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

Diff of /code/trunk/ChangeLog

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

revision 141 by ph10, Fri Mar 30 15:46:27 2007 UTC revision 1311 by ph10, Mon Apr 22 17:35:23 2013 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4  Version 7.1 12-Mar-07  Version 8.33 xx-xxxx-201x
5    -------------------------
6    
7    1.  Added 'U' to some constants that are compared to unsigned integers, to
8        avoid compiler signed/unsigned warnings. Added (int) casts to unsigned
9        variables that are added to signed variables, to ensure the result is
10        signed and can be negated.
11    
12    2.  Applied patch by Daniel Richard G for quashing MSVC warnings to the
13        CMake config files.
14    
15    3.  Revise the creation of config.h.generic so that all boolean macros are
16        #undefined, whereas non-boolean macros are #ifndef/#endif-ed. This makes
17        overriding via -D on the command line possible.
18    
19    4.  Changing the definition of the variable "op" in pcre_exec.c from pcre_uchar
20        to unsigned int is reported to make a quite noticeable speed difference in
21        a specific Windows environment. Testing on Linux did also appear to show
22        some benefit (and it is clearly not harmful). Also fixed the definition of
23        Xop which should be unsigned.
24    
25    5.  Related to (4), changing the definition of the intermediate variable cc
26        in repeated character loops from pcre_uchar to pcre_uint32 also gave speed
27        improvements.
28    
29    6.  Fix forward search in JIT when link size is 3 or greater. Also removed some
30        unnecessary spaces.
31    
32    7.  Adjust autogen.sh and configure.ac to lose warnings given by automake 1.12
33        and later.
34    
35    8.  Fix two buffer over read issues in 16 and 32 bit modes. Affects JIT only.
36    
37    9.  Optimizing fast_forward_start_bits in JIT.
38    
39    10. Adding experimental support for callouts in JIT, and fixing some
40        issues revealed during this work. Namely:
41    
42        (a) Unoptimized capturing brackets incorrectly reset on backtrack.
43    
44        (b) Minimum length was not checked before the matching is started.
45    
46    11. The value of capture_last that is passed to callouts was incorrect in some
47        cases when there was a capture on one path that was subsequently abandoned
48        after a backtrack. Also, the capture_last value is now reset after a
49        recursion, since all captures are also reset in this case.
50    
51    12. The interpreter no longer returns the "too many substrings" error in the
52        case when an overflowing capture is in a branch that is subsequently
53        abandoned after a backtrack.
54    
55    13. In the pathological case when an offset vector of size 2 is used, pcretest
56        now prints out the matched string after a yield of 0 or 1.
57    
58    14. Inlining subpatterns in recursions, when certain conditions are fulfilled.
59        Only supported by the JIT compiler at the moment.
60    
61    15. JIT compiler now supports 32 bit Macs thanks to Lawrence Velazquez.
62    
63    16. Partial matches now set offsets[2] to the "bumpalong" value, that is, the
64        offset of the starting point of the matching process, provided the offsets
65        vector is large enough.
66    
67    17. The \A escape now records a lookbehind value of 1, though its execution
68        does not actually inspect the previous character. This is to ensure that,
69        in partial multi-segment matching, at least one character from the old
70        segment is retained when a new segment is processed. Otherwise, if there
71        are no lookbehinds in the pattern, \A might match incorrectly at the start
72        of a new segment.
73    
74    18. Added some #ifdef __VMS code into pcretest.c to help VMS implementations.
75    
76    19. Redefined some pcre_uchar variables in pcre_exec.c as pcre_uint32; this
77        gives some modest performance improvement in 8-bit mode.
78    
79    20. Added the PCRE-specific property \p{Xuc} for matching characters that can
80        be expressed in certain programming languages using Universal Character
81        Names.
82    
83    21. Unicode validation has been updated in the light of Unicode Corrigendum #9,
84        which points out that "non characters" are not "characters that may not
85        appear in Unicode strings" but rather "characters that are reserved for
86        internal use and have only local meaning".
87    
88    22. When a pattern was compiled with automatic callouts (PCRE_AUTO_CALLOUT) and
89        there was a conditional group that depended on an assertion, if the
90        assertion was false, the callout that immediately followed the alternation
91        in the condition was skipped when pcre_exec() was used for matching.
92    
93    23. Allow an explicit callout to be inserted before an assertion that is the
94        condition for a conditional group, for compatibility with automatic
95        callouts, which always insert a callout at this point.
96    
97    24. In 8.31, (*COMMIT) was confined to within a recursive subpattern. Perl also
98        confines (*SKIP) and (*PRUNE) in the same way, and this has now been done.
99    
100    25. (*PRUNE) is now supported by the JIT compiler.
101    
102    26. Fix infinite loop when /(?<=(*SKIP)ac)a/ is matched against aa.
103    
104    27. Fix the case where there are two or more SKIPs with arguments that may be
105        ignored.
106    
107    28. (*SKIP) is now supported by the JIT compiler.
108    
109    29. (*THEN) is now supported by the JIT compiler.
110    
111    30. Update RunTest with additional test selector options.
112    
113    31. The way PCRE handles backtracking verbs has been changed in two ways.
114    
115        (1) Previously, in something like (*COMMIT)(*SKIP), COMMIT would override
116        SKIP. Now, PCRE acts on whichever backtracking verb is reached first by
117        backtracking. In some cases this makes it more Perl-compatible, but Perl's
118        rather obscure rules do not always do the same thing.
119    
120        (2) Previously, backtracking verbs were confined within assertions. This is
121        no longer the case for positive assertions, except for (*ACCEPT). Again,
122        this sometimes improves Perl compatibility, and sometimes does not.
123    
124    32. A number of tests that were in test 2 because Perl did things differently
125        have been moved to test 1, because either Perl or PCRE has changed, and
126        these tests are now compatible.
127    
128    32. Control verbs are handled in the same way in JIT and interpreter.
129    
130    33. An opening parenthesis in a MARK/PRUNE/SKIP/THEN name in a pattern that
131        contained a forward subroutine reference caused a compile error.
132    
133    34. Auto-detect and optimize limited repetitions in JIT.
134    
135    35. Implement PCRE_NEVER_UTF to lock out the use of UTF, in particular,
136        blocking (*UTF) etc.
137    
138    36. In the interpreter, maximizing pattern repetitions for characters and
139        character types now use tail recursion, which reduces stack usage.
140    
141    
142    Version 8.32 30-November-2012
143    -----------------------------
144    
145    1.  Improved JIT compiler optimizations for first character search and single
146        character iterators.
147    
148    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
149        Patch by Daniel Richard G.
150    
151    3.  Single character iterator optimizations in the JIT compiler.
152    
153    4.  Improved JIT compiler optimizations for character ranges.
154    
155    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
156        Reported by Giuseppe D'Angelo.
157    
158    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
159        of a line, was being set incorrectly in cases where .* appeared inside
160        atomic brackets at the start of a pattern, or where there was a subsequent
161        *PRUNE or *SKIP.
162    
163    7.  Improved instruction cache flush for POWER/PowerPC.
164        Patch by Daniel Richard G.
165    
166    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
167        grep:
168    
169        (a) There is now no limit to the number of patterns to be matched.
170    
171        (b) An error is given if a pattern is too long.
172    
173        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
174            are now supported.
175    
176        (d) --exclude-from and --include-from (multiple use) have been added.
177    
178        (e) Exclusions and inclusions now apply to all files and directories, not
179            just to those obtained from scanning a directory recursively.
180    
181        (f) Multiple uses of -f and --file-list are now supported.
182    
183        (g) In a Windows environment, the default for -d has been changed from
184            "read" (the GNU grep default) to "skip", because otherwise the presence
185            of a directory in the file list provokes an error.
186    
187        (h) The documentation has been revised and clarified in places.
188    
189    9.  Improve the matching speed of capturing brackets.
190    
191    10. Changed the meaning of \X so that it now matches a Unicode extended
192        grapheme cluster.
193    
194    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
195        out POSIX threads when JIT support is configured.
196    
197    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
198    
199    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
200        structure could go wrong in environments where size_t is not the same size
201        as int.
202    
203    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
204        set.
205    
206    15. The EBCDIC support had decayed; later updates to the code had included
207        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
208        general tidy up of EBCDIC-related issues, and the documentation was also
209        not quite right. There is now a test that can be run on ASCII systems to
210        check some of the EBCDIC-related things (but is it not a full test).
211    
212    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
213        in a small tidy to the code.
214    
215    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
216    
217    18. If the --only-matching (-o) option in pcregrep is specified multiple
218        times, each one causes appropriate output. For example, -o1 -o2 outputs the
219        substrings matched by the 1st and 2nd capturing parentheses. A separating
220        string can be specified by --om-separator (default empty).
221    
222    19. Improving the first n character searches.
223    
224    20. Turn case lists for horizontal and vertical white space into macros so that
225        they are defined only once.
226    
227    21. This set of changes together give more compatible Unicode case-folding
228        behaviour for characters that have more than one other case when UCP
229        support is available.
230    
231        (a) The Unicode property table now has offsets into a new table of sets of
232            three or more characters that are case-equivalent. The MultiStage2.py
233            script that generates these tables (the pcre_ucd.c file) now scans
234            CaseFolding.txt instead of UnicodeData.txt for character case
235            information.
236    
237        (b) The code for adding characters or ranges of characters to a character
238            class has been abstracted into a generalized function that also handles
239            case-independence. In UTF-mode with UCP support, this uses the new data
240            to handle characters with more than one other case.
241    
242        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
243            whose other case is greater than 256 are now correctly matched
244            caselessly. Previously, the high codepoint matched the low one, but not
245            vice versa.
246    
247        (d) The processing of \h, \H, \v, and \ in character classes now makes use
248            of the new class addition function, using character lists defined as
249            macros alongside the case definitions of 20 above.
250    
251        (e) Caseless back references now work with characters that have more than
252            one other case.
253    
254        (f) General caseless matching of characters with more than one other case
255            is supported.
256    
257    22. Unicode character properties were updated from Unicode 6.2.0
258    
259    23. Improved CMake support under Windows. Patch by Daniel Richard G.
260    
261    24. Add support for 32-bit character strings, and UTF-32
262    
263    25. Major JIT compiler update (code refactoring and bugfixing).
264        Experimental Sparc 32 support is added.
265    
266    26. Applied a modified version of Daniel Richard G's patch to create
267        pcre.h.generic and config.h.generic by "make" instead of in the
268        PrepareRelease script.
269    
270    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
271        pcre_compile.c when checking for a zero character.
272    
273    28. Introducing a native interface for JIT. Through this interface, the compiled
274        machine code can be directly executed. The purpose of this interface is to
275        provide fast pattern matching, so several sanity checks are not performed.
276        However, feature tests are still performed. The new interface provides
277        1.4x speedup compared to the old one.
278    
279    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
280        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
281        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
282    
283    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
284        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
285        the "old" RFC 2279). Instead, it ended up passing a negative length to
286        pcre_exec().
287    
288    31. Add support for GCC's visibility feature to hide internal functions.
289    
290    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
291        "unknown -C option" after outputting 0 or 1.
292    
293    33. There is now support for generating a code coverage report for the test
294        suite in environments where gcc is the compiler and lcov is installed. This
295        is mainly for the benefit of the developers.
296    
297    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
298        unaddressable using valgrind annotations, allowing valgrind to detect
299        invalid memory accesses. This is mainly for the benefit of the developers.
300    
301    25. (*UTF) can now be used to start a pattern in any of the three libraries.
302    
303    26. Give configure error if --enable-cpp but no C++ compiler found.
304    
305    
306    Version 8.31 06-July-2012
307    -------------------------
308    
309    1.  Fixing a wrong JIT test case and some compiler warnings.
310    
311    2.  Removed a bashism from the RunTest script.
312    
313    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
314        to unsigned type, result still unsigned" that was given by an MS compiler
315        on encountering the code "-sizeof(xxx)".
316    
317    4.  Partial matching support is added to the JIT compiler.
318    
319    5.  Fixed several bugs concerned with partial matching of items that consist
320        of more than one character:
321    
322        (a) /^(..)\1/ did not partially match "aba" because checking references was
323            done on an "all or nothing" basis. This also applied to repeated
324            references.
325    
326        (b) \R did not give a hard partial match if \r was found at the end of the
327            subject.
328    
329        (c) \X did not give a hard partial match after matching one or more
330            characters at the end of the subject.
331    
332        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
333            a partial match for the string "\r".
334    
335        (e) When newline was set to CRLF, the metacharacter "." did not recognize
336            a partial match for a CR character at the end of the subject string.
337    
338    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
339        running pcretest, the text "(JIT)" added to the output whenever JIT is
340        actually used to run the match.
341    
342    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
343        or /S+[+] with a digit between 1 and 7.
344    
345    8.  OP_NOT now supports any UTF character not just single-byte ones.
346    
347    9.  (*MARK) control verb is now supported by the JIT compiler.
348    
349    10. The command "./RunTest list" lists the available tests without actually
350        running any of them. (Because I keep forgetting what they all are.)
351    
352    11. Add PCRE_INFO_MAXLOOKBEHIND.
353    
354    12. Applied a (slightly modified) user-supplied patch that improves performance
355        when the heap is used for recursion (compiled with --disable-stack-for-
356        recursion). Instead of malloc and free for each heap frame each time a
357        logical recursion happens, frames are retained on a chain and re-used where
358        possible. This sometimes gives as much as 30% improvement.
359    
360    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
361        call.
362    
363    14. As documented, (*COMMIT) is now confined to within a positive assertion.
364    
365    15. It is now possible to link pcretest with libedit as an alternative to
366        libreadline.
367    
368    16. (*COMMIT) control verb is now supported by the JIT compiler.
369    
370    17. The Unicode data tables have been updated to Unicode 6.1.0.
371    
372    18. Added --file-list option to pcregrep.
373    
374    19. Added binary file support to pcregrep, including the -a, --binary-files,
375        -I, and --text options.
376    
377    20. The madvise function is renamed for posix_madvise for QNX compatibility
378        reasons. Fixed by Giuseppe D'Angelo.
379    
380    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
381    
382    22. Changed the option for creating symbolic links for 16-bit man pages from
383        -s to -sf so that re-installing does not cause issues.
384    
385    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
386    
387    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
388        matches in certain environments (the workspace was not being correctly
389        retained). Also added to pcre_dfa_exec() a simple plausibility check on
390        some of the workspace data at the beginning of a restart.
391    
392    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
393        was not doing so when it should - probably a typo introduced by SVN 528
394        (change 8.10/14).
395    
396    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
397        \w+ when the character tables indicated that \x{c4} was a word character.
398        There were several related cases, all because the tests for doing a table
399        lookup were testing for characters less than 127 instead of 255.
400    
401    27. If a pattern contains capturing parentheses that are not used in a match,
402        their slots in the ovector are set to -1. For those that are higher than
403        any matched groups, this happens at the end of processing. In the case when
404        there were back references that the ovector was too small to contain
405        (causing temporary malloc'd memory to be used during matching), and the
406        highest capturing number was not used, memory off the end of the ovector
407        was incorrectly being set to -1. (It was using the size of the temporary
408        memory instead of the true size.)
409    
410    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
411        ovector size, it uses memory at the end of the block that it has got.
412    
413    29. Check for an overlong MARK name and give an error at compile time. The
414        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
415    
416    30. JIT compiler update.
417    
418    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
419        Rill for the patch.
420    
421    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
422    
423    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
424    
425    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
426        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
427        was enabled.
428    
429    35. Improve JIT code generation for greedy plus quantifier.
430    
431    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
432        1 to "aa" instead of to an empty string. The bug affected repeated groups
433        that could potentially match an empty string.
434    
435    37. Optimizing single character iterators in JIT.
436    
437    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
438        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
439        codepoints that are too big for the mode are faulted, and in a UTF mode,
440        disallowed codepoints are also faulted.
441    
442    39. If PCRE was compiled with UTF support, in three places in the DFA
443        matcher there was code that should only have been obeyed in UTF mode, but
444        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
445        processing when bytes with values greater than 127 were present. In 16-bit
446        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
447        both cases the values are those that cannot be the first data item in a UTF
448        character. The three items that might have provoked this were recursions,
449        possessively repeated groups, and atomic groups.
450    
451    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
452        and pcregrep, because some OS require shared objects to be explicitly
453        passed to ld, causing the link step to fail if they are not.
454    
455    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
456        mode, patterns that started with \h* or \R* might be incorrectly matched.
457    
458    
459    Version 8.30 04-February-2012
460    -----------------------------
461    
462    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
463        name is defined in ctype.h.
464    
465    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
466        only in quite long subpatterns.
467    
468    3.  Removed the function pcre_info(), which has been obsolete and deprecated
469        since it was replaced by pcre_fullinfo() in February 2000.
470    
471    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
472        match a (*MARK), and the match failed at the start of the subject, a
473        reference to memory before the start of the subject could occur. This bug
474        was introduced by fix 17 of release 8.21.
475    
476    5.  A reference to an unset group with zero minimum repetition was giving
477        totally wrong answers (in non-JavaScript-compatibility mode). For example,
478        /(another)?(\1?)test/ matched against "hello world test". This bug was
479        introduced in release 8.13.
480    
481    6.  Add support for 16-bit character strings (a large amount of work involving
482        many changes and refactorings).
483    
484    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
485        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
486        from a file.
487    
488    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
489        rounding is not applied in this particular case).
490    
491    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
492        if they appear, or are escaped, in patterns.
493    
494    10. Get rid of a number of -Wunused-but-set-variable warnings.
495    
496    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
497        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
498        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
499        also returns the mark "x". This bug applied to capturing parentheses,
500        non-capturing parentheses, and atomic parentheses. It also applied to some
501        assertions.
502    
503    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
504        information out of configure.ac instead of relying on pcre.h.generic, which
505        is not stored in the repository.
506    
507    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
508        -lreadline.
509    
510    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
511    
512    15. Applied Graycode's patch to put the top-level frame on the stack rather
513        than the heap when not using the stack for recursion. This gives a
514        performance improvement in many cases when recursion is not deep.
515    
516    16. Experimental code added to "pcretest -C" to output the stack frame size.
517    
518    
519    Version 8.21 12-Dec-2011
520    ------------------------
521    
522    1.  Updating the JIT compiler.
523    
524    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
525        are added as well.
526    
527    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
528        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
529        calling _pcre_jit_exec. Some extra comments are added.
530    
531    4.  (*MARK) settings inside atomic groups that do not contain any capturing
532        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
533        was introduced by change 18 for 8.20.
534    
535    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
536        ECMA-262 standard.
537    
538    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
539        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
540        This bug was probably introduced by change 9 of 8.13.
541    
542    7.  While fixing 6 above, I noticed that a number of other items were being
543        incorrectly rejected as "not fixed length". This arose partly because newer
544        opcodes had not been added to the fixed-length checking code. I have (a)
545        corrected the bug and added tests for these items, and (b) arranged for an
546        error to occur if an unknown opcode is encountered while checking for fixed
547        length instead of just assuming "not fixed length". The items that were
548        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
549        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
550        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
551    
552    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
553        being incorrectly compiled and would have given unpredicatble results.
554    
555    9.  A possessively repeated subpattern with minimum repeat count greater than
556        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
557        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
558        the first (A) could occur when it should not.
559    
560    10. Add a cast and remove a redundant test from the code.
561    
562    11. JIT should use pcre_malloc/pcre_free for allocation.
563    
564    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
565        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
566        is anything other than /usr, -L is still shown).
567    
568    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
569    
570    14. Perl does not support \N without a following name in a [] class; PCRE now
571        also gives an error.
572    
573    15. If a forward reference was repeated with an upper limit of around 2000,
574        it caused the error "internal error: overran compiling workspace". The
575        maximum number of forward references (including repeats) was limited by the
576        internal workspace, and dependent on the LINK_SIZE. The code has been
577        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
578        the default depends on LINK_SIZE. There is a new upper limit (for safety)
579        of around 200,000 forward references. While doing this, I also speeded up
580        the filling in of repeated forward references.
581    
582    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
583        incorrectly expecting the subject to contain another "a" after the start.
584    
585    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
586        in the match, the SKIP should be ignored. This was not happening; instead
587        the SKIP was being treated as NOMATCH. For patterns such as
588        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
589        tested.
590    
591    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
592        now much more compatible with Perl, in particular in cases where the result
593        is a non-match for a non-anchored pattern. For example, if
594        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
595        "m", where previously it did not return a name. A side effect of this
596        change is that for partial matches, the last encountered mark name is
597        returned, as for non matches. A number of tests that were previously not
598        Perl-compatible have been moved into the Perl-compatible test files. The
599        refactoring has had the pleasing side effect of removing one argument from
600        the match() function, thus reducing its stack requirements.
601    
602    19. If the /S+ option was used in pcretest to study a pattern using JIT,
603        subsequent uses of /S (without +) incorrectly behaved like /S+.
604    
605    21. Retrieve executable code size support for the JIT compiler and fixing
606        some warnings.
607    
608    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
609        not work when the shorter character appeared right at the end of the
610        subject string.
611    
612    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
613        systems.
614    
615    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
616        output it when the /M option is used in pcretest.
617    
618    25. The CheckMan script was not being included in the distribution. Also, added
619        an explicit "perl" to run Perl scripts from the PrepareRelease script
620        because this is reportedly needed in Windows.
621    
622    26. If study data was being save in a file and studying had not found a set of
623        "starts with" bytes for the pattern, the data written to the file (though
624        never used) was taken from uninitialized memory and so caused valgrind to
625        complain.
626    
627    27. Updated RunTest.bat as provided by Sheri Pierce.
628    
629    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
630    
631    29. Computation of memory usage for the table of capturing group names was
632        giving an unnecessarily large value.
633    
634    
635    Version 8.20 21-Oct-2011
636    ------------------------
637    
638    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
639        a POSIX class. After further experiments with Perl, which convinced me that
640        Perl has bugs and confusions, a closing square bracket is no longer allowed
641        in a POSIX name. This bug also affected patterns with classes that started
642        with full stops.
643    
644    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
645        captured substring, but while checking the failing first alternative,
646        substring 1 is temporarily captured. If the output vector supplied to
647        pcre_exec() was not big enough for this capture, the yield of the function
648        was still zero ("insufficient space for captured substrings"). This cannot
649        be totally fixed without adding another stack variable, which seems a lot
650        of expense for a edge case. However, I have improved the situation in cases
651        such as /(a)(b)x|abc/ matched against "abc", where the return code
652        indicates that fewer than the maximum number of slots in the ovector have
653        been set.
654    
655    3.  Related to (2) above: when there are more back references in a pattern than
656        slots in the output vector, pcre_exec() uses temporary memory during
657        matching, and copies in the captures as far as possible afterwards. It was
658        using the entire output vector, but this conflicts with the specification
659        that only 2/3 is used for passing back captured substrings. Now it uses
660        only the first 2/3, for compatibility. This is, of course, another edge
661        case.
662    
663    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
664        main code base, and can be used by building with --enable-jit. When this is
665        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
666        runtime --no-jit option is given.
667    
668    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
669        ovector, the return from the function was zero, implying that there were
670        other matches that did not fit. The correct "exactly full" value is now
671        returned.
672    
673    6.  If a subpattern that was called recursively or as a subroutine contained
674        (*PRUNE) or any other control that caused it to give a non-standard return,
675        invalid errors such as "Error -26 (nested recursion at the same subject
676        position)" or even infinite loops could occur.
677    
678    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
679        computing the minimum length on reaching *ACCEPT, and so ended up with the
680        wrong value of 1 rather than 0. Further investigation indicates that
681        computing a minimum subject length in the presence of *ACCEPT is difficult
682        (think back references, subroutine calls), and so I have changed the code
683        so that no minimum is registered for a pattern that contains *ACCEPT.
684    
685    8.  If (*THEN) was present in the first (true) branch of a conditional group,
686        it was not handled as intended. [But see 16 below.]
687    
688    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
689        Sheri Pierce.
690    
691    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
692        the first byte in a match must be "a".
693    
694    11. Change 17 for 8.13 increased the recursion depth for patterns like
695        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
696        pattern contains any instances of (*THEN). If it does not, the old
697        optimizations are restored. It would be nice to do this on a per-group
698        basis, but at the moment that is not feasible.
699    
700    12. In some environments, the output of pcretest -C is CRLF terminated. This
701        broke RunTest's code that checks for the link size. A single white space
702        character after the value is now allowed for.
703    
704    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
705        For "fr", it uses the Windows-specific input and output files.
706    
707    14. If (*THEN) appeared in a group that was called recursively or as a
708        subroutine, it did not work as intended. [But see next item.]
709    
710    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
711        pattern fragments (but not containing any | characters). If A and B are
712        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
713        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
714        D. In other words, Perl considers parentheses that do not contain any |
715        characters to be part of a surrounding alternative, whereas PCRE was
716        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
717        differently. PCRE now behaves in the same way as Perl, except in the case
718        of subroutine/recursion calls such as (?1) which have in any case always
719        been different (but PCRE had them first :-).
720    
721    16. Related to 15 above: Perl does not treat the | in a conditional group as
722        creating alternatives. Such a group is treated in the same way as an
723        ordinary group without any | characters when processing (*THEN). PCRE has
724        been changed to match Perl's behaviour.
725    
726    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
727        RunGrepTest script failed.
728    
729    18. Change 22 for version 13 caused atomic groups to use more stack. This is
730        inevitable for groups that contain captures, but it can lead to a lot of
731        stack use in large patterns. The old behaviour has been restored for atomic
732        groups that do not contain any capturing parentheses.
733    
734    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
735        suppress the check for a minimum subject length at run time. (If it was
736        given to pcre_exec() or pcre_dfa_exec() it did work.)
737    
738    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
739        fail to work when decoding hex characters in data strings in EBCDIC
740        environments.
741    
742    21. It appears that in at least one Mac OS environment, the isxdigit() function
743        is implemented as a macro that evaluates to its argument more than once,
744        contravening the C 90 Standard (I haven't checked a later standard). There
745        was an instance in pcretest which caused it to go wrong when processing
746        \x{...} escapes in subject strings. The has been rewritten to avoid using
747        things like p++ in the argument of isxdigit().
748    
749    
750    Version 8.13 16-Aug-2011
751    ------------------------
752    
753    1.  The Unicode data tables have been updated to Unicode 6.0.0.
754    
755    2.  Two minor typos in pcre_internal.h have been fixed.
756    
757    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
758        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
759        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
760    
761    4.  There were a number of related bugs in the code for matching backrefences
762        caselessly in UTF-8 mode when codes for the characters concerned were
763        different numbers of bytes. For example, U+023A and U+2C65 are an upper
764        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
765        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
766        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
767        2-byte code at the end of the subject (it thought there wasn't enough data
768        left).
769    
770    5.  Comprehensive information about what went wrong is now returned by
771        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
772        as the output vector has at least 2 elements. The offset of the start of
773        the failing character and a reason code are placed in the vector.
774    
775    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
776        now returned is for the first byte of the failing character, instead of the
777        last byte inspected. This is an incompatible change, but I hope it is small
778        enough not to be a problem. It makes the returned offset consistent with
779        pcre_exec() and pcre_dfa_exec().
780    
781    7.  pcretest now gives a text phrase as well as the error number when
782        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
783        failure, the offset and reason code are output.
784    
785    8.  When \R was used with a maximizing quantifier it failed to skip backwards
786        over a \r\n pair if the subsequent match failed. Instead, it just skipped
787        back over a single character (\n). This seems wrong (because it treated the
788        two characters as a single entity when going forwards), conflicts with the
789        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
790        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
791        has been changed.
792    
793    9.  Some internal refactoring has changed the processing so that the handling
794        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
795        time (the PCRE_DOTALL option was changed this way some time ago: version
796        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
797        which was always a bit of a fudge. It also means that there is one less
798        argument for the match() function, which reduces its stack requirements
799        slightly. This change also fixes an incompatibility with Perl: the pattern
800        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
801    
802    10. More internal refactoring has drastically reduced the number of recursive
803        calls to match() for possessively repeated groups such as (abc)++ when
804        using pcre_exec().
805    
806    11. While implementing 10, a number of bugs in the handling of groups were
807        discovered and fixed:
808    
809        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
810        (a|)*(?1) gave a compile-time internal error.
811        ((a|)+)+  did not notice that the outer group could match an empty string.
812        (^a|^)+   was not marked as anchored.
813        (.*a|.*)+ was not marked as matching at start or after a newline.
814    
815    12. Yet more internal refactoring has removed another argument from the match()
816        function. Special calls to this function are now indicated by setting a
817        value in a variable in the "match data" data block.
818    
819    13. Be more explicit in pcre_study() instead of relying on "default" for
820        opcodes that mean there is no starting character; this means that when new
821        ones are added and accidentally left out of pcre_study(), testing should
822        pick them up.
823    
824    14. The -s option of pcretest has been documented for ages as being an old
825        synonym of -m (show memory usage). I have changed it to mean "force study
826        for every regex", that is, assume /S for every regex. This is similar to -i
827        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
828        using it. It makes it easier to run collections of tests with and without
829        study enabled, and thereby test pcre_study() more easily. All the standard
830        tests are now run with and without -s (but some patterns can be marked as
831        "never study" - see 20 below).
832    
833    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
834        restoration of the capturing data to the outer values was not happening
835        correctly.
836    
837    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
838        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
839        pattern had matched an empty string, and so incorrectly returned a no
840        match.
841    
842    17. There was optimizing code for the last branch of non-capturing parentheses,
843        and also for the obeyed branch of a conditional subexpression, which used
844        tail recursion to cut down on stack usage. Unfortunately, now that there is
845        the possibility of (*THEN) occurring in these branches, tail recursion is
846        no longer possible because the return has to be checked for (*THEN). These
847        two optimizations have therefore been removed. [But see 8.20/11 above.]
848    
849    18. If a pattern containing \R was studied, it was assumed that \R always
850        matched two bytes, thus causing the minimum subject length to be
851        incorrectly computed because \R can also match just one byte.
852    
853    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
854        was incorrectly computed.
855    
856    20. If /S is present twice on a test pattern in pcretest input, it now
857        *disables* studying, thereby overriding the use of -s on the command line
858        (see 14 above). This is necessary for one or two tests to keep the output
859        identical in both cases.
860    
861    21. When (*ACCEPT) was used in an assertion that matched an empty string and
862        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
863    
864    22. When an atomic group that contained a capturing parenthesis was
865        successfully matched, but the branch in which it appeared failed, the
866        capturing was not being forgotten if a higher numbered group was later
867        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
868        group 1 to "a", when in fact it should be unset. This applied to multi-
869        branched capturing and non-capturing groups, repeated or not, and also to
870        positive assertions (capturing in negative assertions does not happen
871        in PCRE) and also to nested atomic groups.
872    
873    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
874        subject after a captured substring, to make it easier to tell which of a
875        number of identical substrings has been captured.
876    
877    24. The way atomic groups are processed by pcre_exec() has been changed so that
878        if they are repeated, backtracking one repetition now resets captured
879        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
880        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
881        "aaa". Previously, it would have been "a". As part of this code
882        refactoring, the way recursive calls are handled has also been changed.
883    
884    25. If an assertion condition captured any substrings, they were not passed
885        back unless some other capturing happened later. For example, if
886        (?(?=(a))a) was matched against "a", no capturing was returned.
887    
888    26. When studying a pattern that contained subroutine calls or assertions,
889        the code for finding the minimum length of a possible match was handling
890        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
891        group 1 called group 2 while simultaneously a separate group 2 called group
892        1). A stack overflow occurred in this case. I have fixed this by limiting
893        the recursion depth to 10.
894    
895    27. Updated RunTest.bat in the distribution to the version supplied by Tom
896        Fortmann. This supports explicit test numbers on the command line, and has
897        argument validation and error reporting.
898    
899    28. An instance of \X with an unlimited repeat could fail if at any point the
900        first character it looked at was a mark character.
901    
902    29. Some minor code refactoring concerning Unicode properties and scripts
903        should reduce the stack requirement of match() slightly.
904    
905    30. Added the '=' option to pcretest to check the setting of unused capturing
906        slots at the end of the pattern, which are documented as being -1, but are
907        not included in the return count.
908    
909    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
910        compiled something random. Now it gives a compile-time error (as does
911        Perl).
912    
913    32. A *MARK encountered during the processing of a positive assertion is now
914        recorded and passed back (compatible with Perl).
915    
916    33. If --only-matching or --colour was set on a pcregrep call whose pattern
917        had alternative anchored branches, the search for a second match in a line
918        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
919        matched the line "0102" twice. The same bug affected patterns that started
920        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
921        twice.
922    
923    34. Previously, PCRE did not allow quantification of assertions. However, Perl
924        does, and because of capturing effects, quantifying parenthesized
925        assertions may at times be useful. Quantifiers are now allowed for
926        parenthesized assertions.
927    
928    35. A minor code tidy in pcre_compile() when checking options for \R usage.
929    
930    36. \g was being checked for fancy things in a character class, when it should
931        just be a literal "g".
932    
933    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
934        appearance of a nested POSIX class supersedes an apparent external class.
935        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
936        unescaped square brackets may also appear as part of class names. For
937        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
938        more like Perl. (But see 8.20/1 above.)
939    
940    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
941        was because it thought it was \N{name}, which is not supported).
942    
943    39. Add minix to OS list not supporting the -S option in pcretest.
944    
945    40. PCRE tries to detect cases of infinite recursion at compile time, but it
946        cannot analyze patterns in sufficient detail to catch mutual recursions
947        such as ((?1))((?2)). There is now a runtime test that gives an error if a
948        subgroup is called recursively as a subpattern for a second time at the
949        same position in the subject string. In previous releases this might have
950        been caught by the recursion limit, or it might have run out of stack.
951    
952    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
953        happen only once. PCRE was, however incorrectly giving a compile time error
954        "recursive call could loop indefinitely" because it cannot analyze the
955        pattern in sufficient detail. The compile time test no longer happens when
956        PCRE is compiling a conditional subpattern, but actual runaway loops are
957        now caught at runtime (see 40 above).
958    
959    42. It seems that Perl allows any characters other than a closing parenthesis
960        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
961        has been changed to be the same.
962    
963    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
964        as not to get warnings when autogen.sh is called. Also changed
965        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
966    
967    44. To help people who use pcregrep to scan files containing exceedingly long
968        lines, the following changes have been made:
969    
970        (a) The default value of the buffer size parameter has been increased from
971            8K to 20K. (The actual buffer used is three times this size.)
972    
973        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
974            PCRE is built.
975    
976        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
977            to be set at run time.
978    
979        (d) Numerical values in pcregrep options can be followed by K or M, for
980            example --buffer-size=50K.
981    
982        (e) If a line being scanned overflows pcregrep's buffer, an error is now
983            given and the return code is set to 2.
984    
985    45. Add a pointer to the latest mark to the callout data block.
986    
987    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
988        partial match of an empty string instead of no match. This was specific to
989        the use of ".".
990    
991    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
992        complete match instead of a partial match. This bug was dependent on both
993        the PCRE_UTF8 and PCRE_DOTALL options being set.
994    
995    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
996        starting byte set, because \b was not being ignored.
997    
998    
999    Version 8.12 15-Jan-2011
1000    ------------------------
1001    
1002    1.  Fixed some typos in the markup of the man pages, and wrote a script that
1003        checks for such things as part of the documentation building process.
1004    
1005    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
1006        --match-limit and --recursion-limit options (added for 8.11). In
1007        particular, this made one of the standard tests fail. (The integer value
1008        went into the wrong half of a long int.)
1009    
1010    3.  If the --colour option was given to pcregrep with -v (invert match), it
1011        did strange things, either producing crazy output, or crashing. It should,
1012        of course, ignore a request for colour when reporting lines that do not
1013        match.
1014    
1015    4.  Another pcregrep bug caused similar problems if --colour was specified with
1016        -M (multiline) and the pattern match finished with a line ending.
1017    
1018    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
1019        matched in multiline mode, the following line was shown as part of the
1020        match. This seems wrong, so I have changed it.
1021    
1022    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
1023        the check for further matches in the same line (so they could be coloured)
1024        to overrun the end of the current line. If another match was found, it was
1025        incorrectly shown (and then shown again when found in the next line).
1026    
1027    7.  If pcregrep was compiled under Windows, there was a reference to the
1028        function pcregrep_exit() before it was defined. I am assuming this was
1029        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
1030        reported by a user. I've moved the definition above the reference.
1031    
1032    
1033    Version 8.11 10-Dec-2010
1034    ------------------------
1035    
1036    1.  (*THEN) was not working properly if there were untried alternatives prior
1037        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
1038        backtracked to try for "b" instead of moving to the next alternative branch
1039        at the same level (in this case, to look for "c"). The Perl documentation
1040        is clear that when (*THEN) is backtracked onto, it goes to the "next
1041        alternative in the innermost enclosing group".
1042    
1043    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
1044        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
1045        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
1046        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
1047        (*THEN).
1048    
1049    3.  If \s appeared in a character class, it removed the VT character from
1050        the class, even if it had been included by some previous item, for example
1051        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
1052        of \s, but is part of the POSIX "space" class.)
1053    
1054    4.  A partial match never returns an empty string (because you can always
1055        match an empty string at the end of the subject); however the checking for
1056        an empty string was starting at the "start of match" point. This has been
1057        changed to the "earliest inspected character" point, because the returned
1058        data for a partial match starts at this character. This means that, for
1059        example, /(?<=abc)def/ gives a partial match for the subject "abc"
1060        (previously it gave "no match").
1061    
1062    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
1063        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
1064        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
1065        has an implication that the given string is incomplete (because a partial
1066        match is preferred over a full match). For this reason, these items now
1067        give a partial match in this situation. [Aside: previously, the one case
1068        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
1069        match rather than a full match, which was wrong by the old rules, but is
1070        now correct.]
1071    
1072    6.  There was a bug in the handling of #-introduced comments, recognized when
1073        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
1074        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
1075        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
1076        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
1077        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
1078        places in pcre_compile().
1079    
1080    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
1081        comments when looking ahead for named forward references to subpatterns,
1082        the only newline sequence it recognized was NL. It now handles newlines
1083        according to the set newline convention.
1084    
1085    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
1086        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
1087        cater for a lack of strerror(). These oversights have been fixed.
1088    
1089    9.  Added --match-limit and --recursion-limit to pcregrep.
1090    
1091    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
1092    
1093    11. When the -o option was used, pcregrep was setting a return code of 1, even
1094        when matches were found, and --line-buffered was not being honoured.
1095    
1096    12. Added an optional parentheses number to the -o and --only-matching options
1097        of pcregrep.
1098    
1099    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
1100        can match an empty string. The code to do it in pcretest and pcredemo
1101        needed fixing:
1102    
1103        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
1104            only one byte after an empty string match just before CRLF (this case
1105            just got forgotten; "any" and "anycrlf" were OK).
1106    
1107        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
1108            mode when an empty string match preceded an ASCII character followed by
1109            a non-ASCII character. (The code for advancing by one character rather
1110            than one byte was nonsense.)
1111    
1112        (c) The pcredemo.c sample program did not have any code at all to handle
1113            the cases when CRLF is a valid newline sequence.
1114    
1115    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
1116        as a starting offset was within the subject string. There is now a new
1117        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
1118        negative or greater than the length of the string. In order to test this,
1119        pcretest is extended to allow the setting of negative starting offsets.
1120    
1121    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
1122        starting offset points to the beginning of a UTF-8 character was
1123        unnecessarily clumsy. I tidied it up.
1124    
1125    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
1126        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
1127    
1128    17. Nobody had reported that the --include_dir option, which was added in
1129        release 7.7 should have been called --include-dir (hyphen, not underscore)
1130        for compatibility with GNU grep. I have changed it to --include-dir, but
1131        left --include_dir as an undocumented synonym, and the same for
1132        --exclude-dir, though that is not available in GNU grep, at least as of
1133        release 2.5.4.
1134    
1135    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
1136        characters from a string of bytes) have been redefined so as not to use
1137        loops, in order to improve performance in some environments. At the same
1138        time, I abstracted some of the common code into auxiliary macros to save
1139        repetition (this should not affect the compiled code).
1140    
1141    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1142        compile-time error is now given if \c is not followed by an ASCII
1143        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1144        different, and any byte value is allowed.)
1145    
1146    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1147        START_OPTIMIZE option, which is now allowed at compile time - but just
1148        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1149        to pcregrep and other applications that have no direct access to PCRE
1150        options. The new /Y option in pcretest sets this option when calling
1151        pcre_compile().
1152    
1153    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1154        back references. Groups containing recursive back references were forced to
1155        be atomic by that change, but in the case of named groups, the amount of
1156        memory required was incorrectly computed, leading to "Failed: internal
1157        error: code overflow". This has been fixed.
1158    
1159    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1160        pcretest.c, to avoid build problems in some Borland environments.
1161    
1162    
1163    Version 8.10 25-Jun-2010
1164    ------------------------
1165    
1166    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1167        THEN.
1168    
1169    2.  (*ACCEPT) was not working when inside an atomic group.
1170    
1171    3.  Inside a character class, \B is treated as a literal by default, but
1172        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1173        causes the error). The code is unchanged, but I tidied the documentation.
1174    
1175    4.  Inside a character class, PCRE always treated \R and \X as literals,
1176        whereas Perl faults them if its -w option is set. I have changed PCRE so
1177        that it faults them when PCRE_EXTRA is set.
1178    
1179    5.  Added support for \N, which always matches any character other than
1180        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1181    
1182    6.  When compiling pcregrep with newer versions of gcc which may have
1183        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1184        declared with attribute warn_unused_result" were given. Just casting the
1185        result to (void) does not stop the warnings; a more elaborate fudge is
1186        needed. I've used a macro to implement this.
1187    
1188    7.  Minor change to pcretest.c to avoid a compiler warning.
1189    
1190    8.  Added four artifical Unicode properties to help with an option to make
1191        \s etc use properties (see next item). The new properties are: Xan
1192        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1193    
1194    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1195        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1196        this option. Modified pcretest to add /W to test this facility. Added
1197        REG_UCP to make it available via the POSIX interface.
1198    
1199    10. Added --line-buffered to pcregrep.
1200    
1201    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1202        studied, and the match started with a letter with a code point greater than
1203        127 whose first byte was different to the first byte of the other case of
1204        the letter, the other case of this starting letter was not recognized
1205        (#976).
1206    
1207    12. If a pattern that was studied started with a repeated Unicode property
1208        test, for example, \p{Nd}+, there was the theoretical possibility of
1209        setting up an incorrect bitmap of starting bytes, but fortunately it could
1210        not have actually happened in practice until change 8 above was made (it
1211        added property types that matched character-matching opcodes).
1212    
1213    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1214        possible starting bytes for non-anchored patterns.
1215    
1216    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1217        \R, and also a number of cases that involve Unicode properties, both
1218        explicit and implicit when PCRE_UCP is set.
1219    
1220    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1221        input, it could crash or give wrong results if characters with values
1222        greater than 0xc0 were present in the subject string. (Detail: it assumed
1223        UTF-8 input when processing these items.)
1224    
1225    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1226        size_t is 64-bit (#991).
1227    
1228    17. Added a check for running out of memory when PCRE is compiled with
1229        --disable-stack-for-recursion (#990).
1230    
1231    18. If the last data line in a file for pcretest does not have a newline on
1232        the end, a newline was missing in the output.
1233    
1234    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1235        less than 128) in its various bitmaps. However, there is a facility for
1236        generating tables according to the current locale when PCRE is compiled. It
1237        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1238        characters, are recognized by isspace() and therefore were getting set in
1239        these tables, and indeed these tables seem to approximate to ISO 8859. This
1240        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1241        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1242        which of course cannot start UTF-8 characters. I have changed the code so
1243        that only real ASCII characters (less than 128) and the correct starting
1244        bytes for UTF-8 encodings are set for characters greater than 127 when in
1245        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1246        altogether.)
1247    
1248    20. Added the /T option to pcretest so as to be able to run tests with non-
1249        standard character tables, thus making it possible to include the tests
1250        used for 19 above in the standard set of tests.
1251    
1252    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1253        reference to a subpattern the other side of a comment that contains an
1254        opening parenthesis caused either an internal compiling error, or a
1255        reference to the wrong subpattern.
1256    
1257    
1258    Version 8.02 19-Mar-2010
1259    ------------------------
1260    
1261    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1262    
1263    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1264        configured.
1265    
1266    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1267        original author of that file, following a query about its status.
1268    
1269    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1270        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1271    
1272    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1273        quantifier applied to a forward-referencing subroutine call, could compile
1274        incorrect code or give the error "internal error: previously-checked
1275        referenced subpattern not found".
1276    
1277    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1278        variables to point to external functions. For these systems, therefore,
1279        pcre_malloc etc. are now initialized to local functions that call the
1280        relevant global functions.
1281    
1282    7.  There were two entries missing in the vectors called coptable and poptable
1283        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1284        I've fixed the data, and added a kludgy way of testing at compile time that
1285        the lengths are correct (equal to the number of opcodes).
1286    
1287    8.  Following on from 7, I added a similar kludge to check the length of the
1288        eint vector in pcreposix.c.
1289    
1290    9.  Error texts for pcre_compile() are held as one long string to avoid too
1291        much relocation at load time. To find a text, the string is searched,
1292        counting zeros. There was no check for running off the end of the string,
1293        which could happen if a new error number was added without updating the
1294        string.
1295    
1296    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1297    
1298    11. \K was not working if it appeared in an atomic group or in a group that
1299        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1300        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1301        the assertion is positive, but not if it is negative.
1302    
1303    12. Change 11 fortuitously reduced the size of the stack frame used in the
1304        "match()" function of pcre_exec.c by one pointer. Forthcoming
1305        implementation of support for (*MARK) will need an extra pointer on the
1306        stack; I have reserved it now, so that the stack frame size does not
1307        decrease.
1308    
1309    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1310        item in branch that calls a recursion is a subroutine call - as in the
1311        second branch in the above example - was incorrectly given the compile-
1312        time error "recursive call could loop indefinitely" because pcre_compile()
1313        was not correctly checking the subroutine for matching a non-empty string.
1314    
1315    14. The checks for overrunning compiling workspace could trigger after an
1316        overrun had occurred. This is a "should never occur" error, but it can be
1317        triggered by pathological patterns such as hundreds of nested parentheses.
1318        The checks now trigger 100 bytes before the end of the workspace.
1319    
1320    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1321    
1322    
1323    Version 8.01 19-Jan-2010
1324    ------------------------
1325    
1326    1.  If a pattern contained a conditional subpattern with only one branch (in
1327        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1328        computed the wrong minimum data length (which is of course zero for such
1329        subpatterns). This could cause incorrect "no match" results.
1330    
1331    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1332        the pattern is reset in the first branch, pcre_compile() failed with
1333        "internal error: code overflow at offset...". This happened only when
1334        the reset was to the original external option setting. (An optimization
1335        abstracts leading options settings into an external setting, which was the
1336        cause of this.)
1337    
1338    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1339        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1340        assertion pattern did not match (meaning that the assertion was true), it
1341        was incorrectly treated as false if the SKIP had been reached during the
1342        matching. This also applied to assertions used as conditions.
1343    
1344    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1345        assertion subpattern, including such a pattern used as a condition,
1346        unpredictable results occurred, instead of the error return
1347        PCRE_ERROR_DFA_UITEM.
1348    
1349    5.  The C++ GlobalReplace function was not working like Perl for the special
1350        situation when an empty string is matched. It now does the fancy magic
1351        stuff that is necessary.
1352    
1353    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1354        removed. (These were left over from very, very early versions of PCRE.)
1355    
1356    7.  Some cosmetic changes to the code to make life easier when compiling it
1357        as part of something else:
1358    
1359        (a) Change DEBUG to PCRE_DEBUG.
1360    
1361        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1362            called "current" as "current_branch", to prevent a collision with the
1363            Linux macro when compiled as a kernel module.
1364    
1365        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1366            prevent a collision with the Linux macro when compiled as a kernel
1367            module.
1368    
1369    8.  In pcre_compile() there are some checks for integer overflows that used to
1370        cast potentially large values to (double). This has been changed to that
1371        when building, a check for int64_t is made, and if it is found, it is used
1372        instead, thus avoiding the use of floating point arithmetic. (There is no
1373        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1374        double.
1375    
1376    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1377        2005 (difference between two addresses compared to an unsigned value).
1378    
1379    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1380        custom one, because of the following reported problem in Windows:
1381    
1382          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1383              under Win32.
1384          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1385              therefore missing the function definition.
1386          - The compiler thus generates a "C" signature for the test function.
1387          - The linker fails to find the "C" function.
1388          - PCRE fails to configure if asked to do so against libbz2.
1389    
1390    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1391        messages were output:
1392    
1393          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1394          rerunning libtoolize, to keep the correct libtool macros in-tree.
1395          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1396    
1397        I have done both of these things.
1398    
1399    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1400        most of the time, it *can* run out if it is given a pattern that contains a
1401        runaway infinite recursion. I updated the discussion in the pcrestack man
1402        page.
1403    
1404    13. Now that we have gone to the x.xx style of version numbers, the minor
1405        version may start with zero. Using 08 or 09 is a bad idea because users
1406        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1407        interpreted as invalid octal numbers. I've updated the previous comment in
1408        configure.ac, and also added a check that gives an error if 08 or 09 are
1409        used.
1410    
1411    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1412        causing partial matching to fail when the end of the subject matched \W
1413        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1414    
1415    15. There were some discrepancies between the declarations in pcre_internal.h
1416        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1417        their definitions. The declarations used "const uschar *" and the
1418        definitions used USPTR. Even though USPTR is normally defined as "const
1419        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1420        reported that: "This difference in casting confuses some C++ compilers, for
1421        example, SunCC recognizes above declarations as different functions and
1422        generates broken code for hbpcre." I have changed the declarations to use
1423        USPTR.
1424    
1425    16. GNU libtool is named differently on some systems. The autogen.sh script now
1426        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1427        (FreeBSD).
1428    
1429    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1430        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1431        comment: "Figure out how to create a longlong from a string: strtoll and
1432        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1433        instance, but it only takes 2 args instead of 3!"
1434    
1435    18. A subtle bug concerned with back references has been fixed by a change of
1436        specification, with a corresponding code fix. A pattern such as
1437        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1438        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1439        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1440        same bug. Such groups have to be quantified to be useful, or contained
1441        inside another quantified group. (If there's no repetition, the reference
1442        can never match.) The problem arises because, having left the group and
1443        moved on to the rest of the pattern, a later failure that backtracks into
1444        the group uses the captured value from the final iteration of the group
1445        rather than the correct earlier one. I have fixed this in PCRE by forcing
1446        any group that contains a reference to itself to be an atomic group; that
1447        is, there cannot be any backtracking into it once it has completed. This is
1448        similar to recursive and subroutine calls.
1449    
1450    
1451    Version 8.00 19-Oct-09
1452    ----------------------
1453    
1454    1.  The table for translating pcre_compile() error codes into POSIX error codes
1455        was out-of-date, and there was no check on the pcre_compile() error code
1456        being within the table. This could lead to an OK return being given in
1457        error.
1458    
1459    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1460        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1461        in a Windows environment.
1462    
1463    3.  The pcregrep --count option prints the count for each file even when it is
1464        zero, as does GNU grep. However, pcregrep was also printing all files when
1465        --files-with-matches was added. Now, when both options are given, it prints
1466        counts only for those files that have at least one match. (GNU grep just
1467        prints the file name in this circumstance, but including the count seems
1468        more useful - otherwise, why use --count?) Also ensured that the
1469        combination -clh just lists non-zero counts, with no names.
1470    
1471    4.  The long form of the pcregrep -F option was incorrectly implemented as
1472        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1473        but it seems right to fix it, and I didn't think it was worth preserving
1474        the old behaviour.
1475    
1476    5.  The command line items --regex=pattern and --regexp=pattern were not
1477        recognized by pcregrep, which required --regex pattern or --regexp pattern
1478        (with a space rather than an '='). The man page documented the '=' forms,
1479        which are compatible with GNU grep; these now work.
1480    
1481    6.  No libpcreposix.pc file was created for pkg-config; there was just
1482        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1483    
1484    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1485        when UCP support is not needed, by modifying the Python script that
1486        generates it from Unicode data files. This should not matter if the module
1487        is correctly used as a library, but I received one complaint about 50K of
1488        unwanted data. My guess is that the person linked everything into his
1489        program rather than using a library. Anyway, it does no harm.
1490    
1491    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1492        was a minimum greater than 1 for a wide character in a possessive
1493        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1494        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1495        character. Chaos in the form of incorrect output or a compiling loop could
1496        result.
1497    
1498    9.  The restrictions on what a pattern can contain when partial matching is
1499        requested for pcre_exec() have been removed. All patterns can now be
1500        partially matched by this function. In addition, if there are at least two
1501        slots in the offset vector, the offset of the earliest inspected character
1502        for the match and the offset of the end of the subject are set in them when
1503        PCRE_ERROR_PARTIAL is returned.
1504    
1505    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1506        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1507        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1508        and may be more useful for multi-segment matching.
1509    
1510    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1511        used to be given if ever the end of the subject was reached; now it is
1512        given only if matching could not proceed because another character was
1513        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1514        string "Z", which now yields "no match" instead of "partial match". In the
1515        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1516        final character ended with (*FAIL).
1517    
1518    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1519        if the pattern had a "must contain" character that was already found in the
1520        earlier partial match, unless partial matching was again requested. For
1521        example, with the pattern /dog.(body)?/, the "must contain" character is
1522        "g". If the first part-match was for the string "dog", restarting with
1523        "sbody" failed. This bug has been fixed.
1524    
1525    13. The string returned by pcre_dfa_exec() after a partial match has been
1526        changed so that it starts at the first inspected character rather than the
1527        first character of the match. This makes a difference only if the pattern
1528        starts with a lookbehind assertion or \b or \B (\K is not supported by
1529        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1530        matching functions compatible, and I think it's the right thing to do.
1531    
1532    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1533        so that the demonstration program is easily available in environments where
1534        PCRE has not been installed from source.
1535    
1536    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1537        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1538        library.
1539    
1540    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1541        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1542        is not the first non-POSIX option to be added. Clearly some people find
1543        these options useful.
1544    
1545    17. If a caller to the POSIX matching function regexec() passes a non-zero
1546        value for nmatch with a NULL value for pmatch, the value of
1547        nmatch is forced to zero.
1548    
1549    18. RunGrepTest did not have a test for the availability of the -u option of
1550        the diff command, as RunTest does. It now checks in the same way as
1551        RunTest, and also checks for the -b option.
1552    
1553    19. If an odd number of negated classes containing just a single character
1554        interposed, within parentheses, between a forward reference to a named
1555        subpattern and the definition of the subpattern, compilation crashed with
1556        an internal error, complaining that it could not find the referenced
1557        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1558        [The bug was that it was starting one character too far in when skipping
1559        over the character class, thus treating the ] as data rather than
1560        terminating the class. This meant it could skip too much.]
1561    
1562    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1563        /g option in pcretest when the pattern contains \K, which makes it possible
1564        to have an empty string match not at the start, even when the pattern is
1565        anchored. Updated pcretest and pcredemo to use this option.
1566    
1567    21. If the maximum number of capturing subpatterns in a recursion was greater
1568        than the maximum at the outer level, the higher number was returned, but
1569        with unset values at the outer level. The correct (outer level) value is
1570        now given.
1571    
1572    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1573        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1574        make it do so. The string so far is captured, making this feature
1575        compatible with Perl.
1576    
1577    23. The tests have been re-organized, adding tests 11 and 12, to make it
1578        possible to check the Perl 5.10 features against Perl 5.10.
1579    
1580    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1581        pattern matches a fixed length string. PCRE did not allow this; now it
1582        does. Neither allows recursion.
1583    
1584    25. I finally figured out how to implement a request to provide the minimum
1585        length of subject string that was needed in order to match a given pattern.
1586        (It was back references and recursion that I had previously got hung up
1587        on.) This code has now been added to pcre_study(); it finds a lower bound
1588        to the length of subject needed. It is not necessarily the greatest lower
1589        bound, but using it to avoid searching strings that are too short does give
1590        some useful speed-ups. The value is available to calling programs via
1591        pcre_fullinfo().
1592    
1593    26. While implementing 25, I discovered to my embarrassment that pcretest had
1594        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1595        study optimizations had never been tested with that matching function.
1596        Oops. What is worse, even when it was passed study data, there was a bug in
1597        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1598        were also very few tests of studied patterns with pcre_dfa_exec().
1599    
1600    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1601        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1602        on the other side of the coin, they are no longer allowed to have different
1603        names, because these cannot be distinguished in PCRE, and this has caused
1604        confusion. (This is a difference from Perl.)
1605    
1606    28. When duplicate subpattern names are present (necessarily with different
1607        numbers, as required by 27 above), and a test is made by name in a
1608        conditional pattern, either for a subpattern having been matched, or for
1609        recursion in such a pattern, all the associated numbered subpatterns are
1610        tested, and the overall condition is true if the condition is true for any
1611        one of them. This is the way Perl works, and is also more like the way
1612        testing by number works.
1613    
1614    
1615    Version 7.9 11-Apr-09
1616    ---------------------
1617    
1618    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1619        (pcretest), all targets were linked against these libraries. This included
1620        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1621        libraries. This caused unwanted dependencies to be created. This problem
1622        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1623        pcretest is linked with readline.
1624    
1625    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1626        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1627        moved outside it again, because FALSE and TRUE are already defined in AIX,
1628        but BOOL is not.
1629    
1630    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1631        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1632    
1633    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1634        hyphens) following file names and line numbers when outputting matching
1635        lines. This is not true; no spaces are inserted. I have also clarified the
1636        wording for the --colour (or --color) option.
1637    
1638    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1639        was not coloured; this is different to GNU grep, so I have changed it to be
1640        the same.
1641    
1642    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1643        each matching line was coloured. Now it goes on to look for further matches
1644        of any of the test patterns, which is the same behaviour as GNU grep.
1645    
1646    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1647        doesn't make sense to accept an empty string match in pcregrep, so I have
1648        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1649        seems to be how GNU grep behaves.
1650    
1651    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1652        start or after a newline", because the conditional assertion was not being
1653        correctly handled. The rule now is that both the assertion and what follows
1654        in the first alternative must satisfy the test.
1655    
1656    9.  If auto-callout was enabled in a pattern with a conditional group whose
1657        condition was an assertion, PCRE could crash during matching, both with
1658        pcre_exec() and pcre_dfa_exec().
1659    
1660    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1661        used for matching.
1662    
1663    11. Unicode property support in character classes was not working for
1664        characters (bytes) greater than 127 when not in UTF-8 mode.
1665    
1666    12. Added the -M command line option to pcretest.
1667    
1668    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1669    
1670    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1671    
1672    16. Added comments and documentation about mis-use of no_arg in the C++
1673        wrapper.
1674    
1675    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1676        from Martin Jerabek that uses macro names for all relevant character and
1677        string constants.
1678    
1679    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1680        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1681        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1682        these, but not everybody uses configure.
1683    
1684    19. A conditional group that had only one branch was not being correctly
1685        recognized as an item that could match an empty string. This meant that an
1686        enclosing group might also not be so recognized, causing infinite looping
1687        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1688        with the subject "ab", where knowledge that the repeated group can match
1689        nothing is needed in order to break the loop.
1690    
1691    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1692        exec(), but without supplying a callout function, matching went wrong.
1693    
1694    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1695        leak if the size of the offset vector was greater than 30. When the vector
1696        is smaller, the saved offsets during recursion go onto a local stack
1697        vector, but for larger vectors malloc() is used. It was failing to free
1698        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1699        error, in fact).
1700    
1701    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1702        heapframe that is used only when UTF-8 support is enabled. This caused no
1703        problem, but was untidy.
1704    
1705    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1706        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1707        included within another project.
1708    
1709    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1710        slightly modified by me:
1711    
1712          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1713              not building pcregrep.
1714    
1715          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1716              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1717    
1718    25. Forward references, both numeric and by name, in patterns that made use of
1719        duplicate group numbers, could behave incorrectly or give incorrect errors,
1720        because when scanning forward to find the reference group, PCRE was not
1721        taking into account the duplicate group numbers. A pattern such as
1722        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1723    
1724    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1725        the feature of a custom pointer more persuasive (as requested by a user).
1726    
1727    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1728        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1729        pre-defined.
1730    
1731    28. Added support for (*UTF8) at the start of a pattern.
1732    
1733    29. Arrange for flags added by the "release type" setting in CMake to be shown
1734        in the configuration summary.
1735    
1736    
1737    Version 7.8 05-Sep-08
1738    ---------------------
1739    
1740    1.  Replaced UCP searching code with optimized version as implemented for Ad
1741        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1742        stage table and inline lookup instead of a function, giving speed ups of 2
1743        to 5 times on some simple patterns that I tested. Permission was given to
1744        distribute the MultiStage2.py script that generates the tables (it's not in
1745        the tarball, but is in the Subversion repository).
1746    
1747    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1748        scripts.
1749    
1750    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1751        a group with a zero qualifier. The result of the study could be incorrect,
1752        or the function might crash, depending on the pattern.
1753    
1754    4.  Caseless matching was not working for non-ASCII characters in back
1755        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1756        It now works when Unicode Property Support is available.
1757    
1758    5.  In pcretest, an escape such as \x{de} in the data was always generating
1759        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1760        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1761        truncation.
1762    
1763    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1764    
1765    7.  Added two (int) casts to pcregrep when printing the difference of two
1766        pointers, in case they are 64-bit values.
1767    
1768    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1769        test 2 if it fails.
1770    
1771    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1772        and a #define of that name to empty if it is not externally set. This is to
1773        allow users of MSVC to set it if necessary.
1774    
1775    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1776        the convenience functions in the pcre_get.c source file.
1777    
1778    11. An option change at the start of a pattern that had top-level alternatives
1779        could cause overwriting and/or a crash. This command provoked a crash in
1780        some environments:
1781    
1782          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1783    
1784        This potential security problem was recorded as CVE-2008-2371.
1785    
1786    12. For a pattern where the match had to start at the beginning or immediately
1787        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1788        pcre_dfa_exec() could read past the end of the passed subject if there was
1789        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1790        pcretest so that it places the subject at the end of its malloc-ed buffer.
1791    
1792    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1793        exec() might read past the end of the data buffer in UTF-8 mode.
1794    
1795    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1796        the data contained the byte 0x85 as part of a UTF-8 character within its
1797        first line. This applied both to normal and DFA matching.
1798    
1799    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1800        /^[^d]*?$/8 failed to match "abc".
1801    
1802    16. Added a missing copyright notice to pcrecpp_internal.h.
1803    
1804    17. Make it more clear in the documentation that values returned from
1805        pcre_exec() in ovector are byte offsets, not character counts.
1806    
1807    18. Tidied a few places to stop certain compilers from issuing warnings.
1808    
1809    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1810        supplied by Stefan Weber. I made a further small update for 7.8 because
1811        there is a change of source arrangements: the pcre_searchfuncs.c module is
1812        replaced by pcre_ucd.c.
1813    
1814    
1815    Version 7.7 07-May-08
1816    ---------------------
1817    
1818    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1819        a string to a long long, pretend we don't even have a long long." This is
1820        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1821    
1822    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1823        pre-7.6 versions, which defined a global no_arg variable instead of putting
1824        it in the RE class. (See also #8 below.)
1825    
1826    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1827        Lopes.
1828    
1829    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1830    
1831        (1) The include/exclude patterns were being applied to the whole pathnames
1832            of files, instead of just to the final components.
1833    
1834        (2) If there was more than one level of directory, the subdirectories were
1835            skipped unless they satisfied the include/exclude conditions. This is
1836            inconsistent with GNU grep (and could even be seen as contrary to the
1837            pcregrep specification - which I improved to make it absolutely clear).
1838            The action now is always to scan all levels of directory, and just
1839            apply the include/exclude patterns to regular files.
1840    
1841    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1842        --exclude_dir in the tests to avoid scanning .svn directories.
1843    
1844    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1845        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1846        doesn't support NULs in patterns.
1847    
1848    7.  Added some missing "const"s to declarations of static tables in
1849        pcre_compile.c and pcre_dfa_exec.c.
1850    
1851    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1852        caused by fix #2  above. (Subsequently also a second patch to fix the
1853        first patch. And a third patch - this was a messy problem.)
1854    
1855    9.  Applied Craig's patch to remove the use of push_back().
1856    
1857    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1858        matching function regexec().
1859    
1860    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1861        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1862        references. PCRE supports relative numbers with this syntax (I don't think
1863        Oniguruma does).
1864    
1865    12. Previously, a group with a zero repeat such as (...){0} was completely
1866        omitted from the compiled regex. However, this means that if the group
1867        was called as a subroutine from elsewhere in the pattern, things went wrong
1868        (an internal error was given). Such groups are now left in the compiled
1869        pattern, with a new opcode that causes them to be skipped at execution
1870        time.
1871    
1872    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1873        to the way PCRE behaves:
1874    
1875        (a) A lone ] character is dis-allowed (Perl treats it as data).
1876    
1877        (b) A back reference to an unmatched subpattern matches an empty string
1878            (Perl fails the current match path).
1879    
1880        (c) A data ] in a character class must be notated as \] because if the
1881            first data character in a class is ], it defines an empty class. (In
1882            Perl it is not possible to have an empty class.) The empty class []
1883            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1884            The negative empty class [^] matches any one character, independently
1885            of the DOTALL setting.
1886    
1887    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1888        non-existent subpattern following a character class starting with ']' and
1889        containing () gave an internal compiling error instead of "reference to
1890        non-existent subpattern". Fortunately, when the pattern did exist, the
1891        compiled code was correct. (When scanning forwards to check for the
1892        existencd of the subpattern, it was treating the data ']' as terminating
1893        the class, so got the count wrong. When actually compiling, the reference
1894        was subsequently set up correctly.)
1895    
1896    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1897        it was being rejected as not supported by pcre_dfa_exec(), even though
1898        other assertions are supported. I have made pcre_dfa_exec() support
1899        (*FAIL).
1900    
1901    16. The implementation of 13c above involved the invention of a new opcode,
1902        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1903        cannot be changed at match time, I realized I could make a small
1904        improvement to matching performance by compiling OP_ALLANY instead of
1905        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1906        on the OP_ANY path.
1907    
1908    17. Compiling pcretest on Windows with readline support failed without the
1909        following two fixes: (1) Make the unistd.h include conditional on
1910        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1911    
1912    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1913        ncurses library to be included for pcretest when ReadLine support is
1914        requested, but also to allow for it to be overridden. This patch came from
1915        Daniel Bergström.
1916    
1917    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1918        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1919        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1920        spotting this.
1921    
1922    
1923    Version 7.6 28-Jan-08
1924    ---------------------
1925    
1926    1.  A character class containing a very large number of characters with
1927        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1928        overflow.
1929    
1930    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1931        HAVE_LONG_LONG is not defined.
1932    
1933    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1934        bring them up to date and include new features. This patch includes:
1935    
1936        - Fixed PH's badly added libz and libbz2 support.
1937        - Fixed a problem with static linking.
1938        - Added pcredemo. [But later removed - see 7 below.]
1939        - Fixed dftables problem and added an option.
1940        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1941            HAVE_LONG_LONG.
1942        - Added readline support for pcretest.
1943        - Added an listing of the option settings after cmake has run.
1944    
1945    4.  A user submitted a patch to Makefile that makes it easy to create
1946        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1947        Makefile.am that cause it to include this special target, without
1948        affecting anything else. Note that the same mingw target plus all
1949        the other distribution libraries and programs are now supported
1950        when configuring with CMake (see 6 below) instead of with
1951        Configure/Make.
1952    
1953    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1954        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1955        exported in the Windows port". It has not yet been confirmed that the patch
1956        solves the problem, but it does no harm.
1957    
1958    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1959        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1960        with CMake, and also correct the comment about stack recursion.
1961    
1962    7.  Remove the automatic building of pcredemo from the ./configure system and
1963        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1964        of a program that users should build themselves after PCRE is installed, so
1965        building it automatically is not really right. What is more, it gave
1966        trouble in some build environments.
1967    
1968    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1969    
1970    
1971    Version 7.5 10-Jan-08
1972    ---------------------
1973    
1974    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1975        values in parens when parsing an RE using the C++ wrapper."
1976    
1977    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1978        Characters greater than 255 were excluded from the class instead of being
1979        included.
1980    
1981    3.  The same bug as (2) above applied to negated POSIX classes such as
1982        [:^space:].
1983    
1984    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1985        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1986        I have changed it.
1987    
1988    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1989        first named subpattern) and a construct such as (?&a) would reference the
1990        first named subpattern whose name started with "a" (in other words, the
1991        length check was missing). Both these problems are fixed. "Subpattern name
1992        expected" is now given for (?&) (a zero-length name), and this patch also
1993        makes it give the same error for \k'' (previously it complained that that
1994        was a reference to a non-existent subpattern).
1995    
1996    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1997        this is right because (?- can be followed by option settings as well as by
1998        digits. I have, however, made the messages clearer.
1999    
2000    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
2001        than the number used in the conditional) now cause a compile-time error.
2002        This is actually not compatible with Perl, which accepts such patterns, but
2003        treats the conditional as always being FALSE (as PCRE used to), but it
2004        seems to me that giving a diagnostic is better.
2005    
2006    8.  Change "alphameric" to the more common word "alphanumeric" in comments
2007        and messages.
2008    
2009    9.  Fix two occurrences of "backslash" in comments that should have been
2010        "backspace".
2011    
2012    10. Remove two redundant lines of code that can never be obeyed (their function
2013        was moved elsewhere).
2014    
2015    11. The program that makes PCRE's Unicode character property table had a bug
2016        which caused it to generate incorrect table entries for sequences of
2017        characters that have the same character type, but are in different scripts.
2018        It amalgamated them into a single range, with the script of the first of
2019        them. In other words, some characters were in the wrong script. There were
2020        thirteen such cases, affecting characters in the following ranges:
2021    
2022          U+002b0 - U+002c1
2023          U+0060c - U+0060d
2024          U+0061e - U+00612
2025          U+0064b - U+0065e
2026          U+0074d - U+0076d
2027          U+01800 - U+01805
2028          U+01d00 - U+01d77
2029          U+01d9b - U+01dbf
2030          U+0200b - U+0200f
2031          U+030fc - U+030fe
2032          U+03260 - U+0327f
2033          U+0fb46 - U+0fbb1
2034          U+10450 - U+1049d
2035    
2036    12. The -o option (show only the matching part of a line) for pcregrep was not
2037        compatible with GNU grep in that, if there was more than one match in a
2038        line, it showed only the first of them. It now behaves in the same way as
2039        GNU grep.
2040    
2041    13. If the -o and -v options were combined for pcregrep, it printed a blank
2042        line for every non-matching line. GNU grep prints nothing, and pcregrep now
2043        does the same. The return code can be used to tell if there were any
2044        non-matching lines.
2045    
2046    14. Added --file-offsets and --line-offsets to pcregrep.
2047    
2048    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
2049        infinitely looping recursion. The bug was that positive lookaheads were not
2050        being skipped when checking for a possible empty match (negative lookaheads
2051        and both kinds of lookbehind were skipped).
2052    
2053    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
2054        inclusion of <windows.h> to before rather than after the definition of
2055        INVALID_FILE_ATTRIBUTES (patch from David Byron).
2056    
2057    17. Specifying a possessive quantifier with a specific limit for a Unicode
2058        character property caused pcre_compile() to compile bad code, which led at
2059        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
2060        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
2061        caused the error; without that there was no problem.
2062    
2063    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
2064    
2065    19. Added --enable-pcretest-libreadline.
2066    
2067    20. In pcrecpp.cc, the variable 'count' was incremented twice in
2068        RE::GlobalReplace(). As a result, the number of replacements returned was
2069        double what it should be. I removed one of the increments, but Craig sent a
2070        later patch that removed the other one (the right fix) and added unit tests
2071        that check the return values (which was not done before).
2072    
2073    21. Several CMake things:
2074    
2075        (1) Arranged that, when cmake is used on Unix, the libraries end up with
2076            the names libpcre and libpcreposix, not just pcre and pcreposix.
2077    
2078        (2) The above change means that pcretest and pcregrep are now correctly
2079            linked with the newly-built libraries, not previously installed ones.
2080    
2081        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
2082    
2083    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
2084        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
2085        UTF-8 newline character). The key issue is that the pattern starts .*;
2086        this means that the match must be either at the beginning, or after a
2087        newline. The bug was in the code for advancing after a failed match and
2088        checking that the new position followed a newline. It was not taking
2089        account of UTF-8 characters correctly.
2090    
2091    23. PCRE was behaving differently from Perl in the way it recognized POSIX
2092        character classes. PCRE was not treating the sequence [:...:] as a
2093        character class unless the ... were all letters. Perl, however, seems to
2094        allow any characters between [: and :], though of course it rejects as
2095        unknown any "names" that contain non-letters, because all the known class
2096        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
2097        for example, whereas PCRE did not - it did not recognize a POSIX character
2098        class. This seemed a bit dangerous, so the code has been changed to be
2099        closer to Perl. The behaviour is not identical to Perl, because PCRE will
2100        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
2101        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
2102        Perl does, and where it didn't before.
2103    
2104    24. Rewrite so as to remove the single use of %n from pcregrep because in some
2105        Windows environments %n is disabled by default.
2106    
2107    
2108    Version 7.4 21-Sep-07
2109    ---------------------
2110    
2111    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
2112        means that a class such as [\s] counted as "explicit reference to CR or
2113        LF". That isn't really right - the whole point of the change was to try to
2114        help when there was an actual mention of one of the two characters. So now
2115        the change happens only if \r or \n (or a literal CR or LF) character is
2116        encountered.
2117    
2118    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
2119        of both had grown to the point where there were only 3 bits left.
2120        Fortunately, there was spare space in the data structure, and so I have
2121        moved the internal flags into a new 16-bit field to free up more option
2122        bits.
2123    
2124    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
2125        but did not set the internal JCHANGED flag - either of these is enough to
2126        control the way the "get" function works - but the PCRE_INFO_JCHANGED
2127        facility is supposed to tell if (?J) was ever used, so now (?J) at the
2128        start sets both bits.
2129    
2130    4.  Added options (at build time, compile time, exec time) to change \R from
2131        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
2132    
2133    5.  doc/pcresyntax.html was missing from the distribution.
2134    
2135    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
2136        compatibility, even though it is no longer used.
2137    
2138    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
2139        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
2140        windows.h file is present (where different names are used). [This was
2141        reversed later after testing - see 16 below.]
2142    
2143    8.  Changed all #include <config.h> to #include "config.h". There were also
2144        some further <pcre.h> cases that I changed to "pcre.h".
2145    
2146    9.  When pcregrep was used with the --colour option, it missed the line ending
2147        sequence off the lines that it output.
2148    
2149    10. It was pointed out to me that arrays of string pointers cause lots of
2150        relocations when a shared library is dynamically loaded. A technique of
2151        using a single long string with a table of offsets can drastically reduce
2152        these. I have refactored PCRE in four places to do this. The result is
2153        dramatic:
2154    
2155          Originally:                          290
2156          After changing UCP table:            187
2157          After changing error message table:   43
2158          After changing table of "verbs"       36
2159          After changing table of Posix names   22
2160    
2161        Thanks to the folks working on Gregex for glib for this insight.
2162    
2163    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
2164        unicode-properties was also set.
2165    
2166    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
2167    
2168    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
2169        checked only for CRLF.
2170    
2171    14. Added casts to pcretest.c to avoid compiler warnings.
2172    
2173    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
2174    
2175    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
2176        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
2177        entirely. This removes changes made in 7 above.
2178    
2179    17. The CMake files have been updated, and there is now more information about
2180        building with CMake in the NON-UNIX-USE document.
2181    
2182    
2183    Version 7.3 28-Aug-07
2184    ---------------------
2185    
2186     1. In the rejigging of the build system that eventually resulted in 7.1, the
2187        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
2188        brackets there is not right, since it causes compilers to look for an
2189        installed pcre.h, not the version that is in the source that is being
2190        compiled (which of course may be different). I have changed it back to:
2191    
2192          #include "pcre.h"
2193    
2194        I have a vague recollection that the change was concerned with compiling in
2195        different directories, but in the new build system, that is taken care of
2196        by the VPATH setting the Makefile.
2197    
2198     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
2199        when the subject happened to end in the byte 0x85 (e.g. if the last
2200        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
2201        characters but of course it shouldn't be taken as a newline when it is part
2202        of another character. The bug was that, for an unlimited repeat of . in
2203        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
2204        characters when looking for a newline.
2205    
2206     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
2207    
2208     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
2209        in debug output.
2210    
2211     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
2212        long printing in the pcrecpp unittest when running under MinGW.
2213    
2214     6. ESC_K was left out of the EBCDIC table.
2215    
2216     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
2217        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
2218        limit also applies to "virtual nesting" when a pattern is recursive, and in
2219        this case 1000 isn't so big. I have been able to remove this limit at the
2220        expense of backing off one optimization in certain circumstances. Normally,
2221        when pcre_exec() would call its internal match() function recursively and
2222        immediately return the result unconditionally, it uses a "tail recursion"
2223        feature to save stack. However, when a subpattern that can match an empty
2224        string has an unlimited repetition quantifier, it no longer makes this
2225        optimization. That gives it a stack frame in which to save the data for
2226        checking that an empty string has been matched. Previously this was taken
2227        from the 1000-entry workspace that had been reserved. So now there is no
2228        explicit limit, but more stack is used.
2229    
2230     8. Applied Daniel's patches to solve problems with the import/export magic
2231        syntax that is required for Windows, and which was going wrong for the
2232        pcreposix and pcrecpp parts of the library. These were overlooked when this
2233        problem was solved for the main library.
2234    
2235     9. There were some crude static tests to avoid integer overflow when computing
2236        the size of patterns that contain repeated groups with explicit upper
2237        limits. As the maximum quantifier is 65535, the maximum group length was
2238        set at 30,000 so that the product of these two numbers did not overflow a
2239        32-bit integer. However, it turns out that people want to use groups that
2240        are longer than 30,000 bytes (though not repeat them that many times).
2241        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
2242        made it possible to implement the integer overflow checks in a much more
2243        dynamic way, which I have now done. The artificial limitation on group
2244        length has been removed - we now have only the limit on the total length of
2245        the compiled pattern, which depends on the LINK_SIZE setting.
2246    
2247    10. Fixed a bug in the documentation for get/copy named substring when
2248        duplicate names are permitted. If none of the named substrings are set, the
2249        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
2250        empty string.
2251    
2252    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
2253        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
2254        because the ] is interpreted as the first data character and the
2255        terminating ] is not found. PCRE has been made compatible with Perl in this
2256        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
2257        cause memory overwriting.
2258    
2259    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
2260        string has been matched (to stop an infinite loop). It was not recognizing
2261        a conditional subpattern that could match an empty string if that
2262        subpattern was within another subpattern. For example, it looped when
2263        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
2264        condition was not nested. This bug has been fixed.
2265    
2266    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
2267        past the start of the subject in the presence of bytes with the top bit
2268        set, for example "\x8aBCD".
2269    
2270    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
2271        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
2272    
2273    14. Optimized (?!) to (*FAIL).
2274    
2275    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
2276        This restricts code points to be within the range 0 to 0x10FFFF, excluding
2277        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
2278        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
2279        does: it's just the validity check that is more restrictive.
2280    
2281    16. Inserted checks for integer overflows during escape sequence (backslash)
2282        processing, and also fixed erroneous offset values for syntax errors during
2283        backslash processing.
2284    
2285    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
2286        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
2287    
2288    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
2289        caused an overrun.
2290    
2291    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
2292        something other than just ASCII characters) inside a group that had an
2293        unlimited repeat caused a loop at compile time (while checking to see
2294        whether the group could match an empty string).
2295    
2296    20. Debugging a pattern containing \p or \P could cause a crash. For example,
2297        [\P{Any}] did so. (Error in the code for printing property names.)
2298    
2299    21. An orphan \E inside a character class could cause a crash.
2300    
2301    22. A repeated capturing bracket such as (A)? could cause a wild memory
2302        reference during compilation.
2303    
2304    23. There are several functions in pcre_compile() that scan along a compiled
2305        expression for various reasons (e.g. to see if it's fixed length for look
2306        behind). There were bugs in these functions when a repeated \p or \P was
2307        present in the pattern. These operators have additional parameters compared
2308        with \d, etc, and these were not being taken into account when moving along
2309        the compiled data. Specifically:
2310    
2311        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
2312            length.
2313    
2314        (b) An item such as \pL+ within a repeated group could cause crashes or
2315            loops.
2316    
2317        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
2318            "reference to non-existent subpattern" error.
2319    
2320        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
2321    
2322    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
2323        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
2324    
2325    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
2326    
2327    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
2328        character were causing crashes (broken optimization).
2329    
2330    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
2331        \p or \P) caused a compile-time loop.
2332    
2333    28. More problems have arisen in unanchored patterns when CRLF is a valid line
2334        break. For example, the unstudied pattern [\r\n]A does not match the string
2335        "\r\nA" because change 7.0/46 below moves the current point on by two
2336        characters after failing to match at the start. However, the pattern \nA
2337        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
2338        the same is true. There doesn't seem any very clean way out of this, but
2339        what I have chosen to do makes the common cases work: PCRE now takes note
2340        of whether there can be an explicit match for \r or \n anywhere in the
2341        pattern, and if so, 7.0/46 no longer applies. As part of this change,
2342        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
2343        pattern has explicit CR or LF references.
2344    
2345    29. Added (*CR) etc for changing newline setting at start of pattern.
2346    
2347    
2348    Version 7.2 19-Jun-07
2349    ---------------------
2350    
2351     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
2352        which is apparently normally available under Windows.
2353    
2354     2. Re-jig the pcregrep tests with different newline settings in an attempt
2355        to make them independent of the local environment's newline setting.
2356    
2357     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
2358    
2359     4. Some of the "internals" tests were previously cut out when the link size
2360        was not 2, because the output contained actual offsets. The recent new
2361        "Z" feature of pcretest means that these can be cut out, making the tests
2362        usable with all link sizes.
2363    
2364     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
2365        stack recursion. This gives a massive performance boost under BSD, but just
2366        a small improvement under Linux. However, it saves one field in the frame
2367        in all cases.
2368    
2369     6. Added more features from the forthcoming Perl 5.10:
2370    
2371        (a) (?-n) (where n is a string of digits) is a relative subroutine or
2372            recursion call. It refers to the nth most recently opened parentheses.
2373    
2374        (b) (?+n) is also a relative subroutine call; it refers to the nth next
2375            to be opened parentheses.
2376    
2377        (c) Conditions that refer to capturing parentheses can be specified
2378            relatively, for example, (?(-2)... or (?(+3)...
2379    
2380        (d) \K resets the start of the current match so that everything before
2381            is not part of it.
2382    
2383        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
2384    
2385        (f) \g{name} is another synonym - part of Perl 5.10's unification of
2386            reference syntax.
2387    
2388        (g) (?| introduces a group in which the numbering of parentheses in each
2389            alternative starts with the same number.
2390    
2391        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
2392    
2393     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
2394        PCRE_INFO_JCHANGED.
2395    
2396     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
2397        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
2398        for detecting groups that can match an empty string.
2399    
2400     9. A pattern with a very large number of alternatives (more than several
2401        hundred) was running out of internal workspace during the pre-compile
2402        phase, where pcre_compile() figures out how much memory will be needed. A
2403        bit of new cunning has reduced the workspace needed for groups with
2404        alternatives. The 1000-alternative test pattern now uses 12 bytes of
2405        workspace instead of running out of the 4096 that are available.
2406    
2407    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
2408    
2409    11. Applied patch from Google to remove an optimization that didn't quite work.
2410        The report of the bug said:
2411    
2412          pcrecpp::RE("a*").FullMatch("aaa") matches, while
2413          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
2414          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
2415    
2416    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
2417        it matched the wrong number of bytes.
2418    
2419    
2420    Version 7.1 24-Apr-07
2421  ---------------------  ---------------------
2422    
2423   1. Applied Bob Rossi and Daniel G's patches to convert the build system to one   1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
# Line 126  Version 7.1 12-Mar-07 Line 2542  Version 7.1 12-Mar-07
2542  16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"  16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
2543      in an attempt to make files that differ only in their line terminators      in an attempt to make files that differ only in their line terminators
2544      compare equal. This works on Linux.      compare equal. This works on Linux.
2545    
2546  17. Under certain error circumstances pcregrep might try to free random memory  17. Under certain error circumstances pcregrep might try to free random memory
2547      as it exited. This is now fixed, thanks to valgrind.      as it exited. This is now fixed, thanks to valgrind.
2548    
2549  19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string  19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
2550      "abc\r\n\r\n", it found an unwanted second match after the second \r. This      "abc\r\n\r\n", it found an unwanted second match after the second \r. This
2551      was because its rules for how to advance for /g after matching an empty      was because its rules for how to advance for /g after matching an empty
2552      string did not allow for this case. They now check for it specially.      string at the end of a line did not allow for this case. They now check for
2553        it specially.
2554    
2555    20. pcretest is supposed to handle patterns and data of any length, by
2556        extending its buffers when necessary. It was getting this wrong when the
2557        buffer for a data line had to be extended.
2558    
2559    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
2560        CRLF as a newline sequence.
2561    
2562    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
2563        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
2564        I have nevertheless tidied it up.
2565    
2566    23. Added some casts to kill warnings from HP-UX ia64 compiler.
2567    
2568    24. Added a man page for pcre-config.
2569    
2570    
2571  Version 7.0 19-Dec-06  Version 7.0 19-Dec-06

Legend:
Removed from v.141  
changed lines
  Added in v.1311

  ViewVC Help
Powered by ViewVC 1.1.5