ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 193 by ph10, Mon Jul 30 11:34:18 2007 UTC revision 1540 by ph10, Sun Mar 29 17:41:16 2015 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4  Version 7.3 05-Jul-07  Version 8.37 xx-xxx-2015
5    ------------------------
7    1.  When an (*ACCEPT) is triggered inside capturing parentheses, it arranges
8        for those parentheses to be closed with whatever has been captured so far.
9        However, it was failing to mark any other groups between the hightest
10        capture so far and the currrent group as "unset". Thus, the ovector for
11        those groups contained whatever was previously there. An example is the
12        pattern /(x)|((*ACCEPT))/ when matched against "abcd".
14    2.  If an assertion condition was quantified with a minimum of zero (an odd
15        thing to do, but it happened), SIGSEGV or other misbehaviour could occur.
17    3.  If a pattern in pcretest input had the P (POSIX) modifier followed by an
18        unrecognized modifier, a crash could occur.
20    4.  An attempt to do global matching in pcretest with a zero-length ovector
21        caused a crash.
23    5.  Fixed a memory leak during matching that could occur for a subpattern
24        subroutine call (recursive or otherwise) if the number of captured groups
25        that had to be saved was greater than ten.
27    6.  Catch a bad opcode during auto-possessification after compiling a bad UTF
28        string with NO_UTF_CHECK. This is a tidyup, not a bug fix, as passing bad
29        UTF with NO_UTF_CHECK is documented as having an undefined outcome.
31    7.  A UTF pattern containing a "not" match of a non-ASCII character and a
32        subroutine reference could loop at compile time. Example: /[^\xff]((?1))/.
34    8. When a pattern is compiled, it remembers the highest back reference so that
35       when matching, if the ovector is too small, extra memory can be obtained to
36       use instead. A conditional subpattern whose condition is a check on a
37       capture having happened, such as, for example in the pattern
38       /^(?:(a)|b)(?(1)A|B)/, is another kind of back reference, but it was not
39       setting the highest backreference number. This mattered only if pcre_exec()
40       was called with an ovector that was too small to hold the capture, and there
41       was no other kind of back reference (a situation which is probably quite
42       rare). The effect of the bug was that the condition was always treated as
43       FALSE when the capture could not be consulted, leading to a incorrect
44       behaviour by pcre2_match(). This bug has been fixed.
46    9. A reference to a duplicated named group (either a back reference or a test
47       for being set in a conditional) that occurred in a part of the pattern where
48       PCRE_DUPNAMES was not set caused the amount of memory needed for the pattern
49       to be incorrectly calculated, leading to overwriting.
51    10. A mutually recursive set of back references such as (\2)(\1) caused a
52        segfault at study time (while trying to find the minimum matching length).
53        The infinite loop is now broken (with the minimum length unset, that is,
54        zero).
56    11. If an assertion that was used as a condition was quantified with a minimum
57        of zero, matching went wrong. In particular, if the whole group had
58        unlimited repetition and could match an empty string, a segfault was
59        likely. The pattern (?(?=0)?)+ is an example that caused this. Perl allows
60        assertions to be quantified, but not if they are being used as conditions,
61        so the above pattern is faulted by Perl. PCRE has now been changed so that
62        it also rejects such patterns.
64    12. A possessive capturing group such as (a)*+ with a minimum repeat of zero
65        failed to allow the zero-repeat case if pcre2_exec() was called with an
66        ovector too small to capture the group.
68    13. Fixed two bugs in pcretest that were discovered by fuzzing and reported by
69        Red Hat Product Security:
71        (a) A crash if /K and /F were both set with the option to save the compiled
72        pattern.
74        (b) Another crash if the option to print captured substrings in a callout
75        was combined with setting a null ovector, for example \O\C+ as a subject
76        string.
78    14. A pattern such as "((?2){0,1999}())?", which has a group containing a
79        forward reference repeated a large (but limited) number of times within a
80        repeated outer group that has a zero minimum quantifier, caused incorrect
81        code to be compiled, leading to the error "internal error:
82        previously-checked referenced subpattern not found" when an incorrect
83        memory address was read. This bug was reported as "heap overflow",
84        discovered by Kai Lu of Fortinet's FortiGuard Labs and given the CVE number
85        CVE-2015-2325.
87    23. A pattern such as "((?+1)(\1))/" containing a forward reference subroutine
88        call within a group that also contained a recursive back reference caused
89        incorrect code to be compiled. This bug was reported as "heap overflow",
90        discovered by Kai Lu of Fortinet's FortiGuard Labs, and given the CVE
91        number CVE-2015-2326.
93    24. Computing the size of the JIT read-only data in advance has been a source
94        of various issues, and new ones are still appear unfortunately. To fix
95        existing and future issues, size computation is eliminated from the code,
96        and replaced by on-demand memory allocation.
98    25. A pattern such as /(?i)[A-`]/, where characters in the other case are
99        adjacent to the end of the range, and the range contained characters with
100        more than one other case, caused incorrect behaviour when compiled in UTF
101        mode. In that example, the range a-j was left out of the class.
103    26. Fix JIT compilation of conditional blocks, which assertion
104        is converted to (*FAIL). E.g: /(?(?!))/.
106    27. The pattern /(?(?!)^)/ caused references to random memory. This bug was
107        discovered by the LLVM fuzzer.
109    28. The assertion (?!) is optimized to (*FAIL). This was not handled correctly
110        when this assertion was used as a condition, for example (?(?!)a|b). In
111        pcre2_match() it worked by luck; in pcre2_dfa_match() it gave an incorrect
112        error about an unsupported item.
114    29. For some types of pattern, for example /Z*(|d*){216}/, the auto-
115        possessification code could take exponential time to complete. A recursion
116        depth limit of 1000 has been imposed to limit the resources used by this
117        optimization.
119    30. A pattern such as /(*UTF)[\S\V\H]/, which contains a negated special class
120        such as \S in non-UCP mode, explicit wide characters (> 255) can be ignored
121        because \S ensures they are all in the class. The code for doing this was
122        interacting badly with the code for computing the amount of space needed to
123        compile the pattern, leading to a buffer overflow. This bug was discovered
124        by the LLVM fuzzer.
126    31. A pattern such as /((?2)+)((?1))/ which has mutual recursion nested inside
127        other kinds of group caused stack overflow at compile time. This bug was
128        discovered by the LLVM fuzzer.
130    32. A pattern such as /(?1)(?#?'){8}(a)/ which had a parenthesized comment
131        between a subroutine call and its quantifier was incorrectly compiled,
132        leading to buffer overflow or other errors. This bug was discovered by the
133        LLVM fuzzer.
135    33. The illegal pattern /(?(?<E>.*!.*)?)/ was not being diagnosed as missing an
136        assertion after (?(. The code was failing to check the character after
137        (?(?< for the ! or = that would indicate a lookbehind assertion. This bug
138        was discovered by the LLVM fuzzer.
140    34. A pattern such as /X((?2)()*+){2}+/ which has a possessive quantifier with
141        a fixed maximum following a group that contains a subroutine reference was
142        incorrectly compiled and could trigger buffer overflow. This bug was
143        discovered by the LLVM fuzzer.
146    Version 8.36 26-September-2014
147    ------------------------------
149    1.  Got rid of some compiler warnings in the C++ modules that were shown up by
150        -Wmissing-field-initializers and -Wunused-parameter.
152    2.  The tests for quantifiers being too big (greater than 65535) were being
153        applied after reading the number, and stupidly assuming that integer
154        overflow would give a negative number. The tests are now applied as the
155        numbers are read.
157    3.  Tidy code in pcre_exec.c where two branches that used to be different are
158        now the same.
160    4.  The JIT compiler did not generate match limit checks for certain
161        bracketed expressions with quantifiers. This may lead to exponential
162        backtracking, instead of returning with PCRE_ERROR_MATCHLIMIT. This
163        issue should be resolved now.
165    5.  Fixed an issue, which occures when nested alternatives are optimized
166        with table jumps.
168    6.  Inserted two casts and changed some ints to size_t in the light of some
169        reported 64-bit compiler warnings (Bugzilla 1477).
171    7.  Fixed a bug concerned with zero-minimum possessive groups that could match
172        an empty string, which sometimes were behaving incorrectly in the
173        interpreter (though correctly in the JIT matcher). This pcretest input is
174        an example:
176          '\A(?:[^"]++|"(?:[^"]*+|"")*+")++'
179        the interpreter was reporting a match of 'NON QUOTED ' only, whereas the
180        JIT matcher and Perl both matched 'NON QUOTED "QUOT""ED" AFTER '. The test
181        for an empty string was breaking the inner loop and carrying on at a lower
182        level, when possessive repeated groups should always return to a higher
183        level as they have no backtrack points in them. The empty string test now
184        occurs at the outer level.
186    8.  Fixed a bug that was incorrectly auto-possessifying \w+ in the pattern
187        ^\w+(?>\s*)(?<=\w) which caused it not to match "test test".
189    9.  Give a compile-time error for \o{} (as Perl does) and for \x{} (which Perl
190        doesn't).
192    10. Change 8.34/15 introduced a bug that caused the amount of memory needed
193        to hold a pattern to be incorrectly computed (too small) when there were
194        named back references to duplicated names. This could cause "internal
195        error: code overflow" or "double free or corruption" or other memory
196        handling errors.
198    11. When named subpatterns had the same prefixes, back references could be
199        confused. For example, in this pattern:
201          /(?P<Name>a)?(?P<Name2>b)?(?(<Name>)c|d)*l/
203        the reference to 'Name' was incorrectly treated as a reference to a
204        duplicate name.
206    12. A pattern such as /^s?c/mi8 where the optional character has more than
207        one "other case" was incorrectly compiled such that it would only try to
208        match starting at "c".
210    13. When a pattern starting with \s was studied, VT was not included in the
211        list of possible starting characters; this should have been part of the
212        8.34/18 patch.
214    14. If a character class started [\Qx]... where x is any character, the class
215        was incorrectly terminated at the ].
217    15. If a pattern that started with a caseless match for a character with more
218        than one "other case" was studied, PCRE did not set up the starting code
219        unit bit map for the list of possible characters. Now it does. This is an
220        optimization improvement, not a bug fix.
222    16. The Unicode data tables have been updated to Unicode 7.0.0.
224    17. Fixed a number of memory leaks in pcregrep.
226    18. Avoid a compiler warning (from some compilers) for a function call with
227        a cast that removes "const" from an lvalue by using an intermediate
228        variable (to which the compiler does not object).
230    19. Incorrect code was compiled if a group that contained an internal recursive
231        back reference was optional (had quantifier with a minimum of zero). This
232        example compiled incorrect code: /(((a\2)|(a*)\g<-1>))*/ and other examples
233        caused segmentation faults because of stack overflows at compile time.
235    20. A pattern such as /((?(R)a|(?1)))+/, which contains a recursion within a
236        group that is quantified with an indefinite repeat, caused a compile-time
237        loop which used up all the system stack and provoked a segmentation fault.
238        This was not the same bug as 19 above.
240    21. Add PCRECPP_EXP_DECL declaration to operator<< in pcre_stringpiece.h.
241        Patch by Mike Frysinger.
244    Version 8.35 04-April-2014
245    --------------------------
247    1.  A new flag is set, when property checks are present in an XCLASS.
248        When this flag is not set, PCRE can perform certain optimizations
249        such as studying these XCLASS-es.
251    2.  The auto-possessification of character sets were improved: a normal
252        and an extended character set can be compared now. Furthermore
253        the JIT compiler optimizes more character set checks.
255    3.  Got rid of some compiler warnings for potentially uninitialized variables
256        that show up only when compiled with -O2.
258    4.  A pattern such as (?=ab\K) that uses \K in an assertion can set the start
259        of a match later then the end of the match. The pcretest program was not
260        handling the case sensibly - it was outputting from the start to the next
261        binary zero. It now reports this situation in a message, and outputs the
262        text from the end to the start.
264    5.  Fast forward search is improved in JIT. Instead of the first three
265        characters, any three characters with fixed position can be searched.
266        Search order: first, last, middle.
268    6.  Improve character range checks in JIT. Characters are read by an inprecise
269        function now, which returns with an unknown value if the character code is
270        above a certain threshold (e.g: 256). The only limitation is that the value
271        must be bigger than the threshold as well. This function is useful when
272        the characters above the threshold are handled in the same way.
274    7.  The macros whose names start with RAWUCHAR are placeholders for a future
275        mode in which only the bottom 21 bits of 32-bit data items are used. To
276        make this more memorable for those maintaining the code, the names have
277        been changed to start with UCHAR21, and an extensive comment has been added
278        to their definition.
280    8.  Add missing (new) files sljitNativeTILEGX.c and sljitNativeTILEGX-encoder.c
281        to the export list in Makefile.am (they were accidentally omitted from the
282        8.34 tarball).
284    9.  The informational output from pcretest used the phrase "starting byte set"
285        which is inappropriate for the 16-bit and 32-bit libraries. As the output
286        for "first char" and "need char" really means "non-UTF-char", I've changed
287        "byte" to "char", and slightly reworded the output. The documentation about
288        these values has also been (I hope) clarified.
290    10. Another JIT related optimization: use table jumps for selecting the correct
291        backtracking path, when more than four alternatives are present inside a
292        bracket.
294    11. Empty match is not possible, when the minimum length is greater than zero,
295        and there is no \K in the pattern. JIT should avoid empty match checks in
296        such cases.
298    12. In a caseless character class with UCP support, when a character with more
299        than one alternative case was not the first character of a range, not all
300        the alternative cases were added to the class. For example, s and \x{17f}
301        are both alternative cases for S: the class [RST] was handled correctly,
302        but [R-T] was not.
304    13. The configure.ac file always checked for pthread support when JIT was
305        enabled. This is not used in Windows, so I have put this test inside a
306        check for the presence of windows.h (which was already tested for).
308    14. Improve pattern prefix search by a simplified Boyer-Moore algorithm in JIT.
309        The algorithm provides a way to skip certain starting offsets, and usually
310        faster than linear prefix searches.
312    15. Change 13 for 8.20 updated RunTest to check for the 'fr' locale as well
313        as for 'fr_FR' and 'french'. For some reason, however, it then used the
314        Windows-specific input and output files, which have 'french' screwed in.
315        So this could never have worked. One of the problems with locales is that
316        they aren't always the same. I have now updated RunTest so that it checks
317        the output of the locale test (test 3) against three different output
318        files, and it allows the test to pass if any one of them matches. With luck
319        this should make the test pass on some versions of Solaris where it was
320        failing. Because of the uncertainty, the script did not used to stop if
321        test 3 failed; it now does. If further versions of a French locale ever
322        come to light, they can now easily be added.
324    16. If --with-pcregrep-bufsize was given a non-integer value such as "50K",
325        there was a message during ./configure, but it did not stop. This now
326        provokes an error. The invalid example in README has been corrected.
327        If a value less than the minimum is given, the minimum value has always
328        been used, but now a warning is given.
330    17. If --enable-bsr-anycrlf was set, the special 16/32-bit test failed. This
331        was a bug in the test system, which is now fixed. Also, the list of various
332        configurations that are tested for each release did not have one with both
333        16/32 bits and --enable-bar-anycrlf. It now does.
335    18. pcretest was missing "-C bsr" for displaying the \R default setting.
337    19. Little endian PowerPC systems are supported now by the JIT compiler.
339    20. The fast forward newline mechanism could enter to an infinite loop on
340        certain invalid UTF-8 input. Although we don't support these cases
341        this issue can be fixed by a performance optimization.
343    21. Change 33 of 8.34 is not sufficient to ensure stack safety because it does
344        not take account if existing stack usage. There is now a new global
345        variable called pcre_stack_guard that can be set to point to an external
346        function to check stack availability. It is called at the start of
347        processing every parenthesized group.
349    22. A typo in the code meant that in ungreedy mode the max/min qualifier
350        behaved like a min-possessive qualifier, and, for example, /a{1,3}b/U did
351        not match "ab".
353    23. When UTF was disabled, the JIT program reported some incorrect compile
354        errors. These messages are silenced now.
356    24. Experimental support for ARM-64 and MIPS-64 has been added to the JIT
357        compiler.
359    25. Change all the temporary files used in RunGrepTest to be different to those
360        used by RunTest so that the tests can be run simultaneously, for example by
361        "make -j check".
364    Version 8.34 15-December-2013
365    -----------------------------
367    1.  Add pcre[16|32]_jit_free_unused_memory to forcibly free unused JIT
368        executable memory. Patch inspired by Carsten Klein.
370    2.  ./configure --enable-coverage defined SUPPORT_GCOV in config.h, although
371        this macro is never tested and has no effect, because the work to support
372        coverage involves only compiling and linking options and special targets in
373        the Makefile. The comment in config.h implied that defining the macro would
374        enable coverage support, which is totally false. There was also support for
375        setting this macro in the CMake files (my fault, I just copied it from
376        configure). SUPPORT_GCOV has now been removed.
378    3.  Make a small performance improvement in strlen16() and strlen32() in
379        pcretest.
381    4.  Change 36 for 8.33 left some unreachable statements in pcre_exec.c,
382        detected by the Solaris compiler (gcc doesn't seem to be able to diagnose
383        these cases). There was also one in pcretest.c.
385    5.  Cleaned up a "may be uninitialized" compiler warning in pcre_exec.c.
387    6.  In UTF mode, the code for checking whether a group could match an empty
388        string (which is used for indefinitely repeated groups to allow for
389        breaking an infinite loop) was broken when the group contained a repeated
390        negated single-character class with a character that occupied more than one
391        data item and had a minimum repetition of zero (for example, [^\x{100}]* in
392        UTF-8 mode). The effect was undefined: the group might or might not be
393        deemed as matching an empty string, or the program might have crashed.
395    7.  The code for checking whether a group could match an empty string was not
396        recognizing that \h, \H, \v, \V, and \R must match a character.
398    8.  Implemented PCRE_INFO_MATCH_EMPTY, which yields 1 if the pattern can match
399        an empty string. If it can, pcretest shows this in its information output.
401    9.  Fixed two related bugs that applied to Unicode extended grapheme clusters
402        that were repeated with a maximizing qualifier (e.g. \X* or \X{2,5}) when
403        matched by pcre_exec() without using JIT:
405        (a) If the rest of the pattern did not match after a maximal run of
406            grapheme clusters, the code for backing up to try with fewer of them
407            did not always back up over a full grapheme when characters that do not
408            have the modifier quality were involved, e.g. Hangul syllables.
410        (b) If the match point in a subject started with modifier character, and
411            there was no match, the code could incorrectly back up beyond the match
412            point, and potentially beyond the first character in the subject,
413            leading to a segfault or an incorrect match result.
415    10. A conditional group with an assertion condition could lead to PCRE
416        recording an incorrect first data item for a match if no other first data
417        item was recorded. For example, the pattern (?(?=ab)ab) recorded "a" as a
418        first data item, and therefore matched "ca" after "c" instead of at the
419        start.
421    11. Change 40 for 8.33 (allowing pcregrep to find empty strings) showed up a
422        bug that caused the command "echo a | ./pcregrep -M '|a'" to loop.
424    12. The source of pcregrep now includes z/OS-specific code so that it can be
425        compiled for z/OS as part of the special z/OS distribution.
427    13. Added the -T and -TM options to pcretest.
429    14. The code in pcre_compile.c for creating the table of named capturing groups
430        has been refactored. Instead of creating the table dynamically during the
431        actual compiling pass, the information is remembered during the pre-compile
432        pass (on the stack unless there are more than 20 named groups, in which
433        case malloc() is used) and the whole table is created before the actual
434        compile happens. This has simplified the code (it is now nearly 150 lines
435        shorter) and prepared the way for better handling of references to groups
436        with duplicate names.
438    15. A back reference to a named subpattern when there is more than one of the
439        same name now checks them in the order in which they appear in the pattern.
440        The first one that is set is used for the reference. Previously only the
441        first one was inspected. This change makes PCRE more compatible with Perl.
443    16. Unicode character properties were updated from Unicode 6.3.0.
445    17. The compile-time code for auto-possessification has been refactored, based
446        on a patch by Zoltan Herczeg. It now happens after instead of during
447        compilation. The code is cleaner, and more cases are handled. The option
448        PCRE_NO_AUTO_POSSESS is added for testing purposes, and the -O and /O
449        options in pcretest are provided to set it. It can also be set by
450        (*NO_AUTO_POSSESS) at the start of a pattern.
452    18. The character VT has been added to the default ("C" locale) set of
453        characters that match \s and are generally treated as white space,
454        following this same change in Perl 5.18. There is now no difference between
455        "Perl space" and "POSIX space". Whether VT is treated as white space in
456        other locales depends on the locale.
458    19. The code for checking named groups as conditions, either for being set or
459        for being recursed, has been refactored (this is related to 14 and 15
460        above). Processing unduplicated named groups should now be as fast at
461        numerical groups, and processing duplicated groups should be faster than
462        before.
464    20. Two patches to the CMake build system, by Alexander Barkov:
466          (1) Replace the "source" command by "." in CMakeLists.txt because
467              "source" is a bash-ism.
469          (2) Add missing HAVE_STDINT_H and HAVE_INTTYPES_H to config-cmake.h.in;
470              without these the CMake build does not work on Solaris.
472    21. Perl has changed its handling of \8 and \9. If there is no previously
473        encountered capturing group of those numbers, they are treated as the
474        literal characters 8 and 9 instead of a binary zero followed by the
475        literals. PCRE now does the same.
477    22. Following Perl, added \o{} to specify codepoints in octal, making it
478        possible to specify values greater than 0777 and also making them
479        unambiguous.
481    23. Perl now gives an error for missing closing braces after \x{... instead of
482        treating the string as literal. PCRE now does the same.
484    24. RunTest used to grumble if an inappropriate test was selected explicitly,
485        but just skip it when running all tests. This make it awkward to run ranges
486        of tests when one of them was inappropriate. Now it just skips any
487        inappropriate tests, as it always did when running all tests.
489    25. If PCRE_AUTO_CALLOUT and PCRE_UCP were set for a pattern that contained
490        character types such as \d or \w, too many callouts were inserted, and the
491        data that they returned was rubbish.
493    26. In UCP mode, \s was not matching two of the characters that Perl matches,
494        namely NEL (U+0085) and MONGOLIAN VOWEL SEPARATOR (U+180E), though they
495        were matched by \h. The code has now been refactored so that the lists of
496        the horizontal and vertical whitespace characters used for \h and \v (which
497        are defined only in one place) are now also used for \s.
499    27. Add JIT support for the 64 bit TileGX architecture.
500        Patch by Jiong Wang (Tilera Corporation).
502    28. Possessive quantifiers for classes (both explicit and automatically
503        generated) now use special opcodes instead of wrapping in ONCE brackets.
505    29. Whereas an item such as A{4}+ ignored the possessivenes of the quantifier
506        (because it's meaningless), this was not happening when PCRE_CASELESS was
507        set. Not wrong, but inefficient.
509    30. Updated perltest.pl to add /u (force Unicode mode) when /W (use Unicode
510        properties for \w, \d, etc) is present in a test regex. Otherwise if the
511        test contains no characters greater than 255, Perl doesn't realise it
512        should be using Unicode semantics.
514    31. Upgraded the handling of the POSIX classes [:graph:], [:print:], and
515        [:punct:] when PCRE_UCP is set so as to include the same characters as Perl
516        does in Unicode mode.
518    32. Added the "forbid" facility to pcretest so that putting tests into the
519        wrong test files can sometimes be quickly detected.
521    33. There is now a limit (default 250) on the depth of nesting of parentheses.
522        This limit is imposed to control the amount of system stack used at compile
523        time. It can be changed at build time by --with-parens-nest-limit=xxx or
524        the equivalent in CMake.
526    34. Character classes such as [A-\d] or [a-[:digit:]] now cause compile-time
527        errors. Perl warns for these when in warning mode, but PCRE has no facility
528        for giving warnings.
530    35. Change 34 for 8.13 allowed quantifiers on assertions, because Perl does.
531        However, this was not working for (?!) because it is optimized to (*FAIL),
532        for which PCRE does not allow quantifiers. The optimization is now disabled
533        when a quantifier follows (?!). I can't see any use for this, but it makes
534        things uniform.
536    36. Perl no longer allows group names to start with digits, so I have made this
537        change also in PCRE. It simplifies the code a bit.
539    37. In extended mode, Perl ignores spaces before a + that indicates a
540        possessive quantifier. PCRE allowed a space before the quantifier, but not
541        before the possessive +. It now does.
543    38. The use of \K (reset reported match start) within a repeated possessive
544        group such as (a\Kb)*+ was not working.
546    40. Document that the same character tables must be used at compile time and
547        run time, and that the facility to pass tables to pcre_exec() and
548        pcre_dfa_exec() is for use only with saved/restored patterns.
550    41. Applied Jeff Trawick's patch CMakeLists.txt, which "provides two new
551        features for Builds with MSVC:
553        1. Support pcre.rc and/or pcreposix.rc (as is already done for MinGW
554           builds). The .rc files can be used to set FileDescription and many other
555           attributes.
557        2. Add an option (-DINSTALL_MSVC_PDB) to enable installation of .pdb files.
558           This allows higher-level build scripts which want .pdb files to avoid
559           hard-coding the exact files needed."
561    42. Added support for [[:<:]] and [[:>:]] as used in the BSD POSIX library to
562        mean "start of word" and "end of word", respectively, as a transition aid.
564    43. A minimizing repeat of a class containing codepoints greater than 255 in
565        non-UTF 16-bit or 32-bit modes caused an internal error when PCRE was
566        compiled to use the heap for recursion.
568    44. Got rid of some compiler warnings for unused variables when UTF but not UCP
569        is configured.
572    Version 8.33 28-May-2013
573    ------------------------
575    1.  Added 'U' to some constants that are compared to unsigned integers, to
576        avoid compiler signed/unsigned warnings. Added (int) casts to unsigned
577        variables that are added to signed variables, to ensure the result is
578        signed and can be negated.
580    2.  Applied patch by Daniel Richard G for quashing MSVC warnings to the
581        CMake config files.
583    3.  Revise the creation of config.h.generic so that all boolean macros are
584        #undefined, whereas non-boolean macros are #ifndef/#endif-ed. This makes
585        overriding via -D on the command line possible.
587    4.  Changing the definition of the variable "op" in pcre_exec.c from pcre_uchar
588        to unsigned int is reported to make a quite noticeable speed difference in
589        a specific Windows environment. Testing on Linux did also appear to show
590        some benefit (and it is clearly not harmful). Also fixed the definition of
591        Xop which should be unsigned.
593    5.  Related to (4), changing the definition of the intermediate variable cc
594        in repeated character loops from pcre_uchar to pcre_uint32 also gave speed
595        improvements.
597    6.  Fix forward search in JIT when link size is 3 or greater. Also removed some
598        unnecessary spaces.
600    7.  Adjust autogen.sh and configure.ac to lose warnings given by automake 1.12
601        and later.
603    8.  Fix two buffer over read issues in 16 and 32 bit modes. Affects JIT only.
605    9.  Optimizing fast_forward_start_bits in JIT.
607    10. Adding support for callouts in JIT, and fixing some issues revealed
608        during this work. Namely:
610        (a) Unoptimized capturing brackets incorrectly reset on backtrack.
612        (b) Minimum length was not checked before the matching is started.
614    11. The value of capture_last that is passed to callouts was incorrect in some
615        cases when there was a capture on one path that was subsequently abandoned
616        after a backtrack. Also, the capture_last value is now reset after a
617        recursion, since all captures are also reset in this case.
619    12. The interpreter no longer returns the "too many substrings" error in the
620        case when an overflowing capture is in a branch that is subsequently
621        abandoned after a backtrack.
623    13. In the pathological case when an offset vector of size 2 is used, pcretest
624        now prints out the matched string after a yield of 0 or 1.
626    14. Inlining subpatterns in recursions, when certain conditions are fulfilled.
627        Only supported by the JIT compiler at the moment.
629    15. JIT compiler now supports 32 bit Macs thanks to Lawrence Velazquez.
631    16. Partial matches now set offsets[2] to the "bumpalong" value, that is, the
632        offset of the starting point of the matching process, provided the offsets
633        vector is large enough.
635    17. The \A escape now records a lookbehind value of 1, though its execution
636        does not actually inspect the previous character. This is to ensure that,
637        in partial multi-segment matching, at least one character from the old
638        segment is retained when a new segment is processed. Otherwise, if there
639        are no lookbehinds in the pattern, \A might match incorrectly at the start
640        of a new segment.
642    18. Added some #ifdef __VMS code into pcretest.c to help VMS implementations.
644    19. Redefined some pcre_uchar variables in pcre_exec.c as pcre_uint32; this
645        gives some modest performance improvement in 8-bit mode.
647    20. Added the PCRE-specific property \p{Xuc} for matching characters that can
648        be expressed in certain programming languages using Universal Character
649        Names.
651    21. Unicode validation has been updated in the light of Unicode Corrigendum #9,
652        which points out that "non characters" are not "characters that may not
653        appear in Unicode strings" but rather "characters that are reserved for
654        internal use and have only local meaning".
656    22. When a pattern was compiled with automatic callouts (PCRE_AUTO_CALLOUT) and
657        there was a conditional group that depended on an assertion, if the
658        assertion was false, the callout that immediately followed the alternation
659        in the condition was skipped when pcre_exec() was used for matching.
661    23. Allow an explicit callout to be inserted before an assertion that is the
662        condition for a conditional group, for compatibility with automatic
663        callouts, which always insert a callout at this point.
665    24. In 8.31, (*COMMIT) was confined to within a recursive subpattern. Perl also
666        confines (*SKIP) and (*PRUNE) in the same way, and this has now been done.
668    25. (*PRUNE) is now supported by the JIT compiler.
670    26. Fix infinite loop when /(?<=(*SKIP)ac)a/ is matched against aa.
672    27. Fix the case where there are two or more SKIPs with arguments that may be
673        ignored.
675    28. (*SKIP) is now supported by the JIT compiler.
677    29. (*THEN) is now supported by the JIT compiler.
679    30. Update RunTest with additional test selector options.
681    31. The way PCRE handles backtracking verbs has been changed in two ways.
683        (1) Previously, in something like (*COMMIT)(*SKIP), COMMIT would override
684        SKIP. Now, PCRE acts on whichever backtracking verb is reached first by
685        backtracking. In some cases this makes it more Perl-compatible, but Perl's
686        rather obscure rules do not always do the same thing.
688        (2) Previously, backtracking verbs were confined within assertions. This is
689        no longer the case for positive assertions, except for (*ACCEPT). Again,
690        this sometimes improves Perl compatibility, and sometimes does not.
692    32. A number of tests that were in test 2 because Perl did things differently
693        have been moved to test 1, because either Perl or PCRE has changed, and
694        these tests are now compatible.
696    32. Backtracking control verbs are now handled in the same way in JIT and
697        interpreter.
699    33. An opening parenthesis in a MARK/PRUNE/SKIP/THEN name in a pattern that
700        contained a forward subroutine reference caused a compile error.
702    34. Auto-detect and optimize limited repetitions in JIT.
704    35. Implement PCRE_NEVER_UTF to lock out the use of UTF, in particular,
705        blocking (*UTF) etc.
707    36. In the interpreter, maximizing pattern repetitions for characters and
708        character types now use tail recursion, which reduces stack usage.
710    37. The value of the max lookbehind was not correctly preserved if a compiled
711        and saved regex was reloaded on a host of different endianness.
713    38. Implemented (*LIMIT_MATCH) and (*LIMIT_RECURSION). As part of the extension
714        of the compiled pattern block, expand the flags field from 16 to 32 bits
715        because it was almost full.
717    39. Try madvise first before posix_madvise.
719    40. Change 7 for PCRE 7.9 made it impossible for pcregrep to find empty lines
720        with a pattern such as ^$. It has taken 4 years for anybody to notice! The
721        original change locked out all matches of empty strings. This has been
722        changed so that one match of an empty string per line is recognized.
723        Subsequent searches on the same line (for colouring or for --only-matching,
724        for example) do not recognize empty strings.
726    41. Applied a user patch to fix a number of spelling mistakes in comments.
728    42. Data lines longer than 65536 caused pcretest to crash.
730    43. Clarified the data type for length and startoffset arguments for pcre_exec
731        and pcre_dfa_exec in the function-specific man pages, where they were
732        explicitly stated to be in bytes, never having been updated. I also added
733        some clarification to the pcreapi man page.
735    44. A call to pcre_dfa_exec() with an output vector size less than 2 caused
736        a segmentation fault.
739    Version 8.32 30-November-2012
740    -----------------------------
742    1.  Improved JIT compiler optimizations for first character search and single
743        character iterators.
745    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
746        Patch by Daniel Richard G.
748    3.  Single character iterator optimizations in the JIT compiler.
750    4.  Improved JIT compiler optimizations for character ranges.
752    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
753        Reported by Giuseppe D'Angelo.
755    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
756        of a line, was being set incorrectly in cases where .* appeared inside
757        atomic brackets at the start of a pattern, or where there was a subsequent
758        *PRUNE or *SKIP.
760    7.  Improved instruction cache flush for POWER/PowerPC.
761        Patch by Daniel Richard G.
763    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
764        grep:
766        (a) There is now no limit to the number of patterns to be matched.
768        (b) An error is given if a pattern is too long.
770        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
771            are now supported.
773        (d) --exclude-from and --include-from (multiple use) have been added.
775        (e) Exclusions and inclusions now apply to all files and directories, not
776            just to those obtained from scanning a directory recursively.
778        (f) Multiple uses of -f and --file-list are now supported.
780        (g) In a Windows environment, the default for -d has been changed from
781            "read" (the GNU grep default) to "skip", because otherwise the presence
782            of a directory in the file list provokes an error.
784        (h) The documentation has been revised and clarified in places.
786    9.  Improve the matching speed of capturing brackets.
788    10. Changed the meaning of \X so that it now matches a Unicode extended
789        grapheme cluster.
791    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
792        out POSIX threads when JIT support is configured.
794    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
796    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
797        structure could go wrong in environments where size_t is not the same size
798        as int.
800    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
801        set.
803    15. The EBCDIC support had decayed; later updates to the code had included
804        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
805        general tidy up of EBCDIC-related issues, and the documentation was also
806        not quite right. There is now a test that can be run on ASCII systems to
807        check some of the EBCDIC-related things (but is it not a full test).
809    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
810        in a small tidy to the code.
812    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
814    18. If the --only-matching (-o) option in pcregrep is specified multiple
815        times, each one causes appropriate output. For example, -o1 -o2 outputs the
816        substrings matched by the 1st and 2nd capturing parentheses. A separating
817        string can be specified by --om-separator (default empty).
819    19. Improving the first n character searches.
821    20. Turn case lists for horizontal and vertical white space into macros so that
822        they are defined only once.
824    21. This set of changes together give more compatible Unicode case-folding
825        behaviour for characters that have more than one other case when UCP
826        support is available.
828        (a) The Unicode property table now has offsets into a new table of sets of
829            three or more characters that are case-equivalent. The MultiStage2.py
830            script that generates these tables (the pcre_ucd.c file) now scans
831            CaseFolding.txt instead of UnicodeData.txt for character case
832            information.
834        (b) The code for adding characters or ranges of characters to a character
835            class has been abstracted into a generalized function that also handles
836            case-independence. In UTF-mode with UCP support, this uses the new data
837            to handle characters with more than one other case.
839        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
840            whose other case is greater than 256 are now correctly matched
841            caselessly. Previously, the high codepoint matched the low one, but not
842            vice versa.
844        (d) The processing of \h, \H, \v, and \ in character classes now makes use
845            of the new class addition function, using character lists defined as
846            macros alongside the case definitions of 20 above.
848        (e) Caseless back references now work with characters that have more than
849            one other case.
851        (f) General caseless matching of characters with more than one other case
852            is supported.
854    22. Unicode character properties were updated from Unicode 6.2.0
856    23. Improved CMake support under Windows. Patch by Daniel Richard G.
858    24. Add support for 32-bit character strings, and UTF-32
860    25. Major JIT compiler update (code refactoring and bugfixing).
861        Experimental Sparc 32 support is added.
863    26. Applied a modified version of Daniel Richard G's patch to create
864        pcre.h.generic and config.h.generic by "make" instead of in the
865        PrepareRelease script.
867    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
868        pcre_compile.c when checking for a zero character.
870    28. Introducing a native interface for JIT. Through this interface, the compiled
871        machine code can be directly executed. The purpose of this interface is to
872        provide fast pattern matching, so several sanity checks are not performed.
873        However, feature tests are still performed. The new interface provides
874        1.4x speedup compared to the old one.
876    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
877        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
878        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
880    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
881        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
882        the "old" RFC 2279). Instead, it ended up passing a negative length to
883        pcre_exec().
885    31. Add support for GCC's visibility feature to hide internal functions.
887    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
888        "unknown -C option" after outputting 0 or 1.
890    33. There is now support for generating a code coverage report for the test
891        suite in environments where gcc is the compiler and lcov is installed. This
892        is mainly for the benefit of the developers.
894    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
895        unaddressable using valgrind annotations, allowing valgrind to detect
896        invalid memory accesses. This is mainly for the benefit of the developers.
898    25. (*UTF) can now be used to start a pattern in any of the three libraries.
900    26. Give configure error if --enable-cpp but no C++ compiler found.
903    Version 8.31 06-July-2012
904    -------------------------
906    1.  Fixing a wrong JIT test case and some compiler warnings.
908    2.  Removed a bashism from the RunTest script.
910    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
911        to unsigned type, result still unsigned" that was given by an MS compiler
912        on encountering the code "-sizeof(xxx)".
914    4.  Partial matching support is added to the JIT compiler.
916    5.  Fixed several bugs concerned with partial matching of items that consist
917        of more than one character:
919        (a) /^(..)\1/ did not partially match "aba" because checking references was
920            done on an "all or nothing" basis. This also applied to repeated
921            references.
923        (b) \R did not give a hard partial match if \r was found at the end of the
924            subject.
926        (c) \X did not give a hard partial match after matching one or more
927            characters at the end of the subject.
929        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
930            a partial match for the string "\r".
932        (e) When newline was set to CRLF, the metacharacter "." did not recognize
933            a partial match for a CR character at the end of the subject string.
935    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
936        running pcretest, the text "(JIT)" added to the output whenever JIT is
937        actually used to run the match.
939    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
940        or /S+[+] with a digit between 1 and 7.
942    8.  OP_NOT now supports any UTF character not just single-byte ones.
944    9.  (*MARK) control verb is now supported by the JIT compiler.
946    10. The command "./RunTest list" lists the available tests without actually
947        running any of them. (Because I keep forgetting what they all are.)
951    12. Applied a (slightly modified) user-supplied patch that improves performance
952        when the heap is used for recursion (compiled with --disable-stack-for-
953        recursion). Instead of malloc and free for each heap frame each time a
954        logical recursion happens, frames are retained on a chain and re-used where
955        possible. This sometimes gives as much as 30% improvement.
957    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
958        call.
960    14. As documented, (*COMMIT) is now confined to within a positive assertion.
962    15. It is now possible to link pcretest with libedit as an alternative to
963        libreadline.
965    16. (*COMMIT) control verb is now supported by the JIT compiler.
967    17. The Unicode data tables have been updated to Unicode 6.1.0.
969    18. Added --file-list option to pcregrep.
971    19. Added binary file support to pcregrep, including the -a, --binary-files,
972        -I, and --text options.
974    20. The madvise function is renamed for posix_madvise for QNX compatibility
975        reasons. Fixed by Giuseppe D'Angelo.
977    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
979    22. Changed the option for creating symbolic links for 16-bit man pages from
980        -s to -sf so that re-installing does not cause issues.
982    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
984    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
985        matches in certain environments (the workspace was not being correctly
986        retained). Also added to pcre_dfa_exec() a simple plausibility check on
987        some of the workspace data at the beginning of a restart.
989    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
990        was not doing so when it should - probably a typo introduced by SVN 528
991        (change 8.10/14).
993    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
994        \w+ when the character tables indicated that \x{c4} was a word character.
995        There were several related cases, all because the tests for doing a table
996        lookup were testing for characters less than 127 instead of 255.
998    27. If a pattern contains capturing parentheses that are not used in a match,
999        their slots in the ovector are set to -1. For those that are higher than
1000        any matched groups, this happens at the end of processing. In the case when
1001        there were back references that the ovector was too small to contain
1002        (causing temporary malloc'd memory to be used during matching), and the
1003        highest capturing number was not used, memory off the end of the ovector
1004        was incorrectly being set to -1. (It was using the size of the temporary
1005        memory instead of the true size.)
1007    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
1008        ovector size, it uses memory at the end of the block that it has got.
1010    29. Check for an overlong MARK name and give an error at compile time. The
1011        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
1013    30. JIT compiler update.
1015    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
1016        Rill for the patch.
1018    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
1020    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
1022    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
1023        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
1024        was enabled.
1026    35. Improve JIT code generation for greedy plus quantifier.
1028    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
1029        1 to "aa" instead of to an empty string. The bug affected repeated groups
1030        that could potentially match an empty string.
1032    37. Optimizing single character iterators in JIT.
1034    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
1035        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
1036        codepoints that are too big for the mode are faulted, and in a UTF mode,
1037        disallowed codepoints are also faulted.
1039    39. If PCRE was compiled with UTF support, in three places in the DFA
1040        matcher there was code that should only have been obeyed in UTF mode, but
1041        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
1042        processing when bytes with values greater than 127 were present. In 16-bit
1043        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
1044        both cases the values are those that cannot be the first data item in a UTF
1045        character. The three items that might have provoked this were recursions,
1046        possessively repeated groups, and atomic groups.
1048    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
1049        and pcregrep, because some OS require shared objects to be explicitly
1050        passed to ld, causing the link step to fail if they are not.
1052    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
1053        mode, patterns that started with \h* or \R* might be incorrectly matched.
1056    Version 8.30 04-February-2012
1057    -----------------------------
1059    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
1060        name is defined in ctype.h.
1062    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
1063        only in quite long subpatterns.
1065    3.  Removed the function pcre_info(), which has been obsolete and deprecated
1066        since it was replaced by pcre_fullinfo() in February 2000.
1068    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
1069        match a (*MARK), and the match failed at the start of the subject, a
1070        reference to memory before the start of the subject could occur. This bug
1071        was introduced by fix 17 of release 8.21.
1073    5.  A reference to an unset group with zero minimum repetition was giving
1074        totally wrong answers (in non-JavaScript-compatibility mode). For example,
1075        /(another)?(\1?)test/ matched against "hello world test". This bug was
1076        introduced in release 8.13.
1078    6.  Add support for 16-bit character strings (a large amount of work involving
1079        many changes and refactorings).
1081    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
1082        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
1083        from a file.
1085    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
1086        rounding is not applied in this particular case).
1088    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
1089        if they appear, or are escaped, in patterns.
1091    10. Get rid of a number of -Wunused-but-set-variable warnings.
1093    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
1094        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
1095        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
1096        also returns the mark "x". This bug applied to capturing parentheses,
1097        non-capturing parentheses, and atomic parentheses. It also applied to some
1098        assertions.
1100    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
1101        information out of configure.ac instead of relying on pcre.h.generic, which
1102        is not stored in the repository.
1104    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
1105        -lreadline.
1107    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
1109    15. Applied Graycode's patch to put the top-level frame on the stack rather
1110        than the heap when not using the stack for recursion. This gives a
1111        performance improvement in many cases when recursion is not deep.
1113    16. Experimental code added to "pcretest -C" to output the stack frame size.
1116    Version 8.21 12-Dec-2011
1117    ------------------------
1119    1.  Updating the JIT compiler.
1121    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
1122        are added as well.
1124    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
1125        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
1126        calling _pcre_jit_exec. Some extra comments are added.
1128    4.  (*MARK) settings inside atomic groups that do not contain any capturing
1129        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
1130        was introduced by change 18 for 8.20.
1132    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
1133        ECMA-262 standard.
1135    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
1136        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
1137        This bug was probably introduced by change 9 of 8.13.
1139    7.  While fixing 6 above, I noticed that a number of other items were being
1140        incorrectly rejected as "not fixed length". This arose partly because newer
1141        opcodes had not been added to the fixed-length checking code. I have (a)
1142        corrected the bug and added tests for these items, and (b) arranged for an
1143        error to occur if an unknown opcode is encountered while checking for fixed
1144        length instead of just assuming "not fixed length". The items that were
1145        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
1146        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
1147        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
1149    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
1150        being incorrectly compiled and would have given unpredicatble results.
1152    9.  A possessively repeated subpattern with minimum repeat count greater than
1153        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
1154        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
1155        the first (A) could occur when it should not.
1157    10. Add a cast and remove a redundant test from the code.
1159    11. JIT should use pcre_malloc/pcre_free for allocation.
1161    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
1162        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
1163        is anything other than /usr, -L is still shown).
1165    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
1167    14. Perl does not support \N without a following name in a [] class; PCRE now
1168        also gives an error.
1170    15. If a forward reference was repeated with an upper limit of around 2000,
1171        it caused the error "internal error: overran compiling workspace". The
1172        maximum number of forward references (including repeats) was limited by the
1173        internal workspace, and dependent on the LINK_SIZE. The code has been
1174        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
1175        the default depends on LINK_SIZE. There is a new upper limit (for safety)
1176        of around 200,000 forward references. While doing this, I also speeded up
1177        the filling in of repeated forward references.
1179    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
1180        incorrectly expecting the subject to contain another "a" after the start.
1182    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
1183        in the match, the SKIP should be ignored. This was not happening; instead
1184        the SKIP was being treated as NOMATCH. For patterns such as
1185        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
1186        tested.
1188    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
1189        now much more compatible with Perl, in particular in cases where the result
1190        is a non-match for a non-anchored pattern. For example, if
1191        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
1192        "m", where previously it did not return a name. A side effect of this
1193        change is that for partial matches, the last encountered mark name is
1194        returned, as for non matches. A number of tests that were previously not
1195        Perl-compatible have been moved into the Perl-compatible test files. The
1196        refactoring has had the pleasing side effect of removing one argument from
1197        the match() function, thus reducing its stack requirements.
1199    19. If the /S+ option was used in pcretest to study a pattern using JIT,
1200        subsequent uses of /S (without +) incorrectly behaved like /S+.
1202    21. Retrieve executable code size support for the JIT compiler and fixing
1203        some warnings.
1205    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
1206        not work when the shorter character appeared right at the end of the
1207        subject string.
1209    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
1210        systems.
1212    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
1213        output it when the /M option is used in pcretest.
1215    25. The CheckMan script was not being included in the distribution. Also, added
1216        an explicit "perl" to run Perl scripts from the PrepareRelease script
1217        because this is reportedly needed in Windows.
1219    26. If study data was being save in a file and studying had not found a set of
1220        "starts with" bytes for the pattern, the data written to the file (though
1221        never used) was taken from uninitialized memory and so caused valgrind to
1222        complain.
1224    27. Updated RunTest.bat as provided by Sheri Pierce.
1226    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
1228    29. Computation of memory usage for the table of capturing group names was
1229        giving an unnecessarily large value.
1232    Version 8.20 21-Oct-2011
1233    ------------------------
1235    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
1236        a POSIX class. After further experiments with Perl, which convinced me that
1237        Perl has bugs and confusions, a closing square bracket is no longer allowed
1238        in a POSIX name. This bug also affected patterns with classes that started
1239        with full stops.
1241    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
1242        captured substring, but while checking the failing first alternative,
1243        substring 1 is temporarily captured. If the output vector supplied to
1244        pcre_exec() was not big enough for this capture, the yield of the function
1245        was still zero ("insufficient space for captured substrings"). This cannot
1246        be totally fixed without adding another stack variable, which seems a lot
1247        of expense for a edge case. However, I have improved the situation in cases
1248        such as /(a)(b)x|abc/ matched against "abc", where the return code
1249        indicates that fewer than the maximum number of slots in the ovector have
1250        been set.
1252    3.  Related to (2) above: when there are more back references in a pattern than
1253        slots in the output vector, pcre_exec() uses temporary memory during
1254        matching, and copies in the captures as far as possible afterwards. It was
1255        using the entire output vector, but this conflicts with the specification
1256        that only 2/3 is used for passing back captured substrings. Now it uses
1257        only the first 2/3, for compatibility. This is, of course, another edge
1258        case.
1260    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
1261        main code base, and can be used by building with --enable-jit. When this is
1262        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
1263        runtime --no-jit option is given.
1265    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
1266        ovector, the return from the function was zero, implying that there were
1267        other matches that did not fit. The correct "exactly full" value is now
1268        returned.
1270    6.  If a subpattern that was called recursively or as a subroutine contained
1271        (*PRUNE) or any other control that caused it to give a non-standard return,
1272        invalid errors such as "Error -26 (nested recursion at the same subject
1273        position)" or even infinite loops could occur.
1275    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
1276        computing the minimum length on reaching *ACCEPT, and so ended up with the
1277        wrong value of 1 rather than 0. Further investigation indicates that
1278        computing a minimum subject length in the presence of *ACCEPT is difficult
1279        (think back references, subroutine calls), and so I have changed the code
1280        so that no minimum is registered for a pattern that contains *ACCEPT.
1282    8.  If (*THEN) was present in the first (true) branch of a conditional group,
1283        it was not handled as intended. [But see 16 below.]
1285    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
1286        Sheri Pierce.
1288    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
1289        the first byte in a match must be "a".
1291    11. Change 17 for 8.13 increased the recursion depth for patterns like
1292        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
1293        pattern contains any instances of (*THEN). If it does not, the old
1294        optimizations are restored. It would be nice to do this on a per-group
1295        basis, but at the moment that is not feasible.
1297    12. In some environments, the output of pcretest -C is CRLF terminated. This
1298        broke RunTest's code that checks for the link size. A single white space
1299        character after the value is now allowed for.
1301    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
1302        For "fr", it uses the Windows-specific input and output files.
1304    14. If (*THEN) appeared in a group that was called recursively or as a
1305        subroutine, it did not work as intended. [But see next item.]
1307    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
1308        pattern fragments (but not containing any | characters). If A and B are
1309        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
1310        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
1311        D. In other words, Perl considers parentheses that do not contain any |
1312        characters to be part of a surrounding alternative, whereas PCRE was
1313        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
1314        differently. PCRE now behaves in the same way as Perl, except in the case
1315        of subroutine/recursion calls such as (?1) which have in any case always
1316        been different (but PCRE had them first :-).
1318    16. Related to 15 above: Perl does not treat the | in a conditional group as
1319        creating alternatives. Such a group is treated in the same way as an
1320        ordinary group without any | characters when processing (*THEN). PCRE has
1321        been changed to match Perl's behaviour.
1323    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
1324        RunGrepTest script failed.
1326    18. Change 22 for version 13 caused atomic groups to use more stack. This is
1327        inevitable for groups that contain captures, but it can lead to a lot of
1328        stack use in large patterns. The old behaviour has been restored for atomic
1329        groups that do not contain any capturing parentheses.
1331    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
1332        suppress the check for a minimum subject length at run time. (If it was
1333        given to pcre_exec() or pcre_dfa_exec() it did work.)
1335    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
1336        fail to work when decoding hex characters in data strings in EBCDIC
1337        environments.
1339    21. It appears that in at least one Mac OS environment, the isxdigit() function
1340        is implemented as a macro that evaluates to its argument more than once,
1341        contravening the C 90 Standard (I haven't checked a later standard). There
1342        was an instance in pcretest which caused it to go wrong when processing
1343        \x{...} escapes in subject strings. The has been rewritten to avoid using
1344        things like p++ in the argument of isxdigit().
1347    Version 8.13 16-Aug-2011
1348    ------------------------
1350    1.  The Unicode data tables have been updated to Unicode 6.0.0.
1352    2.  Two minor typos in pcre_internal.h have been fixed.
1354    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
1355        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
1356        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
1358    4.  There were a number of related bugs in the code for matching backrefences
1359        caselessly in UTF-8 mode when codes for the characters concerned were
1360        different numbers of bytes. For example, U+023A and U+2C65 are an upper
1361        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
1362        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
1363        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
1364        2-byte code at the end of the subject (it thought there wasn't enough data
1365        left).
1367    5.  Comprehensive information about what went wrong is now returned by
1368        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
1369        as the output vector has at least 2 elements. The offset of the start of
1370        the failing character and a reason code are placed in the vector.
1372    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
1373        now returned is for the first byte of the failing character, instead of the
1374        last byte inspected. This is an incompatible change, but I hope it is small
1375        enough not to be a problem. It makes the returned offset consistent with
1376        pcre_exec() and pcre_dfa_exec().
1378    7.  pcretest now gives a text phrase as well as the error number when
1379        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
1380        failure, the offset and reason code are output.
1382    8.  When \R was used with a maximizing quantifier it failed to skip backwards
1383        over a \r\n pair if the subsequent match failed. Instead, it just skipped
1384        back over a single character (\n). This seems wrong (because it treated the
1385        two characters as a single entity when going forwards), conflicts with the
1386        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
1387        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
1388        has been changed.
1390    9.  Some internal refactoring has changed the processing so that the handling
1391        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
1392        time (the PCRE_DOTALL option was changed this way some time ago: version
1393        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
1394        which was always a bit of a fudge. It also means that there is one less
1395        argument for the match() function, which reduces its stack requirements
1396        slightly. This change also fixes an incompatibility with Perl: the pattern
1397        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
1399    10. More internal refactoring has drastically reduced the number of recursive
1400        calls to match() for possessively repeated groups such as (abc)++ when
1401        using pcre_exec().
1403    11. While implementing 10, a number of bugs in the handling of groups were
1404        discovered and fixed:
1406        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
1407        (a|)*(?1) gave a compile-time internal error.
1408        ((a|)+)+  did not notice that the outer group could match an empty string.
1409        (^a|^)+   was not marked as anchored.
1410        (.*a|.*)+ was not marked as matching at start or after a newline.
1412    12. Yet more internal refactoring has removed another argument from the match()
1413        function. Special calls to this function are now indicated by setting a
1414        value in a variable in the "match data" data block.
1416    13. Be more explicit in pcre_study() instead of relying on "default" for
1417        opcodes that mean there is no starting character; this means that when new
1418        ones are added and accidentally left out of pcre_study(), testing should
1419        pick them up.
1421    14. The -s option of pcretest has been documented for ages as being an old
1422        synonym of -m (show memory usage). I have changed it to mean "force study
1423        for every regex", that is, assume /S for every regex. This is similar to -i
1424        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
1425        using it. It makes it easier to run collections of tests with and without
1426        study enabled, and thereby test pcre_study() more easily. All the standard
1427        tests are now run with and without -s (but some patterns can be marked as
1428        "never study" - see 20 below).
1430    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
1431        restoration of the capturing data to the outer values was not happening
1432        correctly.
1434    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
1435        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
1436        pattern had matched an empty string, and so incorrectly returned a no
1437        match.
1439    17. There was optimizing code for the last branch of non-capturing parentheses,
1440        and also for the obeyed branch of a conditional subexpression, which used
1441        tail recursion to cut down on stack usage. Unfortunately, now that there is
1442        the possibility of (*THEN) occurring in these branches, tail recursion is
1443        no longer possible because the return has to be checked for (*THEN). These
1444        two optimizations have therefore been removed. [But see 8.20/11 above.]
1446    18. If a pattern containing \R was studied, it was assumed that \R always
1447        matched two bytes, thus causing the minimum subject length to be
1448        incorrectly computed because \R can also match just one byte.
1450    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
1451        was incorrectly computed.
1453    20. If /S is present twice on a test pattern in pcretest input, it now
1454        *disables* studying, thereby overriding the use of -s on the command line
1455        (see 14 above). This is necessary for one or two tests to keep the output
1456        identical in both cases.
1458    21. When (*ACCEPT) was used in an assertion that matched an empty string and
1459        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
1461    22. When an atomic group that contained a capturing parenthesis was
1462        successfully matched, but the branch in which it appeared failed, the
1463        capturing was not being forgotten if a higher numbered group was later
1464        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
1465        group 1 to "a", when in fact it should be unset. This applied to multi-
1466        branched capturing and non-capturing groups, repeated or not, and also to
1467        positive assertions (capturing in negative assertions does not happen
1468        in PCRE) and also to nested atomic groups.
1470    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
1471        subject after a captured substring, to make it easier to tell which of a
1472        number of identical substrings has been captured.
1474    24. The way atomic groups are processed by pcre_exec() has been changed so that
1475        if they are repeated, backtracking one repetition now resets captured
1476        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
1477        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
1478        "aaa". Previously, it would have been "a". As part of this code
1479        refactoring, the way recursive calls are handled has also been changed.
1481    25. If an assertion condition captured any substrings, they were not passed
1482        back unless some other capturing happened later. For example, if
1483        (?(?=(a))a) was matched against "a", no capturing was returned.
1485    26. When studying a pattern that contained subroutine calls or assertions,
1486        the code for finding the minimum length of a possible match was handling
1487        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
1488        group 1 called group 2 while simultaneously a separate group 2 called group
1489        1). A stack overflow occurred in this case. I have fixed this by limiting
1490        the recursion depth to 10.
1492    27. Updated RunTest.bat in the distribution to the version supplied by Tom
1493        Fortmann. This supports explicit test numbers on the command line, and has
1494        argument validation and error reporting.
1496    28. An instance of \X with an unlimited repeat could fail if at any point the
1497        first character it looked at was a mark character.
1499    29. Some minor code refactoring concerning Unicode properties and scripts
1500        should reduce the stack requirement of match() slightly.
1502    30. Added the '=' option to pcretest to check the setting of unused capturing
1503        slots at the end of the pattern, which are documented as being -1, but are
1504        not included in the return count.
1506    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
1507        compiled something random. Now it gives a compile-time error (as does
1508        Perl).
1510    32. A *MARK encountered during the processing of a positive assertion is now
1511        recorded and passed back (compatible with Perl).
1513    33. If --only-matching or --colour was set on a pcregrep call whose pattern
1514        had alternative anchored branches, the search for a second match in a line
1515        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
1516        matched the line "0102" twice. The same bug affected patterns that started
1517        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
1518        twice.
1520    34. Previously, PCRE did not allow quantification of assertions. However, Perl
1521        does, and because of capturing effects, quantifying parenthesized
1522        assertions may at times be useful. Quantifiers are now allowed for
1523        parenthesized assertions.
1525    35. A minor code tidy in pcre_compile() when checking options for \R usage.
1527    36. \g was being checked for fancy things in a character class, when it should
1528        just be a literal "g".
1530    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
1531        appearance of a nested POSIX class supersedes an apparent external class.
1532        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
1533        unescaped square brackets may also appear as part of class names. For
1534        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
1535        more like Perl. (But see 8.20/1 above.)
1537    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
1538        was because it thought it was \N{name}, which is not supported).
1540    39. Add minix to OS list not supporting the -S option in pcretest.
1542    40. PCRE tries to detect cases of infinite recursion at compile time, but it
1543        cannot analyze patterns in sufficient detail to catch mutual recursions
1544        such as ((?1))((?2)). There is now a runtime test that gives an error if a
1545        subgroup is called recursively as a subpattern for a second time at the
1546        same position in the subject string. In previous releases this might have
1547        been caught by the recursion limit, or it might have run out of stack.
1549    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
1550        happen only once. PCRE was, however incorrectly giving a compile time error
1551        "recursive call could loop indefinitely" because it cannot analyze the
1552        pattern in sufficient detail. The compile time test no longer happens when
1553        PCRE is compiling a conditional subpattern, but actual runaway loops are
1554        now caught at runtime (see 40 above).
1556    42. It seems that Perl allows any characters other than a closing parenthesis
1557        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
1558        has been changed to be the same.
1560    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
1561        as not to get warnings when autogen.sh is called. Also changed
1562        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
1564    44. To help people who use pcregrep to scan files containing exceedingly long
1565        lines, the following changes have been made:
1567        (a) The default value of the buffer size parameter has been increased from
1568            8K to 20K. (The actual buffer used is three times this size.)
1570        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
1571            PCRE is built.
1573        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
1574            to be set at run time.
1576        (d) Numerical values in pcregrep options can be followed by K or M, for
1577            example --buffer-size=50K.
1579        (e) If a line being scanned overflows pcregrep's buffer, an error is now
1580            given and the return code is set to 2.
1582    45. Add a pointer to the latest mark to the callout data block.
1584    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
1585        partial match of an empty string instead of no match. This was specific to
1586        the use of ".".
1588    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
1589        complete match instead of a partial match. This bug was dependent on both
1590        the PCRE_UTF8 and PCRE_DOTALL options being set.
1592    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
1593        starting byte set, because \b was not being ignored.
1596    Version 8.12 15-Jan-2011
1597    ------------------------
1599    1.  Fixed some typos in the markup of the man pages, and wrote a script that
1600        checks for such things as part of the documentation building process.
1602    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
1603        --match-limit and --recursion-limit options (added for 8.11). In
1604        particular, this made one of the standard tests fail. (The integer value
1605        went into the wrong half of a long int.)
1607    3.  If the --colour option was given to pcregrep with -v (invert match), it
1608        did strange things, either producing crazy output, or crashing. It should,
1609        of course, ignore a request for colour when reporting lines that do not
1610        match.
1612    4.  Another pcregrep bug caused similar problems if --colour was specified with
1613        -M (multiline) and the pattern match finished with a line ending.
1615    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
1616        matched in multiline mode, the following line was shown as part of the
1617        match. This seems wrong, so I have changed it.
1619    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
1620        the check for further matches in the same line (so they could be coloured)
1621        to overrun the end of the current line. If another match was found, it was
1622        incorrectly shown (and then shown again when found in the next line).
1624    7.  If pcregrep was compiled under Windows, there was a reference to the
1625        function pcregrep_exit() before it was defined. I am assuming this was
1626        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
1627        reported by a user. I've moved the definition above the reference.
1630    Version 8.11 10-Dec-2010
1631    ------------------------
1633    1.  (*THEN) was not working properly if there were untried alternatives prior
1634        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
1635        backtracked to try for "b" instead of moving to the next alternative branch
1636        at the same level (in this case, to look for "c"). The Perl documentation
1637        is clear that when (*THEN) is backtracked onto, it goes to the "next
1638        alternative in the innermost enclosing group".
1640    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
1641        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
1642        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
1643        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
1644        (*THEN).
1646    3.  If \s appeared in a character class, it removed the VT character from
1647        the class, even if it had been included by some previous item, for example
1648        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
1649        of \s, but is part of the POSIX "space" class.)
1651    4.  A partial match never returns an empty string (because you can always
1652        match an empty string at the end of the subject); however the checking for
1653        an empty string was starting at the "start of match" point. This has been
1654        changed to the "earliest inspected character" point, because the returned
1655        data for a partial match starts at this character. This means that, for
1656        example, /(?<=abc)def/ gives a partial match for the subject "abc"
1657        (previously it gave "no match").
1659    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
1660        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
1661        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
1662        has an implication that the given string is incomplete (because a partial
1663        match is preferred over a full match). For this reason, these items now
1664        give a partial match in this situation. [Aside: previously, the one case
1665        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
1666        match rather than a full match, which was wrong by the old rules, but is
1667        now correct.]
1669    6.  There was a bug in the handling of #-introduced comments, recognized when
1670        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
1671        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
1672        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
1673        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
1674        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
1675        places in pcre_compile().
1677    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
1678        comments when looking ahead for named forward references to subpatterns,
1679        the only newline sequence it recognized was NL. It now handles newlines
1680        according to the set newline convention.
1682    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
1683        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
1684        cater for a lack of strerror(). These oversights have been fixed.
1686    9.  Added --match-limit and --recursion-limit to pcregrep.
1688    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
1690    11. When the -o option was used, pcregrep was setting a return code of 1, even
1691        when matches were found, and --line-buffered was not being honoured.
1693    12. Added an optional parentheses number to the -o and --only-matching options
1694        of pcregrep.
1696    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
1697        can match an empty string. The code to do it in pcretest and pcredemo
1698        needed fixing:
1700        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
1701            only one byte after an empty string match just before CRLF (this case
1702            just got forgotten; "any" and "anycrlf" were OK).
1704        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
1705            mode when an empty string match preceded an ASCII character followed by
1706            a non-ASCII character. (The code for advancing by one character rather
1707            than one byte was nonsense.)
1709        (c) The pcredemo.c sample program did not have any code at all to handle
1710            the cases when CRLF is a valid newline sequence.
1712    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
1713        as a starting offset was within the subject string. There is now a new
1714        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
1715        negative or greater than the length of the string. In order to test this,
1716        pcretest is extended to allow the setting of negative starting offsets.
1718    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
1719        starting offset points to the beginning of a UTF-8 character was
1720        unnecessarily clumsy. I tidied it up.
1722    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
1723        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
1725    17. Nobody had reported that the --include_dir option, which was added in
1726        release 7.7 should have been called --include-dir (hyphen, not underscore)
1727        for compatibility with GNU grep. I have changed it to --include-dir, but
1728        left --include_dir as an undocumented synonym, and the same for
1729        --exclude-dir, though that is not available in GNU grep, at least as of
1730        release 2.5.4.
1732    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
1733        characters from a string of bytes) have been redefined so as not to use
1734        loops, in order to improve performance in some environments. At the same
1735        time, I abstracted some of the common code into auxiliary macros to save
1736        repetition (this should not affect the compiled code).
1738    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1739        compile-time error is now given if \c is not followed by an ASCII
1740        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1741        different, and any byte value is allowed.)
1743    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1744        START_OPTIMIZE option, which is now allowed at compile time - but just
1745        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1746        to pcregrep and other applications that have no direct access to PCRE
1747        options. The new /Y option in pcretest sets this option when calling
1748        pcre_compile().
1750    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1751        back references. Groups containing recursive back references were forced to
1752        be atomic by that change, but in the case of named groups, the amount of
1753        memory required was incorrectly computed, leading to "Failed: internal
1754        error: code overflow". This has been fixed.
1756    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1757        pcretest.c, to avoid build problems in some Borland environments.
1760    Version 8.10 25-Jun-2010
1761    ------------------------
1763    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1764        THEN.
1766    2.  (*ACCEPT) was not working when inside an atomic group.
1768    3.  Inside a character class, \B is treated as a literal by default, but
1769        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1770        causes the error). The code is unchanged, but I tidied the documentation.
1772    4.  Inside a character class, PCRE always treated \R and \X as literals,
1773        whereas Perl faults them if its -w option is set. I have changed PCRE so
1774        that it faults them when PCRE_EXTRA is set.
1776    5.  Added support for \N, which always matches any character other than
1777        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1779    6.  When compiling pcregrep with newer versions of gcc which may have
1780        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1781        declared with attribute warn_unused_result" were given. Just casting the
1782        result to (void) does not stop the warnings; a more elaborate fudge is
1783        needed. I've used a macro to implement this.
1785    7.  Minor change to pcretest.c to avoid a compiler warning.
1787    8.  Added four artifical Unicode properties to help with an option to make
1788        \s etc use properties (see next item). The new properties are: Xan
1789        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1791    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1792        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1793        this option. Modified pcretest to add /W to test this facility. Added
1794        REG_UCP to make it available via the POSIX interface.
1796    10. Added --line-buffered to pcregrep.
1798    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1799        studied, and the match started with a letter with a code point greater than
1800        127 whose first byte was different to the first byte of the other case of
1801        the letter, the other case of this starting letter was not recognized
1802        (#976).
1804    12. If a pattern that was studied started with a repeated Unicode property
1805        test, for example, \p{Nd}+, there was the theoretical possibility of
1806        setting up an incorrect bitmap of starting bytes, but fortunately it could
1807        not have actually happened in practice until change 8 above was made (it
1808        added property types that matched character-matching opcodes).
1810    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1811        possible starting bytes for non-anchored patterns.
1813    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1814        \R, and also a number of cases that involve Unicode properties, both
1815        explicit and implicit when PCRE_UCP is set.
1817    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1818        input, it could crash or give wrong results if characters with values
1819        greater than 0xc0 were present in the subject string. (Detail: it assumed
1820        UTF-8 input when processing these items.)
1822    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1823        size_t is 64-bit (#991).
1825    17. Added a check for running out of memory when PCRE is compiled with
1826        --disable-stack-for-recursion (#990).
1828    18. If the last data line in a file for pcretest does not have a newline on
1829        the end, a newline was missing in the output.
1831    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1832        less than 128) in its various bitmaps. However, there is a facility for
1833        generating tables according to the current locale when PCRE is compiled. It
1834        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1835        characters, are recognized by isspace() and therefore were getting set in
1836        these tables, and indeed these tables seem to approximate to ISO 8859. This
1837        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1838        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1839        which of course cannot start UTF-8 characters. I have changed the code so
1840        that only real ASCII characters (less than 128) and the correct starting
1841        bytes for UTF-8 encodings are set for characters greater than 127 when in
1842        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1843        altogether.)
1845    20. Added the /T option to pcretest so as to be able to run tests with non-
1846        standard character tables, thus making it possible to include the tests
1847        used for 19 above in the standard set of tests.
1849    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1850        reference to a subpattern the other side of a comment that contains an
1851        opening parenthesis caused either an internal compiling error, or a
1852        reference to the wrong subpattern.
1855    Version 8.02 19-Mar-2010
1856    ------------------------
1858    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1860    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1861        configured.
1863    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1864        original author of that file, following a query about its status.
1866    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1867        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1869    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1870        quantifier applied to a forward-referencing subroutine call, could compile
1871        incorrect code or give the error "internal error: previously-checked
1872        referenced subpattern not found".
1874    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1875        variables to point to external functions. For these systems, therefore,
1876        pcre_malloc etc. are now initialized to local functions that call the
1877        relevant global functions.
1879    7.  There were two entries missing in the vectors called coptable and poptable
1880        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1881        I've fixed the data, and added a kludgy way of testing at compile time that
1882        the lengths are correct (equal to the number of opcodes).
1884    8.  Following on from 7, I added a similar kludge to check the length of the
1885        eint vector in pcreposix.c.
1887    9.  Error texts for pcre_compile() are held as one long string to avoid too
1888        much relocation at load time. To find a text, the string is searched,
1889        counting zeros. There was no check for running off the end of the string,
1890        which could happen if a new error number was added without updating the
1891        string.
1893    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1895    11. \K was not working if it appeared in an atomic group or in a group that
1896        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1897        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1898        the assertion is positive, but not if it is negative.
1900    12. Change 11 fortuitously reduced the size of the stack frame used in the
1901        "match()" function of pcre_exec.c by one pointer. Forthcoming
1902        implementation of support for (*MARK) will need an extra pointer on the
1903        stack; I have reserved it now, so that the stack frame size does not
1904        decrease.
1906    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1907        item in branch that calls a recursion is a subroutine call - as in the
1908        second branch in the above example - was incorrectly given the compile-
1909        time error "recursive call could loop indefinitely" because pcre_compile()
1910        was not correctly checking the subroutine for matching a non-empty string.
1912    14. The checks for overrunning compiling workspace could trigger after an
1913        overrun had occurred. This is a "should never occur" error, but it can be
1914        triggered by pathological patterns such as hundreds of nested parentheses.
1915        The checks now trigger 100 bytes before the end of the workspace.
1917    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1920    Version 8.01 19-Jan-2010
1921    ------------------------
1923    1.  If a pattern contained a conditional subpattern with only one branch (in
1924        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1925        computed the wrong minimum data length (which is of course zero for such
1926        subpatterns). This could cause incorrect "no match" results.
1928    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1929        the pattern is reset in the first branch, pcre_compile() failed with
1930        "internal error: code overflow at offset...". This happened only when
1931        the reset was to the original external option setting. (An optimization
1932        abstracts leading options settings into an external setting, which was the
1933        cause of this.)
1935    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1936        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1937        assertion pattern did not match (meaning that the assertion was true), it
1938        was incorrectly treated as false if the SKIP had been reached during the
1939        matching. This also applied to assertions used as conditions.
1941    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1942        assertion subpattern, including such a pattern used as a condition,
1943        unpredictable results occurred, instead of the error return
1946    5.  The C++ GlobalReplace function was not working like Perl for the special
1947        situation when an empty string is matched. It now does the fancy magic
1948        stuff that is necessary.
1950    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1951        removed. (These were left over from very, very early versions of PCRE.)
1953    7.  Some cosmetic changes to the code to make life easier when compiling it
1954        as part of something else:
1956        (a) Change DEBUG to PCRE_DEBUG.
1958        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1959            called "current" as "current_branch", to prevent a collision with the
1960            Linux macro when compiled as a kernel module.
1962        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1963            prevent a collision with the Linux macro when compiled as a kernel
1964            module.
1966    8.  In pcre_compile() there are some checks for integer overflows that used to
1967        cast potentially large values to (double). This has been changed to that
1968        when building, a check for int64_t is made, and if it is found, it is used
1969        instead, thus avoiding the use of floating point arithmetic. (There is no
1970        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1971        double.
1973    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1974        2005 (difference between two addresses compared to an unsigned value).
1976    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1977        custom one, because of the following reported problem in Windows:
1979          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1980              under Win32.
1981          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1982              therefore missing the function definition.
1983          - The compiler thus generates a "C" signature for the test function.
1984          - The linker fails to find the "C" function.
1985          - PCRE fails to configure if asked to do so against libbz2.
1987    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1988        messages were output:
1990          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1991          rerunning libtoolize, to keep the correct libtool macros in-tree.
1992          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1994        I have done both of these things.
1996    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1997        most of the time, it *can* run out if it is given a pattern that contains a
1998        runaway infinite recursion. I updated the discussion in the pcrestack man
1999        page.
2001    13. Now that we have gone to the x.xx style of version numbers, the minor
2002        version may start with zero. Using 08 or 09 is a bad idea because users
2003        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
2004        interpreted as invalid octal numbers. I've updated the previous comment in
2005        configure.ac, and also added a check that gives an error if 08 or 09 are
2006        used.
2008    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
2009        causing partial matching to fail when the end of the subject matched \W
2010        in a UTF-8 pattern where \W was quantified with a minimum of 3.
2012    15. There were some discrepancies between the declarations in pcre_internal.h
2013        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
2014        their definitions. The declarations used "const uschar *" and the
2015        definitions used USPTR. Even though USPTR is normally defined as "const
2016        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
2017        reported that: "This difference in casting confuses some C++ compilers, for
2018        example, SunCC recognizes above declarations as different functions and
2019        generates broken code for hbpcre." I have changed the declarations to use
2020        USPTR.
2022    16. GNU libtool is named differently on some systems. The autogen.sh script now
2023        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
2024        (FreeBSD).
2026    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
2027        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
2028        comment: "Figure out how to create a longlong from a string: strtoll and
2029        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
2030        instance, but it only takes 2 args instead of 3!"
2032    18. A subtle bug concerned with back references has been fixed by a change of
2033        specification, with a corresponding code fix. A pattern such as
2034        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
2035        refers, was giving matches when it shouldn't. For example, xa=xaaa would
2036        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
2037        same bug. Such groups have to be quantified to be useful, or contained
2038        inside another quantified group. (If there's no repetition, the reference
2039        can never match.) The problem arises because, having left the group and
2040        moved on to the rest of the pattern, a later failure that backtracks into
2041        the group uses the captured value from the final iteration of the group
2042        rather than the correct earlier one. I have fixed this in PCRE by forcing
2043        any group that contains a reference to itself to be an atomic group; that
2044        is, there cannot be any backtracking into it once it has completed. This is
2045        similar to recursive and subroutine calls.
2048    Version 8.00 19-Oct-09
2049    ----------------------
2051    1.  The table for translating pcre_compile() error codes into POSIX error codes
2052        was out-of-date, and there was no check on the pcre_compile() error code
2053        being within the table. This could lead to an OK return being given in
2054        error.
2056    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
2057        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
2058        in a Windows environment.
2060    3.  The pcregrep --count option prints the count for each file even when it is
2061        zero, as does GNU grep. However, pcregrep was also printing all files when
2062        --files-with-matches was added. Now, when both options are given, it prints
2063        counts only for those files that have at least one match. (GNU grep just
2064        prints the file name in this circumstance, but including the count seems
2065        more useful - otherwise, why use --count?) Also ensured that the
2066        combination -clh just lists non-zero counts, with no names.
2068    4.  The long form of the pcregrep -F option was incorrectly implemented as
2069        --fixed_strings instead of --fixed-strings. This is an incompatible change,
2070        but it seems right to fix it, and I didn't think it was worth preserving
2071        the old behaviour.
2073    5.  The command line items --regex=pattern and --regexp=pattern were not
2074        recognized by pcregrep, which required --regex pattern or --regexp pattern
2075        (with a space rather than an '='). The man page documented the '=' forms,
2076        which are compatible with GNU grep; these now work.
2078    6.  No libpcreposix.pc file was created for pkg-config; there was just
2079        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
2081    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
2082        when UCP support is not needed, by modifying the Python script that
2083        generates it from Unicode data files. This should not matter if the module
2084        is correctly used as a library, but I received one complaint about 50K of
2085        unwanted data. My guess is that the person linked everything into his
2086        program rather than using a library. Anyway, it does no harm.
2088    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
2089        was a minimum greater than 1 for a wide character in a possessive
2090        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
2091        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
2092        character. Chaos in the form of incorrect output or a compiling loop could
2093        result.
2095    9.  The restrictions on what a pattern can contain when partial matching is
2096        requested for pcre_exec() have been removed. All patterns can now be
2097        partially matched by this function. In addition, if there are at least two
2098        slots in the offset vector, the offset of the earliest inspected character
2099        for the match and the offset of the end of the subject are set in them when
2100        PCRE_ERROR_PARTIAL is returned.
2102    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
2103        synonymous with PCRE_PARTIAL, for backwards compatibility, and
2104        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
2105        and may be more useful for multi-segment matching.
2107    11. Partial matching with pcre_exec() is now more intuitive. A partial match
2108        used to be given if ever the end of the subject was reached; now it is
2109        given only if matching could not proceed because another character was
2110        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
2111        string "Z", which now yields "no match" instead of "partial match". In the
2112        case of pcre_dfa_exec(), "no match" is given if every matching path for the
2113        final character ended with (*FAIL).
2115    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
2116        if the pattern had a "must contain" character that was already found in the
2117        earlier partial match, unless partial matching was again requested. For
2118        example, with the pattern /dog.(body)?/, the "must contain" character is
2119        "g". If the first part-match was for the string "dog", restarting with
2120        "sbody" failed. This bug has been fixed.
2122    13. The string returned by pcre_dfa_exec() after a partial match has been
2123        changed so that it starts at the first inspected character rather than the
2124        first character of the match. This makes a difference only if the pattern
2125        starts with a lookbehind assertion or \b or \B (\K is not supported by
2126        pcre_dfa_exec()). It's an incompatible change, but it makes the two
2127        matching functions compatible, and I think it's the right thing to do.
2129    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
2130        so that the demonstration program is easily available in environments where
2131        PCRE has not been installed from source.
2133    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
2134        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
2135        library.
2137    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
2138        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
2139        is not the first non-POSIX option to be added. Clearly some people find
2140        these options useful.
2142    17. If a caller to the POSIX matching function regexec() passes a non-zero
2143        value for nmatch with a NULL value for pmatch, the value of
2144        nmatch is forced to zero.
2146    18. RunGrepTest did not have a test for the availability of the -u option of
2147        the diff command, as RunTest does. It now checks in the same way as
2148        RunTest, and also checks for the -b option.
2150    19. If an odd number of negated classes containing just a single character
2151        interposed, within parentheses, between a forward reference to a named
2152        subpattern and the definition of the subpattern, compilation crashed with
2153        an internal error, complaining that it could not find the referenced
2154        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
2155        [The bug was that it was starting one character too far in when skipping
2156        over the character class, thus treating the ] as data rather than
2157        terminating the class. This meant it could skip too much.]
2159    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
2160        /g option in pcretest when the pattern contains \K, which makes it possible
2161        to have an empty string match not at the start, even when the pattern is
2162        anchored. Updated pcretest and pcredemo to use this option.
2164    21. If the maximum number of capturing subpatterns in a recursion was greater
2165        than the maximum at the outer level, the higher number was returned, but
2166        with unset values at the outer level. The correct (outer level) value is
2167        now given.
2169    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
2170        PCRE did not set those parentheses (unlike Perl). I have now found a way to
2171        make it do so. The string so far is captured, making this feature
2172        compatible with Perl.
2174    23. The tests have been re-organized, adding tests 11 and 12, to make it
2175        possible to check the Perl 5.10 features against Perl 5.10.
2177    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
2178        pattern matches a fixed length string. PCRE did not allow this; now it
2179        does. Neither allows recursion.
2181    25. I finally figured out how to implement a request to provide the minimum
2182        length of subject string that was needed in order to match a given pattern.
2183        (It was back references and recursion that I had previously got hung up
2184        on.) This code has now been added to pcre_study(); it finds a lower bound
2185        to the length of subject needed. It is not necessarily the greatest lower
2186        bound, but using it to avoid searching strings that are too short does give
2187        some useful speed-ups. The value is available to calling programs via
2188        pcre_fullinfo().
2190    26. While implementing 25, I discovered to my embarrassment that pcretest had
2191        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
2192        study optimizations had never been tested with that matching function.
2193        Oops. What is worse, even when it was passed study data, there was a bug in
2194        pcre_dfa_exec() that meant it never actually used it. Double oops. There
2195        were also very few tests of studied patterns with pcre_dfa_exec().
2197    27. If (?| is used to create subpatterns with duplicate numbers, they are now
2198        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
2199        on the other side of the coin, they are no longer allowed to have different
2200        names, because these cannot be distinguished in PCRE, and this has caused
2201        confusion. (This is a difference from Perl.)
2203    28. When duplicate subpattern names are present (necessarily with different
2204        numbers, as required by 27 above), and a test is made by name in a
2205        conditional pattern, either for a subpattern having been matched, or for
2206        recursion in such a pattern, all the associated numbered subpatterns are
2207        tested, and the overall condition is true if the condition is true for any
2208        one of them. This is the way Perl works, and is also more like the way
2209        testing by number works.
2212    Version 7.9 11-Apr-09
2213    ---------------------
2215    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
2216        (pcretest), all targets were linked against these libraries. This included
2217        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
2218        libraries. This caused unwanted dependencies to be created. This problem
2219        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
2220        pcretest is linked with readline.
2222    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
2223        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
2224        moved outside it again, because FALSE and TRUE are already defined in AIX,
2225        but BOOL is not.
2227    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
2228        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
2230    4.  The pcregrep documentation said spaces were inserted as well as colons (or
2231        hyphens) following file names and line numbers when outputting matching
2232        lines. This is not true; no spaces are inserted. I have also clarified the
2233        wording for the --colour (or --color) option.
2235    5.  In pcregrep, when --colour was used with -o, the list of matching strings
2236        was not coloured; this is different to GNU grep, so I have changed it to be
2237        the same.
2239    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
2240        each matching line was coloured. Now it goes on to look for further matches
2241        of any of the test patterns, which is the same behaviour as GNU grep.
2243    7.  A pattern that could match an empty string could cause pcregrep to loop; it
2244        doesn't make sense to accept an empty string match in pcregrep, so I have
2245        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
2246        seems to be how GNU grep behaves. [But see later change 40 for release
2247        8.33.]
2249    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
2250        start or after a newline", because the conditional assertion was not being
2251        correctly handled. The rule now is that both the assertion and what follows
2252        in the first alternative must satisfy the test.
2254    9.  If auto-callout was enabled in a pattern with a conditional group whose
2255        condition was an assertion, PCRE could crash during matching, both with
2256        pcre_exec() and pcre_dfa_exec().
2258    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
2259        used for matching.
2261    11. Unicode property support in character classes was not working for
2262        characters (bytes) greater than 127 when not in UTF-8 mode.
2264    12. Added the -M command line option to pcretest.
2266    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
2268    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
2270    16. Added comments and documentation about mis-use of no_arg in the C++
2271        wrapper.
2273    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
2274        from Martin Jerabek that uses macro names for all relevant character and
2275        string constants.
2277    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
2278        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
2279        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
2280        these, but not everybody uses configure.
2282    19. A conditional group that had only one branch was not being correctly
2283        recognized as an item that could match an empty string. This meant that an
2284        enclosing group might also not be so recognized, causing infinite looping
2285        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
2286        with the subject "ab", where knowledge that the repeated group can match
2287        nothing is needed in order to break the loop.
2289    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
2290        exec(), but without supplying a callout function, matching went wrong.
2292    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
2293        leak if the size of the offset vector was greater than 30. When the vector
2294        is smaller, the saved offsets during recursion go onto a local stack
2295        vector, but for larger vectors malloc() is used. It was failing to free
2296        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
2297        error, in fact).
2299    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
2300        heapframe that is used only when UTF-8 support is enabled. This caused no
2301        problem, but was untidy.
2303    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
2304        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
2305        included within another project.
2307    24. Steven Van Ingelgem's patches to add more options to the CMake support,
2308        slightly modified by me:
2310          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
2311              not building pcregrep.
2313          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
2314              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
2316    25. Forward references, both numeric and by name, in patterns that made use of
2317        duplicate group numbers, could behave incorrectly or give incorrect errors,
2318        because when scanning forward to find the reference group, PCRE was not
2319        taking into account the duplicate group numbers. A pattern such as
2320        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
2322    26. Changed a few more instances of "const unsigned char *" to USPTR, making
2323        the feature of a custom pointer more persuasive (as requested by a user).
2325    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
2326        pcretest.c, inside #ifndefs, because it seems they are sometimes already
2327        pre-defined.
2329    28. Added support for (*UTF8) at the start of a pattern.
2331    29. Arrange for flags added by the "release type" setting in CMake to be shown
2332        in the configuration summary.
2335    Version 7.8 05-Sep-08
2336    ---------------------
2338    1.  Replaced UCP searching code with optimized version as implemented for Ad
2339        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
2340        stage table and inline lookup instead of a function, giving speed ups of 2
2341        to 5 times on some simple patterns that I tested. Permission was given to
2342        distribute the MultiStage2.py script that generates the tables (it's not in
2343        the tarball, but is in the Subversion repository).
2345    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
2346        scripts.
2348    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
2349        a group with a zero qualifier. The result of the study could be incorrect,
2350        or the function might crash, depending on the pattern.
2352    4.  Caseless matching was not working for non-ASCII characters in back
2353        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
2354        It now works when Unicode Property Support is available.
2356    5.  In pcretest, an escape such as \x{de} in the data was always generating
2357        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
2358        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
2359        truncation.
2361    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
2363    7.  Added two (int) casts to pcregrep when printing the difference of two
2364        pointers, in case they are 64-bit values.
2366    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
2367        test 2 if it fails.
2369    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
2370        and a #define of that name to empty if it is not externally set. This is to
2371        allow users of MSVC to set it if necessary.
2373    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
2374        the convenience functions in the pcre_get.c source file.
2376    11. An option change at the start of a pattern that had top-level alternatives
2377        could cause overwriting and/or a crash. This command provoked a crash in
2378        some environments:
2380          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
2382        This potential security problem was recorded as CVE-2008-2371.
2384    12. For a pattern where the match had to start at the beginning or immediately
2385        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
2386        pcre_dfa_exec() could read past the end of the passed subject if there was
2387        no match. To help with detecting such bugs (e.g. with valgrind), I modified
2388        pcretest so that it places the subject at the end of its malloc-ed buffer.
2390    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
2391        exec() might read past the end of the data buffer in UTF-8 mode.
2393    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
2394        the data contained the byte 0x85 as part of a UTF-8 character within its
2395        first line. This applied both to normal and DFA matching.
2397    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
2398        /^[^d]*?$/8 failed to match "abc".
2400    16. Added a missing copyright notice to pcrecpp_internal.h.
2402    17. Make it more clear in the documentation that values returned from
2403        pcre_exec() in ovector are byte offsets, not character counts.
2405    18. Tidied a few places to stop certain compilers from issuing warnings.
2407    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
2408        supplied by Stefan Weber. I made a further small update for 7.8 because
2409        there is a change of source arrangements: the pcre_searchfuncs.c module is
2410        replaced by pcre_ucd.c.
2413    Version 7.7 07-May-08
2414    ---------------------
2416    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
2417        a string to a long long, pretend we don't even have a long long." This is
2418        done by checking for the strtoq, strtoll, and _strtoi64 functions.
2420    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
2421        pre-7.6 versions, which defined a global no_arg variable instead of putting
2422        it in the RE class. (See also #8 below.)
2424    3.  Remove a line of dead code, identified by coverity and reported by Nuno
2425        Lopes.
2427    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
2429        (1) The include/exclude patterns were being applied to the whole pathnames
2430            of files, instead of just to the final components.
2432        (2) If there was more than one level of directory, the subdirectories were
2433            skipped unless they satisfied the include/exclude conditions. This is
2434            inconsistent with GNU grep (and could even be seen as contrary to the
2435            pcregrep specification - which I improved to make it absolutely clear).
2436            The action now is always to scan all levels of directory, and just
2437            apply the include/exclude patterns to regular files.
2439    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
2440        --exclude_dir in the tests to avoid scanning .svn directories.
2442    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
2443        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
2444        doesn't support NULs in patterns.
2446    7.  Added some missing "const"s to declarations of static tables in
2447        pcre_compile.c and pcre_dfa_exec.c.
2449    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
2450        caused by fix #2  above. (Subsequently also a second patch to fix the
2451        first patch. And a third patch - this was a messy problem.)
2453    9.  Applied Craig's patch to remove the use of push_back().
2455    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
2456        matching function regexec().
2458    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
2459        which, however, unlike Perl's \g{...}, are subroutine calls, not back
2460        references. PCRE supports relative numbers with this syntax (I don't think
2461        Oniguruma does).
2463    12. Previously, a group with a zero repeat such as (...){0} was completely
2464        omitted from the compiled regex. However, this means that if the group
2465        was called as a subroutine from elsewhere in the pattern, things went wrong
2466        (an internal error was given). Such groups are now left in the compiled
2467        pattern, with a new opcode that causes them to be skipped at execution
2468        time.
2470    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
2471        to the way PCRE behaves:
2473        (a) A lone ] character is dis-allowed (Perl treats it as data).
2475        (b) A back reference to an unmatched subpattern matches an empty string
2476            (Perl fails the current match path).
2478        (c) A data ] in a character class must be notated as \] because if the
2479            first data character in a class is ], it defines an empty class. (In
2480            Perl it is not possible to have an empty class.) The empty class []
2481            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
2482            The negative empty class [^] matches any one character, independently
2483            of the DOTALL setting.
2485    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
2486        non-existent subpattern following a character class starting with ']' and
2487        containing () gave an internal compiling error instead of "reference to
2488        non-existent subpattern". Fortunately, when the pattern did exist, the
2489        compiled code was correct. (When scanning forwards to check for the
2490        existence of the subpattern, it was treating the data ']' as terminating
2491        the class, so got the count wrong. When actually compiling, the reference
2492        was subsequently set up correctly.)
2494    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
2495        it was being rejected as not supported by pcre_dfa_exec(), even though
2496        other assertions are supported. I have made pcre_dfa_exec() support
2497        (*FAIL).
2499    16. The implementation of 13c above involved the invention of a new opcode,
2500        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
2501        cannot be changed at match time, I realized I could make a small
2502        improvement to matching performance by compiling OP_ALLANY instead of
2503        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
2504        on the OP_ANY path.
2506    17. Compiling pcretest on Windows with readline support failed without the
2507        following two fixes: (1) Make the unistd.h include conditional on
2508        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
2510    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
2511        ncurses library to be included for pcretest when ReadLine support is
2512        requested, but also to allow for it to be overridden. This patch came from
2513        Daniel Bergström.
2515    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
2516        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
2517        any errors with the current Unicode tables. Thanks to Peter Kankowski for
2518        spotting this.
2521    Version 7.6 28-Jan-08
2522    ---------------------
2524    1.  A character class containing a very large number of characters with
2525        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
2526        overflow.
2528    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
2529        HAVE_LONG_LONG is not defined.
2531    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
2532        bring them up to date and include new features. This patch includes:
2534        - Fixed PH's badly added libz and libbz2 support.
2535        - Fixed a problem with static linking.
2536        - Added pcredemo. [But later removed - see 7 below.]
2537        - Fixed dftables problem and added an option.
2538        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
2539            HAVE_LONG_LONG.
2540        - Added readline support for pcretest.
2541        - Added an listing of the option settings after cmake has run.
2543    4.  A user submitted a patch to Makefile that makes it easy to create
2544        "pcre.dll" under mingw when using Configure/Make. I added stuff to
2545        Makefile.am that cause it to include this special target, without
2546        affecting anything else. Note that the same mingw target plus all
2547        the other distribution libraries and programs are now supported
2548        when configuring with CMake (see 6 below) instead of with
2549        Configure/Make.
2551    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
2552        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
2553        exported in the Windows port". It has not yet been confirmed that the patch
2554        solves the problem, but it does no harm.
2556    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
2557        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
2558        with CMake, and also correct the comment about stack recursion.
2560    7.  Remove the automatic building of pcredemo from the ./configure system and
2561        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
2562        of a program that users should build themselves after PCRE is installed, so
2563        building it automatically is not really right. What is more, it gave
2564        trouble in some build environments.
2566    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
2569    Version 7.5 10-Jan-08
2570    ---------------------
2572    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
2573        values in parens when parsing an RE using the C++ wrapper."
2575    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
2576        Characters greater than 255 were excluded from the class instead of being
2577        included.
2579    3.  The same bug as (2) above applied to negated POSIX classes such as
2580        [:^space:].
2582    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
2583        defined or documented. It seems to have been a typo for PCRE_STATIC, so
2584        I have changed it.
2586    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
2587        first named subpattern) and a construct such as (?&a) would reference the
2588        first named subpattern whose name started with "a" (in other words, the
2589        length check was missing). Both these problems are fixed. "Subpattern name
2590        expected" is now given for (?&) (a zero-length name), and this patch also
2591        makes it give the same error for \k'' (previously it complained that that
2592        was a reference to a non-existent subpattern).
2594    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
2595        this is right because (?- can be followed by option settings as well as by
2596        digits. I have, however, made the messages clearer.
2598    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
2599        than the number used in the conditional) now cause a compile-time error.
2600        This is actually not compatible with Perl, which accepts such patterns, but
2601        treats the conditional as always being FALSE (as PCRE used to), but it
2602        seems to me that giving a diagnostic is better.
2604    8.  Change "alphameric" to the more common word "alphanumeric" in comments
2605        and messages.
2607    9.  Fix two occurrences of "backslash" in comments that should have been
2608        "backspace".
2610    10. Remove two redundant lines of code that can never be obeyed (their function
2611        was moved elsewhere).
2613    11. The program that makes PCRE's Unicode character property table had a bug
2614        which caused it to generate incorrect table entries for sequences of
2615        characters that have the same character type, but are in different scripts.
2616        It amalgamated them into a single range, with the script of the first of
2617        them. In other words, some characters were in the wrong script. There were
2618        thirteen such cases, affecting characters in the following ranges:
2620          U+002b0 - U+002c1
2621          U+0060c - U+0060d
2622          U+0061e - U+00612
2623          U+0064b - U+0065e
2624          U+0074d - U+0076d
2625          U+01800 - U+01805
2626          U+01d00 - U+01d77
2627          U+01d9b - U+01dbf
2628          U+0200b - U+0200f
2629          U+030fc - U+030fe
2630          U+03260 - U+0327f
2631          U+0fb46 - U+0fbb1
2632          U+10450 - U+1049d
2634    12. The -o option (show only the matching part of a line) for pcregrep was not
2635        compatible with GNU grep in that, if there was more than one match in a
2636        line, it showed only the first of them. It now behaves in the same way as
2637        GNU grep.
2639    13. If the -o and -v options were combined for pcregrep, it printed a blank
2640        line for every non-matching line. GNU grep prints nothing, and pcregrep now
2641        does the same. The return code can be used to tell if there were any
2642        non-matching lines.
2644    14. Added --file-offsets and --line-offsets to pcregrep.
2646    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
2647        infinitely looping recursion. The bug was that positive lookaheads were not
2648        being skipped when checking for a possible empty match (negative lookaheads
2649        and both kinds of lookbehind were skipped).
2651    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
2652        inclusion of <windows.h> to before rather than after the definition of
2653        INVALID_FILE_ATTRIBUTES (patch from David Byron).
2655    17. Specifying a possessive quantifier with a specific limit for a Unicode
2656        character property caused pcre_compile() to compile bad code, which led at
2657        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
2658        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
2659        caused the error; without that there was no problem.
2661    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
2663    19. Added --enable-pcretest-libreadline.
2665    20. In pcrecpp.cc, the variable 'count' was incremented twice in
2666        RE::GlobalReplace(). As a result, the number of replacements returned was
2667        double what it should be. I removed one of the increments, but Craig sent a
2668        later patch that removed the other one (the right fix) and added unit tests
2669        that check the return values (which was not done before).
2671    21. Several CMake things:
2673        (1) Arranged that, when cmake is used on Unix, the libraries end up with
2674            the names libpcre and libpcreposix, not just pcre and pcreposix.
2676        (2) The above change means that pcretest and pcregrep are now correctly
2677            linked with the newly-built libraries, not previously installed ones.
2681    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
2682        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
2683        UTF-8 newline character). The key issue is that the pattern starts .*;
2684        this means that the match must be either at the beginning, or after a
2685        newline. The bug was in the code for advancing after a failed match and
2686        checking that the new position followed a newline. It was not taking
2687        account of UTF-8 characters correctly.
2689    23. PCRE was behaving differently from Perl in the way it recognized POSIX
2690        character classes. PCRE was not treating the sequence [:...:] as a
2691        character class unless the ... were all letters. Perl, however, seems to
2692        allow any characters between [: and :], though of course it rejects as
2693        unknown any "names" that contain non-letters, because all the known class
2694        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
2695        for example, whereas PCRE did not - it did not recognize a POSIX character
2696        class. This seemed a bit dangerous, so the code has been changed to be
2697        closer to Perl. The behaviour is not identical to Perl, because PCRE will
2698        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
2699        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
2700        Perl does, and where it didn't before.
2702    24. Rewrite so as to remove the single use of %n from pcregrep because in some
2703        Windows environments %n is disabled by default.
2706    Version 7.4 21-Sep-07
2707    ---------------------
2709    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
2710        means that a class such as [\s] counted as "explicit reference to CR or
2711        LF". That isn't really right - the whole point of the change was to try to
2712        help when there was an actual mention of one of the two characters. So now
2713        the change happens only if \r or \n (or a literal CR or LF) character is
2714        encountered.
2716    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
2717        of both had grown to the point where there were only 3 bits left.
2718        Fortunately, there was spare space in the data structure, and so I have
2719        moved the internal flags into a new 16-bit field to free up more option
2720        bits.
2722    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
2723        but did not set the internal JCHANGED flag - either of these is enough to
2724        control the way the "get" function works - but the PCRE_INFO_JCHANGED
2725        facility is supposed to tell if (?J) was ever used, so now (?J) at the
2726        start sets both bits.
2728    4.  Added options (at build time, compile time, exec time) to change \R from
2729        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
2731    5.  doc/pcresyntax.html was missing from the distribution.
2733    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
2734        compatibility, even though it is no longer used.
2736    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
2737        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
2738        windows.h file is present (where different names are used). [This was
2739        reversed later after testing - see 16 below.]
2741    8.  Changed all #include <config.h> to #include "config.h". There were also
2742        some further <pcre.h> cases that I changed to "pcre.h".
2744    9.  When pcregrep was used with the --colour option, it missed the line ending
2745        sequence off the lines that it output.
2747    10. It was pointed out to me that arrays of string pointers cause lots of
2748        relocations when a shared library is dynamically loaded. A technique of
2749        using a single long string with a table of offsets can drastically reduce
2750        these. I have refactored PCRE in four places to do this. The result is
2751        dramatic:
2753          Originally:                          290
2754          After changing UCP table:            187
2755          After changing error message table:   43
2756          After changing table of "verbs"       36
2757          After changing table of Posix names   22
2759        Thanks to the folks working on Gregex for glib for this insight.
2761    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
2762        unicode-properties was also set.
2764    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
2766    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
2767        checked only for CRLF.
2769    14. Added casts to pcretest.c to avoid compiler warnings.
2771    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
2773    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
2774        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
2775        entirely. This removes changes made in 7 above.
2777    17. The CMake files have been updated, and there is now more information about
2778        building with CMake in the NON-UNIX-USE document.
2781    Version 7.3 28-Aug-07
2782  ---------------------  ---------------------
2784   1. In the rejigging of the build system that eventually resulted in 7.1, the   1. In the rejigging of the build system that eventually resulted in 7.1, the
# Line 9  Version 7.3 05-Jul-07 Line 2786  Version 7.3 05-Jul-07
2786      brackets there is not right, since it causes compilers to look for an      brackets there is not right, since it causes compilers to look for an
2787      installed pcre.h, not the version that is in the source that is being      installed pcre.h, not the version that is in the source that is being
2788      compiled (which of course may be different). I have changed it back to:      compiled (which of course may be different). I have changed it back to:
2790        #include "pcre.h"        #include "pcre.h"
2792      I have a vague recollection that the change was concerned with compiling in      I have a vague recollection that the change was concerned with compiling in
2793      different directories, but in the new build system, that is taken care of      different directories, but in the new build system, that is taken care of
2794      by the VPATH setting the Makefile.      by the VPATH setting the Makefile.
2796   2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed   2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
2797      when the subject happened to end in the byte 0x85 (e.g. if the last      when the subject happened to end in the byte 0x85 (e.g. if the last
2798      character was \x{1ec5}). *Character* 0x85 is one of the "any" newline      character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
# Line 23  Version 7.3 05-Jul-07 Line 2800  Version 7.3 05-Jul-07
2800      of another character. The bug was that, for an unlimited repeat of . in      of another character. The bug was that, for an unlimited repeat of . in
2801      not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by      not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
2802      characters when looking for a newline.      characters when looking for a newline.
  3. A small performance improvement in the DOTALL UTF-8 mode .* case.  
  4. Debugging: adjusted the names of opcodes for different kinds of parentheses  
     in debug output.  
  5. Arrange to use "%I64d" instead of "%lld" for long printing in the pcrecpp  
     unittest when running under MinGW.  
2804     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
2806     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
2807        in debug output.
2809     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
2810        long printing in the pcrecpp unittest when running under MinGW.
2812     6. ESC_K was left out of the EBCDIC table.
2814     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
2815        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
2816        limit also applies to "virtual nesting" when a pattern is recursive, and in
2817        this case 1000 isn't so big. I have been able to remove this limit at the
2818        expense of backing off one optimization in certain circumstances. Normally,
2819        when pcre_exec() would call its internal match() function recursively and
2820        immediately return the result unconditionally, it uses a "tail recursion"
2821        feature to save stack. However, when a subpattern that can match an empty
2822        string has an unlimited repetition quantifier, it no longer makes this
2823        optimization. That gives it a stack frame in which to save the data for
2824        checking that an empty string has been matched. Previously this was taken
2825        from the 1000-entry workspace that had been reserved. So now there is no
2826        explicit limit, but more stack is used.
2828     8. Applied Daniel's patches to solve problems with the import/export magic
2829        syntax that is required for Windows, and which was going wrong for the
2830        pcreposix and pcrecpp parts of the library. These were overlooked when this
2831        problem was solved for the main library.
2833     9. There were some crude static tests to avoid integer overflow when computing
2834        the size of patterns that contain repeated groups with explicit upper
2835        limits. As the maximum quantifier is 65535, the maximum group length was
2836        set at 30,000 so that the product of these two numbers did not overflow a
2837        32-bit integer. However, it turns out that people want to use groups that
2838        are longer than 30,000 bytes (though not repeat them that many times).
2839        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
2840        made it possible to implement the integer overflow checks in a much more
2841        dynamic way, which I have now done. The artificial limitation on group
2842        length has been removed - we now have only the limit on the total length of
2843        the compiled pattern, which depends on the LINK_SIZE setting.
2845    10. Fixed a bug in the documentation for get/copy named substring when
2846        duplicate names are permitted. If none of the named substrings are set, the
2847        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
2848        empty string.
2850    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
2851        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
2852        because the ] is interpreted as the first data character and the
2853        terminating ] is not found. PCRE has been made compatible with Perl in this
2854        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
2855        cause memory overwriting.
2857    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
2858        string has been matched (to stop an infinite loop). It was not recognizing
2859        a conditional subpattern that could match an empty string if that
2860        subpattern was within another subpattern. For example, it looped when
2861        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
2862        condition was not nested. This bug has been fixed.
2864    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
2865        past the start of the subject in the presence of bytes with the top bit
2866        set, for example "\x8aBCD".
2868    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
2869        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
2871    14. Optimized (?!) to (*FAIL).
2873    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
2874        This restricts code points to be within the range 0 to 0x10FFFF, excluding
2875        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
2876        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
2877        does: it's just the validity check that is more restrictive.
2879    16. Inserted checks for integer overflows during escape sequence (backslash)
2880        processing, and also fixed erroneous offset values for syntax errors during
2881        backslash processing.
2883    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
2884        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
2886    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
2887        caused an overrun.
2889    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
2890        something other than just ASCII characters) inside a group that had an
2891        unlimited repeat caused a loop at compile time (while checking to see
2892        whether the group could match an empty string).
2894    20. Debugging a pattern containing \p or \P could cause a crash. For example,
2895        [\P{Any}] did so. (Error in the code for printing property names.)
2897    21. An orphan \E inside a character class could cause a crash.
2899    22. A repeated capturing bracket such as (A)? could cause a wild memory
2900        reference during compilation.
2902    23. There are several functions in pcre_compile() that scan along a compiled
2903        expression for various reasons (e.g. to see if it's fixed length for look
2904        behind). There were bugs in these functions when a repeated \p or \P was
2905        present in the pattern. These operators have additional parameters compared
2906        with \d, etc, and these were not being taken into account when moving along
2907        the compiled data. Specifically:
2909        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
2910            length.
2912        (b) An item such as \pL+ within a repeated group could cause crashes or
2913            loops.
2915        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
2916            "reference to non-existent subpattern" error.
2918        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
2920    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
2921        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
2923    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
2925    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
2926        character were causing crashes (broken optimization).
2928    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
2929        \p or \P) caused a compile-time loop.
2931    28. More problems have arisen in unanchored patterns when CRLF is a valid line
2932        break. For example, the unstudied pattern [\r\n]A does not match the string
2933        "\r\nA" because change 7.0/46 below moves the current point on by two
2934        characters after failing to match at the start. However, the pattern \nA
2935        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
2936        the same is true. There doesn't seem any very clean way out of this, but
2937        what I have chosen to do makes the common cases work: PCRE now takes note
2938        of whether there can be an explicit match for \r or \n anywhere in the
2939        pattern, and if so, 7.0/46 no longer applies. As part of this change,
2940        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
2941        pattern has explicit CR or LF references.
2943    29. Added (*CR) etc for changing newline setting at start of pattern.
2946  Version 7.2 19-Jun-07  Version 7.2 19-Jun-07

Removed from v.193  
changed lines
  Added in v.1540

  ViewVC Help
Powered by ViewVC 1.1.5