        match is preferred over a full match). For this reason, these items now
929        give a partial match in this situation. [Aside: previously, the one case
930        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
931        match rather than a full match, which was wrong by the old rules, but is
932        now correct.]
933    
934    6.  There was a bug in the handling of #-introduced comments, recognized when
935        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
936        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
937        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
938        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
939        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
940        places in pcre_compile().
941    
942    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
943        comments when looking ahead for named forward references to subpatterns,
944        the only newline sequence it recognized was NL. It now handles newlines
945        according to the set newline convention.
946    
947    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
948        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
949        cater for a lack of strerror(). These oversights have been fixed.
950    
951    9.  Added --match-limit and --recursion-limit to pcregrep.
952    
953    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
954    
955    11. When the -o option was used, pcregrep was setting a return code of 1, even
956        when matches were found, and --line-buffered was not being honoured.
957    
958    12. Added an optional parentheses number to the -o and --only-matching options
959        of pcregrep.
960    
961    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
962        can match an empty string. The code to do it in pcretest and pcredemo
963        needed fixing:
964    
965        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
966            only one byte after an empty string match just before CRLF (this case
967            just got forgotten; "any" and "anycrlf" were OK).
968    
969        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
970            mode when an empty string match preceded an ASCII character followed by
971            a non-ASCII character. (The code for advancing by one character rather
972            than one byte was nonsense.)
973    
974        (c) The pcredemo.c sample program did not have any code at all to handle
975            the cases when CRLF is a valid newline sequence.
976    
977    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
978        as a starting offset was within the subject string. There is now a new
979        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
980        negative or greater than the length of the string. In order to test this,
981        pcretest is extended to allow the setting of negative starting offsets.
982    
983    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
984        starting offset points to the beginning of a UTF-8 character was
985        unnecessarily clumsy. I tidied it up.
986    
987    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
988        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
989    
990    17. Nobody had reported that the --include_dir option, which was added in
991        release 7.7 should have been called --include-dir (hyphen, not underscore)
992        for compatibility with GNU grep. I have changed it to --include-dir, but
993        left --include_dir as an undocumented synonym, and the same for
994        --exclude-dir, though that is not available in GNU grep, at least as of
995        release 2.5.4.
996    
997    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
998        characters from a string of bytes) have been redefined so as not to use
999        loops, in order to improve performance in some environments. At the same
1000        time, I abstracted some of the common code into auxiliary macros to save
1001        repetition (this should not affect the compiled code).
1002    
1003    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1004        compile-time error is now given if \c is not followed by an ASCII
1005        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1006        different, and any byte value is allowed.)
1007    
1008    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1009        START_OPTIMIZE option, which is now allowed at compile time - but just
1010        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1011        to pcregrep and other applications that have no direct access to PCRE
1012        options. The new /Y option in pcretest sets this option when calling
1013        pcre_compile().
1014    
1015    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1016        back references. Groups containing recursive back references were forced to
1017        be atomic by that change, but in the case of named groups, the amount of
1018        memory required was incorrectly computed, leading to "Failed: internal
1019        error: code overflow". This has been fixed.
1020    
1021    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1022        pcretest.c, to avoid build problems in some Borland environments.
1023    
1024    
1025    Version 8.10 25-Jun-2010
1026    ------------------------
1027    
1028    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1029        THEN.
1030    
1031    2.  (*ACCEPT) was not working when inside an atomic group.
1032    
1033    3.  Inside a character class, \B is treated as a literal by default, but
1034        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1035        causes the error). The code is unchanged, but I tidied the documentation.
1036    
1037    4.  Inside a character class, PCRE always treated \R and \X as literals,
1038        whereas Perl faults them if its -w option is set. I have changed PCRE so
1039        that it faults them when PCRE_EXTRA is set.
1040    
1041    5.  Added support for \N, which always matches any character other than
1042        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1043    
1044    6.  When compiling pcregrep with newer versions of gcc which may have
1045        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1046        declared with attribute warn_unused_result" were given. Just casting the
1047        result to (void) does not stop the warnings; a more elaborate fudge is
1048        needed. I've used a macro to implement this.
1049    
1050    7.  Minor change to pcretest.c to avoid a compiler warning.
1051    
1052    8.  Added four artifical Unicode properties to help with an option to make
1053        \s etc use properties (see next item). The new properties are: Xan
1054        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1055    
1056    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1057        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1058        this option. Modified pcretest to add /W to test this facility. Added
1059        REG_UCP to make it available via the POSIX interface.
1060    
1061    10. Added --line-buffered to pcregrep.
1062    
1063    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1064        studied, and the match started with a letter with a code point greater than
1065        127 whose first byte was different to the first byte of the other case of
1066        the letter, the other case of this starting letter was not recognized
1067        (#976).
1068    
1069    12. If a pattern that was studied started with a repeated Unicode property
1070        test, for example, \p{Nd}+, there was the theoretical possibility of
1071        setting up an incorrect bitmap of starting bytes, but fortunately it could
1072        not have actually happened in practice until change 8 above was made (it
1073        added property types that matched character-matching opcodes).
1074    
1075    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1076        possible starting bytes for non-anchored patterns.
1077    
1078    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1079        \R, and also a number of cases that involve Unicode properties, both
1080        explicit and implicit when PCRE_UCP is set.
1081    
1082    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1083        input, it could crash or give wrong results if characters with values
1084        greater than 0xc0 were present in the subject string. (Detail: it assumed
1085        UTF-8 input when processing these items.)
1086    
1087    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1088        size_t is 64-bit (#991).
1089    
1090    17. Added a check for running out of memory when PCRE is compiled with
1091        --disable-stack-for-recursion (#990).
1092    
1093    18. If the last data line in a file for pcretest does not have a newline on
1094        the end, a newline was missing in the output.
1095    
1096    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1097        less than 128) in its various bitmaps. However, there is a facility for
1098        generating tables according to the current locale when PCRE is compiled. It
1099        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1100        characters, are recognized by isspace() and therefore were getting set in
1101        these tables, and indeed these tables seem to approximate to ISO 8859. This
1102        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1103        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1104        which of course cannot start UTF-8 characters. I have changed the code so
1105        that only real ASCII characters (less than 128) and the correct starting
1106        bytes for UTF-8 encodings are set for characters greater than 127 when in
1107        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1108        altogether.)
1109    
1110    20. Added the /T option to pcretest so as to be able to run tests with non-
1111        standard character tables, thus making it possible to include the tests
1112        used for 19 above in the standard set of tests.
1113    
1114    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1115        reference to a subpattern the other side of a comment that contains an
1116        opening parenthesis caused either an internal compiling error, or a
1117        reference to the wrong subpattern.
1118    
1119    
1120    Version 8.02 19-Mar-2010
1121    ------------------------
1122    
1123    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1124    
1125    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1126        configured.
1127    
1128    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1129        original author of that file, following a query about its status.
1130    
1131    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1132        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1133    
1134    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1135        quantifier applied to a forward-referencing subroutine call, could compile
1136        incorrect code or give the error "internal error: previously-checked
1137        referenced subpattern not found".
1138    
1139    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1140        variables to point to external functions. For these systems, therefore,
1141        pcre_malloc etc. are now initialized to local functions that call the
1142        relevant global functions.
1143    
1144    7.  There were two entries missing in the vectors called coptable and poptable
1145        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1146        I've fixed the data, and added a kludgy way of testing at compile time that
1147        the lengths are correct (equal to the number of opcodes).
1148    
1149    8.  Following on from 7, I added a similar kludge to check the length of the
1150        eint vector in pcreposix.c.
1151    
1152    9.  Error texts for pcre_compile() are held as one long string to avoid too
1153        much relocation at load time. To find a text, the string is searched,
1154        counting zeros. There was no check for running off the end of the string,
1155        which could happen if a new error number was added without updating the
1156        string.
1157    
1158    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1159    
1160    11. \K was not working if it appeared in an atomic group or in a group that
1161        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1162        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1163        the assertion is positive, but not if it is negative.
1164    
1165    12. Change 11 fortuitously reduced the size of the stack frame used in the
1166        "match()" function of pcre_exec.c by one pointer. Forthcoming
1167        implementation of support for (*MARK) will need an extra pointer on the
1168        stack; I have reserved it now, so that the stack frame size does not
1169        decrease.
1170    
1171    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1172        item in branch that calls a recursion is a subroutine call - as in the
1173        second branch in the above example - was incorrectly given the compile-
1174        time error "recursive call could loop indefinitely" because pcre_compile()
1175        was not correctly checking the subroutine for matching a non-empty string.
1176    
1177    14. The checks for overrunning compiling workspace could trigger after an
1178        overrun had occurred. This is a "should never occur" error, but it can be
1179        triggered by pathological patterns such as hundreds of nested parentheses.
1180        The checks now trigger 100 bytes before the end of the workspace.
1181    
1182    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1183    
1184    
1185    Version 8.01 19-Jan-2010
1186    ------------------------
1187    
1188    1.  If a pattern contained a conditional subpattern with only one branch (in
1189        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1190        computed the wrong minimum data length (which is of course zero for such
1191        subpatterns). This could cause incorrect "no match" results.
1192    
1193    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1194        the pattern is reset in the first branch, pcre_compile() failed with
1195        "internal error: code overflow at offset...". This happened only when
1196        the reset was to the original external option setting. (An optimization
1197        abstracts leading options settings into an external setting, which was the
1198        cause of this.)
1199    
1200    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1201        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1202        assertion pattern did not match (meaning that the assertion was true), it
1203        was incorrectly treated as false if the SKIP had been reached during the
1204        matching. This also applied to assertions used as conditions.
1205    
1206    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1207        assertion subpattern, including such a pattern used as a condition,
1208        unpredictable results occurred, instead of the error return
1209        PCRE_ERROR_DFA_UITEM.
1210    
1211    5.  The C++ GlobalReplace function was not working like Perl for the special
1212        situation when an empty string is matched. It now does the fancy magic
1213        stuff that is necessary.
1214    
1215    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1216        removed. (These were left over from very, very early versions of PCRE.)
1217    
1218    7.  Some cosmetic changes to the code to make life easier when compiling it
1219        as part of something else:
1220    
1221        (a) Change DEBUG to PCRE_DEBUG.
1222    
1223        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1224            called "current" as "current_branch", to prevent a collision with the
1225            Linux macro when compiled as a kernel module.
1226    
1227        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1228            prevent a collision with the Linux macro when compiled as a kernel
1229            module.
1230    
1231    8.  In pcre_compile() there are some checks for integer overflows that used to
1232        cast potentially large values to (double). This has been changed to that
1233        when building, a check for int64_t is made, and if it is found, it is used
1234        instead, thus avoiding the use of floating point arithmetic. (There is no
1235        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1236        double.
1237    
1238    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1239        2005 (difference between two addresses compared to an unsigned value).
1240    
1241    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1242        custom one, because of the following reported problem in Windows:
1243    
1244          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1245              under Win32.
1246          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1247              therefore missing the function definition.
1248          - The compiler thus generates a "C" signature for the test function.
1249          - The linker fails to find the "C" function.
1250          - PCRE fails to configure if asked to do so against libbz2.
1251    
1252    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1253        messages were output:
1254    
1255          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1256          rerunning libtoolize, to keep the correct libtool macros in-tree.
1257          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1258    
1259        I have done both of these things.
1260    
1261    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1262        most of the time, it *can* run out if it is given a pattern that contains a
1263        runaway infinite recursion. I updated the discussion in the pcrestack man
1264        page.
1265    
1266    13. Now that we have gone to the x.xx style of version numbers, the minor
1267        version may start with zero. Using 08 or 09 is a bad idea because users
1268        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1269        interpreted as invalid octal numbers. I've updated the previous comment in
1270        configure.ac, and also added a check that gives an error if 08 or 09 are
1271        used.
1272    
1273    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1274        causing partial matching to fail when the end of the subject matched \W
1275        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1276    
1277    15. There were some discrepancies between the declarations in pcre_internal.h
1278        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1279        their definitions. The declarations used "const uschar *" and the
1280        definitions used USPTR. Even though USPTR is normally defined as "const
1281        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1282        reported that: "This difference in casting confuses some C++ compilers, for
1283        example, SunCC recognizes above declarations as different functions and
1284        generates broken code for hbpcre." I have changed the declarations to use
1285        USPTR.
1286    
1287    16. GNU libtool is named differently on some systems. The autogen.sh script now
1288        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1289        (FreeBSD).
1290    
1291    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1292        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1293        comment: "Figure out how to create a longlong from a string: strtoll and
1294        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1295        instance, but it only takes 2 args instead of 3!"
1296    
1297    18. A subtle bug concerned with back references has been fixed by a change of
1298        specification, with a corresponding code fix. A pattern such as
1299        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1300        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1301        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1302        same bug. Such groups have to be quantified to be useful, or contained
1303        inside another quantified group. (If there's no repetition, the reference
1304        can never match.) The problem arises because, having left the group and
1305        moved on to the rest of the pattern, a later failure that backtracks into
1306        the group uses the captured value from the final iteration of the group
1307        rather than the correct earlier one. I have fixed this in PCRE by forcing
1308        any group that contains a reference to itself to be an atomic group; that
1309        is, there cannot be any backtracking into it once it has completed. This is
1310        similar to recursive and subroutine calls.
1311    
1312    
1313    Version 8.00 19-Oct-09
1314    ----------------------
1315    
1316    1.  The table for translating pcre_compile() error codes into POSIX error codes
1317        was out-of-date, and there was no check on the pcre_compile() error code
1318        being within the table. This could lead to an OK return being given in
1319        error.
1320    
1321    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1322        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1323        in a Windows environment.
1324    
1325    3.  The pcregrep --count option prints the count for each file even when it is
1326        zero, as does GNU grep. However, pcregrep was also printing all files when
1327        --files-with-matches was added. Now, when both options are given, it prints
1328        counts only for those files that have at least one match. (GNU grep just
1329        prints the file name in this circumstance, but including the count seems
1330        more useful - otherwise, why use --count?) Also ensured that the
1331        combination -clh just lists non-zero counts, with no names.
1332    
1333    4.  The long form of the pcregrep -F option was incorrectly implemented as
1334        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1335        but it seems right to fix it, and I didn't think it was worth preserving
1336        the old behaviour.
1337    
1338    5.  The command line items --regex=pattern and --regexp=pattern were not
1339        recognized by pcregrep, which required --regex pattern or --regexp pattern
1340        (with a space rather than an '='). The man page documented the '=' forms,
1341        which are compatible with GNU grep; these now work.
1342    
1343    6.  No libpcreposix.pc file was created for pkg-config; there was just
1344        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1345    
1346    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1347        when UCP support is not needed, by modifying the Python script that
1348        generates it from Unicode data files. This should not matter if the module
1349        is correctly used as a library, but I received one complaint about 50K of
1350        unwanted data. My guess is that the person linked everything into his
1351        program rather than using a library. Anyway, it does no harm.
1352    
1353    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1354        was a minimum greater than 1 for a wide character in a possessive
1355        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1356        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1357        character. Chaos in the form of incorrect output or a compiling loop could
1358        result.
1359    
1360    9.  The restrictions on what a pattern can contain when partial matching is
1361        requested for pcre_exec() have been removed. All patterns can now be
1362        partially matched by this function. In addition, if there are at least two
1363        slots in the offset vector, the offset of the earliest inspected character
1364        for the match and the offset of the end of the subject are set in them when
1365        PCRE_ERROR_PARTIAL is returned.
1366    
1367    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1368        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1369        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1370        and may be more useful for multi-segment matching.
1371    
1372    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1373        used to be given if ever the end of the subject was reached; now it is
1374        given only if matching could not proceed because another character was
1375        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1376        string "Z", which now yields "no match" instead of "partial match". In the
1377        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1378        final character ended with (*FAIL).
1379    
1380    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1381        if the pattern had a "must contain" character that was already found in the
1382        earlier partial match, unless partial matching was again requested. For
1383        example, with the pattern /dog.(body)?/, the "must contain" character is
1384        "g". If the first part-match was for the string "dog", restarting with
1385        "sbody" failed. This bug has been fixed.
1386    
1387    13. The string returned by pcre_dfa_exec() after a partial match has been
1388        changed so that it starts at the first inspected character rather than the
1389        first character of the match. This makes a difference only if the pattern
1390        starts with a lookbehind assertion or \b or \B (\K is not supported by
1391        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1392        matching functions compatible, and I think it's the right thing to do.
1393    
1394    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1395        so that the demonstration program is easily available in environments where
1396        PCRE has not been installed from source.
1397    
1398    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1399        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1400        library.
1401    
1402    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1403        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1404        is not the first non-POSIX option to be added. Clearly some people find
1405        these options useful.
1406    
1407    17. If a caller to the POSIX matching function regexec() passes a non-zero
1408        value for nmatch with a NULL value for pmatch, the value of
1409        nmatch is forced to zero.
1410    
1411    18. RunGrepTest did not have a test for the availability of the -u option of
1412        the diff command, as RunTest does. It now checks in the same way as
1413        RunTest, and also checks for the -b option.
1414    
1415    19. If an odd number of negated classes containing just a single character
1416        interposed, within parentheses, between a forward reference to a named
1417        subpattern and the definition of the subpattern, compilation crashed with
1418        an internal error, complaining that it could not find the referenced
1419        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1420        [The bug was that it was starting one character too far in when skipping
1421        over the character class, thus treating the ] as data rather than
1422        terminating the class. This meant it could skip too much.]
1423    
1424    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1425        /g option in pcretest when the pattern contains \K, which makes it possible
1426        to have an empty string match not at the start, even when the pattern is
1427        anchored. Updated pcretest and pcredemo to use this option.
1428    
1429    21. If the maximum number of capturing subpatterns in a recursion was greater
1430        than the maximum at the outer level, the higher number was returned, but
1431        with unset values at the outer level. The correct (outer level) value is
1432        now given.
1433    
1434    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1435        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1436        make it do so. The string so far is captured, making this feature
1437        compatible with Perl.
1438    
1439    23. The tests have been re-organized, adding tests 11 and 12, to make it
1440        possible to check the Perl 5.10 features against Perl 5.10.
1441    
1442    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1443        pattern matches a fixed length string. PCRE did not allow this; now it
1444        does. Neither allows recursion.
1445    
1446    25. I finally figured out how to implement a request to provide the minimum
1447        length of subject string that was needed in order to match a given pattern.
1448        (It was back references and recursion that I had previously got hung up
1449        on.) This code has now been added to pcre_study(); it finds a lower bound
1450        to the length of subject needed. It is not necessarily the greatest lower
1451        bound, but using it to avoid searching strings that are too short does give
1452        some useful speed-ups. The value is available to calling programs via
1453        pcre_fullinfo().
1454    
1455    26. While implementing 25, I discovered to my embarrassment that pcretest had
1456        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1457        study optimizations had never been tested with that matching function.
1458        Oops. What is worse, even when it was passed study data, there was a bug in
1459        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1460        were also very few tests of studied patterns with pcre_dfa_exec().
1461    
1462    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1463        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1464        on the other side of the coin, they are no longer allowed to have different
1465        names, because these cannot be distinguished in PCRE, and this has caused
1466        confusion. (This is a difference from Perl.)
1467    
1468    28. When duplicate subpattern names are present (necessarily with different
1469        numbers, as required by 27 above), and a test is made by name in a
1470        conditional pattern, either for a subpattern having been matched, or for
1471        recursion in such a pattern, all the associated numbered subpatterns are
1472        tested, and the overall condition is true if the condition is true for any
1473        one of them. This is the way Perl works, and is also more like the way
1474        testing by number works.
1475    
1476    
1477    Version 7.9 11-Apr-09
1478    ---------------------
1479    
1480    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1481        (pcretest), all targets were linked against these libraries. This included
1482        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1483        libraries. This caused unwanted dependencies to be created. This problem
1484        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1485        pcretest is linked with readline.
1486    
1487    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1488        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1489        moved outside it again, because FALSE and TRUE are already defined in AIX,
1490        but BOOL is not.
1491    
1492    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1493        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1494    
1495    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1496        hyphens) following file names and line numbers when outputting matching
1497        lines. This is not true; no spaces are inserted. I have also clarified the
1498        wording for the --colour (or --color) option.
1499    
1500    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1501        was not coloured; this is different to GNU grep, so I have changed it to be
1502        the same.
1503    
1504    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1505        each matching line was coloured. Now it goes on to look for further matches
1506        of any of the test patterns, which is the same behaviour as GNU grep.
1507    
1508    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1509        doesn't make sense to accept an empty string match in pcregrep, so I have
1510        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1511        seems to be how GNU grep behaves.
1512    
1513    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1514        start or after a newline", because the conditional assertion was not being
1515        correctly handled. The rule now is that both the assertion and what follows
1516        in the first alternative must satisfy the test.
1517    
1518    9.  If auto-callout was enabled in a pattern with a conditional group whose
1519        condition was an assertion, PCRE could crash during matching, both with
1520        pcre_exec() and pcre_dfa_exec().
1521    
1522    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1523        used for matching.
1524    
1525    11. Unicode property support in character classes was not working for
1526        characters (bytes) greater than 127 when not in UTF-8 mode.
1527    
1528    12. Added the -M command line option to pcretest.
1529    
1530    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1531    
1532    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1533    
1534    16. Added comments and documentation about mis-use of no_arg in the C++
1535        wrapper.
1536    
1537    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1538        from Martin Jerabek that uses macro names for all relevant character and
1539        string constants.
1540    
1541    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1542        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1543        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1544        these, but not everybody uses configure.
1545    
1546    19. A conditional group that had only one branch was not being correctly
1547        recognized as an item that could match an empty string. This meant that an
1548        enclosing group might also not be so recognized, causing infinite looping
1549        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1550        with the subject "ab", where knowledge that the repeated group can match
1551        nothing is needed in order to break the loop.
1552    
1553    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1554        exec(), but without supplying a callout function, matching went wrong.
1555    
1556    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1557        leak if the size of the offset vector was greater than 30. When the vector
1558        is smaller, the saved offsets during recursion go onto a local stack
1559        vector, but for larger vectors malloc() is used. It was failing to free
1560        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1561        error, in fact).
1562    
1563    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1564        heapframe that is used only when UTF-8 support is enabled. This caused no
1565        problem, but was untidy.
1566    
1567    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1568        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1569        included within another project.
1570    
1571    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1572        slightly modified by me:
1573    
1574          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1575              not building pcregrep.
1576    
1577          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1578              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1579    
1580    25. Forward references, both numeric and by name, in patterns that made use of
1581        duplicate group numbers, could behave incorrectly or give incorrect errors,
1582        because when scanning forward to find the reference group, PCRE was not
1583        taking into account the duplicate group numbers. A pattern such as
1584        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1585    
1586    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1587        the feature of a custom pointer more persuasive (as requested by a user).
1588    
1589    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1590        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1591        pre-defined.
1592    
1593    28. Added support for (*UTF8) at the start of a pattern.
1594    
1595    29. Arrange for flags added by the "release type" setting in CMake to be shown
1596        in the configuration summary.
1597    
1598    
1599    Version 7.8 05-Sep-08
1600    ---------------------
1601    
1602    1.  Replaced UCP searching code with optimized version as implemented for Ad
1603        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1604        stage table and inline lookup instead of a function, giving speed ups of 2
1605        to 5 times on some simple patterns that I tested. Permission was given to
1606        distribute the MultiStage2.py script that generates the tables (it's not in
1607        the tarball, but is in the Subversion repository).
1608    
1609    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1610        scripts.
1611    
1612    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1613        a group with a zero qualifier. The result of the study could be incorrect,
1614        or the function might crash, depending on the pattern.
1615    
1616    4.  Caseless matching was not working for non-ASCII characters in back
1617        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1618        It now works when Unicode Property Support is available.
1619    
1620    5.  In pcretest, an escape such as \x{de} in the data was always generating
1621        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1622        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1623        truncation.
1624    
1625    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1626    
1627    7.  Added two (int) casts to pcregrep when printing the difference of two
1628        pointers, in case they are 64-bit values.
1629    
1630    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1631        test 2 if it fails.
1632    
1633    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1634        and a #define of that name to empty if it is not externally set. This is to
1635        allow users of MSVC to set it if necessary.
1636    
1637    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1638        the convenience functions in the pcre_get.c source file.
1639    
1640    11. An option change at the start of a pattern that had top-level alternatives
1641        could cause overwriting and/or a crash. This command provoked a crash in
1642        some environments:
1643    
1644          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1645    
1646        This potential security problem was recorded as CVE-2008-2371.
1647    
1648    12. For a pattern where the match had to start at the beginning or immediately
1649        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1650        pcre_dfa_exec() could read past the end of the passed subject if there was
1651        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1652        pcretest so that it places the subject at the end of its malloc-ed buffer.
1653    
1654    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1655        exec() might read past the end of the data buffer in UTF-8 mode.
1656    
1657    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1658        the data contained the byte 0x85 as part of a UTF-8 character within its
1659        first line. This applied both to normal and DFA matching.
1660    
1661    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1662        /^[^d]*?$/8 failed to match "abc".
1663    
1664    16. Added a missing copyright notice to pcrecpp_internal.h.
1665    
1666    17. Make it more clear in the documentation that values returned from
1667        pcre_exec() in ovector are byte offsets, not character counts.
1668    
1669    18. Tidied a few places to stop certain compilers from issuing warnings.
1670    
1671    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1672        supplied by Stefan Weber. I made a further small update for 7.8 because
1673        there is a change of source arrangements: the pcre_searchfuncs.c module is
1674        replaced by pcre_ucd.c.
1675    
1676    
1677    Version 7.7 07-May-08
1678    ---------------------
1679    
1680    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1681        a string to a long long, pretend we don't even have a long long." This is
1682        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1683    
1684    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1685        pre-7.6 versions, which defined a global no_arg variable instead of putting
1686        it in the RE class. (See also #8 below.)
1687    
1688    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1689        Lopes.
1690    
1691    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1692    
1693        (1) The include/exclude patterns were being applied to the whole pathnames
1694            of files, instead of just to the final components.
1695    
1696        (2) If there was more than one level of directory, the subdirectories were
1697            skipped unless they satisfied the include/exclude conditions. This is
1698            inconsistent with GNU grep (and could even be seen as contrary to the
1699            pcregrep specification - which I improved to make it absolutely clear).
1700            The action now is always to scan all levels of directory, and just
1701            apply the include/exclude patterns to regular files.
1702    
1703    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1704        --exclude_dir in the tests to avoid scanning .svn directories.
1705    
1706    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1707        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1708        doesn't support NULs in patterns.
1709    
1710    7.  Added some missing "const"s to declarations of static tables in
1711        pcre_compile.c and pcre_dfa_exec.c.
1712    
1713    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1714        caused by fix #2  above. (Subsequently also a second patch to fix the
1715        first patch. And a third patch - this was a messy problem.)
1716    
1717    9.  Applied Craig's patch to remove the use of push_back().
1718    
1719    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1720        matching function regexec().
1721    
1722    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1723        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1724        references. PCRE supports relative numbers with this syntax (I don't think
1725        Oniguruma does).
1726    
1727    12. Previously, a group with a zero repeat such as (...){0} was completely
1728        omitted from the compiled regex. However, this means that if the group
1729        was called as a subroutine from elsewhere in the pattern, things went wrong
1730        (an internal error was given). Such groups are now left in the compiled
1731        pattern, with a new opcode that causes them to be skipped at execution
1732        time.
1733    
1734    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1735        to the way PCRE behaves:
1736    
1737        (a) A lone ] character is dis-allowed (Perl treats it as data).
1738    
1739        (b) A back reference to an unmatched subpattern matches an empty string
1740            (Perl fails the current match path).
1741    
1742        (c) A data ] in a character class must be notated as \] because if the
1743            first data character in a class is ], it defines an empty class. (In
1744            Perl it is not possible to have an empty class.) The empty class []
1745            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1746            The negative empty class [^] matches any one character, independently
1747            of the DOTALL setting.
1748    
1749    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1750        non-existent subpattern following a character class starting with ']' and
1751        containing () gave an internal compiling error instead of "reference to
1752        non-existent subpattern". Fortunately, when the pattern did exist, the
1753        compiled code was correct. (When scanning forwards to check for the
1754        existencd of the subpattern, it was treating the data ']' as terminating
1755        the class, so got the count wrong. When actually compiling, the reference
1756        was subsequently set up correctly.)
1757    
1758    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1759        it was being rejected as not supported by pcre_dfa_exec(), even though
1760        other assertions are supported. I have made pcre_dfa_exec() support
1761        (*FAIL).
1762    
1763    16. The implementation of 13c above involved the invention of a new opcode,
1764        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1765        cannot be changed at match time, I realized I could make a small
1766        improvement to matching performance by compiling OP_ALLANY instead of
1767        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1768        on the OP_ANY path.
1769    
1770    17. Compiling pcretest on Windows with readline support failed without the
1771        following two fixes: (1) Make the unistd.h include conditional on
1772        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1773    
1774    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1775        ncurses library to be included for pcretest when ReadLine support is
1776        requested, but also to allow for it to be overridden. This patch came from
1777        Daniel Bergström.
1778    
1779    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1780        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1781        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1782        spotting this.
1783    
1784    
1785    Version 7.6 28-Jan-08
1786    ---------------------
1787    
1788    1.  A character class containing a very large number of characters with
1789        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1790        overflow.
1791    
1792    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1793        HAVE_LONG_LONG is not defined.
1794    
1795    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1796        bring them up to date and include new features. This patch includes:
1797    
1798        - Fixed PH's badly added libz and libbz2 support.
1799        - Fixed a problem with static linking.
1800        - Added pcredemo. [But later removed - see 7 below.]
1801        - Fixed dftables problem and added an option.
1802        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1803            HAVE_LONG_LONG.
1804        - Added readline support for pcretest.
1805        - Added an listing of the option settings after cmake has run.
1806    
1807    4.  A user submitted a patch to Makefile that makes it easy to create
1808        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1809        Makefile.am that cause it to include this special target, without
1810        affecting anything else. Note that the same mingw target plus all
1811        the other distribution libraries and programs are now supported
1812        when configuring with CMake (see 6 below) instead of with
1813        Configure/Make.
1814    
1815    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1816        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1817        exported in the Windows port". It has not yet been confirmed that the patch
1818        solves the problem, but it does no harm.
1819    
1820    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1821        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1822        with CMake, and also correct the comment about stack recursion.
1823    
1824    7.  Remove the automatic building of pcredemo from the ./configure system and
1825        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1826        of a program that users should build themselves after PCRE is installed, so
1827        building it automatically is not really right. What is more, it gave
1828        trouble in some build environments.
1829    
1830    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1831    
1832    
1833    Version 7.5 10-Jan-08
1834    ---------------------
1835    
1836    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1837        values in parens when parsing an RE using the C++ wrapper."
1838    
1839    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1840        Characters greater than 255 were excluded from the class instead of being
1841        included.
1842    
1843    3.  The same bug as (2) above applied to negated POSIX classes such as
1844        [:^space:].
1845    
1846    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1847        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1848        I have changed it.
1849    
1850    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1851        first named subpattern) and a construct such as (?&a) would reference the
1852        first named subpattern whose name started with "a" (in other words, the
1853        length check was missing). Both these problems are fixed. "Subpattern name
1854        expected" is now given for (?&) (a zero-length name), and this patch also
1855        makes it give the same error for \k'' (previously it complained that that
1856        was a reference to a non-existent subpattern).
1857    
1858    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1859        this is right because (?- can be followed by option settings as well as by
1860        digits. I have, however, made the messages clearer.
1861    
1862    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1863        than the number used in the conditional) now cause a compile-time error.
1864        This is actually not compatible with Perl, which accepts such patterns, but
1865        treats the conditional as always being FALSE (as PCRE used to), but it
1866        seems to me that giving a diagnostic is better.
1867    
1868    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1869        and messages.
1870    
1871    9.  Fix two occurrences of "backslash" in comments that should have been
1872        "backspace".
1873    
1874    10. Remove two redundant lines of code that can never be obeyed (their function
1875        was moved elsewhere).
1876    
1877    11. The program that makes PCRE's Unicode character property table had a bug
1878        which caused it to generate incorrect table entries for sequences of
1879        characters that have the same character type, but are in different scripts.
1880        It amalgamated them into a single range, with the script of the first of
1881        them. In other words, some characters were in the wrong script. There were
1882        thirteen such cases, affecting characters in the following ranges:
1883    
1884          U+002b0 - U+002c1
1885          U+0060c - U+0060d
1886          U+0061e - U+00612
1887          U+0064b - U+0065e
1888          U+0074d - U+0076d
1889          U+01800 - U+01805
1890          U+01d00 - U+01d77
1891          U+01d9b - U+01dbf
1892          U+0200b - U+0200f
1893          U+030fc - U+030fe
1894          U+03260 - U+0327f
1895          U+0fb46 - U+0fbb1
1896          U+10450 - U+1049d
1897    
1898    12. The -o option (show only the matching part of a line) for pcregrep was not
1899        compatible with GNU grep in that, if there was more than one match in a
1900        line, it showed only the first of them. It now behaves in the same way as
1901        GNU grep.
1902    
1903    13. If the -o and -v options were combined for pcregrep, it printed a blank
1904        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1905        does the same. The return code can be used to tell if there were any
1906        non-matching lines.
1907    
1908    14. Added --file-offsets and --line-offsets to pcregrep.
1909    
1910    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1911        infinitely looping recursion. The bug was that positive lookaheads were not
1912        being skipped when checking for a possible empty match (negative lookaheads
1913        and both kinds of lookbehind were skipped).
1914    
1915    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1916        inclusion of <windows.h> to before rather than after the definition of
1917        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1918    
1919    17. Specifying a possessive quantifier with a specific limit for a Unicode
1920        character property caused pcre_compile() to compile bad code, which led at
1921        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1922        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1923        caused the error; without that there was no problem.
1924    
1925    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1926    
1927    19. Added --enable-pcretest-libreadline.
1928    
1929    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1930        RE::GlobalReplace(). As a result, the number of replacements returned was
1931        double what it should be. I removed one of the increments, but Craig sent a
1932        later patch that removed the other one (the right fix) and added unit tests
1933        that check the return values (which was not done before).
1934    
1935    21. Several CMake things:
1936    
1937        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1938            the names libpcre and libpcreposix, not just pcre and pcreposix.
1939    
1940        (2) The above change means that pcretest and pcregrep are now correctly
1941            linked with the newly-built libraries, not previously installed ones.
1942    
1943        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
1944    
1945    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1946        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1947        UTF-8 newline character). The key issue is that the pattern starts .*;
1948        this means that the match must be either at the beginning, or after a
1949        newline. The bug was in the code for advancing after a failed match and
1950        checking that the new position followed a newline. It was not taking
1951        account of UTF-8 characters correctly.
1952    
1953    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1954        character classes. PCRE was not treating the sequence [:...:] as a
1955        character class unless the ... were all letters. Perl, however, seems to
1956        allow any characters between [: and :], though of course it rejects as
1957        unknown any "names" that contain non-letters, because all the known class
1958        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1959        for example, whereas PCRE did not - it did not recognize a POSIX character
1960        class. This seemed a bit dangerous, so the code has been changed to be
1961        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1962        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1963        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1964        Perl does, and where it didn't before.
1965    
1966    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1967        Windows environments %n is disabled by default.
1968    
1969    
1970    Version 7.4 21-Sep-07
1971    ---------------------
1972    
1973    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1974        means that a class such as [\s] counted as "explicit reference to CR or
1975        LF". That isn't really right - the whole point of the change was to try to
1976        help when there was an actual mention of one of the two characters. So now
1977        the change happens only if \r or \n (or a literal CR or LF) character is
1978        encountered.
1979    
1980    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1981        of both had grown to the point where there were only 3 bits left.
1982        Fortunately, there was spare space in the data structure, and so I have
1983        moved the internal flags into a new 16-bit field to free up more option
1984        bits.
1985    
1986    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1987        but did not set the internal JCHANGED flag - either of these is enough to
1988        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1989        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1990        start sets both bits.
1991    
1992    4.  Added options (at build time, compile time, exec time) to change \R from
1993        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1994    
1995    5.  doc/pcresyntax.html was missing from the distribution.
1996    
1997    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1998        compatibility, even though it is no longer used.
1999    
2000    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
2001        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
2002        windows.h file is present (where different names are used). [This was
2003        reversed later after testing - see 16 below.]
2004    
2005    8.  Changed all #include <config.h> to #include "config.h". There were also
2006        some further <pcre.h> cases that I changed to "pcre.h".
2007    
2008    9.  When pcregrep was used with the --colour option, it missed the line ending
2009        sequence off the lines that it output.
2010    
2011    10. It was pointed out to me that arrays of string pointers cause lots of
2012        relocations when a shared library is dynamically loaded. A technique of
2013        using a single long string with a table of offsets can drastically reduce
2014        these. I have refactored PCRE in four places to do this. The result is
2015        dramatic:
2016    
2017          Originally:                          290
2018          After changing UCP table:            187
2019          After changing error message table:   43
2020          After changing table of "verbs"       36
2021          After changing table of Posix names   22
2022    
2023        Thanks to the folks working on Gregex for glib for this insight.
2024    
2025    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
2026        unicode-properties was also set.
2027    
2028    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
2029    
2030    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
2031        checked only for CRLF.
2032    
2033    14. Added casts to pcretest.c to avoid compiler warnings.
2034    
2035    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
2036    
2037    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
2038        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
2039        entirely. This removes changes made in 7 above.
2040    
2041    17. The CMake files have been updated, and there is now more information about
2042        building with CMake in the NON-UNIX-USE document.
2043    
2044    
2045    Version 7.3 28-Aug-07
2046    ---------------------
2047    
2048     1. In the rejigging of the build system that eventually resulted in 7.1, the
2049        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
2050        brackets there is not right, since it causes compilers to look for an
2051        installed pcre.h, not the version that is in the source that is being
2052        compiled (which of course may be different). I have changed it back to:
2053    
2054          #include "pcre.h"
2055    
2056        I have a vague recollection that the change was concerned with compiling in
2057        different directories, but in the new build system, that is taken care of
2058        by the VPATH setting the Makefile.
2059    
2060     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
2061        when the subject happened to end in the byte 0x85 (e.g. if the last
2062        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
2063        characters but of course it shouldn't be taken as a newline when it is part
2064        of another character. The bug was that, for an unlimited repeat of . in
2065        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
2066        characters when looking for a newline.
2067    
2068     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
2069    
2070     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
2071        in debug output.
2072    
2073     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
2074        long printing in the pcrecpp unittest when running under MinGW.
2075    
2076     6. ESC_K was left out of the EBCDIC table.
2077    
2078     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
2079        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
2080        limit also applies to "virtual nesting" when a pattern is recursive, and in
2081        this case 1000 isn't so big. I have been able to remove this limit at the
2082        expense of backing off one optimization in certain circumstances. Normally,
2083        when pcre_exec() would call its internal match() function recursively and
2084        immediately return the result unconditionally, it uses a "tail recursion"
2085        feature to save stack. However, when a subpattern that can match an empty
2086        string has an unlimited repetition quantifier, it no longer makes this
2087        optimization. That gives it a stack frame in which to save the data for
2088        checking that an empty string has been matched. Previously this was taken
2089        from the 1000-entry workspace that had been reserved. So now there is no
2090        explicit limit, but more stack is used.
2091    
2092     8. Applied Daniel's patches to solve problems with the import/export magic
2093        syntax that is required for Windows, and which was going wrong for the
2094        pcreposix and pcrecpp parts of the library. These were overlooked when this
2095        problem was solved for the main library.
2096    
2097     9. There were some crude static tests to avoid integer overflow when computing
2098        the size of patterns that contain repeated groups with explicit upper
2099        limits. As the maximum quantifier is 65535, the maximum group length was
2100        set at 30,000 so that the product of these two numbers did not overflow a
2101        32-bit integer. However, it turns out that people want to use groups that
2102        are longer than 30,000 bytes (though not repeat them that many times).
2103        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
2104        made it possible to implement the integer overflow checks in a much more
2105        dynamic way, which I have now done. The artificial limitation on group
2106        length has been removed - we now have only the limit on the total length of
2107        the compiled pattern, which depends on the LINK_SIZE setting.
2108    
2109    10. Fixed a bug in the documentation for get/copy named substring when
2110        duplicate names are permitted. If none of the named substrings are set, the
2111        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
2112        empty string.
2113    
2114    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
2115        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
2116        because the ] is interpreted as the first data character and the
2117        terminating ] is not found. PCRE has been made compatible with Perl in this
2118        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
2119        cause memory overwriting.
2120    
2121    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
2122        string has been matched (to stop an infinite loop). It was not recognizing
2123        a conditional subpattern that could match an empty string if that
2124        subpattern was within another subpattern. For example, it looped when
2125        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
2126        condition was not nested. This bug has been fixed.
2127    
2128    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
2129        past the start of the subject in the presence of bytes with the top bit
2130        set, for example "\x8aBCD".
2131    
2132    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
2133        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
2134    
2135    14. Optimized (?!) to (*FAIL).
2136    
2137    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
2138        This restricts code points to be within the range 0 to 0x10FFFF, excluding
2139        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
2140        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
2141        does: it's just the validity check that is more restrictive.
2142    
2143    16. Inserted checks for integer overflows during escape sequence (backslash)
2144        processing, and also fixed erroneous offset values for syntax errors during
2145        backslash processing.
2146    
2147    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
2148        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
2149    
2150    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
2151        caused an overrun.
2152    
2153    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
2154        something other than just ASCII characters) inside a group that had an
2155        unlimited repeat caused a loop at compile time (while checking to see
2156        whether the group could match an empty string).
2157    
2158    20. Debugging a pattern containing \p or \P could cause a crash. For example,
2159        [\P{Any}] did so. (Error in the code for printing property names.)
2160    
2161    21. An orphan \E inside a character class could cause a crash.
2162    
2163    22. A repeated capturing bracket such as (A)? could cause a wild memory
2164        reference during compilation.
2165    
2166    23. There are several functions in pcre_compile() that scan along a compiled
2167        expression for various reasons (e.g. to see if it's fixed length for look
2168        behind). There were bugs in these functions when a repeated \p or \P was
2169        present in the pattern. These operators have additional parameters compared
2170        with \d, etc, and these were not being taken into account when moving along
2171        the compiled data. Specifically:
2172    
2173        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
2174            length.
2175    
2176        (b) An item such as \pL+ within a repeated group could cause crashes or
2177            loops.
2178    
2179        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
2180            "reference to non-existent subpattern" error.
2181    
2182        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
2183    
2184    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
2185        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
2186    
2187    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
2188    
2189    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
2190        character were causing crashes (broken optimization).
2191    
2192    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
2193        \p or \P) caused a compile-time loop.
2194    
2195    28. More problems have arisen in unanchored patterns when CRLF is a valid line
2196        break. For example, the unstudied pattern [\r\n]A does not match the string
2197        "\r\nA" because change 7.0/46 below moves the current point on by two
2198        characters after failing to match at the start. However, the pattern \nA
2199        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
2200        the same is true. There doesn't seem any very clean way out of this, but
2201        what I have chosen to do makes the common cases work: PCRE now takes note
2202        of whether there can be an explicit match for \r or \n anywhere in the
2203        pattern, and if so, 7.0/46 no longer applies. As part of this change,
2204        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
2205        pattern has explicit CR or LF references.
2206    
2207    29. Added (*CR) etc for changing newline setting at start of pattern.
2208    
2209    
2210    Version 7.2 19-Jun-07
2211  ---------------------  ---------------------
2212    
2213   1. If the fr_FR locale cannot be found for test 3, try the "french" locale,   1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
2214      which is apparently normally available under Windows.      which is apparently normally available under Windows.
2215    
2216   2. Re-jig the pcregrep tests with different newline settings in an attempt   2. Re-jig the pcregrep tests with different newline settings in an attempt
2217      to make them independent of the local environment's newline setting.      to make them independent of the local environment's newline setting.
2218    
2219     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
2220    
2221     4. Some of the "internals" tests were previously cut out when the link size
2222        was not 2, because the output contained actual offsets. The recent new
2223        "Z" feature of pcretest means that these can be cut out, making the tests
2224        usable with all link sizes.
2225    
2226     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
2227        stack recursion. This gives a massive performance boost under BSD, but just
2228        a small improvement under Linux. However, it saves one field in the frame
2229        in all cases.
2230    
2231     6. Added more features from the forthcoming Perl 5.10:
2232    
2233        (a) (?-n) (where n is a string of digits) is a relative subroutine or
2234            recursion call. It refers to the nth most recently opened parentheses.
2235    
2236        (b) (?+n) is also a relative subroutine call; it refers to the nth next
2237            to be opened parentheses.
2238    
2239        (c) Conditions that refer to capturing parentheses can be specified
2240            relatively, for example, (?(-2)... or (?(+3)...
2241    
2242        (d) \K resets the start of the current match so that everything before
2243            is not part of it.
2244    
2245        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
2246    
2247        (f) \g{name} is another synonym - part of Perl 5.10's unification of
2248            reference syntax.
2249    
2250        (g) (?| introduces a group in which the numbering of parentheses in each
2251            alternative starts with the same number.
2252    
2253        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
2254    
2255     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
2256        PCRE_INFO_JCHANGED.
2257    
2258     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
2259        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
2260        for detecting groups that can match an empty string.
2261    
2262     9. A pattern with a very large number of alternatives (more than several
2263        hundred) was running out of internal workspace during the pre-compile
2264        phase, where pcre_compile() figures out how much memory will be needed. A
2265        bit of new cunning has reduced the workspace needed for groups with
2266        alternatives. The 1000-alternative test pattern now uses 12 bytes of
2267        workspace instead of running out of the 4096 that are available.
2268    
2269    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
2270    
2271    11. Applied patch from Google to remove an optimization that didn't quite work.
2272        The report of the bug said:
2273    
2274          pcrecpp::RE("a*").FullMatch("aaa") matches, while
2275          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
2276          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
2277    
2278    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
2279        it matched the wrong number of bytes.
2280    
2281    
2282  Version 7.1 24-Apr-07  Version 7.1 24-Apr-07
2283  ---------------------  ---------------------

Legend:
Removed from v.158  
changed lines
  Added in v.1226

  ViewVC Help
Powered by ViewVC 1.1.5