ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 425 by ph10, Mon Aug 17 14:48:48 2009 UTC revision 1581 by ph10, Wed Jul 22 09:06:15 2015 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4  Version 8.00 ??-???-??  Note that the PCRE 8.xx series (PCRE1) is now in a bugfix-only state. All
5    development is happening in the PCRE2 10.xx series.
7    Version 8.38 xx-xxx-xxxx
8    ------------------------
10    1.  If a group that contained a recursive back reference also contained a
11        forward reference subroutine call followed by a non-forward-reference
12        subroutine call, for example /.((?2)(?R)\1)()/, pcre2_compile() failed to
13        compile correct code, leading to undefined behaviour or an internally
14        detected error. This bug was discovered by the LLVM fuzzer.
16    2.  Quantification of certain items (e.g. atomic back references) could cause
17        incorrect code to be compiled when recursive forward references were
18        involved. For example, in this pattern: /(?1)()((((((\1++))\x85)+)|))/.
19        This bug was discovered by the LLVM fuzzer.
21    3.  A repeated conditional group whose condition was a reference by name caused
22        a buffer overflow if there was more than one group with the given name.
23        This bug was discovered by the LLVM fuzzer.
25    4.  A recursive back reference by name within a group that had the same name as
26        another group caused a buffer overflow. For example:
27        /(?J)(?'d'(?'d'\g{d}))/. This bug was discovered by the LLVM fuzzer.
29    5.  A forward reference by name to a group whose number is the same as the
30        current group, for example in this pattern: /(?|(\k'Pm')|(?'Pm'))/, caused
31        a buffer overflow at compile time. This bug was discovered by the LLVM
32        fuzzer.
34    6.  A lookbehind assertion within a set of mutually recursive subpatterns could
35        provoke a buffer overflow. This bug was discovered by the LLVM fuzzer.
37    7.  Another buffer overflow bug involved duplicate named groups with a
38        reference between their definition, with a group that reset capture
39        numbers, for example: /(?J:(?|(?'R')(\k'R')|((?'R'))))/. This has been
40        fixed by always allowing for more memory, even if not needed. (A proper fix
41        is implemented in PCRE2, but it involves more refactoring.)
43    8.  There was no check for integer overflow in subroutine calls such as (?123).
45    9.  The table entry for \l in EBCDIC environments was incorrect, leading to its
46        being treated as a literal 'l' instead of causing an error.
48    10. There was a buffer overflow if pcre_exec() was called with an ovector of
49        size 1. This bug was found by american fuzzy lop.
51    11. If a non-capturing group containing a conditional group that could match
52        an empty string was repeated, it was not identified as matching an empty
53        string itself. For example: /^(?:(?(1)x|)+)+$()/.
55    12. In an EBCDIC environment, pcretest was mishandling the escape sequences
56        \a and \e in test subject lines.
58    13. In an EBCDIC environment, \a in a pattern was converted to the ASCII
59        instead of the EBCDIC value.
61    14. The handling of \c in an EBCDIC environment has been revised so that it is
62        now compatible with the specification in Perl's perlebcdic page.
64    15. The EBCDIC character 0x41 is a non-breaking space, equivalent to 0xa0 in
65        ASCII/Unicode. This has now been added to the list of characters that are
66        recognized as white space in EBCDIC.
68    16. When PCRE was compiled without UCP support, the use of \p and \P gave an
69        error (correctly) when used outside a class, but did not give an error
70        within a class.
72    17. \h within a class was incorrectly compiled in EBCDIC environments.
74    18. A pattern with an unmatched closing parenthesis that contained a backward
75        assertion which itself contained a forward reference caused buffer
76        overflow. And example pattern is: /(?=di(?<=(?1))|(?=(.))))/.
78    19. JIT should return with error when the compiled pattern requires more stack
79        space than the maximum.
81    20. A possessively repeated conditional group that could match an empty string,
82        for example, /(?(R))*+/, was incorrectly compiled.
84    21. Fix infinite recursion in the JIT compiler when certain patterns such as
85        /(?:|a|){100}x/ are analysed.
87    22. Some patterns with character classes involving [: and \\ were incorrectly
88        compiled and could cause reading from uninitialized memory or an incorrect
89        error diagnosis.
91    16. Pathological patterns containing many nested occurrences of [: caused
92        pcre_compile() to run for a very long time.
94    17. A conditional group with only one branch has an implicit empty alternative
95        branch and must therefore be treated as potentially matching an empty
96        string.
99    Version 8.37 28-April-2015
100    --------------------------
102    1.  When an (*ACCEPT) is triggered inside capturing parentheses, it arranges
103        for those parentheses to be closed with whatever has been captured so far.
104        However, it was failing to mark any other groups between the hightest
105        capture so far and the currrent group as "unset". Thus, the ovector for
106        those groups contained whatever was previously there. An example is the
107        pattern /(x)|((*ACCEPT))/ when matched against "abcd".
109    2.  If an assertion condition was quantified with a minimum of zero (an odd
110        thing to do, but it happened), SIGSEGV or other misbehaviour could occur.
112    3.  If a pattern in pcretest input had the P (POSIX) modifier followed by an
113        unrecognized modifier, a crash could occur.
115    4.  An attempt to do global matching in pcretest with a zero-length ovector
116        caused a crash.
118    5.  Fixed a memory leak during matching that could occur for a subpattern
119        subroutine call (recursive or otherwise) if the number of captured groups
120        that had to be saved was greater than ten.
122    6.  Catch a bad opcode during auto-possessification after compiling a bad UTF
123        string with NO_UTF_CHECK. This is a tidyup, not a bug fix, as passing bad
124        UTF with NO_UTF_CHECK is documented as having an undefined outcome.
126    7.  A UTF pattern containing a "not" match of a non-ASCII character and a
127        subroutine reference could loop at compile time. Example: /[^\xff]((?1))/.
129    8. When a pattern is compiled, it remembers the highest back reference so that
130       when matching, if the ovector is too small, extra memory can be obtained to
131       use instead. A conditional subpattern whose condition is a check on a
132       capture having happened, such as, for example in the pattern
133       /^(?:(a)|b)(?(1)A|B)/, is another kind of back reference, but it was not
134       setting the highest backreference number. This mattered only if pcre_exec()
135       was called with an ovector that was too small to hold the capture, and there
136       was no other kind of back reference (a situation which is probably quite
137       rare). The effect of the bug was that the condition was always treated as
138       FALSE when the capture could not be consulted, leading to a incorrect
139       behaviour by pcre_exec(). This bug has been fixed.
141    9. A reference to a duplicated named group (either a back reference or a test
142       for being set in a conditional) that occurred in a part of the pattern where
143       PCRE_DUPNAMES was not set caused the amount of memory needed for the pattern
144       to be incorrectly calculated, leading to overwriting.
146    10. A mutually recursive set of back references such as (\2)(\1) caused a
147        segfault at study time (while trying to find the minimum matching length).
148        The infinite loop is now broken (with the minimum length unset, that is,
149        zero).
151    11. If an assertion that was used as a condition was quantified with a minimum
152        of zero, matching went wrong. In particular, if the whole group had
153        unlimited repetition and could match an empty string, a segfault was
154        likely. The pattern (?(?=0)?)+ is an example that caused this. Perl allows
155        assertions to be quantified, but not if they are being used as conditions,
156        so the above pattern is faulted by Perl. PCRE has now been changed so that
157        it also rejects such patterns.
159    12. A possessive capturing group such as (a)*+ with a minimum repeat of zero
160        failed to allow the zero-repeat case if pcre2_exec() was called with an
161        ovector too small to capture the group.
163    13. Fixed two bugs in pcretest that were discovered by fuzzing and reported by
164        Red Hat Product Security:
166        (a) A crash if /K and /F were both set with the option to save the compiled
167        pattern.
169        (b) Another crash if the option to print captured substrings in a callout
170        was combined with setting a null ovector, for example \O\C+ as a subject
171        string.
173    14. A pattern such as "((?2){0,1999}())?", which has a group containing a
174        forward reference repeated a large (but limited) number of times within a
175        repeated outer group that has a zero minimum quantifier, caused incorrect
176        code to be compiled, leading to the error "internal error:
177        previously-checked referenced subpattern not found" when an incorrect
178        memory address was read. This bug was reported as "heap overflow",
179        discovered by Kai Lu of Fortinet's FortiGuard Labs and given the CVE number
180        CVE-2015-2325.
182    23. A pattern such as "((?+1)(\1))/" containing a forward reference subroutine
183        call within a group that also contained a recursive back reference caused
184        incorrect code to be compiled. This bug was reported as "heap overflow",
185        discovered by Kai Lu of Fortinet's FortiGuard Labs, and given the CVE
186        number CVE-2015-2326.
188    24. Computing the size of the JIT read-only data in advance has been a source
189        of various issues, and new ones are still appear unfortunately. To fix
190        existing and future issues, size computation is eliminated from the code,
191        and replaced by on-demand memory allocation.
193    25. A pattern such as /(?i)[A-`]/, where characters in the other case are
194        adjacent to the end of the range, and the range contained characters with
195        more than one other case, caused incorrect behaviour when compiled in UTF
196        mode. In that example, the range a-j was left out of the class.
198    26. Fix JIT compilation of conditional blocks, which assertion
199        is converted to (*FAIL). E.g: /(?(?!))/.
201    27. The pattern /(?(?!)^)/ caused references to random memory. This bug was
202        discovered by the LLVM fuzzer.
204    28. The assertion (?!) is optimized to (*FAIL). This was not handled correctly
205        when this assertion was used as a condition, for example (?(?!)a|b). In
206        pcre2_match() it worked by luck; in pcre2_dfa_match() it gave an incorrect
207        error about an unsupported item.
209    29. For some types of pattern, for example /Z*(|d*){216}/, the auto-
210        possessification code could take exponential time to complete. A recursion
211        depth limit of 1000 has been imposed to limit the resources used by this
212        optimization.
214    30. A pattern such as /(*UTF)[\S\V\H]/, which contains a negated special class
215        such as \S in non-UCP mode, explicit wide characters (> 255) can be ignored
216        because \S ensures they are all in the class. The code for doing this was
217        interacting badly with the code for computing the amount of space needed to
218        compile the pattern, leading to a buffer overflow. This bug was discovered
219        by the LLVM fuzzer.
221    31. A pattern such as /((?2)+)((?1))/ which has mutual recursion nested inside
222        other kinds of group caused stack overflow at compile time. This bug was
223        discovered by the LLVM fuzzer.
225    32. A pattern such as /(?1)(?#?'){8}(a)/ which had a parenthesized comment
226        between a subroutine call and its quantifier was incorrectly compiled,
227        leading to buffer overflow or other errors. This bug was discovered by the
228        LLVM fuzzer.
230    33. The illegal pattern /(?(?<E>.*!.*)?)/ was not being diagnosed as missing an
231        assertion after (?(. The code was failing to check the character after
232        (?(?< for the ! or = that would indicate a lookbehind assertion. This bug
233        was discovered by the LLVM fuzzer.
235    34. A pattern such as /X((?2)()*+){2}+/ which has a possessive quantifier with
236        a fixed maximum following a group that contains a subroutine reference was
237        incorrectly compiled and could trigger buffer overflow. This bug was
238        discovered by the LLVM fuzzer.
240    35. A mutual recursion within a lookbehind assertion such as (?<=((?2))((?1)))
241        caused a stack overflow instead of the diagnosis of a non-fixed length
242        lookbehind assertion. This bug was discovered by the LLVM fuzzer.
244    36. The use of \K in a positive lookbehind assertion in a non-anchored pattern
245        (e.g. /(?<=\Ka)/) could make pcregrep loop.
247    37. There was a similar problem to 36 in pcretest for global matches.
249    38. If a greedy quantified \X was preceded by \C in UTF mode (e.g. \C\X*),
250        and a subsequent item in the pattern caused a non-match, backtracking over
251        the repeated \X did not stop, but carried on past the start of the subject,
252        causing reference to random memory and/or a segfault. There were also some
253        other cases where backtracking after \C could crash. This set of bugs was
254        discovered by the LLVM fuzzer.
256    39. The function for finding the minimum length of a matching string could take
257        a very long time if mutual recursion was present many times in a pattern,
258        for example, /((?2){73}(?2))((?1))/. A better mutual recursion detection
259        method has been implemented. This infelicity was discovered by the LLVM
260        fuzzer.
262    40. Static linking against the PCRE library using the pkg-config module was
263        failing on missing pthread symbols.
266    Version 8.36 26-September-2014
267    ------------------------------
269    1.  Got rid of some compiler warnings in the C++ modules that were shown up by
270        -Wmissing-field-initializers and -Wunused-parameter.
272    2.  The tests for quantifiers being too big (greater than 65535) were being
273        applied after reading the number, and stupidly assuming that integer
274        overflow would give a negative number. The tests are now applied as the
275        numbers are read.
277    3.  Tidy code in pcre_exec.c where two branches that used to be different are
278        now the same.
280    4.  The JIT compiler did not generate match limit checks for certain
281        bracketed expressions with quantifiers. This may lead to exponential
282        backtracking, instead of returning with PCRE_ERROR_MATCHLIMIT. This
283        issue should be resolved now.
285    5.  Fixed an issue, which occures when nested alternatives are optimized
286        with table jumps.
288    6.  Inserted two casts and changed some ints to size_t in the light of some
289        reported 64-bit compiler warnings (Bugzilla 1477).
291    7.  Fixed a bug concerned with zero-minimum possessive groups that could match
292        an empty string, which sometimes were behaving incorrectly in the
293        interpreter (though correctly in the JIT matcher). This pcretest input is
294        an example:
296          '\A(?:[^"]++|"(?:[^"]*+|"")*+")++'
299        the interpreter was reporting a match of 'NON QUOTED ' only, whereas the
300        JIT matcher and Perl both matched 'NON QUOTED "QUOT""ED" AFTER '. The test
301        for an empty string was breaking the inner loop and carrying on at a lower
302        level, when possessive repeated groups should always return to a higher
303        level as they have no backtrack points in them. The empty string test now
304        occurs at the outer level.
306    8.  Fixed a bug that was incorrectly auto-possessifying \w+ in the pattern
307        ^\w+(?>\s*)(?<=\w) which caused it not to match "test test".
309    9.  Give a compile-time error for \o{} (as Perl does) and for \x{} (which Perl
310        doesn't).
312    10. Change 8.34/15 introduced a bug that caused the amount of memory needed
313        to hold a pattern to be incorrectly computed (too small) when there were
314        named back references to duplicated names. This could cause "internal
315        error: code overflow" or "double free or corruption" or other memory
316        handling errors.
318    11. When named subpatterns had the same prefixes, back references could be
319        confused. For example, in this pattern:
321          /(?P<Name>a)?(?P<Name2>b)?(?(<Name>)c|d)*l/
323        the reference to 'Name' was incorrectly treated as a reference to a
324        duplicate name.
326    12. A pattern such as /^s?c/mi8 where the optional character has more than
327        one "other case" was incorrectly compiled such that it would only try to
328        match starting at "c".
330    13. When a pattern starting with \s was studied, VT was not included in the
331        list of possible starting characters; this should have been part of the
332        8.34/18 patch.
334    14. If a character class started [\Qx]... where x is any character, the class
335        was incorrectly terminated at the ].
337    15. If a pattern that started with a caseless match for a character with more
338        than one "other case" was studied, PCRE did not set up the starting code
339        unit bit map for the list of possible characters. Now it does. This is an
340        optimization improvement, not a bug fix.
342    16. The Unicode data tables have been updated to Unicode 7.0.0.
344    17. Fixed a number of memory leaks in pcregrep.
346    18. Avoid a compiler warning (from some compilers) for a function call with
347        a cast that removes "const" from an lvalue by using an intermediate
348        variable (to which the compiler does not object).
350    19. Incorrect code was compiled if a group that contained an internal recursive
351        back reference was optional (had quantifier with a minimum of zero). This
352        example compiled incorrect code: /(((a\2)|(a*)\g<-1>))*/ and other examples
353        caused segmentation faults because of stack overflows at compile time.
355    20. A pattern such as /((?(R)a|(?1)))+/, which contains a recursion within a
356        group that is quantified with an indefinite repeat, caused a compile-time
357        loop which used up all the system stack and provoked a segmentation fault.
358        This was not the same bug as 19 above.
360    21. Add PCRECPP_EXP_DECL declaration to operator<< in pcre_stringpiece.h.
361        Patch by Mike Frysinger.
364    Version 8.35 04-April-2014
365    --------------------------
367    1.  A new flag is set, when property checks are present in an XCLASS.
368        When this flag is not set, PCRE can perform certain optimizations
369        such as studying these XCLASS-es.
371    2.  The auto-possessification of character sets were improved: a normal
372        and an extended character set can be compared now. Furthermore
373        the JIT compiler optimizes more character set checks.
375    3.  Got rid of some compiler warnings for potentially uninitialized variables
376        that show up only when compiled with -O2.
378    4.  A pattern such as (?=ab\K) that uses \K in an assertion can set the start
379        of a match later then the end of the match. The pcretest program was not
380        handling the case sensibly - it was outputting from the start to the next
381        binary zero. It now reports this situation in a message, and outputs the
382        text from the end to the start.
384    5.  Fast forward search is improved in JIT. Instead of the first three
385        characters, any three characters with fixed position can be searched.
386        Search order: first, last, middle.
388    6.  Improve character range checks in JIT. Characters are read by an inprecise
389        function now, which returns with an unknown value if the character code is
390        above a certain threshold (e.g: 256). The only limitation is that the value
391        must be bigger than the threshold as well. This function is useful when
392        the characters above the threshold are handled in the same way.
394    7.  The macros whose names start with RAWUCHAR are placeholders for a future
395        mode in which only the bottom 21 bits of 32-bit data items are used. To
396        make this more memorable for those maintaining the code, the names have
397        been changed to start with UCHAR21, and an extensive comment has been added
398        to their definition.
400    8.  Add missing (new) files sljitNativeTILEGX.c and sljitNativeTILEGX-encoder.c
401        to the export list in Makefile.am (they were accidentally omitted from the
402        8.34 tarball).
404    9.  The informational output from pcretest used the phrase "starting byte set"
405        which is inappropriate for the 16-bit and 32-bit libraries. As the output
406        for "first char" and "need char" really means "non-UTF-char", I've changed
407        "byte" to "char", and slightly reworded the output. The documentation about
408        these values has also been (I hope) clarified.
410    10. Another JIT related optimization: use table jumps for selecting the correct
411        backtracking path, when more than four alternatives are present inside a
412        bracket.
414    11. Empty match is not possible, when the minimum length is greater than zero,
415        and there is no \K in the pattern. JIT should avoid empty match checks in
416        such cases.
418    12. In a caseless character class with UCP support, when a character with more
419        than one alternative case was not the first character of a range, not all
420        the alternative cases were added to the class. For example, s and \x{17f}
421        are both alternative cases for S: the class [RST] was handled correctly,
422        but [R-T] was not.
424    13. The configure.ac file always checked for pthread support when JIT was
425        enabled. This is not used in Windows, so I have put this test inside a
426        check for the presence of windows.h (which was already tested for).
428    14. Improve pattern prefix search by a simplified Boyer-Moore algorithm in JIT.
429        The algorithm provides a way to skip certain starting offsets, and usually
430        faster than linear prefix searches.
432    15. Change 13 for 8.20 updated RunTest to check for the 'fr' locale as well
433        as for 'fr_FR' and 'french'. For some reason, however, it then used the
434        Windows-specific input and output files, which have 'french' screwed in.
435        So this could never have worked. One of the problems with locales is that
436        they aren't always the same. I have now updated RunTest so that it checks
437        the output of the locale test (test 3) against three different output
438        files, and it allows the test to pass if any one of them matches. With luck
439        this should make the test pass on some versions of Solaris where it was
440        failing. Because of the uncertainty, the script did not used to stop if
441        test 3 failed; it now does. If further versions of a French locale ever
442        come to light, they can now easily be added.
444    16. If --with-pcregrep-bufsize was given a non-integer value such as "50K",
445        there was a message during ./configure, but it did not stop. This now
446        provokes an error. The invalid example in README has been corrected.
447        If a value less than the minimum is given, the minimum value has always
448        been used, but now a warning is given.
450    17. If --enable-bsr-anycrlf was set, the special 16/32-bit test failed. This
451        was a bug in the test system, which is now fixed. Also, the list of various
452        configurations that are tested for each release did not have one with both
453        16/32 bits and --enable-bar-anycrlf. It now does.
455    18. pcretest was missing "-C bsr" for displaying the \R default setting.
457    19. Little endian PowerPC systems are supported now by the JIT compiler.
459    20. The fast forward newline mechanism could enter to an infinite loop on
460        certain invalid UTF-8 input. Although we don't support these cases
461        this issue can be fixed by a performance optimization.
463    21. Change 33 of 8.34 is not sufficient to ensure stack safety because it does
464        not take account if existing stack usage. There is now a new global
465        variable called pcre_stack_guard that can be set to point to an external
466        function to check stack availability. It is called at the start of
467        processing every parenthesized group.
469    22. A typo in the code meant that in ungreedy mode the max/min qualifier
470        behaved like a min-possessive qualifier, and, for example, /a{1,3}b/U did
471        not match "ab".
473    23. When UTF was disabled, the JIT program reported some incorrect compile
474        errors. These messages are silenced now.
476    24. Experimental support for ARM-64 and MIPS-64 has been added to the JIT
477        compiler.
479    25. Change all the temporary files used in RunGrepTest to be different to those
480        used by RunTest so that the tests can be run simultaneously, for example by
481        "make -j check".
484    Version 8.34 15-December-2013
485    -----------------------------
487    1.  Add pcre[16|32]_jit_free_unused_memory to forcibly free unused JIT
488        executable memory. Patch inspired by Carsten Klein.
490    2.  ./configure --enable-coverage defined SUPPORT_GCOV in config.h, although
491        this macro is never tested and has no effect, because the work to support
492        coverage involves only compiling and linking options and special targets in
493        the Makefile. The comment in config.h implied that defining the macro would
494        enable coverage support, which is totally false. There was also support for
495        setting this macro in the CMake files (my fault, I just copied it from
496        configure). SUPPORT_GCOV has now been removed.
498    3.  Make a small performance improvement in strlen16() and strlen32() in
499        pcretest.
501    4.  Change 36 for 8.33 left some unreachable statements in pcre_exec.c,
502        detected by the Solaris compiler (gcc doesn't seem to be able to diagnose
503        these cases). There was also one in pcretest.c.
505    5.  Cleaned up a "may be uninitialized" compiler warning in pcre_exec.c.
507    6.  In UTF mode, the code for checking whether a group could match an empty
508        string (which is used for indefinitely repeated groups to allow for
509        breaking an infinite loop) was broken when the group contained a repeated
510        negated single-character class with a character that occupied more than one
511        data item and had a minimum repetition of zero (for example, [^\x{100}]* in
512        UTF-8 mode). The effect was undefined: the group might or might not be
513        deemed as matching an empty string, or the program might have crashed.
515    7.  The code for checking whether a group could match an empty string was not
516        recognizing that \h, \H, \v, \V, and \R must match a character.
518    8.  Implemented PCRE_INFO_MATCH_EMPTY, which yields 1 if the pattern can match
519        an empty string. If it can, pcretest shows this in its information output.
521    9.  Fixed two related bugs that applied to Unicode extended grapheme clusters
522        that were repeated with a maximizing qualifier (e.g. \X* or \X{2,5}) when
523        matched by pcre_exec() without using JIT:
525        (a) If the rest of the pattern did not match after a maximal run of
526            grapheme clusters, the code for backing up to try with fewer of them
527            did not always back up over a full grapheme when characters that do not
528            have the modifier quality were involved, e.g. Hangul syllables.
530        (b) If the match point in a subject started with modifier character, and
531            there was no match, the code could incorrectly back up beyond the match
532            point, and potentially beyond the first character in the subject,
533            leading to a segfault or an incorrect match result.
535    10. A conditional group with an assertion condition could lead to PCRE
536        recording an incorrect first data item for a match if no other first data
537        item was recorded. For example, the pattern (?(?=ab)ab) recorded "a" as a
538        first data item, and therefore matched "ca" after "c" instead of at the
539        start.
541    11. Change 40 for 8.33 (allowing pcregrep to find empty strings) showed up a
542        bug that caused the command "echo a | ./pcregrep -M '|a'" to loop.
544    12. The source of pcregrep now includes z/OS-specific code so that it can be
545        compiled for z/OS as part of the special z/OS distribution.
547    13. Added the -T and -TM options to pcretest.
549    14. The code in pcre_compile.c for creating the table of named capturing groups
550        has been refactored. Instead of creating the table dynamically during the
551        actual compiling pass, the information is remembered during the pre-compile
552        pass (on the stack unless there are more than 20 named groups, in which
553        case malloc() is used) and the whole table is created before the actual
554        compile happens. This has simplified the code (it is now nearly 150 lines
555        shorter) and prepared the way for better handling of references to groups
556        with duplicate names.
558    15. A back reference to a named subpattern when there is more than one of the
559        same name now checks them in the order in which they appear in the pattern.
560        The first one that is set is used for the reference. Previously only the
561        first one was inspected. This change makes PCRE more compatible with Perl.
563    16. Unicode character properties were updated from Unicode 6.3.0.
565    17. The compile-time code for auto-possessification has been refactored, based
566        on a patch by Zoltan Herczeg. It now happens after instead of during
567        compilation. The code is cleaner, and more cases are handled. The option
568        PCRE_NO_AUTO_POSSESS is added for testing purposes, and the -O and /O
569        options in pcretest are provided to set it. It can also be set by
570        (*NO_AUTO_POSSESS) at the start of a pattern.
572    18. The character VT has been added to the default ("C" locale) set of
573        characters that match \s and are generally treated as white space,
574        following this same change in Perl 5.18. There is now no difference between
575        "Perl space" and "POSIX space". Whether VT is treated as white space in
576        other locales depends on the locale.
578    19. The code for checking named groups as conditions, either for being set or
579        for being recursed, has been refactored (this is related to 14 and 15
580        above). Processing unduplicated named groups should now be as fast at
581        numerical groups, and processing duplicated groups should be faster than
582        before.
584    20. Two patches to the CMake build system, by Alexander Barkov:
586          (1) Replace the "source" command by "." in CMakeLists.txt because
587              "source" is a bash-ism.
589          (2) Add missing HAVE_STDINT_H and HAVE_INTTYPES_H to config-cmake.h.in;
590              without these the CMake build does not work on Solaris.
592    21. Perl has changed its handling of \8 and \9. If there is no previously
593        encountered capturing group of those numbers, they are treated as the
594        literal characters 8 and 9 instead of a binary zero followed by the
595        literals. PCRE now does the same.
597    22. Following Perl, added \o{} to specify codepoints in octal, making it
598        possible to specify values greater than 0777 and also making them
599        unambiguous.
601    23. Perl now gives an error for missing closing braces after \x{... instead of
602        treating the string as literal. PCRE now does the same.
604    24. RunTest used to grumble if an inappropriate test was selected explicitly,
605        but just skip it when running all tests. This make it awkward to run ranges
606        of tests when one of them was inappropriate. Now it just skips any
607        inappropriate tests, as it always did when running all tests.
609    25. If PCRE_AUTO_CALLOUT and PCRE_UCP were set for a pattern that contained
610        character types such as \d or \w, too many callouts were inserted, and the
611        data that they returned was rubbish.
613    26. In UCP mode, \s was not matching two of the characters that Perl matches,
614        namely NEL (U+0085) and MONGOLIAN VOWEL SEPARATOR (U+180E), though they
615        were matched by \h. The code has now been refactored so that the lists of
616        the horizontal and vertical whitespace characters used for \h and \v (which
617        are defined only in one place) are now also used for \s.
619    27. Add JIT support for the 64 bit TileGX architecture.
620        Patch by Jiong Wang (Tilera Corporation).
622    28. Possessive quantifiers for classes (both explicit and automatically
623        generated) now use special opcodes instead of wrapping in ONCE brackets.
625    29. Whereas an item such as A{4}+ ignored the possessivenes of the quantifier
626        (because it's meaningless), this was not happening when PCRE_CASELESS was
627        set. Not wrong, but inefficient.
629    30. Updated perltest.pl to add /u (force Unicode mode) when /W (use Unicode
630        properties for \w, \d, etc) is present in a test regex. Otherwise if the
631        test contains no characters greater than 255, Perl doesn't realise it
632        should be using Unicode semantics.
634    31. Upgraded the handling of the POSIX classes [:graph:], [:print:], and
635        [:punct:] when PCRE_UCP is set so as to include the same characters as Perl
636        does in Unicode mode.
638    32. Added the "forbid" facility to pcretest so that putting tests into the
639        wrong test files can sometimes be quickly detected.
641    33. There is now a limit (default 250) on the depth of nesting of parentheses.
642        This limit is imposed to control the amount of system stack used at compile
643        time. It can be changed at build time by --with-parens-nest-limit=xxx or
644        the equivalent in CMake.
646    34. Character classes such as [A-\d] or [a-[:digit:]] now cause compile-time
647        errors. Perl warns for these when in warning mode, but PCRE has no facility
648        for giving warnings.
650    35. Change 34 for 8.13 allowed quantifiers on assertions, because Perl does.
651        However, this was not working for (?!) because it is optimized to (*FAIL),
652        for which PCRE does not allow quantifiers. The optimization is now disabled
653        when a quantifier follows (?!). I can't see any use for this, but it makes
654        things uniform.
656    36. Perl no longer allows group names to start with digits, so I have made this
657        change also in PCRE. It simplifies the code a bit.
659    37. In extended mode, Perl ignores spaces before a + that indicates a
660        possessive quantifier. PCRE allowed a space before the quantifier, but not
661        before the possessive +. It now does.
663    38. The use of \K (reset reported match start) within a repeated possessive
664        group such as (a\Kb)*+ was not working.
666    40. Document that the same character tables must be used at compile time and
667        run time, and that the facility to pass tables to pcre_exec() and
668        pcre_dfa_exec() is for use only with saved/restored patterns.
670    41. Applied Jeff Trawick's patch CMakeLists.txt, which "provides two new
671        features for Builds with MSVC:
673        1. Support pcre.rc and/or pcreposix.rc (as is already done for MinGW
674           builds). The .rc files can be used to set FileDescription and many other
675           attributes.
677        2. Add an option (-DINSTALL_MSVC_PDB) to enable installation of .pdb files.
678           This allows higher-level build scripts which want .pdb files to avoid
679           hard-coding the exact files needed."
681    42. Added support for [[:<:]] and [[:>:]] as used in the BSD POSIX library to
682        mean "start of word" and "end of word", respectively, as a transition aid.
684    43. A minimizing repeat of a class containing codepoints greater than 255 in
685        non-UTF 16-bit or 32-bit modes caused an internal error when PCRE was
686        compiled to use the heap for recursion.
688    44. Got rid of some compiler warnings for unused variables when UTF but not UCP
689        is configured.
692    Version 8.33 28-May-2013
693    ------------------------
695    1.  Added 'U' to some constants that are compared to unsigned integers, to
696        avoid compiler signed/unsigned warnings. Added (int) casts to unsigned
697        variables that are added to signed variables, to ensure the result is
698        signed and can be negated.
700    2.  Applied patch by Daniel Richard G for quashing MSVC warnings to the
701        CMake config files.
703    3.  Revise the creation of config.h.generic so that all boolean macros are
704        #undefined, whereas non-boolean macros are #ifndef/#endif-ed. This makes
705        overriding via -D on the command line possible.
707    4.  Changing the definition of the variable "op" in pcre_exec.c from pcre_uchar
708        to unsigned int is reported to make a quite noticeable speed difference in
709        a specific Windows environment. Testing on Linux did also appear to show
710        some benefit (and it is clearly not harmful). Also fixed the definition of
711        Xop which should be unsigned.
713    5.  Related to (4), changing the definition of the intermediate variable cc
714        in repeated character loops from pcre_uchar to pcre_uint32 also gave speed
715        improvements.
717    6.  Fix forward search in JIT when link size is 3 or greater. Also removed some
718        unnecessary spaces.
720    7.  Adjust autogen.sh and configure.ac to lose warnings given by automake 1.12
721        and later.
723    8.  Fix two buffer over read issues in 16 and 32 bit modes. Affects JIT only.
725    9.  Optimizing fast_forward_start_bits in JIT.
727    10. Adding support for callouts in JIT, and fixing some issues revealed
728        during this work. Namely:
730        (a) Unoptimized capturing brackets incorrectly reset on backtrack.
732        (b) Minimum length was not checked before the matching is started.
734    11. The value of capture_last that is passed to callouts was incorrect in some
735        cases when there was a capture on one path that was subsequently abandoned
736        after a backtrack. Also, the capture_last value is now reset after a
737        recursion, since all captures are also reset in this case.
739    12. The interpreter no longer returns the "too many substrings" error in the
740        case when an overflowing capture is in a branch that is subsequently
741        abandoned after a backtrack.
743    13. In the pathological case when an offset vector of size 2 is used, pcretest
744        now prints out the matched string after a yield of 0 or 1.
746    14. Inlining subpatterns in recursions, when certain conditions are fulfilled.
747        Only supported by the JIT compiler at the moment.
749    15. JIT compiler now supports 32 bit Macs thanks to Lawrence Velazquez.
751    16. Partial matches now set offsets[2] to the "bumpalong" value, that is, the
752        offset of the starting point of the matching process, provided the offsets
753        vector is large enough.
755    17. The \A escape now records a lookbehind value of 1, though its execution
756        does not actually inspect the previous character. This is to ensure that,
757        in partial multi-segment matching, at least one character from the old
758        segment is retained when a new segment is processed. Otherwise, if there
759        are no lookbehinds in the pattern, \A might match incorrectly at the start
760        of a new segment.
762    18. Added some #ifdef __VMS code into pcretest.c to help VMS implementations.
764    19. Redefined some pcre_uchar variables in pcre_exec.c as pcre_uint32; this
765        gives some modest performance improvement in 8-bit mode.
767    20. Added the PCRE-specific property \p{Xuc} for matching characters that can
768        be expressed in certain programming languages using Universal Character
769        Names.
771    21. Unicode validation has been updated in the light of Unicode Corrigendum #9,
772        which points out that "non characters" are not "characters that may not
773        appear in Unicode strings" but rather "characters that are reserved for
774        internal use and have only local meaning".
776    22. When a pattern was compiled with automatic callouts (PCRE_AUTO_CALLOUT) and
777        there was a conditional group that depended on an assertion, if the
778        assertion was false, the callout that immediately followed the alternation
779        in the condition was skipped when pcre_exec() was used for matching.
781    23. Allow an explicit callout to be inserted before an assertion that is the
782        condition for a conditional group, for compatibility with automatic
783        callouts, which always insert a callout at this point.
785    24. In 8.31, (*COMMIT) was confined to within a recursive subpattern. Perl also
786        confines (*SKIP) and (*PRUNE) in the same way, and this has now been done.
788    25. (*PRUNE) is now supported by the JIT compiler.
790    26. Fix infinite loop when /(?<=(*SKIP)ac)a/ is matched against aa.
792    27. Fix the case where there are two or more SKIPs with arguments that may be
793        ignored.
795    28. (*SKIP) is now supported by the JIT compiler.
797    29. (*THEN) is now supported by the JIT compiler.
799    30. Update RunTest with additional test selector options.
801    31. The way PCRE handles backtracking verbs has been changed in two ways.
803        (1) Previously, in something like (*COMMIT)(*SKIP), COMMIT would override
804        SKIP. Now, PCRE acts on whichever backtracking verb is reached first by
805        backtracking. In some cases this makes it more Perl-compatible, but Perl's
806        rather obscure rules do not always do the same thing.
808        (2) Previously, backtracking verbs were confined within assertions. This is
809        no longer the case for positive assertions, except for (*ACCEPT). Again,
810        this sometimes improves Perl compatibility, and sometimes does not.
812    32. A number of tests that were in test 2 because Perl did things differently
813        have been moved to test 1, because either Perl or PCRE has changed, and
814        these tests are now compatible.
816    32. Backtracking control verbs are now handled in the same way in JIT and
817        interpreter.
819    33. An opening parenthesis in a MARK/PRUNE/SKIP/THEN name in a pattern that
820        contained a forward subroutine reference caused a compile error.
822    34. Auto-detect and optimize limited repetitions in JIT.
824    35. Implement PCRE_NEVER_UTF to lock out the use of UTF, in particular,
825        blocking (*UTF) etc.
827    36. In the interpreter, maximizing pattern repetitions for characters and
828        character types now use tail recursion, which reduces stack usage.
830    37. The value of the max lookbehind was not correctly preserved if a compiled
831        and saved regex was reloaded on a host of different endianness.
833    38. Implemented (*LIMIT_MATCH) and (*LIMIT_RECURSION). As part of the extension
834        of the compiled pattern block, expand the flags field from 16 to 32 bits
835        because it was almost full.
837    39. Try madvise first before posix_madvise.
839    40. Change 7 for PCRE 7.9 made it impossible for pcregrep to find empty lines
840        with a pattern such as ^$. It has taken 4 years for anybody to notice! The
841        original change locked out all matches of empty strings. This has been
842        changed so that one match of an empty string per line is recognized.
843        Subsequent searches on the same line (for colouring or for --only-matching,
844        for example) do not recognize empty strings.
846    41. Applied a user patch to fix a number of spelling mistakes in comments.
848    42. Data lines longer than 65536 caused pcretest to crash.
850    43. Clarified the data type for length and startoffset arguments for pcre_exec
851        and pcre_dfa_exec in the function-specific man pages, where they were
852        explicitly stated to be in bytes, never having been updated. I also added
853        some clarification to the pcreapi man page.
855    44. A call to pcre_dfa_exec() with an output vector size less than 2 caused
856        a segmentation fault.
859    Version 8.32 30-November-2012
860    -----------------------------
862    1.  Improved JIT compiler optimizations for first character search and single
863        character iterators.
865    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
866        Patch by Daniel Richard G.
868    3.  Single character iterator optimizations in the JIT compiler.
870    4.  Improved JIT compiler optimizations for character ranges.
872    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
873        Reported by Giuseppe D'Angelo.
875    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
876        of a line, was being set incorrectly in cases where .* appeared inside
877        atomic brackets at the start of a pattern, or where there was a subsequent
878        *PRUNE or *SKIP.
880    7.  Improved instruction cache flush for POWER/PowerPC.
881        Patch by Daniel Richard G.
883    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
884        grep:
886        (a) There is now no limit to the number of patterns to be matched.
888        (b) An error is given if a pattern is too long.
890        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
891            are now supported.
893        (d) --exclude-from and --include-from (multiple use) have been added.
895        (e) Exclusions and inclusions now apply to all files and directories, not
896            just to those obtained from scanning a directory recursively.
898        (f) Multiple uses of -f and --file-list are now supported.
900        (g) In a Windows environment, the default for -d has been changed from
901            "read" (the GNU grep default) to "skip", because otherwise the presence
902            of a directory in the file list provokes an error.
904        (h) The documentation has been revised and clarified in places.
906    9.  Improve the matching speed of capturing brackets.
908    10. Changed the meaning of \X so that it now matches a Unicode extended
909        grapheme cluster.
911    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
912        out POSIX threads when JIT support is configured.
914    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
916    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
917        structure could go wrong in environments where size_t is not the same size
918        as int.
920    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
921        set.
923    15. The EBCDIC support had decayed; later updates to the code had included
924        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
925        general tidy up of EBCDIC-related issues, and the documentation was also
926        not quite right. There is now a test that can be run on ASCII systems to
927        check some of the EBCDIC-related things (but is it not a full test).
929    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
930        in a small tidy to the code.
932    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
934    18. If the --only-matching (-o) option in pcregrep is specified multiple
935        times, each one causes appropriate output. For example, -o1 -o2 outputs the
936        substrings matched by the 1st and 2nd capturing parentheses. A separating
937        string can be specified by --om-separator (default empty).
939    19. Improving the first n character searches.
941    20. Turn case lists for horizontal and vertical white space into macros so that
942        they are defined only once.
944    21. This set of changes together give more compatible Unicode case-folding
945        behaviour for characters that have more than one other case when UCP
946        support is available.
948        (a) The Unicode property table now has offsets into a new table of sets of
949            three or more characters that are case-equivalent. The MultiStage2.py
950            script that generates these tables (the pcre_ucd.c file) now scans
951            CaseFolding.txt instead of UnicodeData.txt for character case
952            information.
954        (b) The code for adding characters or ranges of characters to a character
955            class has been abstracted into a generalized function that also handles
956            case-independence. In UTF-mode with UCP support, this uses the new data
957            to handle characters with more than one other case.
959        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
960            whose other case is greater than 256 are now correctly matched
961            caselessly. Previously, the high codepoint matched the low one, but not
962            vice versa.
964        (d) The processing of \h, \H, \v, and \ in character classes now makes use
965            of the new class addition function, using character lists defined as
966            macros alongside the case definitions of 20 above.
968        (e) Caseless back references now work with characters that have more than
969            one other case.
971        (f) General caseless matching of characters with more than one other case
972            is supported.
974    22. Unicode character properties were updated from Unicode 6.2.0
976    23. Improved CMake support under Windows. Patch by Daniel Richard G.
978    24. Add support for 32-bit character strings, and UTF-32
980    25. Major JIT compiler update (code refactoring and bugfixing).
981        Experimental Sparc 32 support is added.
983    26. Applied a modified version of Daniel Richard G's patch to create
984        pcre.h.generic and config.h.generic by "make" instead of in the
985        PrepareRelease script.
987    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
988        pcre_compile.c when checking for a zero character.
990    28. Introducing a native interface for JIT. Through this interface, the compiled
991        machine code can be directly executed. The purpose of this interface is to
992        provide fast pattern matching, so several sanity checks are not performed.
993        However, feature tests are still performed. The new interface provides
994        1.4x speedup compared to the old one.
996    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
997        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
998        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
1000    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
1001        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
1002        the "old" RFC 2279). Instead, it ended up passing a negative length to
1003        pcre_exec().
1005    31. Add support for GCC's visibility feature to hide internal functions.
1007    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
1008        "unknown -C option" after outputting 0 or 1.
1010    33. There is now support for generating a code coverage report for the test
1011        suite in environments where gcc is the compiler and lcov is installed. This
1012        is mainly for the benefit of the developers.
1014    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
1015        unaddressable using valgrind annotations, allowing valgrind to detect
1016        invalid memory accesses. This is mainly for the benefit of the developers.
1018    25. (*UTF) can now be used to start a pattern in any of the three libraries.
1020    26. Give configure error if --enable-cpp but no C++ compiler found.
1023    Version 8.31 06-July-2012
1024    -------------------------
1026    1.  Fixing a wrong JIT test case and some compiler warnings.
1028    2.  Removed a bashism from the RunTest script.
1030    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
1031        to unsigned type, result still unsigned" that was given by an MS compiler
1032        on encountering the code "-sizeof(xxx)".
1034    4.  Partial matching support is added to the JIT compiler.
1036    5.  Fixed several bugs concerned with partial matching of items that consist
1037        of more than one character:
1039        (a) /^(..)\1/ did not partially match "aba" because checking references was
1040            done on an "all or nothing" basis. This also applied to repeated
1041            references.
1043        (b) \R did not give a hard partial match if \r was found at the end of the
1044            subject.
1046        (c) \X did not give a hard partial match after matching one or more
1047            characters at the end of the subject.
1049        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
1050            a partial match for the string "\r".
1052        (e) When newline was set to CRLF, the metacharacter "." did not recognize
1053            a partial match for a CR character at the end of the subject string.
1055    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
1056        running pcretest, the text "(JIT)" added to the output whenever JIT is
1057        actually used to run the match.
1059    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
1060        or /S+[+] with a digit between 1 and 7.
1062    8.  OP_NOT now supports any UTF character not just single-byte ones.
1064    9.  (*MARK) control verb is now supported by the JIT compiler.
1066    10. The command "./RunTest list" lists the available tests without actually
1067        running any of them. (Because I keep forgetting what they all are.)
1071    12. Applied a (slightly modified) user-supplied patch that improves performance
1072        when the heap is used for recursion (compiled with --disable-stack-for-
1073        recursion). Instead of malloc and free for each heap frame each time a
1074        logical recursion happens, frames are retained on a chain and re-used where
1075        possible. This sometimes gives as much as 30% improvement.
1077    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
1078        call.
1080    14. As documented, (*COMMIT) is now confined to within a positive assertion.
1082    15. It is now possible to link pcretest with libedit as an alternative to
1083        libreadline.
1085    16. (*COMMIT) control verb is now supported by the JIT compiler.
1087    17. The Unicode data tables have been updated to Unicode 6.1.0.
1089    18. Added --file-list option to pcregrep.
1091    19. Added binary file support to pcregrep, including the -a, --binary-files,
1092        -I, and --text options.
1094    20. The madvise function is renamed for posix_madvise for QNX compatibility
1095        reasons. Fixed by Giuseppe D'Angelo.
1097    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
1099    22. Changed the option for creating symbolic links for 16-bit man pages from
1100        -s to -sf so that re-installing does not cause issues.
1102    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
1104    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
1105        matches in certain environments (the workspace was not being correctly
1106        retained). Also added to pcre_dfa_exec() a simple plausibility check on
1107        some of the workspace data at the beginning of a restart.
1109    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
1110        was not doing so when it should - probably a typo introduced by SVN 528
1111        (change 8.10/14).
1113    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
1114        \w+ when the character tables indicated that \x{c4} was a word character.
1115        There were several related cases, all because the tests for doing a table
1116        lookup were testing for characters less than 127 instead of 255.
1118    27. If a pattern contains capturing parentheses that are not used in a match,
1119        their slots in the ovector are set to -1. For those that are higher than
1120        any matched groups, this happens at the end of processing. In the case when
1121        there were back references that the ovector was too small to contain
1122        (causing temporary malloc'd memory to be used during matching), and the
1123        highest capturing number was not used, memory off the end of the ovector
1124        was incorrectly being set to -1. (It was using the size of the temporary
1125        memory instead of the true size.)
1127    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
1128        ovector size, it uses memory at the end of the block that it has got.
1130    29. Check for an overlong MARK name and give an error at compile time. The
1131        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
1133    30. JIT compiler update.
1135    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
1136        Rill for the patch.
1138    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
1140    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
1142    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
1143        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
1144        was enabled.
1146    35. Improve JIT code generation for greedy plus quantifier.
1148    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
1149        1 to "aa" instead of to an empty string. The bug affected repeated groups
1150        that could potentially match an empty string.
1152    37. Optimizing single character iterators in JIT.
1154    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
1155        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
1156        codepoints that are too big for the mode are faulted, and in a UTF mode,
1157        disallowed codepoints are also faulted.
1159    39. If PCRE was compiled with UTF support, in three places in the DFA
1160        matcher there was code that should only have been obeyed in UTF mode, but
1161        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
1162        processing when bytes with values greater than 127 were present. In 16-bit
1163        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
1164        both cases the values are those that cannot be the first data item in a UTF
1165        character. The three items that might have provoked this were recursions,
1166        possessively repeated groups, and atomic groups.
1168    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
1169        and pcregrep, because some OS require shared objects to be explicitly
1170        passed to ld, causing the link step to fail if they are not.
1172    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
1173        mode, patterns that started with \h* or \R* might be incorrectly matched.
1176    Version 8.30 04-February-2012
1177    -----------------------------
1179    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
1180        name is defined in ctype.h.
1182    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
1183        only in quite long subpatterns.
1185    3.  Removed the function pcre_info(), which has been obsolete and deprecated
1186        since it was replaced by pcre_fullinfo() in February 2000.
1188    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
1189        match a (*MARK), and the match failed at the start of the subject, a
1190        reference to memory before the start of the subject could occur. This bug
1191        was introduced by fix 17 of release 8.21.
1193    5.  A reference to an unset group with zero minimum repetition was giving
1194        totally wrong answers (in non-JavaScript-compatibility mode). For example,
1195        /(another)?(\1?)test/ matched against "hello world test". This bug was
1196        introduced in release 8.13.
1198    6.  Add support for 16-bit character strings (a large amount of work involving
1199        many changes and refactorings).
1201    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
1202        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
1203        from a file.
1205    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
1206        rounding is not applied in this particular case).
1208    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
1209        if they appear, or are escaped, in patterns.
1211    10. Get rid of a number of -Wunused-but-set-variable warnings.
1213    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
1214        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
1215        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
1216        also returns the mark "x". This bug applied to capturing parentheses,
1217        non-capturing parentheses, and atomic parentheses. It also applied to some
1218        assertions.
1220    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
1221        information out of configure.ac instead of relying on pcre.h.generic, which
1222        is not stored in the repository.
1224    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
1225        -lreadline.
1227    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
1229    15. Applied Graycode's patch to put the top-level frame on the stack rather
1230        than the heap when not using the stack for recursion. This gives a
1231        performance improvement in many cases when recursion is not deep.
1233    16. Experimental code added to "pcretest -C" to output the stack frame size.
1236    Version 8.21 12-Dec-2011
1237    ------------------------
1239    1.  Updating the JIT compiler.
1241    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
1242        are added as well.
1244    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
1245        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
1246        calling _pcre_jit_exec. Some extra comments are added.
1248    4.  (*MARK) settings inside atomic groups that do not contain any capturing
1249        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
1250        was introduced by change 18 for 8.20.
1252    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
1253        ECMA-262 standard.
1255    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
1256        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
1257        This bug was probably introduced by change 9 of 8.13.
1259    7.  While fixing 6 above, I noticed that a number of other items were being
1260        incorrectly rejected as "not fixed length". This arose partly because newer
1261        opcodes had not been added to the fixed-length checking code. I have (a)
1262        corrected the bug and added tests for these items, and (b) arranged for an
1263        error to occur if an unknown opcode is encountered while checking for fixed
1264        length instead of just assuming "not fixed length". The items that were
1265        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
1266        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
1267        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
1269    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
1270        being incorrectly compiled and would have given unpredicatble results.
1272    9.  A possessively repeated subpattern with minimum repeat count greater than
1273        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
1274        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
1275        the first (A) could occur when it should not.
1277    10. Add a cast and remove a redundant test from the code.
1279    11. JIT should use pcre_malloc/pcre_free for allocation.
1281    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
1282        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
1283        is anything other than /usr, -L is still shown).
1285    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
1287    14. Perl does not support \N without a following name in a [] class; PCRE now
1288        also gives an error.
1290    15. If a forward reference was repeated with an upper limit of around 2000,
1291        it caused the error "internal error: overran compiling workspace". The
1292        maximum number of forward references (including repeats) was limited by the
1293        internal workspace, and dependent on the LINK_SIZE. The code has been
1294        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
1295        the default depends on LINK_SIZE. There is a new upper limit (for safety)
1296        of around 200,000 forward references. While doing this, I also speeded up
1297        the filling in of repeated forward references.
1299    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
1300        incorrectly expecting the subject to contain another "a" after the start.
1302    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
1303        in the match, the SKIP should be ignored. This was not happening; instead
1304        the SKIP was being treated as NOMATCH. For patterns such as
1305        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
1306        tested.
1308    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
1309        now much more compatible with Perl, in particular in cases where the result
1310        is a non-match for a non-anchored pattern. For example, if
1311        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
1312        "m", where previously it did not return a name. A side effect of this
1313        change is that for partial matches, the last encountered mark name is
1314        returned, as for non matches. A number of tests that were previously not
1315        Perl-compatible have been moved into the Perl-compatible test files. The
1316        refactoring has had the pleasing side effect of removing one argument from
1317        the match() function, thus reducing its stack requirements.
1319    19. If the /S+ option was used in pcretest to study a pattern using JIT,
1320        subsequent uses of /S (without +) incorrectly behaved like /S+.
1322    21. Retrieve executable code size support for the JIT compiler and fixing
1323        some warnings.
1325    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
1326        not work when the shorter character appeared right at the end of the
1327        subject string.
1329    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
1330        systems.
1332    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
1333        output it when the /M option is used in pcretest.
1335    25. The CheckMan script was not being included in the distribution. Also, added
1336        an explicit "perl" to run Perl scripts from the PrepareRelease script
1337        because this is reportedly needed in Windows.
1339    26. If study data was being save in a file and studying had not found a set of
1340        "starts with" bytes for the pattern, the data written to the file (though
1341        never used) was taken from uninitialized memory and so caused valgrind to
1342        complain.
1344    27. Updated RunTest.bat as provided by Sheri Pierce.
1346    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
1348    29. Computation of memory usage for the table of capturing group names was
1349        giving an unnecessarily large value.
1352    Version 8.20 21-Oct-2011
1353    ------------------------
1355    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
1356        a POSIX class. After further experiments with Perl, which convinced me that
1357        Perl has bugs and confusions, a closing square bracket is no longer allowed
1358        in a POSIX name. This bug also affected patterns with classes that started
1359        with full stops.
1361    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
1362        captured substring, but while checking the failing first alternative,
1363        substring 1 is temporarily captured. If the output vector supplied to
1364        pcre_exec() was not big enough for this capture, the yield of the function
1365        was still zero ("insufficient space for captured substrings"). This cannot
1366        be totally fixed without adding another stack variable, which seems a lot
1367        of expense for a edge case. However, I have improved the situation in cases
1368        such as /(a)(b)x|abc/ matched against "abc", where the return code
1369        indicates that fewer than the maximum number of slots in the ovector have
1370        been set.
1372    3.  Related to (2) above: when there are more back references in a pattern than
1373        slots in the output vector, pcre_exec() uses temporary memory during
1374        matching, and copies in the captures as far as possible afterwards. It was
1375        using the entire output vector, but this conflicts with the specification
1376        that only 2/3 is used for passing back captured substrings. Now it uses
1377        only the first 2/3, for compatibility. This is, of course, another edge
1378        case.
1380    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
1381        main code base, and can be used by building with --enable-jit. When this is
1382        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
1383        runtime --no-jit option is given.
1385    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
1386        ovector, the return from the function was zero, implying that there were
1387        other matches that did not fit. The correct "exactly full" value is now
1388        returned.
1390    6.  If a subpattern that was called recursively or as a subroutine contained
1391        (*PRUNE) or any other control that caused it to give a non-standard return,
1392        invalid errors such as "Error -26 (nested recursion at the same subject
1393        position)" or even infinite loops could occur.
1395    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
1396        computing the minimum length on reaching *ACCEPT, and so ended up with the
1397        wrong value of 1 rather than 0. Further investigation indicates that
1398        computing a minimum subject length in the presence of *ACCEPT is difficult
1399        (think back references, subroutine calls), and so I have changed the code
1400        so that no minimum is registered for a pattern that contains *ACCEPT.
1402    8.  If (*THEN) was present in the first (true) branch of a conditional group,
1403        it was not handled as intended. [But see 16 below.]
1405    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
1406        Sheri Pierce.
1408    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
1409        the first byte in a match must be "a".
1411    11. Change 17 for 8.13 increased the recursion depth for patterns like
1412        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
1413        pattern contains any instances of (*THEN). If it does not, the old
1414        optimizations are restored. It would be nice to do this on a per-group
1415        basis, but at the moment that is not feasible.
1417    12. In some environments, the output of pcretest -C is CRLF terminated. This
1418        broke RunTest's code that checks for the link size. A single white space
1419        character after the value is now allowed for.
1421    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
1422        For "fr", it uses the Windows-specific input and output files.
1424    14. If (*THEN) appeared in a group that was called recursively or as a
1425        subroutine, it did not work as intended. [But see next item.]
1427    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
1428        pattern fragments (but not containing any | characters). If A and B are
1429        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
1430        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
1431        D. In other words, Perl considers parentheses that do not contain any |
1432        characters to be part of a surrounding alternative, whereas PCRE was
1433        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
1434        differently. PCRE now behaves in the same way as Perl, except in the case
1435        of subroutine/recursion calls such as (?1) which have in any case always
1436        been different (but PCRE had them first :-).
1438    16. Related to 15 above: Perl does not treat the | in a conditional group as
1439        creating alternatives. Such a group is treated in the same way as an
1440        ordinary group without any | characters when processing (*THEN). PCRE has
1441        been changed to match Perl's behaviour.
1443    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
1444        RunGrepTest script failed.
1446    18. Change 22 for version 13 caused atomic groups to use more stack. This is
1447        inevitable for groups that contain captures, but it can lead to a lot of
1448        stack use in large patterns. The old behaviour has been restored for atomic
1449        groups that do not contain any capturing parentheses.
1451    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
1452        suppress the check for a minimum subject length at run time. (If it was
1453        given to pcre_exec() or pcre_dfa_exec() it did work.)
1455    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
1456        fail to work when decoding hex characters in data strings in EBCDIC
1457        environments.
1459    21. It appears that in at least one Mac OS environment, the isxdigit() function
1460        is implemented as a macro that evaluates to its argument more than once,
1461        contravening the C 90 Standard (I haven't checked a later standard). There
1462        was an instance in pcretest which caused it to go wrong when processing
1463        \x{...} escapes in subject strings. The has been rewritten to avoid using
1464        things like p++ in the argument of isxdigit().
1467    Version 8.13 16-Aug-2011
1468    ------------------------
1470    1.  The Unicode data tables have been updated to Unicode 6.0.0.
1472    2.  Two minor typos in pcre_internal.h have been fixed.
1474    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
1475        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
1476        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
1478    4.  There were a number of related bugs in the code for matching backrefences
1479        caselessly in UTF-8 mode when codes for the characters concerned were
1480        different numbers of bytes. For example, U+023A and U+2C65 are an upper
1481        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
1482        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
1483        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
1484        2-byte code at the end of the subject (it thought there wasn't enough data
1485        left).
1487    5.  Comprehensive information about what went wrong is now returned by
1488        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
1489        as the output vector has at least 2 elements. The offset of the start of
1490        the failing character and a reason code are placed in the vector.
1492    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
1493        now returned is for the first byte of the failing character, instead of the
1494        last byte inspected. This is an incompatible change, but I hope it is small
1495        enough not to be a problem. It makes the returned offset consistent with
1496        pcre_exec() and pcre_dfa_exec().
1498    7.  pcretest now gives a text phrase as well as the error number when
1499        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
1500        failure, the offset and reason code are output.
1502    8.  When \R was used with a maximizing quantifier it failed to skip backwards
1503        over a \r\n pair if the subsequent match failed. Instead, it just skipped
1504        back over a single character (\n). This seems wrong (because it treated the
1505        two characters as a single entity when going forwards), conflicts with the
1506        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
1507        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
1508        has been changed.
1510    9.  Some internal refactoring has changed the processing so that the handling
1511        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
1512        time (the PCRE_DOTALL option was changed this way some time ago: version
1513        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
1514        which was always a bit of a fudge. It also means that there is one less
1515        argument for the match() function, which reduces its stack requirements
1516        slightly. This change also fixes an incompatibility with Perl: the pattern
1517        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
1519    10. More internal refactoring has drastically reduced the number of recursive
1520        calls to match() for possessively repeated groups such as (abc)++ when
1521        using pcre_exec().
1523    11. While implementing 10, a number of bugs in the handling of groups were
1524        discovered and fixed:
1526        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
1527        (a|)*(?1) gave a compile-time internal error.
1528        ((a|)+)+  did not notice that the outer group could match an empty string.
1529        (^a|^)+   was not marked as anchored.
1530        (.*a|.*)+ was not marked as matching at start or after a newline.
1532    12. Yet more internal refactoring has removed another argument from the match()
1533        function. Special calls to this function are now indicated by setting a
1534        value in a variable in the "match data" data block.
1536    13. Be more explicit in pcre_study() instead of relying on "default" for
1537        opcodes that mean there is no starting character; this means that when new
1538        ones are added and accidentally left out of pcre_study(), testing should
1539        pick them up.
1541    14. The -s option of pcretest has been documented for ages as being an old
1542        synonym of -m (show memory usage). I have changed it to mean "force study
1543        for every regex", that is, assume /S for every regex. This is similar to -i
1544        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
1545        using it. It makes it easier to run collections of tests with and without
1546        study enabled, and thereby test pcre_study() more easily. All the standard
1547        tests are now run with and without -s (but some patterns can be marked as
1548        "never study" - see 20 below).
1550    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
1551        restoration of the capturing data to the outer values was not happening
1552        correctly.
1554    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
1555        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
1556        pattern had matched an empty string, and so incorrectly returned a no
1557        match.
1559    17. There was optimizing code for the last branch of non-capturing parentheses,
1560        and also for the obeyed branch of a conditional subexpression, which used
1561        tail recursion to cut down on stack usage. Unfortunately, now that there is
1562        the possibility of (*THEN) occurring in these branches, tail recursion is
1563        no longer possible because the return has to be checked for (*THEN). These
1564        two optimizations have therefore been removed. [But see 8.20/11 above.]
1566    18. If a pattern containing \R was studied, it was assumed that \R always
1567        matched two bytes, thus causing the minimum subject length to be
1568        incorrectly computed because \R can also match just one byte.
1570    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
1571        was incorrectly computed.
1573    20. If /S is present twice on a test pattern in pcretest input, it now
1574        *disables* studying, thereby overriding the use of -s on the command line
1575        (see 14 above). This is necessary for one or two tests to keep the output
1576        identical in both cases.
1578    21. When (*ACCEPT) was used in an assertion that matched an empty string and
1579        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
1581    22. When an atomic group that contained a capturing parenthesis was
1582        successfully matched, but the branch in which it appeared failed, the
1583        capturing was not being forgotten if a higher numbered group was later
1584        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
1585        group 1 to "a", when in fact it should be unset. This applied to multi-
1586        branched capturing and non-capturing groups, repeated or not, and also to
1587        positive assertions (capturing in negative assertions does not happen
1588        in PCRE) and also to nested atomic groups.
1590    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
1591        subject after a captured substring, to make it easier to tell which of a
1592        number of identical substrings has been captured.
1594    24. The way atomic groups are processed by pcre_exec() has been changed so that
1595        if they are repeated, backtracking one repetition now resets captured
1596        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
1597        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
1598        "aaa". Previously, it would have been "a". As part of this code
1599        refactoring, the way recursive calls are handled has also been changed.
1601    25. If an assertion condition captured any substrings, they were not passed
1602        back unless some other capturing happened later. For example, if
1603        (?(?=(a))a) was matched against "a", no capturing was returned.
1605    26. When studying a pattern that contained subroutine calls or assertions,
1606        the code for finding the minimum length of a possible match was handling
1607        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
1608        group 1 called group 2 while simultaneously a separate group 2 called group
1609        1). A stack overflow occurred in this case. I have fixed this by limiting
1610        the recursion depth to 10.
1612    27. Updated RunTest.bat in the distribution to the version supplied by Tom
1613        Fortmann. This supports explicit test numbers on the command line, and has
1614        argument validation and error reporting.
1616    28. An instance of \X with an unlimited repeat could fail if at any point the
1617        first character it looked at was a mark character.
1619    29. Some minor code refactoring concerning Unicode properties and scripts
1620        should reduce the stack requirement of match() slightly.
1622    30. Added the '=' option to pcretest to check the setting of unused capturing
1623        slots at the end of the pattern, which are documented as being -1, but are
1624        not included in the return count.
1626    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
1627        compiled something random. Now it gives a compile-time error (as does
1628        Perl).
1630    32. A *MARK encountered during the processing of a positive assertion is now
1631        recorded and passed back (compatible with Perl).
1633    33. If --only-matching or --colour was set on a pcregrep call whose pattern
1634        had alternative anchored branches, the search for a second match in a line
1635        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
1636        matched the line "0102" twice. The same bug affected patterns that started
1637        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
1638        twice.
1640    34. Previously, PCRE did not allow quantification of assertions. However, Perl
1641        does, and because of capturing effects, quantifying parenthesized
1642        assertions may at times be useful. Quantifiers are now allowed for
1643        parenthesized assertions.
1645    35. A minor code tidy in pcre_compile() when checking options for \R usage.
1647    36. \g was being checked for fancy things in a character class, when it should
1648        just be a literal "g".
1650    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
1651        appearance of a nested POSIX class supersedes an apparent external class.
1652        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
1653        unescaped square brackets may also appear as part of class names. For
1654        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
1655        more like Perl. (But see 8.20/1 above.)
1657    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
1658        was because it thought it was \N{name}, which is not supported).
1660    39. Add minix to OS list not supporting the -S option in pcretest.
1662    40. PCRE tries to detect cases of infinite recursion at compile time, but it
1663        cannot analyze patterns in sufficient detail to catch mutual recursions
1664        such as ((?1))((?2)). There is now a runtime test that gives an error if a
1665        subgroup is called recursively as a subpattern for a second time at the
1666        same position in the subject string. In previous releases this might have
1667        been caught by the recursion limit, or it might have run out of stack.
1669    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
1670        happen only once. PCRE was, however incorrectly giving a compile time error
1671        "recursive call could loop indefinitely" because it cannot analyze the
1672        pattern in sufficient detail. The compile time test no longer happens when
1673        PCRE is compiling a conditional subpattern, but actual runaway loops are
1674        now caught at runtime (see 40 above).
1676    42. It seems that Perl allows any characters other than a closing parenthesis
1677        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
1678        has been changed to be the same.
1680    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
1681        as not to get warnings when autogen.sh is called. Also changed
1682        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
1684    44. To help people who use pcregrep to scan files containing exceedingly long
1685        lines, the following changes have been made:
1687        (a) The default value of the buffer size parameter has been increased from
1688            8K to 20K. (The actual buffer used is three times this size.)
1690        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
1691            PCRE is built.
1693        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
1694            to be set at run time.
1696        (d) Numerical values in pcregrep options can be followed by K or M, for
1697            example --buffer-size=50K.
1699        (e) If a line being scanned overflows pcregrep's buffer, an error is now
1700            given and the return code is set to 2.
1702    45. Add a pointer to the latest mark to the callout data block.
1704    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
1705        partial match of an empty string instead of no match. This was specific to
1706        the use of ".".
1708    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
1709        complete match instead of a partial match. This bug was dependent on both
1710        the PCRE_UTF8 and PCRE_DOTALL options being set.
1712    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
1713        starting byte set, because \b was not being ignored.
1716    Version 8.12 15-Jan-2011
1717    ------------------------
1719    1.  Fixed some typos in the markup of the man pages, and wrote a script that
1720        checks for such things as part of the documentation building process.
1722    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
1723        --match-limit and --recursion-limit options (added for 8.11). In
1724        particular, this made one of the standard tests fail. (The integer value
1725        went into the wrong half of a long int.)
1727    3.  If the --colour option was given to pcregrep with -v (invert match), it
1728        did strange things, either producing crazy output, or crashing. It should,
1729        of course, ignore a request for colour when reporting lines that do not
1730        match.
1732    4.  Another pcregrep bug caused similar problems if --colour was specified with
1733        -M (multiline) and the pattern match finished with a line ending.
1735    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
1736        matched in multiline mode, the following line was shown as part of the
1737        match. This seems wrong, so I have changed it.
1739    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
1740        the check for further matches in the same line (so they could be coloured)
1741        to overrun the end of the current line. If another match was found, it was
1742        incorrectly shown (and then shown again when found in the next line).
1744    7.  If pcregrep was compiled under Windows, there was a reference to the
1745        function pcregrep_exit() before it was defined. I am assuming this was
1746        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
1747        reported by a user. I've moved the definition above the reference.
1750    Version 8.11 10-Dec-2010
1751    ------------------------
1753    1.  (*THEN) was not working properly if there were untried alternatives prior
1754        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
1755        backtracked to try for "b" instead of moving to the next alternative branch
1756        at the same level (in this case, to look for "c"). The Perl documentation
1757        is clear that when (*THEN) is backtracked onto, it goes to the "next
1758        alternative in the innermost enclosing group".
1760    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
1761        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
1762        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
1763        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
1764        (*THEN).
1766    3.  If \s appeared in a character class, it removed the VT character from
1767        the class, even if it had been included by some previous item, for example
1768        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
1769        of \s, but is part of the POSIX "space" class.)
1771    4.  A partial match never returns an empty string (because you can always
1772        match an empty string at the end of the subject); however the checking for
1773        an empty string was starting at the "start of match" point. This has been
1774        changed to the "earliest inspected character" point, because the returned
1775        data for a partial match starts at this character. This means that, for
1776        example, /(?<=abc)def/ gives a partial match for the subject "abc"
1777        (previously it gave "no match").
1779    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
1780        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
1781        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
1782        has an implication that the given string is incomplete (because a partial
1783        match is preferred over a full match). For this reason, these items now
1784        give a partial match in this situation. [Aside: previously, the one case
1785        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
1786        match rather than a full match, which was wrong by the old rules, but is
1787        now correct.]
1789    6.  There was a bug in the handling of #-introduced comments, recognized when
1790        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
1791        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
1792        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
1793        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
1794        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
1795        places in pcre_compile().
1797    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
1798        comments when looking ahead for named forward references to subpatterns,
1799        the only newline sequence it recognized was NL. It now handles newlines
1800        according to the set newline convention.
1802    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
1803        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
1804        cater for a lack of strerror(). These oversights have been fixed.
1806    9.  Added --match-limit and --recursion-limit to pcregrep.
1808    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
1810    11. When the -o option was used, pcregrep was setting a return code of 1, even
1811        when matches were found, and --line-buffered was not being honoured.
1813    12. Added an optional parentheses number to the -o and --only-matching options
1814        of pcregrep.
1816    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
1817        can match an empty string. The code to do it in pcretest and pcredemo
1818        needed fixing:
1820        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
1821            only one byte after an empty string match just before CRLF (this case
1822            just got forgotten; "any" and "anycrlf" were OK).
1824        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
1825            mode when an empty string match preceded an ASCII character followed by
1826            a non-ASCII character. (The code for advancing by one character rather
1827            than one byte was nonsense.)
1829        (c) The pcredemo.c sample program did not have any code at all to handle
1830            the cases when CRLF is a valid newline sequence.
1832    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
1833        as a starting offset was within the subject string. There is now a new
1834        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
1835        negative or greater than the length of the string. In order to test this,
1836        pcretest is extended to allow the setting of negative starting offsets.
1838    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
1839        starting offset points to the beginning of a UTF-8 character was
1840        unnecessarily clumsy. I tidied it up.
1842    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
1843        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
1845    17. Nobody had reported that the --include_dir option, which was added in
1846        release 7.7 should have been called --include-dir (hyphen, not underscore)
1847        for compatibility with GNU grep. I have changed it to --include-dir, but
1848        left --include_dir as an undocumented synonym, and the same for
1849        --exclude-dir, though that is not available in GNU grep, at least as of
1850        release 2.5.4.
1852    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
1853        characters from a string of bytes) have been redefined so as not to use
1854        loops, in order to improve performance in some environments. At the same
1855        time, I abstracted some of the common code into auxiliary macros to save
1856        repetition (this should not affect the compiled code).
1858    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1859        compile-time error is now given if \c is not followed by an ASCII
1860        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1861        different, and any byte value is allowed.)
1863    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1864        START_OPTIMIZE option, which is now allowed at compile time - but just
1865        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1866        to pcregrep and other applications that have no direct access to PCRE
1867        options. The new /Y option in pcretest sets this option when calling
1868        pcre_compile().
1870    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1871        back references. Groups containing recursive back references were forced to
1872        be atomic by that change, but in the case of named groups, the amount of
1873        memory required was incorrectly computed, leading to "Failed: internal
1874        error: code overflow". This has been fixed.
1876    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1877        pcretest.c, to avoid build problems in some Borland environments.
1880    Version 8.10 25-Jun-2010
1881    ------------------------
1883    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1884        THEN.
1886    2.  (*ACCEPT) was not working when inside an atomic group.
1888    3.  Inside a character class, \B is treated as a literal by default, but
1889        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1890        causes the error). The code is unchanged, but I tidied the documentation.
1892    4.  Inside a character class, PCRE always treated \R and \X as literals,
1893        whereas Perl faults them if its -w option is set. I have changed PCRE so
1894        that it faults them when PCRE_EXTRA is set.
1896    5.  Added support for \N, which always matches any character other than
1897        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1899    6.  When compiling pcregrep with newer versions of gcc which may have
1900        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1901        declared with attribute warn_unused_result" were given. Just casting the
1902        result to (void) does not stop the warnings; a more elaborate fudge is
1903        needed. I've used a macro to implement this.
1905    7.  Minor change to pcretest.c to avoid a compiler warning.
1907    8.  Added four artifical Unicode properties to help with an option to make
1908        \s etc use properties (see next item). The new properties are: Xan
1909        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1911    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1912        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1913        this option. Modified pcretest to add /W to test this facility. Added
1914        REG_UCP to make it available via the POSIX interface.
1916    10. Added --line-buffered to pcregrep.
1918    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1919        studied, and the match started with a letter with a code point greater than
1920        127 whose first byte was different to the first byte of the other case of
1921        the letter, the other case of this starting letter was not recognized
1922        (#976).
1924    12. If a pattern that was studied started with a repeated Unicode property
1925        test, for example, \p{Nd}+, there was the theoretical possibility of
1926        setting up an incorrect bitmap of starting bytes, but fortunately it could
1927        not have actually happened in practice until change 8 above was made (it
1928        added property types that matched character-matching opcodes).
1930    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1931        possible starting bytes for non-anchored patterns.
1933    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1934        \R, and also a number of cases that involve Unicode properties, both
1935        explicit and implicit when PCRE_UCP is set.
1937    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1938        input, it could crash or give wrong results if characters with values
1939        greater than 0xc0 were present in the subject string. (Detail: it assumed
1940        UTF-8 input when processing these items.)
1942    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1943        size_t is 64-bit (#991).
1945    17. Added a check for running out of memory when PCRE is compiled with
1946        --disable-stack-for-recursion (#990).
1948    18. If the last data line in a file for pcretest does not have a newline on
1949        the end, a newline was missing in the output.
1951    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1952        less than 128) in its various bitmaps. However, there is a facility for
1953        generating tables according to the current locale when PCRE is compiled. It
1954        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1955        characters, are recognized by isspace() and therefore were getting set in
1956        these tables, and indeed these tables seem to approximate to ISO 8859. This
1957        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1958        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1959        which of course cannot start UTF-8 characters. I have changed the code so
1960        that only real ASCII characters (less than 128) and the correct starting
1961        bytes for UTF-8 encodings are set for characters greater than 127 when in
1962        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1963        altogether.)
1965    20. Added the /T option to pcretest so as to be able to run tests with non-
1966        standard character tables, thus making it possible to include the tests
1967        used for 19 above in the standard set of tests.
1969    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1970        reference to a subpattern the other side of a comment that contains an
1971        opening parenthesis caused either an internal compiling error, or a
1972        reference to the wrong subpattern.
1975    Version 8.02 19-Mar-2010
1976    ------------------------
1978    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1980    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1981        configured.
1983    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1984        original author of that file, following a query about its status.
1986    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1987        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1989    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1990        quantifier applied to a forward-referencing subroutine call, could compile
1991        incorrect code or give the error "internal error: previously-checked
1992        referenced subpattern not found".
1994    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1995        variables to point to external functions. For these systems, therefore,
1996        pcre_malloc etc. are now initialized to local functions that call the
1997        relevant global functions.
1999    7.  There were two entries missing in the vectors called coptable and poptable
2000        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
2001        I've fixed the data, and added a kludgy way of testing at compile time that
2002        the lengths are correct (equal to the number of opcodes).
2004    8.  Following on from 7, I added a similar kludge to check the length of the
2005        eint vector in pcreposix.c.
2007    9.  Error texts for pcre_compile() are held as one long string to avoid too
2008        much relocation at load time. To find a text, the string is searched,
2009        counting zeros. There was no check for running off the end of the string,
2010        which could happen if a new error number was added without updating the
2011        string.
2013    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
2015    11. \K was not working if it appeared in an atomic group or in a group that
2016        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
2017        \K is "not well defined" if used in an assertion. PCRE now accepts it if
2018        the assertion is positive, but not if it is negative.
2020    12. Change 11 fortuitously reduced the size of the stack frame used in the
2021        "match()" function of pcre_exec.c by one pointer. Forthcoming
2022        implementation of support for (*MARK) will need an extra pointer on the
2023        stack; I have reserved it now, so that the stack frame size does not
2024        decrease.
2026    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
2027        item in branch that calls a recursion is a subroutine call - as in the
2028        second branch in the above example - was incorrectly given the compile-
2029        time error "recursive call could loop indefinitely" because pcre_compile()
2030        was not correctly checking the subroutine for matching a non-empty string.
2032    14. The checks for overrunning compiling workspace could trigger after an
2033        overrun had occurred. This is a "should never occur" error, but it can be
2034        triggered by pathological patterns such as hundreds of nested parentheses.
2035        The checks now trigger 100 bytes before the end of the workspace.
2037    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
2040    Version 8.01 19-Jan-2010
2041    ------------------------
2043    1.  If a pattern contained a conditional subpattern with only one branch (in
2044        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
2045        computed the wrong minimum data length (which is of course zero for such
2046        subpatterns). This could cause incorrect "no match" results.
2048    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
2049        the pattern is reset in the first branch, pcre_compile() failed with
2050        "internal error: code overflow at offset...". This happened only when
2051        the reset was to the original external option setting. (An optimization
2052        abstracts leading options settings into an external setting, which was the
2053        cause of this.)
2055    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
2056        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
2057        assertion pattern did not match (meaning that the assertion was true), it
2058        was incorrectly treated as false if the SKIP had been reached during the
2059        matching. This also applied to assertions used as conditions.
2061    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
2062        assertion subpattern, including such a pattern used as a condition,
2063        unpredictable results occurred, instead of the error return
2066    5.  The C++ GlobalReplace function was not working like Perl for the special
2067        situation when an empty string is matched. It now does the fancy magic
2068        stuff that is necessary.
2070    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
2071        removed. (These were left over from very, very early versions of PCRE.)
2073    7.  Some cosmetic changes to the code to make life easier when compiling it
2074        as part of something else:
2076        (a) Change DEBUG to PCRE_DEBUG.
2078        (b) In pcre_compile(), rename the member of the "branch_chain" structure
2079            called "current" as "current_branch", to prevent a collision with the
2080            Linux macro when compiled as a kernel module.
2082        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
2083            prevent a collision with the Linux macro when compiled as a kernel
2084            module.
2086    8.  In pcre_compile() there are some checks for integer overflows that used to
2087        cast potentially large values to (double). This has been changed to that
2088        when building, a check for int64_t is made, and if it is found, it is used
2089        instead, thus avoiding the use of floating point arithmetic. (There is no
2090        other use of FP in PCRE.) If int64_t is not found, the fallback is to
2091        double.
2093    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
2094        2005 (difference between two addresses compared to an unsigned value).
2096    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
2097        custom one, because of the following reported problem in Windows:
2099          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
2100              under Win32.
2101          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
2102              therefore missing the function definition.
2103          - The compiler thus generates a "C" signature for the test function.
2104          - The linker fails to find the "C" function.
2105          - PCRE fails to configure if asked to do so against libbz2.
2107    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
2108        messages were output:
2110          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
2111          rerunning libtoolize, to keep the correct libtool macros in-tree.
2112          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
2114        I have done both of these things.
2116    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
2117        most of the time, it *can* run out if it is given a pattern that contains a
2118        runaway infinite recursion. I updated the discussion in the pcrestack man
2119        page.
2121    13. Now that we have gone to the x.xx style of version numbers, the minor
2122        version may start with zero. Using 08 or 09 is a bad idea because users
2123        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
2124        interpreted as invalid octal numbers. I've updated the previous comment in
2125        configure.ac, and also added a check that gives an error if 08 or 09 are
2126        used.
2128    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
2129        causing partial matching to fail when the end of the subject matched \W
2130        in a UTF-8 pattern where \W was quantified with a minimum of 3.
2132    15. There were some discrepancies between the declarations in pcre_internal.h
2133        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
2134        their definitions. The declarations used "const uschar *" and the
2135        definitions used USPTR. Even though USPTR is normally defined as "const
2136        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
2137        reported that: "This difference in casting confuses some C++ compilers, for
2138        example, SunCC recognizes above declarations as different functions and
2139        generates broken code for hbpcre." I have changed the declarations to use
2140        USPTR.
2142    16. GNU libtool is named differently on some systems. The autogen.sh script now
2143        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
2144        (FreeBSD).
2146    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
2147        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
2148        comment: "Figure out how to create a longlong from a string: strtoll and
2149        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
2150        instance, but it only takes 2 args instead of 3!"
2152    18. A subtle bug concerned with back references has been fixed by a change of
2153        specification, with a corresponding code fix. A pattern such as
2154        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
2155        refers, was giving matches when it shouldn't. For example, xa=xaaa would
2156        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
2157        same bug. Such groups have to be quantified to be useful, or contained
2158        inside another quantified group. (If there's no repetition, the reference
2159        can never match.) The problem arises because, having left the group and
2160        moved on to the rest of the pattern, a later failure that backtracks into
2161        the group uses the captured value from the final iteration of the group
2162        rather than the correct earlier one. I have fixed this in PCRE by forcing
2163        any group that contains a reference to itself to be an atomic group; that
2164        is, there cannot be any backtracking into it once it has completed. This is
2165        similar to recursive and subroutine calls.
2168    Version 8.00 19-Oct-09
2169  ----------------------  ----------------------
2171  1.  The table for translating pcre_compile() error codes into POSIX error codes  1.  The table for translating pcre_compile() error codes into POSIX error codes
2172      was out-of-date, and there was no check on the pcre_compile() error code      was out-of-date, and there was no check on the pcre_compile() error code
2173      being within the table. This could lead to an OK return being given in      being within the table. This could lead to an OK return being given in
2174      error.      error.
2176  2.  Changed the call to open a subject file in pcregrep from fopen(pathname,  2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
2177      "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests      "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
2178      in a Windows environment.      in a Windows environment.
2180  3.  The pcregrep --count option prints the count for each file even when it is  3.  The pcregrep --count option prints the count for each file even when it is
2181      zero, as does GNU grep. However, pcregrep was also printing all files when      zero, as does GNU grep. However, pcregrep was also printing all files when
2182      --files-with-matches was added. Now, when both options are given, it prints      --files-with-matches was added. Now, when both options are given, it prints
2183      counts only for those files that have at least one match. (GNU grep just      counts only for those files that have at least one match. (GNU grep just
2184      prints the file name in this circumstance, but including the count seems      prints the file name in this circumstance, but including the count seems
2185      more useful - otherwise, why use --count?) Also ensured that the      more useful - otherwise, why use --count?) Also ensured that the
2186      combination -clh just lists non-zero counts, with no names.      combination -clh just lists non-zero counts, with no names.
2188  4.  The long form of the pcregrep -F option was incorrectly implemented as  4.  The long form of the pcregrep -F option was incorrectly implemented as
2189      --fixed_strings instead of --fixed-strings. This is an incompatible change,      --fixed_strings instead of --fixed-strings. This is an incompatible change,
2190      but it seems right to fix it, and I didn't think it was worth preserving      but it seems right to fix it, and I didn't think it was worth preserving
2191      the old behaviour.      the old behaviour.
2193  5.  The command line items --regex=pattern and --regexp=pattern were not  5.  The command line items --regex=pattern and --regexp=pattern were not
2194      recognized by pcregrep, which required --regex pattern or --regexp pattern      recognized by pcregrep, which required --regex pattern or --regexp pattern
2195      (with a space rather than an '='). The man page documented the '=' forms,      (with a space rather than an '='). The man page documented the '=' forms,
2196      which are compatible with GNU grep; these now work.      which are compatible with GNU grep; these now work.
2198  6.  No libpcreposix.pc file was created for pkg-config; there was just  6.  No libpcreposix.pc file was created for pkg-config; there was just
2199      libpcre.pc and libpcrecpp.pc. The omission has been rectified.      libpcre.pc and libpcrecpp.pc. The omission has been rectified.
2201  7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size  7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
2202      when UCP support is not needed, by modifying the Python script that      when UCP support is not needed, by modifying the Python script that
2203      generates it from Unicode data files. This should not matter if the module      generates it from Unicode data files. This should not matter if the module
2204      is correctly used as a library, but I received one complaint about 50K of      is correctly used as a library, but I received one complaint about 50K of
2205      unwanted data. My guess is that the person linked everything into his      unwanted data. My guess is that the person linked everything into his
2206      program rather than using a library. Anyway, it does no harm.      program rather than using a library. Anyway, it does no harm.
2208    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
2209        was a minimum greater than 1 for a wide character in a possessive
2210        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
2211        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
2212        character. Chaos in the form of incorrect output or a compiling loop could
2213        result.
2215    9.  The restrictions on what a pattern can contain when partial matching is
2216        requested for pcre_exec() have been removed. All patterns can now be
2217        partially matched by this function. In addition, if there are at least two
2218        slots in the offset vector, the offset of the earliest inspected character
2219        for the match and the offset of the end of the subject are set in them when
2220        PCRE_ERROR_PARTIAL is returned.
2222    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
2223        synonymous with PCRE_PARTIAL, for backwards compatibility, and
2224        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
2225        and may be more useful for multi-segment matching.
2227    11. Partial matching with pcre_exec() is now more intuitive. A partial match
2228        used to be given if ever the end of the subject was reached; now it is
2229        given only if matching could not proceed because another character was
2230        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
2231        string "Z", which now yields "no match" instead of "partial match". In the
2232        case of pcre_dfa_exec(), "no match" is given if every matching path for the
2233        final character ended with (*FAIL).
2235    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
2236        if the pattern had a "must contain" character that was already found in the
2237        earlier partial match, unless partial matching was again requested. For
2238        example, with the pattern /dog.(body)?/, the "must contain" character is
2239        "g". If the first part-match was for the string "dog", restarting with
2240        "sbody" failed. This bug has been fixed.
2242    13. The string returned by pcre_dfa_exec() after a partial match has been
2243        changed so that it starts at the first inspected character rather than the
2244        first character of the match. This makes a difference only if the pattern
2245        starts with a lookbehind assertion or \b or \B (\K is not supported by
2246        pcre_dfa_exec()). It's an incompatible change, but it makes the two
2247        matching functions compatible, and I think it's the right thing to do.
2249    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
2250        so that the demonstration program is easily available in environments where
2251        PCRE has not been installed from source.
2253    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
2254        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
2255        library.
2257    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
2258        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
2259        is not the first non-POSIX option to be added. Clearly some people find
2260        these options useful.
2262    17. If a caller to the POSIX matching function regexec() passes a non-zero
2263        value for nmatch with a NULL value for pmatch, the value of
2264        nmatch is forced to zero.
2266    18. RunGrepTest did not have a test for the availability of the -u option of
2267        the diff command, as RunTest does. It now checks in the same way as
2268        RunTest, and also checks for the -b option.
2270    19. If an odd number of negated classes containing just a single character
2271        interposed, within parentheses, between a forward reference to a named
2272        subpattern and the definition of the subpattern, compilation crashed with
2273        an internal error, complaining that it could not find the referenced
2274        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
2275        [The bug was that it was starting one character too far in when skipping
2276        over the character class, thus treating the ] as data rather than
2277        terminating the class. This meant it could skip too much.]
2279    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
2280        /g option in pcretest when the pattern contains \K, which makes it possible
2281        to have an empty string match not at the start, even when the pattern is
2282        anchored. Updated pcretest and pcredemo to use this option.
2284    21. If the maximum number of capturing subpatterns in a recursion was greater
2285        than the maximum at the outer level, the higher number was returned, but
2286        with unset values at the outer level. The correct (outer level) value is
2287        now given.
2289    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
2290        PCRE did not set those parentheses (unlike Perl). I have now found a way to
2291        make it do so. The string so far is captured, making this feature
2292        compatible with Perl.
2294    23. The tests have been re-organized, adding tests 11 and 12, to make it
2295        possible to check the Perl 5.10 features against Perl 5.10.
2297    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
2298        pattern matches a fixed length string. PCRE did not allow this; now it
2299        does. Neither allows recursion.
2301    25. I finally figured out how to implement a request to provide the minimum
2302        length of subject string that was needed in order to match a given pattern.
2303        (It was back references and recursion that I had previously got hung up
2304        on.) This code has now been added to pcre_study(); it finds a lower bound
2305        to the length of subject needed. It is not necessarily the greatest lower
2306        bound, but using it to avoid searching strings that are too short does give
2307        some useful speed-ups. The value is available to calling programs via
2308        pcre_fullinfo().
2310    26. While implementing 25, I discovered to my embarrassment that pcretest had
2311        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
2312        study optimizations had never been tested with that matching function.
2313        Oops. What is worse, even when it was passed study data, there was a bug in
2314        pcre_dfa_exec() that meant it never actually used it. Double oops. There
2315        were also very few tests of studied patterns with pcre_dfa_exec().
2317    27. If (?| is used to create subpatterns with duplicate numbers, they are now
2318        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
2319        on the other side of the coin, they are no longer allowed to have different
2320        names, because these cannot be distinguished in PCRE, and this has caused
2321        confusion. (This is a difference from Perl.)
2323    28. When duplicate subpattern names are present (necessarily with different
2324        numbers, as required by 27 above), and a test is made by name in a
2325        conditional pattern, either for a subpattern having been matched, or for
2326        recursion in such a pattern, all the associated numbered subpatterns are
2327        tested, and the overall condition is true if the condition is true for any
2328        one of them. This is the way Perl works, and is also more like the way
2329        testing by number works.
2332  Version 7.9 11-Apr-09  Version 7.9 11-Apr-09
2333  ---------------------  ---------------------
# Line 76  Version 7.9 11-Apr-09 Line 2363  Version 7.9 11-Apr-09
2363  7.  A pattern that could match an empty string could cause pcregrep to loop; it  7.  A pattern that could match an empty string could cause pcregrep to loop; it
2364      doesn't make sense to accept an empty string match in pcregrep, so I have      doesn't make sense to accept an empty string match in pcregrep, so I have
2365      locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this      locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
2366      seems to be how GNU grep behaves.      seems to be how GNU grep behaves. [But see later change 40 for release
2367        8.33.]
2369  8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at  8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
2370      start or after a newline", because the conditional assertion was not being      start or after a newline", because the conditional assertion was not being
# Line 319  Version 7.7 07-May-08 Line 2607  Version 7.7 07-May-08
2607      containing () gave an internal compiling error instead of "reference to      containing () gave an internal compiling error instead of "reference to
2608      non-existent subpattern". Fortunately, when the pattern did exist, the      non-existent subpattern". Fortunately, when the pattern did exist, the
2609      compiled code was correct. (When scanning forwards to check for the      compiled code was correct. (When scanning forwards to check for the
2610      existencd of the subpattern, it was treating the data ']' as terminating      existence of the subpattern, it was treating the data ']' as terminating
2611      the class, so got the count wrong. When actually compiling, the reference      the class, so got the count wrong. When actually compiling, the reference
2612      was subsequently set up correctly.)      was subsequently set up correctly.)

Removed from v.425  
changed lines
  Added in v.1581

  ViewVC Help
Powered by ViewVC 1.1.5