ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 45 by nigel, Sat Feb 24 21:39:25 2007 UTC revision 1578 by ph10, Mon Jul 20 16:27:31 2015 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Note that the PCRE 8.xx series (PCRE1) is now in a bugfix-only state. All
5    development is happening in the PCRE2 10.xx series.
7    Version 8.38 xx-xxx-xxxx
8    ------------------------
10    1.  If a group that contained a recursive back reference also contained a
11        forward reference subroutine call followed by a non-forward-reference
12        subroutine call, for example /.((?2)(?R)\1)()/, pcre2_compile() failed to
13        compile correct code, leading to undefined behaviour or an internally
14        detected error. This bug was discovered by the LLVM fuzzer.
16    2.  Quantification of certain items (e.g. atomic back references) could cause
17        incorrect code to be compiled when recursive forward references were
18        involved. For example, in this pattern: /(?1)()((((((\1++))\x85)+)|))/.
19        This bug was discovered by the LLVM fuzzer.
21    3.  A repeated conditional group whose condition was a reference by name caused
22        a buffer overflow if there was more than one group with the given name.
23        This bug was discovered by the LLVM fuzzer.
25    4.  A recursive back reference by name within a group that had the same name as
26        another group caused a buffer overflow. For example:
27        /(?J)(?'d'(?'d'\g{d}))/. This bug was discovered by the LLVM fuzzer.
29    5.  A forward reference by name to a group whose number is the same as the
30        current group, for example in this pattern: /(?|(\k'Pm')|(?'Pm'))/, caused
31        a buffer overflow at compile time. This bug was discovered by the LLVM
32        fuzzer.
34    6.  A lookbehind assertion within a set of mutually recursive subpatterns could
35        provoke a buffer overflow. This bug was discovered by the LLVM fuzzer.
37    7.  Another buffer overflow bug involved duplicate named groups with a
38        reference between their definition, with a group that reset capture
39        numbers, for example: /(?J:(?|(?'R')(\k'R')|((?'R'))))/. This has been
40        fixed by always allowing for more memory, even if not needed. (A proper fix
41        is implemented in PCRE2, but it involves more refactoring.)
43    8.  There was no check for integer overflow in subroutine calls such as (?123).
45    9.  The table entry for \l in EBCDIC environments was incorrect, leading to its
46        being treated as a literal 'l' instead of causing an error.
48    10. There was a buffer overflow if pcre_exec() was called with an ovector of
49        size 1. This bug was found by american fuzzy lop.
51    11. If a non-capturing group containing a conditional group that could match
52        an empty string was repeated, it was not identified as matching an empty
53        string itself. For example: /^(?:(?(1)x|)+)+$()/.
55    12. In an EBCDIC environment, pcretest was mishandling the escape sequences
56        \a and \e in test subject lines.
58    13. In an EBCDIC environment, \a in a pattern was converted to the ASCII
59        instead of the EBCDIC value.
61    14. The handling of \c in an EBCDIC environment has been revised so that it is
62        now compatible with the specification in Perl's perlebcdic page.
64    15. The EBCDIC character 0x41 is a non-breaking space, equivalent to 0xa0 in
65        ASCII/Unicode. This has now been added to the list of characters that are
66        recognized as white space in EBCDIC.
68    16. When PCRE was compiled without UCP support, the use of \p and \P gave an
69        error (correctly) when used outside a class, but did not give an error
70        within a class.
72    17. \h within a class was incorrectly compiled in EBCDIC environments.
74    18. A pattern with an unmatched closing parenthesis that contained a backward
75        assertion which itself contained a forward reference caused buffer
76        overflow. And example pattern is: /(?=di(?<=(?1))|(?=(.))))/.
78    19. JIT should return with error when the compiled pattern requires more stack
79        space than the maximum.
81    20. A possessively repeated conditional group that could match an empty string,
82        for example, /(?(R))*+/, was incorrectly compiled.
84    21. Fix infinite recursion in the JIT compiler when certain patterns such as
85        /(?:|a|){100}x/ are analysed.
87    22. Some patterns with character classes involving [: and \\ were incorrectly
88        compiled and could cause reading from uninitialized memory or an incorrect
89        error diagnosis.
92    Version 8.37 28-April-2015
93    --------------------------
95    1.  When an (*ACCEPT) is triggered inside capturing parentheses, it arranges
96        for those parentheses to be closed with whatever has been captured so far.
97        However, it was failing to mark any other groups between the hightest
98        capture so far and the currrent group as "unset". Thus, the ovector for
99        those groups contained whatever was previously there. An example is the
100        pattern /(x)|((*ACCEPT))/ when matched against "abcd".
102    2.  If an assertion condition was quantified with a minimum of zero (an odd
103        thing to do, but it happened), SIGSEGV or other misbehaviour could occur.
105    3.  If a pattern in pcretest input had the P (POSIX) modifier followed by an
106        unrecognized modifier, a crash could occur.
108    4.  An attempt to do global matching in pcretest with a zero-length ovector
109        caused a crash.
111    5.  Fixed a memory leak during matching that could occur for a subpattern
112        subroutine call (recursive or otherwise) if the number of captured groups
113        that had to be saved was greater than ten.
115    6.  Catch a bad opcode during auto-possessification after compiling a bad UTF
116        string with NO_UTF_CHECK. This is a tidyup, not a bug fix, as passing bad
117        UTF with NO_UTF_CHECK is documented as having an undefined outcome.
119    7.  A UTF pattern containing a "not" match of a non-ASCII character and a
120        subroutine reference could loop at compile time. Example: /[^\xff]((?1))/.
122    8. When a pattern is compiled, it remembers the highest back reference so that
123       when matching, if the ovector is too small, extra memory can be obtained to
124       use instead. A conditional subpattern whose condition is a check on a
125       capture having happened, such as, for example in the pattern
126       /^(?:(a)|b)(?(1)A|B)/, is another kind of back reference, but it was not
127       setting the highest backreference number. This mattered only if pcre_exec()
128       was called with an ovector that was too small to hold the capture, and there
129       was no other kind of back reference (a situation which is probably quite
130       rare). The effect of the bug was that the condition was always treated as
131       FALSE when the capture could not be consulted, leading to a incorrect
132       behaviour by pcre_exec(). This bug has been fixed.
134    9. A reference to a duplicated named group (either a back reference or a test
135       for being set in a conditional) that occurred in a part of the pattern where
136       PCRE_DUPNAMES was not set caused the amount of memory needed for the pattern
137       to be incorrectly calculated, leading to overwriting.
139    10. A mutually recursive set of back references such as (\2)(\1) caused a
140        segfault at study time (while trying to find the minimum matching length).
141        The infinite loop is now broken (with the minimum length unset, that is,
142        zero).
144    11. If an assertion that was used as a condition was quantified with a minimum
145        of zero, matching went wrong. In particular, if the whole group had
146        unlimited repetition and could match an empty string, a segfault was
147        likely. The pattern (?(?=0)?)+ is an example that caused this. Perl allows
148        assertions to be quantified, but not if they are being used as conditions,
149        so the above pattern is faulted by Perl. PCRE has now been changed so that
150        it also rejects such patterns.
152    12. A possessive capturing group such as (a)*+ with a minimum repeat of zero
153        failed to allow the zero-repeat case if pcre2_exec() was called with an
154        ovector too small to capture the group.
156    13. Fixed two bugs in pcretest that were discovered by fuzzing and reported by
157        Red Hat Product Security:
159        (a) A crash if /K and /F were both set with the option to save the compiled
160        pattern.
162        (b) Another crash if the option to print captured substrings in a callout
163        was combined with setting a null ovector, for example \O\C+ as a subject
164        string.
166    14. A pattern such as "((?2){0,1999}())?", which has a group containing a
167        forward reference repeated a large (but limited) number of times within a
168        repeated outer group that has a zero minimum quantifier, caused incorrect
169        code to be compiled, leading to the error "internal error:
170        previously-checked referenced subpattern not found" when an incorrect
171        memory address was read. This bug was reported as "heap overflow",
172        discovered by Kai Lu of Fortinet's FortiGuard Labs and given the CVE number
173        CVE-2015-2325.
175    23. A pattern such as "((?+1)(\1))/" containing a forward reference subroutine
176        call within a group that also contained a recursive back reference caused
177        incorrect code to be compiled. This bug was reported as "heap overflow",
178        discovered by Kai Lu of Fortinet's FortiGuard Labs, and given the CVE
179        number CVE-2015-2326.
181    24. Computing the size of the JIT read-only data in advance has been a source
182        of various issues, and new ones are still appear unfortunately. To fix
183        existing and future issues, size computation is eliminated from the code,
184        and replaced by on-demand memory allocation.
186    25. A pattern such as /(?i)[A-`]/, where characters in the other case are
187        adjacent to the end of the range, and the range contained characters with
188        more than one other case, caused incorrect behaviour when compiled in UTF
189        mode. In that example, the range a-j was left out of the class.
191    26. Fix JIT compilation of conditional blocks, which assertion
192        is converted to (*FAIL). E.g: /(?(?!))/.
194    27. The pattern /(?(?!)^)/ caused references to random memory. This bug was
195        discovered by the LLVM fuzzer.
197    28. The assertion (?!) is optimized to (*FAIL). This was not handled correctly
198        when this assertion was used as a condition, for example (?(?!)a|b). In
199        pcre2_match() it worked by luck; in pcre2_dfa_match() it gave an incorrect
200        error about an unsupported item.
202    29. For some types of pattern, for example /Z*(|d*){216}/, the auto-
203        possessification code could take exponential time to complete. A recursion
204        depth limit of 1000 has been imposed to limit the resources used by this
205        optimization.
207    30. A pattern such as /(*UTF)[\S\V\H]/, which contains a negated special class
208        such as \S in non-UCP mode, explicit wide characters (> 255) can be ignored
209        because \S ensures they are all in the class. The code for doing this was
210        interacting badly with the code for computing the amount of space needed to
211        compile the pattern, leading to a buffer overflow. This bug was discovered
212        by the LLVM fuzzer.
214    31. A pattern such as /((?2)+)((?1))/ which has mutual recursion nested inside
215        other kinds of group caused stack overflow at compile time. This bug was
216        discovered by the LLVM fuzzer.
218    32. A pattern such as /(?1)(?#?'){8}(a)/ which had a parenthesized comment
219        between a subroutine call and its quantifier was incorrectly compiled,
220        leading to buffer overflow or other errors. This bug was discovered by the
221        LLVM fuzzer.
223    33. The illegal pattern /(?(?<E>.*!.*)?)/ was not being diagnosed as missing an
224        assertion after (?(. The code was failing to check the character after
225        (?(?< for the ! or = that would indicate a lookbehind assertion. This bug
226        was discovered by the LLVM fuzzer.
228    34. A pattern such as /X((?2)()*+){2}+/ which has a possessive quantifier with
229        a fixed maximum following a group that contains a subroutine reference was
230        incorrectly compiled and could trigger buffer overflow. This bug was
231        discovered by the LLVM fuzzer.
233    35. A mutual recursion within a lookbehind assertion such as (?<=((?2))((?1)))
234        caused a stack overflow instead of the diagnosis of a non-fixed length
235        lookbehind assertion. This bug was discovered by the LLVM fuzzer.
237    36. The use of \K in a positive lookbehind assertion in a non-anchored pattern
238        (e.g. /(?<=\Ka)/) could make pcregrep loop.
240    37. There was a similar problem to 36 in pcretest for global matches.
242    38. If a greedy quantified \X was preceded by \C in UTF mode (e.g. \C\X*),
243        and a subsequent item in the pattern caused a non-match, backtracking over
244        the repeated \X did not stop, but carried on past the start of the subject,
245        causing reference to random memory and/or a segfault. There were also some
246        other cases where backtracking after \C could crash. This set of bugs was
247        discovered by the LLVM fuzzer.
249    39. The function for finding the minimum length of a matching string could take
250        a very long time if mutual recursion was present many times in a pattern,
251        for example, /((?2){73}(?2))((?1))/. A better mutual recursion detection
252        method has been implemented. This infelicity was discovered by the LLVM
253        fuzzer.
255    40. Static linking against the PCRE library using the pkg-config module was
256        failing on missing pthread symbols.
259    Version 8.36 26-September-2014
260    ------------------------------
262    1.  Got rid of some compiler warnings in the C++ modules that were shown up by
263        -Wmissing-field-initializers and -Wunused-parameter.
265    2.  The tests for quantifiers being too big (greater than 65535) were being
266        applied after reading the number, and stupidly assuming that integer
267        overflow would give a negative number. The tests are now applied as the
268        numbers are read.
270    3.  Tidy code in pcre_exec.c where two branches that used to be different are
271        now the same.
273    4.  The JIT compiler did not generate match limit checks for certain
274        bracketed expressions with quantifiers. This may lead to exponential
275        backtracking, instead of returning with PCRE_ERROR_MATCHLIMIT. This
276        issue should be resolved now.
278    5.  Fixed an issue, which occures when nested alternatives are optimized
279        with table jumps.
281    6.  Inserted two casts and changed some ints to size_t in the light of some
282        reported 64-bit compiler warnings (Bugzilla 1477).
284    7.  Fixed a bug concerned with zero-minimum possessive groups that could match
285        an empty string, which sometimes were behaving incorrectly in the
286        interpreter (though correctly in the JIT matcher). This pcretest input is
287        an example:
289          '\A(?:[^"]++|"(?:[^"]*+|"")*+")++'
292        the interpreter was reporting a match of 'NON QUOTED ' only, whereas the
293        JIT matcher and Perl both matched 'NON QUOTED "QUOT""ED" AFTER '. The test
294        for an empty string was breaking the inner loop and carrying on at a lower
295        level, when possessive repeated groups should always return to a higher
296        level as they have no backtrack points in them. The empty string test now
297        occurs at the outer level.
299    8.  Fixed a bug that was incorrectly auto-possessifying \w+ in the pattern
300        ^\w+(?>\s*)(?<=\w) which caused it not to match "test test".
302    9.  Give a compile-time error for \o{} (as Perl does) and for \x{} (which Perl
303        doesn't).
305    10. Change 8.34/15 introduced a bug that caused the amount of memory needed
306        to hold a pattern to be incorrectly computed (too small) when there were
307        named back references to duplicated names. This could cause "internal
308        error: code overflow" or "double free or corruption" or other memory
309        handling errors.
311    11. When named subpatterns had the same prefixes, back references could be
312        confused. For example, in this pattern:
314          /(?P<Name>a)?(?P<Name2>b)?(?(<Name>)c|d)*l/
316        the reference to 'Name' was incorrectly treated as a reference to a
317        duplicate name.
319    12. A pattern such as /^s?c/mi8 where the optional character has more than
320        one "other case" was incorrectly compiled such that it would only try to
321        match starting at "c".
323    13. When a pattern starting with \s was studied, VT was not included in the
324        list of possible starting characters; this should have been part of the
325        8.34/18 patch.
327    14. If a character class started [\Qx]... where x is any character, the class
328        was incorrectly terminated at the ].
330    15. If a pattern that started with a caseless match for a character with more
331        than one "other case" was studied, PCRE did not set up the starting code
332        unit bit map for the list of possible characters. Now it does. This is an
333        optimization improvement, not a bug fix.
335    16. The Unicode data tables have been updated to Unicode 7.0.0.
337    17. Fixed a number of memory leaks in pcregrep.
339    18. Avoid a compiler warning (from some compilers) for a function call with
340        a cast that removes "const" from an lvalue by using an intermediate
341        variable (to which the compiler does not object).
343    19. Incorrect code was compiled if a group that contained an internal recursive
344        back reference was optional (had quantifier with a minimum of zero). This
345        example compiled incorrect code: /(((a\2)|(a*)\g<-1>))*/ and other examples
346        caused segmentation faults because of stack overflows at compile time.
348    20. A pattern such as /((?(R)a|(?1)))+/, which contains a recursion within a
349        group that is quantified with an indefinite repeat, caused a compile-time
350        loop which used up all the system stack and provoked a segmentation fault.
351        This was not the same bug as 19 above.
353    21. Add PCRECPP_EXP_DECL declaration to operator<< in pcre_stringpiece.h.
354        Patch by Mike Frysinger.
357    Version 8.35 04-April-2014
358    --------------------------
360    1.  A new flag is set, when property checks are present in an XCLASS.
361        When this flag is not set, PCRE can perform certain optimizations
362        such as studying these XCLASS-es.
364    2.  The auto-possessification of character sets were improved: a normal
365        and an extended character set can be compared now. Furthermore
366        the JIT compiler optimizes more character set checks.
368    3.  Got rid of some compiler warnings for potentially uninitialized variables
369        that show up only when compiled with -O2.
371    4.  A pattern such as (?=ab\K) that uses \K in an assertion can set the start
372        of a match later then the end of the match. The pcretest program was not
373        handling the case sensibly - it was outputting from the start to the next
374        binary zero. It now reports this situation in a message, and outputs the
375        text from the end to the start.
377    5.  Fast forward search is improved in JIT. Instead of the first three
378        characters, any three characters with fixed position can be searched.
379        Search order: first, last, middle.
381    6.  Improve character range checks in JIT. Characters are read by an inprecise
382        function now, which returns with an unknown value if the character code is
383        above a certain threshold (e.g: 256). The only limitation is that the value
384        must be bigger than the threshold as well. This function is useful when
385        the characters above the threshold are handled in the same way.
387    7.  The macros whose names start with RAWUCHAR are placeholders for a future
388        mode in which only the bottom 21 bits of 32-bit data items are used. To
389        make this more memorable for those maintaining the code, the names have
390        been changed to start with UCHAR21, and an extensive comment has been added
391        to their definition.
393    8.  Add missing (new) files sljitNativeTILEGX.c and sljitNativeTILEGX-encoder.c
394        to the export list in Makefile.am (they were accidentally omitted from the
395        8.34 tarball).
397    9.  The informational output from pcretest used the phrase "starting byte set"
398        which is inappropriate for the 16-bit and 32-bit libraries. As the output
399        for "first char" and "need char" really means "non-UTF-char", I've changed
400        "byte" to "char", and slightly reworded the output. The documentation about
401        these values has also been (I hope) clarified.
403    10. Another JIT related optimization: use table jumps for selecting the correct
404        backtracking path, when more than four alternatives are present inside a
405        bracket.
407    11. Empty match is not possible, when the minimum length is greater than zero,
408        and there is no \K in the pattern. JIT should avoid empty match checks in
409        such cases.
411    12. In a caseless character class with UCP support, when a character with more
412        than one alternative case was not the first character of a range, not all
413        the alternative cases were added to the class. For example, s and \x{17f}
414        are both alternative cases for S: the class [RST] was handled correctly,
415        but [R-T] was not.
417    13. The configure.ac file always checked for pthread support when JIT was
418        enabled. This is not used in Windows, so I have put this test inside a
419        check for the presence of windows.h (which was already tested for).
421    14. Improve pattern prefix search by a simplified Boyer-Moore algorithm in JIT.
422        The algorithm provides a way to skip certain starting offsets, and usually
423        faster than linear prefix searches.
425    15. Change 13 for 8.20 updated RunTest to check for the 'fr' locale as well
426        as for 'fr_FR' and 'french'. For some reason, however, it then used the
427        Windows-specific input and output files, which have 'french' screwed in.
428        So this could never have worked. One of the problems with locales is that
429        they aren't always the same. I have now updated RunTest so that it checks
430        the output of the locale test (test 3) against three different output
431        files, and it allows the test to pass if any one of them matches. With luck
432        this should make the test pass on some versions of Solaris where it was
433        failing. Because of the uncertainty, the script did not used to stop if
434        test 3 failed; it now does. If further versions of a French locale ever
435        come to light, they can now easily be added.
437    16. If --with-pcregrep-bufsize was given a non-integer value such as "50K",
438        there was a message during ./configure, but it did not stop. This now
439        provokes an error. The invalid example in README has been corrected.
440        If a value less than the minimum is given, the minimum value has always
441        been used, but now a warning is given.
443    17. If --enable-bsr-anycrlf was set, the special 16/32-bit test failed. This
444        was a bug in the test system, which is now fixed. Also, the list of various
445        configurations that are tested for each release did not have one with both
446        16/32 bits and --enable-bar-anycrlf. It now does.
448    18. pcretest was missing "-C bsr" for displaying the \R default setting.
450    19. Little endian PowerPC systems are supported now by the JIT compiler.
452    20. The fast forward newline mechanism could enter to an infinite loop on
453        certain invalid UTF-8 input. Although we don't support these cases
454        this issue can be fixed by a performance optimization.
456    21. Change 33 of 8.34 is not sufficient to ensure stack safety because it does
457        not take account if existing stack usage. There is now a new global
458        variable called pcre_stack_guard that can be set to point to an external
459        function to check stack availability. It is called at the start of
460        processing every parenthesized group.
462    22. A typo in the code meant that in ungreedy mode the max/min qualifier
463        behaved like a min-possessive qualifier, and, for example, /a{1,3}b/U did
464        not match "ab".
466    23. When UTF was disabled, the JIT program reported some incorrect compile
467        errors. These messages are silenced now.
469    24. Experimental support for ARM-64 and MIPS-64 has been added to the JIT
470        compiler.
472    25. Change all the temporary files used in RunGrepTest to be different to those
473        used by RunTest so that the tests can be run simultaneously, for example by
474        "make -j check".
477    Version 8.34 15-December-2013
478    -----------------------------
480    1.  Add pcre[16|32]_jit_free_unused_memory to forcibly free unused JIT
481        executable memory. Patch inspired by Carsten Klein.
483    2.  ./configure --enable-coverage defined SUPPORT_GCOV in config.h, although
484        this macro is never tested and has no effect, because the work to support
485        coverage involves only compiling and linking options and special targets in
486        the Makefile. The comment in config.h implied that defining the macro would
487        enable coverage support, which is totally false. There was also support for
488        setting this macro in the CMake files (my fault, I just copied it from
489        configure). SUPPORT_GCOV has now been removed.
491    3.  Make a small performance improvement in strlen16() and strlen32() in
492        pcretest.
494    4.  Change 36 for 8.33 left some unreachable statements in pcre_exec.c,
495        detected by the Solaris compiler (gcc doesn't seem to be able to diagnose
496        these cases). There was also one in pcretest.c.
498    5.  Cleaned up a "may be uninitialized" compiler warning in pcre_exec.c.
500    6.  In UTF mode, the code for checking whether a group could match an empty
501        string (which is used for indefinitely repeated groups to allow for
502        breaking an infinite loop) was broken when the group contained a repeated
503        negated single-character class with a character that occupied more than one
504        data item and had a minimum repetition of zero (for example, [^\x{100}]* in
505        UTF-8 mode). The effect was undefined: the group might or might not be
506        deemed as matching an empty string, or the program might have crashed.
508    7.  The code for checking whether a group could match an empty string was not
509        recognizing that \h, \H, \v, \V, and \R must match a character.
511    8.  Implemented PCRE_INFO_MATCH_EMPTY, which yields 1 if the pattern can match
512        an empty string. If it can, pcretest shows this in its information output.
514    9.  Fixed two related bugs that applied to Unicode extended grapheme clusters
515        that were repeated with a maximizing qualifier (e.g. \X* or \X{2,5}) when
516        matched by pcre_exec() without using JIT:
518        (a) If the rest of the pattern did not match after a maximal run of
519            grapheme clusters, the code for backing up to try with fewer of them
520            did not always back up over a full grapheme when characters that do not
521            have the modifier quality were involved, e.g. Hangul syllables.
523        (b) If the match point in a subject started with modifier character, and
524            there was no match, the code could incorrectly back up beyond the match
525            point, and potentially beyond the first character in the subject,
526            leading to a segfault or an incorrect match result.
528    10. A conditional group with an assertion condition could lead to PCRE
529        recording an incorrect first data item for a match if no other first data
530        item was recorded. For example, the pattern (?(?=ab)ab) recorded "a" as a
531        first data item, and therefore matched "ca" after "c" instead of at the
532        start.
534    11. Change 40 for 8.33 (allowing pcregrep to find empty strings) showed up a
535        bug that caused the command "echo a | ./pcregrep -M '|a'" to loop.
537    12. The source of pcregrep now includes z/OS-specific code so that it can be
538        compiled for z/OS as part of the special z/OS distribution.
540    13. Added the -T and -TM options to pcretest.
542    14. The code in pcre_compile.c for creating the table of named capturing groups
543        has been refactored. Instead of creating the table dynamically during the
544        actual compiling pass, the information is remembered during the pre-compile
545        pass (on the stack unless there are more than 20 named groups, in which
546        case malloc() is used) and the whole table is created before the actual
547        compile happens. This has simplified the code (it is now nearly 150 lines
548        shorter) and prepared the way for better handling of references to groups
549        with duplicate names.
551    15. A back reference to a named subpattern when there is more than one of the
552        same name now checks them in the order in which they appear in the pattern.
553        The first one that is set is used for the reference. Previously only the
554        first one was inspected. This change makes PCRE more compatible with Perl.
556    16. Unicode character properties were updated from Unicode 6.3.0.
558    17. The compile-time code for auto-possessification has been refactored, based
559        on a patch by Zoltan Herczeg. It now happens after instead of during
560        compilation. The code is cleaner, and more cases are handled. The option
561        PCRE_NO_AUTO_POSSESS is added for testing purposes, and the -O and /O
562        options in pcretest are provided to set it. It can also be set by
563        (*NO_AUTO_POSSESS) at the start of a pattern.
565    18. The character VT has been added to the default ("C" locale) set of
566        characters that match \s and are generally treated as white space,
567        following this same change in Perl 5.18. There is now no difference between
568        "Perl space" and "POSIX space". Whether VT is treated as white space in
569        other locales depends on the locale.
571    19. The code for checking named groups as conditions, either for being set or
572        for being recursed, has been refactored (this is related to 14 and 15
573        above). Processing unduplicated named groups should now be as fast at
574        numerical groups, and processing duplicated groups should be faster than
575        before.
577    20. Two patches to the CMake build system, by Alexander Barkov:
579          (1) Replace the "source" command by "." in CMakeLists.txt because
580              "source" is a bash-ism.
582          (2) Add missing HAVE_STDINT_H and HAVE_INTTYPES_H to config-cmake.h.in;
583              without these the CMake build does not work on Solaris.
585    21. Perl has changed its handling of \8 and \9. If there is no previously
586        encountered capturing group of those numbers, they are treated as the
587        literal characters 8 and 9 instead of a binary zero followed by the
588        literals. PCRE now does the same.
590    22. Following Perl, added \o{} to specify codepoints in octal, making it
591        possible to specify values greater than 0777 and also making them
592        unambiguous.
594    23. Perl now gives an error for missing closing braces after \x{... instead of
595        treating the string as literal. PCRE now does the same.
597    24. RunTest used to grumble if an inappropriate test was selected explicitly,
598        but just skip it when running all tests. This make it awkward to run ranges
599        of tests when one of them was inappropriate. Now it just skips any
600        inappropriate tests, as it always did when running all tests.
602    25. If PCRE_AUTO_CALLOUT and PCRE_UCP were set for a pattern that contained
603        character types such as \d or \w, too many callouts were inserted, and the
604        data that they returned was rubbish.
606    26. In UCP mode, \s was not matching two of the characters that Perl matches,
607        namely NEL (U+0085) and MONGOLIAN VOWEL SEPARATOR (U+180E), though they
608        were matched by \h. The code has now been refactored so that the lists of
609        the horizontal and vertical whitespace characters used for \h and \v (which
610        are defined only in one place) are now also used for \s.
612    27. Add JIT support for the 64 bit TileGX architecture.
613        Patch by Jiong Wang (Tilera Corporation).
615    28. Possessive quantifiers for classes (both explicit and automatically
616        generated) now use special opcodes instead of wrapping in ONCE brackets.
618    29. Whereas an item such as A{4}+ ignored the possessivenes of the quantifier
619        (because it's meaningless), this was not happening when PCRE_CASELESS was
620        set. Not wrong, but inefficient.
622    30. Updated perltest.pl to add /u (force Unicode mode) when /W (use Unicode
623        properties for \w, \d, etc) is present in a test regex. Otherwise if the
624        test contains no characters greater than 255, Perl doesn't realise it
625        should be using Unicode semantics.
627    31. Upgraded the handling of the POSIX classes [:graph:], [:print:], and
628        [:punct:] when PCRE_UCP is set so as to include the same characters as Perl
629        does in Unicode mode.
631    32. Added the "forbid" facility to pcretest so that putting tests into the
632        wrong test files can sometimes be quickly detected.
634    33. There is now a limit (default 250) on the depth of nesting of parentheses.
635        This limit is imposed to control the amount of system stack used at compile
636        time. It can be changed at build time by --with-parens-nest-limit=xxx or
637        the equivalent in CMake.
639    34. Character classes such as [A-\d] or [a-[:digit:]] now cause compile-time
640        errors. Perl warns for these when in warning mode, but PCRE has no facility
641        for giving warnings.
643    35. Change 34 for 8.13 allowed quantifiers on assertions, because Perl does.
644        However, this was not working for (?!) because it is optimized to (*FAIL),
645        for which PCRE does not allow quantifiers. The optimization is now disabled
646        when a quantifier follows (?!). I can't see any use for this, but it makes
647        things uniform.
649    36. Perl no longer allows group names to start with digits, so I have made this
650        change also in PCRE. It simplifies the code a bit.
652    37. In extended mode, Perl ignores spaces before a + that indicates a
653        possessive quantifier. PCRE allowed a space before the quantifier, but not
654        before the possessive +. It now does.
656    38. The use of \K (reset reported match start) within a repeated possessive
657        group such as (a\Kb)*+ was not working.
659    40. Document that the same character tables must be used at compile time and
660        run time, and that the facility to pass tables to pcre_exec() and
661        pcre_dfa_exec() is for use only with saved/restored patterns.
663    41. Applied Jeff Trawick's patch CMakeLists.txt, which "provides two new
664        features for Builds with MSVC:
666        1. Support pcre.rc and/or pcreposix.rc (as is already done for MinGW
667           builds). The .rc files can be used to set FileDescription and many other
668           attributes.
670        2. Add an option (-DINSTALL_MSVC_PDB) to enable installation of .pdb files.
671           This allows higher-level build scripts which want .pdb files to avoid
672           hard-coding the exact files needed."
674    42. Added support for [[:<:]] and [[:>:]] as used in the BSD POSIX library to
675        mean "start of word" and "end of word", respectively, as a transition aid.
677    43. A minimizing repeat of a class containing codepoints greater than 255 in
678        non-UTF 16-bit or 32-bit modes caused an internal error when PCRE was
679        compiled to use the heap for recursion.
681    44. Got rid of some compiler warnings for unused variables when UTF but not UCP
682        is configured.
685    Version 8.33 28-May-2013
686    ------------------------
688    1.  Added 'U' to some constants that are compared to unsigned integers, to
689        avoid compiler signed/unsigned warnings. Added (int) casts to unsigned
690        variables that are added to signed variables, to ensure the result is
691        signed and can be negated.
693    2.  Applied patch by Daniel Richard G for quashing MSVC warnings to the
694        CMake config files.
696    3.  Revise the creation of config.h.generic so that all boolean macros are
697        #undefined, whereas non-boolean macros are #ifndef/#endif-ed. This makes
698        overriding via -D on the command line possible.
700    4.  Changing the definition of the variable "op" in pcre_exec.c from pcre_uchar
701        to unsigned int is reported to make a quite noticeable speed difference in
702        a specific Windows environment. Testing on Linux did also appear to show
703        some benefit (and it is clearly not harmful). Also fixed the definition of
704        Xop which should be unsigned.
706    5.  Related to (4), changing the definition of the intermediate variable cc
707        in repeated character loops from pcre_uchar to pcre_uint32 also gave speed
708        improvements.
710    6.  Fix forward search in JIT when link size is 3 or greater. Also removed some
711        unnecessary spaces.
713    7.  Adjust autogen.sh and configure.ac to lose warnings given by automake 1.12
714        and later.
716    8.  Fix two buffer over read issues in 16 and 32 bit modes. Affects JIT only.
718    9.  Optimizing fast_forward_start_bits in JIT.
720    10. Adding support for callouts in JIT, and fixing some issues revealed
721        during this work. Namely:
723        (a) Unoptimized capturing brackets incorrectly reset on backtrack.
725        (b) Minimum length was not checked before the matching is started.
727    11. The value of capture_last that is passed to callouts was incorrect in some
728        cases when there was a capture on one path that was subsequently abandoned
729        after a backtrack. Also, the capture_last value is now reset after a
730        recursion, since all captures are also reset in this case.
732    12. The interpreter no longer returns the "too many substrings" error in the
733        case when an overflowing capture is in a branch that is subsequently
734        abandoned after a backtrack.
736    13. In the pathological case when an offset vector of size 2 is used, pcretest
737        now prints out the matched string after a yield of 0 or 1.
739    14. Inlining subpatterns in recursions, when certain conditions are fulfilled.
740        Only supported by the JIT compiler at the moment.
742    15. JIT compiler now supports 32 bit Macs thanks to Lawrence Velazquez.
744    16. Partial matches now set offsets[2] to the "bumpalong" value, that is, the
745        offset of the starting point of the matching process, provided the offsets
746        vector is large enough.
748    17. The \A escape now records a lookbehind value of 1, though its execution
749        does not actually inspect the previous character. This is to ensure that,
750        in partial multi-segment matching, at least one character from the old
751        segment is retained when a new segment is processed. Otherwise, if there
752        are no lookbehinds in the pattern, \A might match incorrectly at the start
753        of a new segment.
755    18. Added some #ifdef __VMS code into pcretest.c to help VMS implementations.
757    19. Redefined some pcre_uchar variables in pcre_exec.c as pcre_uint32; this
758        gives some modest performance improvement in 8-bit mode.
760    20. Added the PCRE-specific property \p{Xuc} for matching characters that can
761        be expressed in certain programming languages using Universal Character
762        Names.
764    21. Unicode validation has been updated in the light of Unicode Corrigendum #9,
765        which points out that "non characters" are not "characters that may not
766        appear in Unicode strings" but rather "characters that are reserved for
767        internal use and have only local meaning".
769    22. When a pattern was compiled with automatic callouts (PCRE_AUTO_CALLOUT) and
770        there was a conditional group that depended on an assertion, if the
771        assertion was false, the callout that immediately followed the alternation
772        in the condition was skipped when pcre_exec() was used for matching.
774    23. Allow an explicit callout to be inserted before an assertion that is the
775        condition for a conditional group, for compatibility with automatic
776        callouts, which always insert a callout at this point.
778    24. In 8.31, (*COMMIT) was confined to within a recursive subpattern. Perl also
779        confines (*SKIP) and (*PRUNE) in the same way, and this has now been done.
781    25. (*PRUNE) is now supported by the JIT compiler.
783    26. Fix infinite loop when /(?<=(*SKIP)ac)a/ is matched against aa.
785    27. Fix the case where there are two or more SKIPs with arguments that may be
786        ignored.
788    28. (*SKIP) is now supported by the JIT compiler.
790    29. (*THEN) is now supported by the JIT compiler.
792    30. Update RunTest with additional test selector options.
794    31. The way PCRE handles backtracking verbs has been changed in two ways.
796        (1) Previously, in something like (*COMMIT)(*SKIP), COMMIT would override
797        SKIP. Now, PCRE acts on whichever backtracking verb is reached first by
798        backtracking. In some cases this makes it more Perl-compatible, but Perl's
799        rather obscure rules do not always do the same thing.
801        (2) Previously, backtracking verbs were confined within assertions. This is
802        no longer the case for positive assertions, except for (*ACCEPT). Again,
803        this sometimes improves Perl compatibility, and sometimes does not.
805    32. A number of tests that were in test 2 because Perl did things differently
806        have been moved to test 1, because either Perl or PCRE has changed, and
807        these tests are now compatible.
809    32. Backtracking control verbs are now handled in the same way in JIT and
810        interpreter.
812    33. An opening parenthesis in a MARK/PRUNE/SKIP/THEN name in a pattern that
813        contained a forward subroutine reference caused a compile error.
815    34. Auto-detect and optimize limited repetitions in JIT.
817    35. Implement PCRE_NEVER_UTF to lock out the use of UTF, in particular,
818        blocking (*UTF) etc.
820    36. In the interpreter, maximizing pattern repetitions for characters and
821        character types now use tail recursion, which reduces stack usage.
823    37. The value of the max lookbehind was not correctly preserved if a compiled
824        and saved regex was reloaded on a host of different endianness.
826    38. Implemented (*LIMIT_MATCH) and (*LIMIT_RECURSION). As part of the extension
827        of the compiled pattern block, expand the flags field from 16 to 32 bits
828        because it was almost full.
830    39. Try madvise first before posix_madvise.
832    40. Change 7 for PCRE 7.9 made it impossible for pcregrep to find empty lines
833        with a pattern such as ^$. It has taken 4 years for anybody to notice! The
834        original change locked out all matches of empty strings. This has been
835        changed so that one match of an empty string per line is recognized.
836        Subsequent searches on the same line (for colouring or for --only-matching,
837        for example) do not recognize empty strings.
839    41. Applied a user patch to fix a number of spelling mistakes in comments.
841    42. Data lines longer than 65536 caused pcretest to crash.
843    43. Clarified the data type for length and startoffset arguments for pcre_exec
844        and pcre_dfa_exec in the function-specific man pages, where they were
845        explicitly stated to be in bytes, never having been updated. I also added
846        some clarification to the pcreapi man page.
848    44. A call to pcre_dfa_exec() with an output vector size less than 2 caused
849        a segmentation fault.
852    Version 8.32 30-November-2012
853    -----------------------------
855    1.  Improved JIT compiler optimizations for first character search and single
856        character iterators.
858    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
859        Patch by Daniel Richard G.
861    3.  Single character iterator optimizations in the JIT compiler.
863    4.  Improved JIT compiler optimizations for character ranges.
865    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
866        Reported by Giuseppe D'Angelo.
868    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
869        of a line, was being set incorrectly in cases where .* appeared inside
870        atomic brackets at the start of a pattern, or where there was a subsequent
871        *PRUNE or *SKIP.
873    7.  Improved instruction cache flush for POWER/PowerPC.
874        Patch by Daniel Richard G.
876    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
877        grep:
879        (a) There is now no limit to the number of patterns to be matched.
881        (b) An error is given if a pattern is too long.
883        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
884            are now supported.
886        (d) --exclude-from and --include-from (multiple use) have been added.
888        (e) Exclusions and inclusions now apply to all files and directories, not
889            just to those obtained from scanning a directory recursively.
891        (f) Multiple uses of -f and --file-list are now supported.
893        (g) In a Windows environment, the default for -d has been changed from
894            "read" (the GNU grep default) to "skip", because otherwise the presence
895            of a directory in the file list provokes an error.
897        (h) The documentation has been revised and clarified in places.
899    9.  Improve the matching speed of capturing brackets.
901    10. Changed the meaning of \X so that it now matches a Unicode extended
902        grapheme cluster.
904    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
905        out POSIX threads when JIT support is configured.
907    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
909    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
910        structure could go wrong in environments where size_t is not the same size
911        as int.
913    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
914        set.
916    15. The EBCDIC support had decayed; later updates to the code had included
917        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
918        general tidy up of EBCDIC-related issues, and the documentation was also
919        not quite right. There is now a test that can be run on ASCII systems to
920        check some of the EBCDIC-related things (but is it not a full test).
922    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
923        in a small tidy to the code.
925    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
927    18. If the --only-matching (-o) option in pcregrep is specified multiple
928        times, each one causes appropriate output. For example, -o1 -o2 outputs the
929        substrings matched by the 1st and 2nd capturing parentheses. A separating
930        string can be specified by --om-separator (default empty).
932    19. Improving the first n character searches.
934    20. Turn case lists for horizontal and vertical white space into macros so that
935        they are defined only once.
937    21. This set of changes together give more compatible Unicode case-folding
938        behaviour for characters that have more than one other case when UCP
939        support is available.
941        (a) The Unicode property table now has offsets into a new table of sets of
942            three or more characters that are case-equivalent. The MultiStage2.py
943            script that generates these tables (the pcre_ucd.c file) now scans
944            CaseFolding.txt instead of UnicodeData.txt for character case
945            information.
947        (b) The code for adding characters or ranges of characters to a character
948            class has been abstracted into a generalized function that also handles
949            case-independence. In UTF-mode with UCP support, this uses the new data
950            to handle characters with more than one other case.
952        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
953            whose other case is greater than 256 are now correctly matched
954            caselessly. Previously, the high codepoint matched the low one, but not
955            vice versa.
957        (d) The processing of \h, \H, \v, and \ in character classes now makes use
958            of the new class addition function, using character lists defined as
959            macros alongside the case definitions of 20 above.
961        (e) Caseless back references now work with characters that have more than
962            one other case.
964        (f) General caseless matching of characters with more than one other case
965            is supported.
967    22. Unicode character properties were updated from Unicode 6.2.0
969    23. Improved CMake support under Windows. Patch by Daniel Richard G.
971    24. Add support for 32-bit character strings, and UTF-32
973    25. Major JIT compiler update (code refactoring and bugfixing).
974        Experimental Sparc 32 support is added.
976    26. Applied a modified version of Daniel Richard G's patch to create
977        pcre.h.generic and config.h.generic by "make" instead of in the
978        PrepareRelease script.
980    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
981        pcre_compile.c when checking for a zero character.
983    28. Introducing a native interface for JIT. Through this interface, the compiled
984        machine code can be directly executed. The purpose of this interface is to
985        provide fast pattern matching, so several sanity checks are not performed.
986        However, feature tests are still performed. The new interface provides
987        1.4x speedup compared to the old one.
989    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
990        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
991        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
993    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
994        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
995        the "old" RFC 2279). Instead, it ended up passing a negative length to
996        pcre_exec().
998    31. Add support for GCC's visibility feature to hide internal functions.
1000    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
1001        "unknown -C option" after outputting 0 or 1.
1003    33. There is now support for generating a code coverage report for the test
1004        suite in environments where gcc is the compiler and lcov is installed. This
1005        is mainly for the benefit of the developers.
1007    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
1008        unaddressable using valgrind annotations, allowing valgrind to detect
1009        invalid memory accesses. This is mainly for the benefit of the developers.
1011    25. (*UTF) can now be used to start a pattern in any of the three libraries.
1013    26. Give configure error if --enable-cpp but no C++ compiler found.
1016    Version 8.31 06-July-2012
1017    -------------------------
1019    1.  Fixing a wrong JIT test case and some compiler warnings.
1021    2.  Removed a bashism from the RunTest script.
1023    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
1024        to unsigned type, result still unsigned" that was given by an MS compiler
1025        on encountering the code "-sizeof(xxx)".
1027    4.  Partial matching support is added to the JIT compiler.
1029    5.  Fixed several bugs concerned with partial matching of items that consist
1030        of more than one character:
1032        (a) /^(..)\1/ did not partially match "aba" because checking references was
1033            done on an "all or nothing" basis. This also applied to repeated
1034            references.
1036        (b) \R did not give a hard partial match if \r was found at the end of the
1037            subject.
1039        (c) \X did not give a hard partial match after matching one or more
1040            characters at the end of the subject.
1042        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
1043            a partial match for the string "\r".
1045        (e) When newline was set to CRLF, the metacharacter "." did not recognize
1046            a partial match for a CR character at the end of the subject string.
1048    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
1049        running pcretest, the text "(JIT)" added to the output whenever JIT is
1050        actually used to run the match.
1052    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
1053        or /S+[+] with a digit between 1 and 7.
1055    8.  OP_NOT now supports any UTF character not just single-byte ones.
1057    9.  (*MARK) control verb is now supported by the JIT compiler.
1059    10. The command "./RunTest list" lists the available tests without actually
1060        running any of them. (Because I keep forgetting what they all are.)
1064    12. Applied a (slightly modified) user-supplied patch that improves performance
1065        when the heap is used for recursion (compiled with --disable-stack-for-
1066        recursion). Instead of malloc and free for each heap frame each time a
1067        logical recursion happens, frames are retained on a chain and re-used where
1068        possible. This sometimes gives as much as 30% improvement.
1070    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
1071        call.
1073    14. As documented, (*COMMIT) is now confined to within a positive assertion.
1075    15. It is now possible to link pcretest with libedit as an alternative to
1076        libreadline.
1078    16. (*COMMIT) control verb is now supported by the JIT compiler.
1080    17. The Unicode data tables have been updated to Unicode 6.1.0.
1082    18. Added --file-list option to pcregrep.
1084    19. Added binary file support to pcregrep, including the -a, --binary-files,
1085        -I, and --text options.
1087    20. The madvise function is renamed for posix_madvise for QNX compatibility
1088        reasons. Fixed by Giuseppe D'Angelo.
1090    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
1092    22. Changed the option for creating symbolic links for 16-bit man pages from
1093        -s to -sf so that re-installing does not cause issues.
1095    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
1097    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
1098        matches in certain environments (the workspace was not being correctly
1099        retained). Also added to pcre_dfa_exec() a simple plausibility check on
1100        some of the workspace data at the beginning of a restart.
1102    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
1103        was not doing so when it should - probably a typo introduced by SVN 528
1104        (change 8.10/14).
1106    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
1107        \w+ when the character tables indicated that \x{c4} was a word character.
1108        There were several related cases, all because the tests for doing a table
1109        lookup were testing for characters less than 127 instead of 255.
1111    27. If a pattern contains capturing parentheses that are not used in a match,
1112        their slots in the ovector are set to -1. For those that are higher than
1113        any matched groups, this happens at the end of processing. In the case when
1114        there were back references that the ovector was too small to contain
1115        (causing temporary malloc'd memory to be used during matching), and the
1116        highest capturing number was not used, memory off the end of the ovector
1117        was incorrectly being set to -1. (It was using the size of the temporary
1118        memory instead of the true size.)
1120    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
1121        ovector size, it uses memory at the end of the block that it has got.
1123    29. Check for an overlong MARK name and give an error at compile time. The
1124        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
1126    30. JIT compiler update.
1128    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
1129        Rill for the patch.
1131    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
1133    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
1135    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
1136        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
1137        was enabled.
1139    35. Improve JIT code generation for greedy plus quantifier.
1141    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
1142        1 to "aa" instead of to an empty string. The bug affected repeated groups
1143        that could potentially match an empty string.
1145    37. Optimizing single character iterators in JIT.
1147    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
1148        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
1149        codepoints that are too big for the mode are faulted, and in a UTF mode,
1150        disallowed codepoints are also faulted.
1152    39. If PCRE was compiled with UTF support, in three places in the DFA
1153        matcher there was code that should only have been obeyed in UTF mode, but
1154        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
1155        processing when bytes with values greater than 127 were present. In 16-bit
1156        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
1157        both cases the values are those that cannot be the first data item in a UTF
1158        character. The three items that might have provoked this were recursions,
1159        possessively repeated groups, and atomic groups.
1161    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
1162        and pcregrep, because some OS require shared objects to be explicitly
1163        passed to ld, causing the link step to fail if they are not.
1165    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
1166        mode, patterns that started with \h* or \R* might be incorrectly matched.
1169    Version 8.30 04-February-2012
1170    -----------------------------
1172    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
1173        name is defined in ctype.h.
1175    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
1176        only in quite long subpatterns.
1178    3.  Removed the function pcre_info(), which has been obsolete and deprecated
1179        since it was replaced by pcre_fullinfo() in February 2000.
1181    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
1182        match a (*MARK), and the match failed at the start of the subject, a
1183        reference to memory before the start of the subject could occur. This bug
1184        was introduced by fix 17 of release 8.21.
1186    5.  A reference to an unset group with zero minimum repetition was giving
1187        totally wrong answers (in non-JavaScript-compatibility mode). For example,
1188        /(another)?(\1?)test/ matched against "hello world test". This bug was
1189        introduced in release 8.13.
1191    6.  Add support for 16-bit character strings (a large amount of work involving
1192        many changes and refactorings).
1194    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
1195        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
1196        from a file.
1198    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
1199        rounding is not applied in this particular case).
1201    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
1202        if they appear, or are escaped, in patterns.
1204    10. Get rid of a number of -Wunused-but-set-variable warnings.
1206    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
1207        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
1208        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
1209        also returns the mark "x". This bug applied to capturing parentheses,
1210        non-capturing parentheses, and atomic parentheses. It also applied to some
1211        assertions.
1213    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
1214        information out of configure.ac instead of relying on pcre.h.generic, which
1215        is not stored in the repository.
1217    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
1218        -lreadline.
1220    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
1222    15. Applied Graycode's patch to put the top-level frame on the stack rather
1223        than the heap when not using the stack for recursion. This gives a
1224        performance improvement in many cases when recursion is not deep.
1226    16. Experimental code added to "pcretest -C" to output the stack frame size.
1229    Version 8.21 12-Dec-2011
1230    ------------------------
1232    1.  Updating the JIT compiler.
1234    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
1235        are added as well.
1237    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
1238        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
1239        calling _pcre_jit_exec. Some extra comments are added.
1241    4.  (*MARK) settings inside atomic groups that do not contain any capturing
1242        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
1243        was introduced by change 18 for 8.20.
1245    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
1246        ECMA-262 standard.
1248    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
1249        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
1250        This bug was probably introduced by change 9 of 8.13.
1252    7.  While fixing 6 above, I noticed that a number of other items were being
1253        incorrectly rejected as "not fixed length". This arose partly because newer
1254        opcodes had not been added to the fixed-length checking code. I have (a)
1255        corrected the bug and added tests for these items, and (b) arranged for an
1256        error to occur if an unknown opcode is encountered while checking for fixed
1257        length instead of just assuming "not fixed length". The items that were
1258        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
1259        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
1260        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
1262    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
1263        being incorrectly compiled and would have given unpredicatble results.
1265    9.  A possessively repeated subpattern with minimum repeat count greater than
1266        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
1267        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
1268        the first (A) could occur when it should not.
1270    10. Add a cast and remove a redundant test from the code.
1272    11. JIT should use pcre_malloc/pcre_free for allocation.
1274    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
1275        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
1276        is anything other than /usr, -L is still shown).
1278    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
1280    14. Perl does not support \N without a following name in a [] class; PCRE now
1281        also gives an error.
1283    15. If a forward reference was repeated with an upper limit of around 2000,
1284        it caused the error "internal error: overran compiling workspace". The
1285        maximum number of forward references (including repeats) was limited by the
1286        internal workspace, and dependent on the LINK_SIZE. The code has been
1287        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
1288        the default depends on LINK_SIZE. There is a new upper limit (for safety)
1289        of around 200,000 forward references. While doing this, I also speeded up
1290        the filling in of repeated forward references.
1292    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
1293        incorrectly expecting the subject to contain another "a" after the start.
1295    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
1296        in the match, the SKIP should be ignored. This was not happening; instead
1297        the SKIP was being treated as NOMATCH. For patterns such as
1298        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
1299        tested.
1301    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
1302        now much more compatible with Perl, in particular in cases where the result
1303        is a non-match for a non-anchored pattern. For example, if
1304        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
1305        "m", where previously it did not return a name. A side effect of this
1306        change is that for partial matches, the last encountered mark name is
1307        returned, as for non matches. A number of tests that were previously not
1308        Perl-compatible have been moved into the Perl-compatible test files. The
1309        refactoring has had the pleasing side effect of removing one argument from
1310        the match() function, thus reducing its stack requirements.
1312    19. If the /S+ option was used in pcretest to study a pattern using JIT,
1313        subsequent uses of /S (without +) incorrectly behaved like /S+.
1315    21. Retrieve executable code size support for the JIT compiler and fixing
1316        some warnings.
1318    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
1319        not work when the shorter character appeared right at the end of the
1320        subject string.
1322    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
1323        systems.
1325    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
1326        output it when the /M option is used in pcretest.
1328    25. The CheckMan script was not being included in the distribution. Also, added
1329        an explicit "perl" to run Perl scripts from the PrepareRelease script
1330        because this is reportedly needed in Windows.
1332    26. If study data was being save in a file and studying had not found a set of
1333        "starts with" bytes for the pattern, the data written to the file (though
1334        never used) was taken from uninitialized memory and so caused valgrind to
1335        complain.
1337    27. Updated RunTest.bat as provided by Sheri Pierce.
1339    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
1341    29. Computation of memory usage for the table of capturing group names was
1342        giving an unnecessarily large value.
1345    Version 8.20 21-Oct-2011
1346    ------------------------
1348    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
1349        a POSIX class. After further experiments with Perl, which convinced me that
1350        Perl has bugs and confusions, a closing square bracket is no longer allowed
1351        in a POSIX name. This bug also affected patterns with classes that started
1352        with full stops.
1354    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
1355        captured substring, but while checking the failing first alternative,
1356        substring 1 is temporarily captured. If the output vector supplied to
1357        pcre_exec() was not big enough for this capture, the yield of the function
1358        was still zero ("insufficient space for captured substrings"). This cannot
1359        be totally fixed without adding another stack variable, which seems a lot
1360        of expense for a edge case. However, I have improved the situation in cases
1361        such as /(a)(b)x|abc/ matched against "abc", where the return code
1362        indicates that fewer than the maximum number of slots in the ovector have
1363        been set.
1365    3.  Related to (2) above: when there are more back references in a pattern than
1366        slots in the output vector, pcre_exec() uses temporary memory during
1367        matching, and copies in the captures as far as possible afterwards. It was
1368        using the entire output vector, but this conflicts with the specification
1369        that only 2/3 is used for passing back captured substrings. Now it uses
1370        only the first 2/3, for compatibility. This is, of course, another edge
1371        case.
1373    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
1374        main code base, and can be used by building with --enable-jit. When this is
1375        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
1376        runtime --no-jit option is given.
1378    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
1379        ovector, the return from the function was zero, implying that there were
1380        other matches that did not fit. The correct "exactly full" value is now
1381        returned.
1383    6.  If a subpattern that was called recursively or as a subroutine contained
1384        (*PRUNE) or any other control that caused it to give a non-standard return,
1385        invalid errors such as "Error -26 (nested recursion at the same subject
1386        position)" or even infinite loops could occur.
1388    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
1389        computing the minimum length on reaching *ACCEPT, and so ended up with the
1390        wrong value of 1 rather than 0. Further investigation indicates that
1391        computing a minimum subject length in the presence of *ACCEPT is difficult
1392        (think back references, subroutine calls), and so I have changed the code
1393        so that no minimum is registered for a pattern that contains *ACCEPT.
1395    8.  If (*THEN) was present in the first (true) branch of a conditional group,
1396        it was not handled as intended. [But see 16 below.]
1398    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
1399        Sheri Pierce.
1401    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
1402        the first byte in a match must be "a".
1404    11. Change 17 for 8.13 increased the recursion depth for patterns like
1405        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
1406        pattern contains any instances of (*THEN). If it does not, the old
1407        optimizations are restored. It would be nice to do this on a per-group
1408        basis, but at the moment that is not feasible.
1410    12. In some environments, the output of pcretest -C is CRLF terminated. This
1411        broke RunTest's code that checks for the link size. A single white space
1412        character after the value is now allowed for.
1414    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
1415        For "fr", it uses the Windows-specific input and output files.
1417    14. If (*THEN) appeared in a group that was called recursively or as a
1418        subroutine, it did not work as intended. [But see next item.]
1420    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
1421        pattern fragments (but not containing any | characters). If A and B are
1422        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
1423        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
1424        D. In other words, Perl considers parentheses that do not contain any |
1425        characters to be part of a surrounding alternative, whereas PCRE was
1426        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
1427        differently. PCRE now behaves in the same way as Perl, except in the case
1428        of subroutine/recursion calls such as (?1) which have in any case always
1429        been different (but PCRE had them first :-).
1431    16. Related to 15 above: Perl does not treat the | in a conditional group as
1432        creating alternatives. Such a group is treated in the same way as an
1433        ordinary group without any | characters when processing (*THEN). PCRE has
1434        been changed to match Perl's behaviour.
1436    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
1437        RunGrepTest script failed.
1439    18. Change 22 for version 13 caused atomic groups to use more stack. This is
1440        inevitable for groups that contain captures, but it can lead to a lot of
1441        stack use in large patterns. The old behaviour has been restored for atomic
1442        groups that do not contain any capturing parentheses.
1444    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
1445        suppress the check for a minimum subject length at run time. (If it was
1446        given to pcre_exec() or pcre_dfa_exec() it did work.)
1448    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
1449        fail to work when decoding hex characters in data strings in EBCDIC
1450        environments.
1452    21. It appears that in at least one Mac OS environment, the isxdigit() function
1453        is implemented as a macro that evaluates to its argument more than once,
1454        contravening the C 90 Standard (I haven't checked a later standard). There
1455        was an instance in pcretest which caused it to go wrong when processing
1456        \x{...} escapes in subject strings. The has been rewritten to avoid using
1457        things like p++ in the argument of isxdigit().
1460    Version 8.13 16-Aug-2011
1461    ------------------------
1463    1.  The Unicode data tables have been updated to Unicode 6.0.0.
1465    2.  Two minor typos in pcre_internal.h have been fixed.
1467    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
1468        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
1469        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
1471    4.  There were a number of related bugs in the code for matching backrefences
1472        caselessly in UTF-8 mode when codes for the characters concerned were
1473        different numbers of bytes. For example, U+023A and U+2C65 are an upper
1474        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
1475        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
1476        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
1477        2-byte code at the end of the subject (it thought there wasn't enough data
1478        left).
1480    5.  Comprehensive information about what went wrong is now returned by
1481        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
1482        as the output vector has at least 2 elements. The offset of the start of
1483        the failing character and a reason code are placed in the vector.
1485    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
1486        now returned is for the first byte of the failing character, instead of the
1487        last byte inspected. This is an incompatible change, but I hope it is small
1488        enough not to be a problem. It makes the returned offset consistent with
1489        pcre_exec() and pcre_dfa_exec().
1491    7.  pcretest now gives a text phrase as well as the error number when
1492        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
1493        failure, the offset and reason code are output.
1495    8.  When \R was used with a maximizing quantifier it failed to skip backwards
1496        over a \r\n pair if the subsequent match failed. Instead, it just skipped
1497        back over a single character (\n). This seems wrong (because it treated the
1498        two characters as a single entity when going forwards), conflicts with the
1499        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
1500        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
1501        has been changed.
1503    9.  Some internal refactoring has changed the processing so that the handling
1504        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
1505        time (the PCRE_DOTALL option was changed this way some time ago: version
1506        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
1507        which was always a bit of a fudge. It also means that there is one less
1508        argument for the match() function, which reduces its stack requirements
1509        slightly. This change also fixes an incompatibility with Perl: the pattern
1510        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
1512    10. More internal refactoring has drastically reduced the number of recursive
1513        calls to match() for possessively repeated groups such as (abc)++ when
1514        using pcre_exec().
1516    11. While implementing 10, a number of bugs in the handling of groups were
1517        discovered and fixed:
1519        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
1520        (a|)*(?1) gave a compile-time internal error.
1521        ((a|)+)+  did not notice that the outer group could match an empty string.
1522        (^a|^)+   was not marked as anchored.
1523        (.*a|.*)+ was not marked as matching at start or after a newline.
1525    12. Yet more internal refactoring has removed another argument from the match()
1526        function. Special calls to this function are now indicated by setting a
1527        value in a variable in the "match data" data block.
1529    13. Be more explicit in pcre_study() instead of relying on "default" for
1530        opcodes that mean there is no starting character; this means that when new
1531        ones are added and accidentally left out of pcre_study(), testing should
1532        pick them up.
1534    14. The -s option of pcretest has been documented for ages as being an old
1535        synonym of -m (show memory usage). I have changed it to mean "force study
1536        for every regex", that is, assume /S for every regex. This is similar to -i
1537        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
1538        using it. It makes it easier to run collections of tests with and without
1539        study enabled, and thereby test pcre_study() more easily. All the standard
1540        tests are now run with and without -s (but some patterns can be marked as
1541        "never study" - see 20 below).
1543    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
1544        restoration of the capturing data to the outer values was not happening
1545        correctly.
1547    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
1548        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
1549        pattern had matched an empty string, and so incorrectly returned a no
1550        match.
1552    17. There was optimizing code for the last branch of non-capturing parentheses,
1553        and also for the obeyed branch of a conditional subexpression, which used
1554        tail recursion to cut down on stack usage. Unfortunately, now that there is
1555        the possibility of (*THEN) occurring in these branches, tail recursion is
1556        no longer possible because the return has to be checked for (*THEN). These
1557        two optimizations have therefore been removed. [But see 8.20/11 above.]
1559    18. If a pattern containing \R was studied, it was assumed that \R always
1560        matched two bytes, thus causing the minimum subject length to be
1561        incorrectly computed because \R can also match just one byte.
1563    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
1564        was incorrectly computed.
1566    20. If /S is present twice on a test pattern in pcretest input, it now
1567        *disables* studying, thereby overriding the use of -s on the command line
1568        (see 14 above). This is necessary for one or two tests to keep the output
1569        identical in both cases.
1571    21. When (*ACCEPT) was used in an assertion that matched an empty string and
1572        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
1574    22. When an atomic group that contained a capturing parenthesis was
1575        successfully matched, but the branch in which it appeared failed, the
1576        capturing was not being forgotten if a higher numbered group was later
1577        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
1578        group 1 to "a", when in fact it should be unset. This applied to multi-
1579        branched capturing and non-capturing groups, repeated or not, and also to
1580        positive assertions (capturing in negative assertions does not happen
1581        in PCRE) and also to nested atomic groups.
1583    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
1584        subject after a captured substring, to make it easier to tell which of a
1585        number of identical substrings has been captured.
1587    24. The way atomic groups are processed by pcre_exec() has been changed so that
1588        if they are repeated, backtracking one repetition now resets captured
1589        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
1590        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
1591        "aaa". Previously, it would have been "a". As part of this code
1592        refactoring, the way recursive calls are handled has also been changed.
1594    25. If an assertion condition captured any substrings, they were not passed
1595        back unless some other capturing happened later. For example, if
1596        (?(?=(a))a) was matched against "a", no capturing was returned.
1598    26. When studying a pattern that contained subroutine calls or assertions,
1599        the code for finding the minimum length of a possible match was handling
1600        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
1601        group 1 called group 2 while simultaneously a separate group 2 called group
1602        1). A stack overflow occurred in this case. I have fixed this by limiting
1603        the recursion depth to 10.
1605    27. Updated RunTest.bat in the distribution to the version supplied by Tom
1606        Fortmann. This supports explicit test numbers on the command line, and has
1607        argument validation and error reporting.
1609    28. An instance of \X with an unlimited repeat could fail if at any point the
1610        first character it looked at was a mark character.
1612    29. Some minor code refactoring concerning Unicode properties and scripts
1613        should reduce the stack requirement of match() slightly.
1615    30. Added the '=' option to pcretest to check the setting of unused capturing
1616        slots at the end of the pattern, which are documented as being -1, but are
1617        not included in the return count.
1619    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
1620        compiled something random. Now it gives a compile-time error (as does
1621        Perl).
1623    32. A *MARK encountered during the processing of a positive assertion is now
1624        recorded and passed back (compatible with Perl).
1626    33. If --only-matching or --colour was set on a pcregrep call whose pattern
1627        had alternative anchored branches, the search for a second match in a line
1628        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
1629        matched the line "0102" twice. The same bug affected patterns that started
1630        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
1631        twice.
1633    34. Previously, PCRE did not allow quantification of assertions. However, Perl
1634        does, and because of capturing effects, quantifying parenthesized
1635        assertions may at times be useful. Quantifiers are now allowed for
1636        parenthesized assertions.
1638    35. A minor code tidy in pcre_compile() when checking options for \R usage.
1640    36. \g was being checked for fancy things in a character class, when it should
1641        just be a literal "g".
1643    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
1644        appearance of a nested POSIX class supersedes an apparent external class.
1645        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
1646        unescaped square brackets may also appear as part of class names. For
1647        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
1648        more like Perl. (But see 8.20/1 above.)
1650    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
1651        was because it thought it was \N{name}, which is not supported).
1653    39. Add minix to OS list not supporting the -S option in pcretest.
1655    40. PCRE tries to detect cases of infinite recursion at compile time, but it
1656        cannot analyze patterns in sufficient detail to catch mutual recursions
1657        such as ((?1))((?2)). There is now a runtime test that gives an error if a
1658        subgroup is called recursively as a subpattern for a second time at the
1659        same position in the subject string. In previous releases this might have
1660        been caught by the recursion limit, or it might have run out of stack.
1662    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
1663        happen only once. PCRE was, however incorrectly giving a compile time error
1664        "recursive call could loop indefinitely" because it cannot analyze the
1665        pattern in sufficient detail. The compile time test no longer happens when
1666        PCRE is compiling a conditional subpattern, but actual runaway loops are
1667        now caught at runtime (see 40 above).
1669    42. It seems that Perl allows any characters other than a closing parenthesis
1670        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
1671        has been changed to be the same.
1673    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
1674        as not to get warnings when autogen.sh is called. Also changed
1675        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
1677    44. To help people who use pcregrep to scan files containing exceedingly long
1678        lines, the following changes have been made:
1680        (a) The default value of the buffer size parameter has been increased from
1681            8K to 20K. (The actual buffer used is three times this size.)
1683        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
1684            PCRE is built.
1686        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
1687            to be set at run time.
1689        (d) Numerical values in pcregrep options can be followed by K or M, for
1690            example --buffer-size=50K.
1692        (e) If a line being scanned overflows pcregrep's buffer, an error is now
1693            given and the return code is set to 2.
1695    45. Add a pointer to the latest mark to the callout data block.
1697    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
1698        partial match of an empty string instead of no match. This was specific to
1699        the use of ".".
1701    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
1702        complete match instead of a partial match. This bug was dependent on both
1703        the PCRE_UTF8 and PCRE_DOTALL options being set.
1705    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
1706        starting byte set, because \b was not being ignored.
1709    Version 8.12 15-Jan-2011
1710    ------------------------
1712    1.  Fixed some typos in the markup of the man pages, and wrote a script that
1713        checks for such things as part of the documentation building process.
1715    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
1716        --match-limit and --recursion-limit options (added for 8.11). In
1717        particular, this made one of the standard tests fail. (The integer value
1718        went into the wrong half of a long int.)
1720    3.  If the --colour option was given to pcregrep with -v (invert match), it
1721        did strange things, either producing crazy output, or crashing. It should,
1722        of course, ignore a request for colour when reporting lines that do not
1723        match.
1725    4.  Another pcregrep bug caused similar problems if --colour was specified with
1726        -M (multiline) and the pattern match finished with a line ending.
1728    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
1729        matched in multiline mode, the following line was shown as part of the
1730        match. This seems wrong, so I have changed it.
1732    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
1733        the check for further matches in the same line (so they could be coloured)
1734        to overrun the end of the current line. If another match was found, it was
1735        incorrectly shown (and then shown again when found in the next line).
1737    7.  If pcregrep was compiled under Windows, there was a reference to the
1738        function pcregrep_exit() before it was defined. I am assuming this was
1739        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
1740        reported by a user. I've moved the definition above the reference.
1743    Version 8.11 10-Dec-2010
1744    ------------------------
1746    1.  (*THEN) was not working properly if there were untried alternatives prior
1747        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
1748        backtracked to try for "b" instead of moving to the next alternative branch
1749        at the same level (in this case, to look for "c"). The Perl documentation
1750        is clear that when (*THEN) is backtracked onto, it goes to the "next
1751        alternative in the innermost enclosing group".
1753    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
1754        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
1755        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
1756        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
1757        (*THEN).
1759    3.  If \s appeared in a character class, it removed the VT character from
1760        the class, even if it had been included by some previous item, for example
1761        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
1762        of \s, but is part of the POSIX "space" class.)
1764    4.  A partial match never returns an empty string (because you can always
1765        match an empty string at the end of the subject); however the checking for
1766        an empty string was starting at the "start of match" point. This has been
1767        changed to the "earliest inspected character" point, because the returned
1768        data for a partial match starts at this character. This means that, for
1769        example, /(?<=abc)def/ gives a partial match for the subject "abc"
1770        (previously it gave "no match").
1772    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
1773        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
1774        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
1775        has an implication that the given string is incomplete (because a partial
1776        match is preferred over a full match). For this reason, these items now
1777        give a partial match in this situation. [Aside: previously, the one case
1778        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
1779        match rather than a full match, which was wrong by the old rules, but is
1780        now correct.]
1782    6.  There was a bug in the handling of #-introduced comments, recognized when
1783        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
1784        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
1785        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
1786        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
1787        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
1788        places in pcre_compile().
1790    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
1791        comments when looking ahead for named forward references to subpatterns,
1792        the only newline sequence it recognized was NL. It now handles newlines
1793        according to the set newline convention.
1795    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
1796        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
1797        cater for a lack of strerror(). These oversights have been fixed.
1799    9.  Added --match-limit and --recursion-limit to pcregrep.
1801    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
1803    11. When the -o option was used, pcregrep was setting a return code of 1, even
1804        when matches were found, and --line-buffered was not being honoured.
1806    12. Added an optional parentheses number to the -o and --only-matching options
1807        of pcregrep.
1809    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
1810        can match an empty string. The code to do it in pcretest and pcredemo
1811        needed fixing:
1813        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
1814            only one byte after an empty string match just before CRLF (this case
1815            just got forgotten; "any" and "anycrlf" were OK).
1817        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
1818            mode when an empty string match preceded an ASCII character followed by
1819            a non-ASCII character. (The code for advancing by one character rather
1820            than one byte was nonsense.)
1822        (c) The pcredemo.c sample program did not have any code at all to handle
1823            the cases when CRLF is a valid newline sequence.
1825    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
1826        as a starting offset was within the subject string. There is now a new
1827        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
1828        negative or greater than the length of the string. In order to test this,
1829        pcretest is extended to allow the setting of negative starting offsets.
1831    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
1832        starting offset points to the beginning of a UTF-8 character was
1833        unnecessarily clumsy. I tidied it up.
1835    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
1836        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
1838    17. Nobody had reported that the --include_dir option, which was added in
1839        release 7.7 should have been called --include-dir (hyphen, not underscore)
1840        for compatibility with GNU grep. I have changed it to --include-dir, but
1841        left --include_dir as an undocumented synonym, and the same for
1842        --exclude-dir, though that is not available in GNU grep, at least as of
1843        release 2.5.4.
1845    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
1846        characters from a string of bytes) have been redefined so as not to use
1847        loops, in order to improve performance in some environments. At the same
1848        time, I abstracted some of the common code into auxiliary macros to save
1849        repetition (this should not affect the compiled code).
1851    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1852        compile-time error is now given if \c is not followed by an ASCII
1853        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1854        different, and any byte value is allowed.)
1856    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1857        START_OPTIMIZE option, which is now allowed at compile time - but just
1858        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1859        to pcregrep and other applications that have no direct access to PCRE
1860        options. The new /Y option in pcretest sets this option when calling
1861        pcre_compile().
1863    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1864        back references. Groups containing recursive back references were forced to
1865        be atomic by that change, but in the case of named groups, the amount of
1866        memory required was incorrectly computed, leading to "Failed: internal
1867        error: code overflow". This has been fixed.
1869    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1870        pcretest.c, to avoid build problems in some Borland environments.
1873    Version 8.10 25-Jun-2010
1874    ------------------------
1876    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1877        THEN.
1879    2.  (*ACCEPT) was not working when inside an atomic group.
1881    3.  Inside a character class, \B is treated as a literal by default, but
1882        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1883        causes the error). The code is unchanged, but I tidied the documentation.
1885    4.  Inside a character class, PCRE always treated \R and \X as literals,
1886        whereas Perl faults them if its -w option is set. I have changed PCRE so
1887        that it faults them when PCRE_EXTRA is set.
1889    5.  Added support for \N, which always matches any character other than
1890        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1892    6.  When compiling pcregrep with newer versions of gcc which may have
1893        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1894        declared with attribute warn_unused_result" were given. Just casting the
1895        result to (void) does not stop the warnings; a more elaborate fudge is
1896        needed. I've used a macro to implement this.
1898    7.  Minor change to pcretest.c to avoid a compiler warning.
1900    8.  Added four artifical Unicode properties to help with an option to make
1901        \s etc use properties (see next item). The new properties are: Xan
1902        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1904    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1905        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1906        this option. Modified pcretest to add /W to test this facility. Added
1907        REG_UCP to make it available via the POSIX interface.
1909    10. Added --line-buffered to pcregrep.
1911    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1912        studied, and the match started with a letter with a code point greater than
1913        127 whose first byte was different to the first byte of the other case of
1914        the letter, the other case of this starting letter was not recognized
1915        (#976).
1917    12. If a pattern that was studied started with a repeated Unicode property
1918        test, for example, \p{Nd}+, there was the theoretical possibility of
1919        setting up an incorrect bitmap of starting bytes, but fortunately it could
1920        not have actually happened in practice until change 8 above was made (it
1921        added property types that matched character-matching opcodes).
1923    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1924        possible starting bytes for non-anchored patterns.
1926    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1927        \R, and also a number of cases that involve Unicode properties, both
1928        explicit and implicit when PCRE_UCP is set.
1930    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1931        input, it could crash or give wrong results if characters with values
1932        greater than 0xc0 were present in the subject string. (Detail: it assumed
1933        UTF-8 input when processing these items.)
1935    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1936        size_t is 64-bit (#991).
1938    17. Added a check for running out of memory when PCRE is compiled with
1939        --disable-stack-for-recursion (#990).
1941    18. If the last data line in a file for pcretest does not have a newline on
1942        the end, a newline was missing in the output.
1944    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1945        less than 128) in its various bitmaps. However, there is a facility for
1946        generating tables according to the current locale when PCRE is compiled. It
1947        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1948        characters, are recognized by isspace() and therefore were getting set in
1949        these tables, and indeed these tables seem to approximate to ISO 8859. This
1950        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1951        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1952        which of course cannot start UTF-8 characters. I have changed the code so
1953        that only real ASCII characters (less than 128) and the correct starting
1954        bytes for UTF-8 encodings are set for characters greater than 127 when in
1955        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1956        altogether.)
1958    20. Added the /T option to pcretest so as to be able to run tests with non-
1959        standard character tables, thus making it possible to include the tests
1960        used for 19 above in the standard set of tests.
1962    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1963        reference to a subpattern the other side of a comment that contains an
1964        opening parenthesis caused either an internal compiling error, or a
1965        reference to the wrong subpattern.
1968    Version 8.02 19-Mar-2010
1969    ------------------------
1971    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1973    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1974        configured.
1976    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1977        original author of that file, following a query about its status.
1979    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1980        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1982    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1983        quantifier applied to a forward-referencing subroutine call, could compile
1984        incorrect code or give the error "internal error: previously-checked
1985        referenced subpattern not found".
1987    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1988        variables to point to external functions. For these systems, therefore,
1989        pcre_malloc etc. are now initialized to local functions that call the
1990        relevant global functions.
1992    7.  There were two entries missing in the vectors called coptable and poptable
1993        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1994        I've fixed the data, and added a kludgy way of testing at compile time that
1995        the lengths are correct (equal to the number of opcodes).
1997    8.  Following on from 7, I added a similar kludge to check the length of the
1998        eint vector in pcreposix.c.
2000    9.  Error texts for pcre_compile() are held as one long string to avoid too
2001        much relocation at load time. To find a text, the string is searched,
2002        counting zeros. There was no check for running off the end of the string,
2003        which could happen if a new error number was added without updating the
2004        string.
2006    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
2008    11. \K was not working if it appeared in an atomic group or in a group that
2009        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
2010        \K is "not well defined" if used in an assertion. PCRE now accepts it if
2011        the assertion is positive, but not if it is negative.
2013    12. Change 11 fortuitously reduced the size of the stack frame used in the
2014        "match()" function of pcre_exec.c by one pointer. Forthcoming
2015        implementation of support for (*MARK) will need an extra pointer on the
2016        stack; I have reserved it now, so that the stack frame size does not
2017        decrease.
2019    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
2020        item in branch that calls a recursion is a subroutine call - as in the
2021        second branch in the above example - was incorrectly given the compile-
2022        time error "recursive call could loop indefinitely" because pcre_compile()
2023        was not correctly checking the subroutine for matching a non-empty string.
2025    14. The checks for overrunning compiling workspace could trigger after an
2026        overrun had occurred. This is a "should never occur" error, but it can be
2027        triggered by pathological patterns such as hundreds of nested parentheses.
2028        The checks now trigger 100 bytes before the end of the workspace.
2030    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
2033    Version 8.01 19-Jan-2010
2034    ------------------------
2036    1.  If a pattern contained a conditional subpattern with only one branch (in
2037        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
2038        computed the wrong minimum data length (which is of course zero for such
2039        subpatterns). This could cause incorrect "no match" results.
2041    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
2042        the pattern is reset in the first branch, pcre_compile() failed with
2043        "internal error: code overflow at offset...". This happened only when
2044        the reset was to the original external option setting. (An optimization
2045        abstracts leading options settings into an external setting, which was the
2046        cause of this.)
2048    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
2049        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
2050        assertion pattern did not match (meaning that the assertion was true), it
2051        was incorrectly treated as false if the SKIP had been reached during the
2052        matching. This also applied to assertions used as conditions.
2054    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
2055        assertion subpattern, including such a pattern used as a condition,
2056        unpredictable results occurred, instead of the error return
2059    5.  The C++ GlobalReplace function was not working like Perl for the special
2060        situation when an empty string is matched. It now does the fancy magic
2061        stuff that is necessary.
2063    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
2064        removed. (These were left over from very, very early versions of PCRE.)
2066    7.  Some cosmetic changes to the code to make life easier when compiling it
2067        as part of something else:
2069        (a) Change DEBUG to PCRE_DEBUG.
2071        (b) In pcre_compile(), rename the member of the "branch_chain" structure
2072            called "current" as "current_branch", to prevent a collision with the
2073            Linux macro when compiled as a kernel module.
2075        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
2076            prevent a collision with the Linux macro when compiled as a kernel
2077            module.
2079    8.  In pcre_compile() there are some checks for integer overflows that used to
2080        cast potentially large values to (double). This has been changed to that
2081        when building, a check for int64_t is made, and if it is found, it is used
2082        instead, thus avoiding the use of floating point arithmetic. (There is no
2083        other use of FP in PCRE.) If int64_t is not found, the fallback is to
2084        double.
2086    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
2087        2005 (difference between two addresses compared to an unsigned value).
2089    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
2090        custom one, because of the following reported problem in Windows:
2092          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
2093              under Win32.
2094          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
2095              therefore missing the function definition.
2096          - The compiler thus generates a "C" signature for the test function.
2097          - The linker fails to find the "C" function.
2098          - PCRE fails to configure if asked to do so against libbz2.
2100    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
2101        messages were output:
2103          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
2104          rerunning libtoolize, to keep the correct libtool macros in-tree.
2105          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
2107        I have done both of these things.
2109    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
2110        most of the time, it *can* run out if it is given a pattern that contains a
2111        runaway infinite recursion. I updated the discussion in the pcrestack man
2112        page.
2114    13. Now that we have gone to the x.xx style of version numbers, the minor
2115        version may start with zero. Using 08 or 09 is a bad idea because users
2116        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
2117        interpreted as invalid octal numbers. I've updated the previous comment in
2118        configure.ac, and also added a check that gives an error if 08 or 09 are
2119        used.
2121    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
2122        causing partial matching to fail when the end of the subject matched \W
2123        in a UTF-8 pattern where \W was quantified with a minimum of 3.
2125    15. There were some discrepancies between the declarations in pcre_internal.h
2126        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
2127        their definitions. The declarations used "const uschar *" and the
2128        definitions used USPTR. Even though USPTR is normally defined as "const
2129        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
2130        reported that: "This difference in casting confuses some C++ compilers, for
2131        example, SunCC recognizes above declarations as different functions and
2132        generates broken code for hbpcre." I have changed the declarations to use
2133        USPTR.
2135    16. GNU libtool is named differently on some systems. The autogen.sh script now
2136        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
2137        (FreeBSD).
2139    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
2140        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
2141        comment: "Figure out how to create a longlong from a string: strtoll and
2142        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
2143        instance, but it only takes 2 args instead of 3!"
2145    18. A subtle bug concerned with back references has been fixed by a change of
2146        specification, with a corresponding code fix. A pattern such as
2147        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
2148        refers, was giving matches when it shouldn't. For example, xa=xaaa would
2149        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
2150        same bug. Such groups have to be quantified to be useful, or contained
2151        inside another quantified group. (If there's no repetition, the reference
2152        can never match.) The problem arises because, having left the group and
2153        moved on to the rest of the pattern, a later failure that backtracks into
2154        the group uses the captured value from the final iteration of the group
2155        rather than the correct earlier one. I have fixed this in PCRE by forcing
2156        any group that contains a reference to itself to be an atomic group; that
2157        is, there cannot be any backtracking into it once it has completed. This is
2158        similar to recursive and subroutine calls.
2161    Version 8.00 19-Oct-09
2162    ----------------------
2164    1.  The table for translating pcre_compile() error codes into POSIX error codes
2165        was out-of-date, and there was no check on the pcre_compile() error code
2166        being within the table. This could lead to an OK return being given in
2167        error.
2169    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
2170        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
2171        in a Windows environment.
2173    3.  The pcregrep --count option prints the count for each file even when it is
2174        zero, as does GNU grep. However, pcregrep was also printing all files when
2175        --files-with-matches was added. Now, when both options are given, it prints
2176        counts only for those files that have at least one match. (GNU grep just
2177        prints the file name in this circumstance, but including the count seems
2178        more useful - otherwise, why use --count?) Also ensured that the
2179        combination -clh just lists non-zero counts, with no names.
2181    4.  The long form of the pcregrep -F option was incorrectly implemented as
2182        --fixed_strings instead of --fixed-strings. This is an incompatible change,
2183        but it seems right to fix it, and I didn't think it was worth preserving
2184        the old behaviour.
2186    5.  The command line items --regex=pattern and --regexp=pattern were not
2187        recognized by pcregrep, which required --regex pattern or --regexp pattern
2188        (with a space rather than an '='). The man page documented the '=' forms,
2189        which are compatible with GNU grep; these now work.
2191    6.  No libpcreposix.pc file was created for pkg-config; there was just
2192        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
2194    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
2195        when UCP support is not needed, by modifying the Python script that
2196        generates it from Unicode data files. This should not matter if the module
2197        is correctly used as a library, but I received one complaint about 50K of
2198        unwanted data. My guess is that the person linked everything into his
2199        program rather than using a library. Anyway, it does no harm.
2201    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
2202        was a minimum greater than 1 for a wide character in a possessive
2203        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
2204        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
2205        character. Chaos in the form of incorrect output or a compiling loop could
2206        result.
2208    9.  The restrictions on what a pattern can contain when partial matching is
2209        requested for pcre_exec() have been removed. All patterns can now be
2210        partially matched by this function. In addition, if there are at least two
2211        slots in the offset vector, the offset of the earliest inspected character
2212        for the match and the offset of the end of the subject are set in them when
2213        PCRE_ERROR_PARTIAL is returned.
2215    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
2216        synonymous with PCRE_PARTIAL, for backwards compatibility, and
2217        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
2218        and may be more useful for multi-segment matching.
2220    11. Partial matching with pcre_exec() is now more intuitive. A partial match
2221        used to be given if ever the end of the subject was reached; now it is
2222        given only if matching could not proceed because another character was
2223        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
2224        string "Z", which now yields "no match" instead of "partial match". In the
2225        case of pcre_dfa_exec(), "no match" is given if every matching path for the
2226        final character ended with (*FAIL).
2228    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
2229        if the pattern had a "must contain" character that was already found in the
2230        earlier partial match, unless partial matching was again requested. For
2231        example, with the pattern /dog.(body)?/, the "must contain" character is
2232        "g". If the first part-match was for the string "dog", restarting with
2233        "sbody" failed. This bug has been fixed.
2235    13. The string returned by pcre_dfa_exec() after a partial match has been
2236        changed so that it starts at the first inspected character rather than the
2237        first character of the match. This makes a difference only if the pattern
2238        starts with a lookbehind assertion or \b or \B (\K is not supported by
2239        pcre_dfa_exec()). It's an incompatible change, but it makes the two
2240        matching functions compatible, and I think it's the right thing to do.
2242    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
2243        so that the demonstration program is easily available in environments where
2244        PCRE has not been installed from source.
2246    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
2247        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
2248        library.
2250    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
2251        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
2252        is not the first non-POSIX option to be added. Clearly some people find
2253        these options useful.
2255    17. If a caller to the POSIX matching function regexec() passes a non-zero
2256        value for nmatch with a NULL value for pmatch, the value of
2257        nmatch is forced to zero.
2259    18. RunGrepTest did not have a test for the availability of the -u option of
2260        the diff command, as RunTest does. It now checks in the same way as
2261        RunTest, and also checks for the -b option.
2263    19. If an odd number of negated classes containing just a single character
2264        interposed, within parentheses, between a forward reference to a named
2265        subpattern and the definition of the subpattern, compilation crashed with
2266        an internal error, complaining that it could not find the referenced
2267        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
2268        [The bug was that it was starting one character too far in when skipping
2269        over the character class, thus treating the ] as data rather than
2270        terminating the class. This meant it could skip too much.]
2272    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
2273        /g option in pcretest when the pattern contains \K, which makes it possible
2274        to have an empty string match not at the start, even when the pattern is
2275        anchored. Updated pcretest and pcredemo to use this option.
2277    21. If the maximum number of capturing subpatterns in a recursion was greater
2278        than the maximum at the outer level, the higher number was returned, but
2279        with unset values at the outer level. The correct (outer level) value is
2280        now given.
2282    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
2283        PCRE did not set those parentheses (unlike Perl). I have now found a way to
2284        make it do so. The string so far is captured, making this feature
2285        compatible with Perl.
2287    23. The tests have been re-organized, adding tests 11 and 12, to make it
2288        possible to check the Perl 5.10 features against Perl 5.10.
2290    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
2291        pattern matches a fixed length string. PCRE did not allow this; now it
2292        does. Neither allows recursion.
2294    25. I finally figured out how to implement a request to provide the minimum
2295        length of subject string that was needed in order to match a given pattern.
2296        (It was back references and recursion that I had previously got hung up
2297        on.) This code has now been added to pcre_study(); it finds a lower bound
2298        to the length of subject needed. It is not necessarily the greatest lower
2299        bound, but using it to avoid searching strings that are too short does give
2300        some useful speed-ups. The value is available to calling programs via
2301        pcre_fullinfo().
2303    26. While implementing 25, I discovered to my embarrassment that pcretest had
2304        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
2305        study optimizations had never been tested with that matching function.
2306        Oops. What is worse, even when it was passed study data, there was a bug in
2307        pcre_dfa_exec() that meant it never actually used it. Double oops. There
2308        were also very few tests of studied patterns with pcre_dfa_exec().
2310    27. If (?| is used to create subpatterns with duplicate numbers, they are now
2311        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
2312        on the other side of the coin, they are no longer allowed to have different
2313        names, because these cannot be distinguished in PCRE, and this has caused
2314        confusion. (This is a difference from Perl.)
2316    28. When duplicate subpattern names are present (necessarily with different
2317        numbers, as required by 27 above), and a test is made by name in a
2318        conditional pattern, either for a subpattern having been matched, or for
2319        recursion in such a pattern, all the associated numbered subpatterns are
2320        tested, and the overall condition is true if the condition is true for any
2321        one of them. This is the way Perl works, and is also more like the way
2322        testing by number works.
2325    Version 7.9 11-Apr-09
2326    ---------------------
2328    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
2329        (pcretest), all targets were linked against these libraries. This included
2330        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
2331        libraries. This caused unwanted dependencies to be created. This problem
2332        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
2333        pcretest is linked with readline.
2335    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
2336        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
2337        moved outside it again, because FALSE and TRUE are already defined in AIX,
2338        but BOOL is not.
2340    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
2341        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
2343    4.  The pcregrep documentation said spaces were inserted as well as colons (or
2344        hyphens) following file names and line numbers when outputting matching
2345        lines. This is not true; no spaces are inserted. I have also clarified the
2346        wording for the --colour (or --color) option.
2348    5.  In pcregrep, when --colour was used with -o, the list of matching strings
2349        was not coloured; this is different to GNU grep, so I have changed it to be
2350        the same.
2352    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
2353        each matching line was coloured. Now it goes on to look for further matches
2354        of any of the test patterns, which is the same behaviour as GNU grep.
2356    7.  A pattern that could match an empty string could cause pcregrep to loop; it
2357        doesn't make sense to accept an empty string match in pcregrep, so I have
2358        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
2359        seems to be how GNU grep behaves. [But see later change 40 for release
2360        8.33.]
2362    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
2363        start or after a newline", because the conditional assertion was not being
2364        correctly handled. The rule now is that both the assertion and what follows
2365        in the first alternative must satisfy the test.
2367    9.  If auto-callout was enabled in a pattern with a conditional group whose
2368        condition was an assertion, PCRE could crash during matching, both with
2369        pcre_exec() and pcre_dfa_exec().
2371    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
2372        used for matching.
2374    11. Unicode property support in character classes was not working for
2375        characters (bytes) greater than 127 when not in UTF-8 mode.
2377    12. Added the -M command line option to pcretest.
2379    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
2381    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
2383    16. Added comments and documentation about mis-use of no_arg in the C++
2384        wrapper.
2386    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
2387        from Martin Jerabek that uses macro names for all relevant character and
2388        string constants.
2390    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
2391        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
2392        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
2393        these, but not everybody uses configure.
2395    19. A conditional group that had only one branch was not being correctly
2396        recognized as an item that could match an empty string. This meant that an
2397        enclosing group might also not be so recognized, causing infinite looping
2398        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
2399        with the subject "ab", where knowledge that the repeated group can match
2400        nothing is needed in order to break the loop.
2402    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
2403        exec(), but without supplying a callout function, matching went wrong.
2405    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
2406        leak if the size of the offset vector was greater than 30. When the vector
2407        is smaller, the saved offsets during recursion go onto a local stack
2408        vector, but for larger vectors malloc() is used. It was failing to free
2409        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
2410        error, in fact).
2412    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
2413        heapframe that is used only when UTF-8 support is enabled. This caused no
2414        problem, but was untidy.
2416    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
2417        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
2418        included within another project.
2420    24. Steven Van Ingelgem's patches to add more options to the CMake support,
2421        slightly modified by me:
2423          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
2424              not building pcregrep.
2426          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
2427              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
2429    25. Forward references, both numeric and by name, in patterns that made use of
2430        duplicate group numbers, could behave incorrectly or give incorrect errors,
2431        because when scanning forward to find the reference group, PCRE was not
2432        taking into account the duplicate group numbers. A pattern such as
2433        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
2435    26. Changed a few more instances of "const unsigned char *" to USPTR, making
2436        the feature of a custom pointer more persuasive (as requested by a user).
2438    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
2439        pcretest.c, inside #ifndefs, because it seems they are sometimes already
2440        pre-defined.
2442    28. Added support for (*UTF8) at the start of a pattern.
2444    29. Arrange for flags added by the "release type" setting in CMake to be shown
2445        in the configuration summary.
2448    Version 7.8 05-Sep-08
2449    ---------------------
2451    1.  Replaced UCP searching code with optimized version as implemented for Ad
2452        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
2453        stage table and inline lookup instead of a function, giving speed ups of 2
2454        to 5 times on some simple patterns that I tested. Permission was given to
2455        distribute the MultiStage2.py script that generates the tables (it's not in
2456        the tarball, but is in the Subversion repository).
2458    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
2459        scripts.
2461    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
2462        a group with a zero qualifier. The result of the study could be incorrect,
2463        or the function might crash, depending on the pattern.
2465    4.  Caseless matching was not working for non-ASCII characters in back
2466        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
2467        It now works when Unicode Property Support is available.
2469    5.  In pcretest, an escape such as \x{de} in the data was always generating
2470        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
2471        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
2472        truncation.
2474    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
2476    7.  Added two (int) casts to pcregrep when printing the difference of two
2477        pointers, in case they are 64-bit values.
2479    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
2480        test 2 if it fails.
2482    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
2483        and a #define of that name to empty if it is not externally set. This is to
2484        allow users of MSVC to set it if necessary.
2486    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
2487        the convenience functions in the pcre_get.c source file.
2489    11. An option change at the start of a pattern that had top-level alternatives
2490        could cause overwriting and/or a crash. This command provoked a crash in
2491        some environments:
2493          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
2495        This potential security problem was recorded as CVE-2008-2371.
2497    12. For a pattern where the match had to start at the beginning or immediately
2498        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
2499        pcre_dfa_exec() could read past the end of the passed subject if there was
2500        no match. To help with detecting such bugs (e.g. with valgrind), I modified
2501        pcretest so that it places the subject at the end of its malloc-ed buffer.
2503    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
2504        exec() might read past the end of the data buffer in UTF-8 mode.
2506    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
2507        the data contained the byte 0x85 as part of a UTF-8 character within its
2508        first line. This applied both to normal and DFA matching.
2510    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
2511        /^[^d]*?$/8 failed to match "abc".
2513    16. Added a missing copyright notice to pcrecpp_internal.h.
2515    17. Make it more clear in the documentation that values returned from
2516        pcre_exec() in ovector are byte offsets, not character counts.
2518    18. Tidied a few places to stop certain compilers from issuing warnings.
2520    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
2521        supplied by Stefan Weber. I made a further small update for 7.8 because
2522        there is a change of source arrangements: the pcre_searchfuncs.c module is
2523        replaced by pcre_ucd.c.
2526    Version 7.7 07-May-08
2527    ---------------------
2529    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
2530        a string to a long long, pretend we don't even have a long long." This is
2531        done by checking for the strtoq, strtoll, and _strtoi64 functions.
2533    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
2534        pre-7.6 versions, which defined a global no_arg variable instead of putting
2535        it in the RE class. (See also #8 below.)
2537    3.  Remove a line of dead code, identified by coverity and reported by Nuno
2538        Lopes.
2540    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
2542        (1) The include/exclude patterns were being applied to the whole pathnames
2543            of files, instead of just to the final components.
2545        (2) If there was more than one level of directory, the subdirectories were
2546            skipped unless they satisfied the include/exclude conditions. This is
2547            inconsistent with GNU grep (and could even be seen as contrary to the
2548            pcregrep specification - which I improved to make it absolutely clear).
2549            The action now is always to scan all levels of directory, and just
2550            apply the include/exclude patterns to regular files.
2552    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
2553        --exclude_dir in the tests to avoid scanning .svn directories.
2555    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
2556        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
2557        doesn't support NULs in patterns.
2559    7.  Added some missing "const"s to declarations of static tables in
2560        pcre_compile.c and pcre_dfa_exec.c.
2562    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
2563        caused by fix #2  above. (Subsequently also a second patch to fix the
2564        first patch. And a third patch - this was a messy problem.)
2566    9.  Applied Craig's patch to remove the use of push_back().
2568    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
2569        matching function regexec().
2571    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
2572        which, however, unlike Perl's \g{...}, are subroutine calls, not back
2573        references. PCRE supports relative numbers with this syntax (I don't think
2574        Oniguruma does).
2576    12. Previously, a group with a zero repeat such as (...){0} was completely
2577        omitted from the compiled regex. However, this means that if the group
2578        was called as a subroutine from elsewhere in the pattern, things went wrong
2579        (an internal error was given). Such groups are now left in the compiled
2580        pattern, with a new opcode that causes them to be skipped at execution
2581        time.
2583    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
2584        to the way PCRE behaves:
2586        (a) A lone ] character is dis-allowed (Perl treats it as data).
2588        (b) A back reference to an unmatched subpattern matches an empty string
2589            (Perl fails the current match path).
2591        (c) A data ] in a character class must be notated as \] because if the
2592            first data character in a class is ], it defines an empty class. (In
2593            Perl it is not possible to have an empty class.) The empty class []
2594            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
2595            The negative empty class [^] matches any one character, independently
2596            of the DOTALL setting.
2598    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
2599        non-existent subpattern following a character class starting with ']' and
2600        containing () gave an internal compiling error instead of "reference to
2601        non-existent subpattern". Fortunately, when the pattern did exist, the
2602        compiled code was correct. (When scanning forwards to check for the
2603        existence of the subpattern, it was treating the data ']' as terminating
2604        the class, so got the count wrong. When actually compiling, the reference
2605        was subsequently set up correctly.)
2607    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
2608        it was being rejected as not supported by pcre_dfa_exec(), even though
2609        other assertions are supported. I have made pcre_dfa_exec() support
2610        (*FAIL).
2612    16. The implementation of 13c above involved the invention of a new opcode,
2613        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
2614        cannot be changed at match time, I realized I could make a small
2615        improvement to matching performance by compiling OP_ALLANY instead of
2616        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
2617        on the OP_ANY path.
2619    17. Compiling pcretest on Windows with readline support failed without the
2620        following two fixes: (1) Make the unistd.h include conditional on
2621        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
2623    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
2624        ncurses library to be included for pcretest when ReadLine support is
2625        requested, but also to allow for it to be overridden. This patch came from
2626        Daniel Bergström.
2628    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
2629        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
2630        any errors with the current Unicode tables. Thanks to Peter Kankowski for
2631        spotting this.
2634    Version 7.6 28-Jan-08
2635    ---------------------
2637    1.  A character class containing a very large number of characters with
2638        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
2639        overflow.
2641    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
2642        HAVE_LONG_LONG is not defined.
2644    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
2645        bring them up to date and include new features. This patch includes:
2647        - Fixed PH's badly added libz and libbz2 support.
2648        - Fixed a problem with static linking.
2649        - Added pcredemo. [But later removed - see 7 below.]
2650        - Fixed dftables problem and added an option.
2651        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
2652            HAVE_LONG_LONG.
2653        - Added readline support for pcretest.
2654        - Added an listing of the option settings after cmake has run.
2656    4.  A user submitted a patch to Makefile that makes it easy to create
2657        "pcre.dll" under mingw when using Configure/Make. I added stuff to
2658        Makefile.am that cause it to include this special target, without
2659        affecting anything else. Note that the same mingw target plus all
2660        the other distribution libraries and programs are now supported
2661        when configuring with CMake (see 6 below) instead of with
2662        Configure/Make.
2664    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
2665        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
2666        exported in the Windows port". It has not yet been confirmed that the patch
2667        solves the problem, but it does no harm.
2669    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
2670        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
2671        with CMake, and also correct the comment about stack recursion.
2673    7.  Remove the automatic building of pcredemo from the ./configure system and
2674        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
2675        of a program that users should build themselves after PCRE is installed, so
2676        building it automatically is not really right. What is more, it gave
2677        trouble in some build environments.
2679    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
2682    Version 7.5 10-Jan-08
2683    ---------------------
2685    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
2686        values in parens when parsing an RE using the C++ wrapper."
2688    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
2689        Characters greater than 255 were excluded from the class instead of being
2690        included.
2692    3.  The same bug as (2) above applied to negated POSIX classes such as
2693        [:^space:].
2695    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
2696        defined or documented. It seems to have been a typo for PCRE_STATIC, so
2697        I have changed it.
2699    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
2700        first named subpattern) and a construct such as (?&a) would reference the
2701        first named subpattern whose name started with "a" (in other words, the
2702        length check was missing). Both these problems are fixed. "Subpattern name
2703        expected" is now given for (?&) (a zero-length name), and this patch also
2704        makes it give the same error for \k'' (previously it complained that that
2705        was a reference to a non-existent subpattern).
2707    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
2708        this is right because (?- can be followed by option settings as well as by
2709        digits. I have, however, made the messages clearer.
2711    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
2712        than the number used in the conditional) now cause a compile-time error.
2713        This is actually not compatible with Perl, which accepts such patterns, but
2714        treats the conditional as always being FALSE (as PCRE used to), but it
2715        seems to me that giving a diagnostic is better.
2717    8.  Change "alphameric" to the more common word "alphanumeric" in comments
2718        and messages.
2720    9.  Fix two occurrences of "backslash" in comments that should have been
2721        "backspace".
2723    10. Remove two redundant lines of code that can never be obeyed (their function
2724        was moved elsewhere).
2726    11. The program that makes PCRE's Unicode character property table had a bug
2727        which caused it to generate incorrect table entries for sequences of
2728        characters that have the same character type, but are in different scripts.
2729        It amalgamated them into a single range, with the script of the first of
2730        them. In other words, some characters were in the wrong script. There were
2731        thirteen such cases, affecting characters in the following ranges:
2733          U+002b0 - U+002c1
2734          U+0060c - U+0060d
2735          U+0061e - U+00612
2736          U+0064b - U+0065e
2737          U+0074d - U+0076d
2738          U+01800 - U+01805
2739          U+01d00 - U+01d77
2740          U+01d9b - U+01dbf
2741          U+0200b - U+0200f
2742          U+030fc - U+030fe
2743          U+03260 - U+0327f
2744          U+0fb46 - U+0fbb1
2745          U+10450 - U+1049d
2747    12. The -o option (show only the matching part of a line) for pcregrep was not
2748        compatible with GNU grep in that, if there was more than one match in a
2749        line, it showed only the first of them. It now behaves in the same way as
2750        GNU grep.
2752    13. If the -o and -v options were combined for pcregrep, it printed a blank
2753        line for every non-matching line. GNU grep prints nothing, and pcregrep now
2754        does the same. The return code can be used to tell if there were any
2755        non-matching lines.
2757    14. Added --file-offsets and --line-offsets to pcregrep.
2759    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
2760        infinitely looping recursion. The bug was that positive lookaheads were not
2761        being skipped when checking for a possible empty match (negative lookaheads
2762        and both kinds of lookbehind were skipped).
2764    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
2765        inclusion of <windows.h> to before rather than after the definition of
2766        INVALID_FILE_ATTRIBUTES (patch from David Byron).
2768    17. Specifying a possessive quantifier with a specific limit for a Unicode
2769        character property caused pcre_compile() to compile bad code, which led at
2770        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
2771        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
2772        caused the error; without that there was no problem.
2774    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
2776    19. Added --enable-pcretest-libreadline.
2778    20. In pcrecpp.cc, the variable 'count' was incremented twice in
2779        RE::GlobalReplace(). As a result, the number of replacements returned was
2780        double what it should be. I removed one of the increments, but Craig sent a
2781        later patch that removed the other one (the right fix) and added unit tests
2782        that check the return values (which was not done before).
2784    21. Several CMake things:
2786        (1) Arranged that, when cmake is used on Unix, the libraries end up with
2787            the names libpcre and libpcreposix, not just pcre and pcreposix.
2789        (2) The above change means that pcretest and pcregrep are now correctly
2790            linked with the newly-built libraries, not previously installed ones.
2794    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
2795        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
2796        UTF-8 newline character). The key issue is that the pattern starts .*;
2797        this means that the match must be either at the beginning, or after a
2798        newline. The bug was in the code for advancing after a failed match and
2799        checking that the new position followed a newline. It was not taking
2800        account of UTF-8 characters correctly.
2802    23. PCRE was behaving differently from Perl in the way it recognized POSIX
2803        character classes. PCRE was not treating the sequence [:...:] as a
2804        character class unless the ... were all letters. Perl, however, seems to
2805        allow any characters between [: and :], though of course it rejects as
2806        unknown any "names" that contain non-letters, because all the known class
2807        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
2808        for example, whereas PCRE did not - it did not recognize a POSIX character
2809        class. This seemed a bit dangerous, so the code has been changed to be
2810        closer to Perl. The behaviour is not identical to Perl, because PCRE will
2811        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
2812        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
2813        Perl does, and where it didn't before.
2815    24. Rewrite so as to remove the single use of %n from pcregrep because in some
2816        Windows environments %n is disabled by default.
2819    Version 7.4 21-Sep-07
2820    ---------------------
2822    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
2823        means that a class such as [\s] counted as "explicit reference to CR or
2824        LF". That isn't really right - the whole point of the change was to try to
2825        help when there was an actual mention of one of the two characters. So now
2826        the change happens only if \r or \n (or a literal CR or LF) character is
2827        encountered.
2829    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
2830        of both had grown to the point where there were only 3 bits left.
2831        Fortunately, there was spare space in the data structure, and so I have
2832        moved the internal flags into a new 16-bit field to free up more option
2833        bits.
2835    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
2836        but did not set the internal JCHANGED flag - either of these is enough to
2837        control the way the "get" function works - but the PCRE_INFO_JCHANGED
2838        facility is supposed to tell if (?J) was ever used, so now (?J) at the
2839        start sets both bits.
2841    4.  Added options (at build time, compile time, exec time) to change \R from
2842        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
2844    5.  doc/pcresyntax.html was missing from the distribution.
2846    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
2847        compatibility, even though it is no longer used.
2849    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
2850        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
2851        windows.h file is present (where different names are used). [This was
2852        reversed later after testing - see 16 below.]
2854    8.  Changed all #include <config.h> to #include "config.h". There were also
2855        some further <pcre.h> cases that I changed to "pcre.h".
2857    9.  When pcregrep was used with the --colour option, it missed the line ending
2858        sequence off the lines that it output.
2860    10. It was pointed out to me that arrays of string pointers cause lots of
2861        relocations when a shared library is dynamically loaded. A technique of
2862        using a single long string with a table of offsets can drastically reduce
2863        these. I have refactored PCRE in four places to do this. The result is
2864        dramatic:
2866          Originally:                          290
2867          After changing UCP table:            187
2868          After changing error message table:   43
2869          After changing table of "verbs"       36
2870          After changing table of Posix names   22
2872        Thanks to the folks working on Gregex for glib for this insight.
2874    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
2875        unicode-properties was also set.
2877    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
2879    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
2880        checked only for CRLF.
2882    14. Added casts to pcretest.c to avoid compiler warnings.
2884    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
2886    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
2887        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
2888        entirely. This removes changes made in 7 above.
2890    17. The CMake files have been updated, and there is now more information about
2891        building with CMake in the NON-UNIX-USE document.
2894    Version 7.3 28-Aug-07
2895    ---------------------
2897     1. In the rejigging of the build system that eventually resulted in 7.1, the
2898        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
2899        brackets there is not right, since it causes compilers to look for an
2900        installed pcre.h, not the version that is in the source that is being
2901        compiled (which of course may be different). I have changed it back to:
2903          #include "pcre.h"
2905        I have a vague recollection that the change was concerned with compiling in
2906        different directories, but in the new build system, that is taken care of
2907        by the VPATH setting the Makefile.
2909     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
2910        when the subject happened to end in the byte 0x85 (e.g. if the last
2911        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
2912        characters but of course it shouldn't be taken as a newline when it is part
2913        of another character. The bug was that, for an unlimited repeat of . in
2914        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
2915        characters when looking for a newline.
2917     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
2919     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
2920        in debug output.
2922     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
2923        long printing in the pcrecpp unittest when running under MinGW.
2925     6. ESC_K was left out of the EBCDIC table.
2927     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
2928        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
2929        limit also applies to "virtual nesting" when a pattern is recursive, and in
2930        this case 1000 isn't so big. I have been able to remove this limit at the
2931        expense of backing off one optimization in certain circumstances. Normally,
2932        when pcre_exec() would call its internal match() function recursively and
2933        immediately return the result unconditionally, it uses a "tail recursion"
2934        feature to save stack. However, when a subpattern that can match an empty
2935        string has an unlimited repetition quantifier, it no longer makes this
2936        optimization. That gives it a stack frame in which to save the data for
2937        checking that an empty string has been matched. Previously this was taken
2938        from the 1000-entry workspace that had been reserved. So now there is no
2939        explicit limit, but more stack is used.
2941     8. Applied Daniel's patches to solve problems with the import/export magic
2942        syntax that is required for Windows, and which was going wrong for the
2943        pcreposix and pcrecpp parts of the library. These were overlooked when this
2944        problem was solved for the main library.
2946     9. There were some crude static tests to avoid integer overflow when computing
2947        the size of patterns that contain repeated groups with explicit upper
2948        limits. As the maximum quantifier is 65535, the maximum group length was
2949        set at 30,000 so that the product of these two numbers did not overflow a
2950        32-bit integer. However, it turns out that people want to use groups that
2951        are longer than 30,000 bytes (though not repeat them that many times).
2952        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
2953        made it possible to implement the integer overflow checks in a much more
2954        dynamic way, which I have now done. The artificial limitation on group
2955        length has been removed - we now have only the limit on the total length of
2956        the compiled pattern, which depends on the LINK_SIZE setting.
2958    10. Fixed a bug in the documentation for get/copy named substring when
2959        duplicate names are permitted. If none of the named substrings are set, the
2960        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
2961        empty string.
2963    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
2964        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
2965        because the ] is interpreted as the first data character and the
2966        terminating ] is not found. PCRE has been made compatible with Perl in this
2967        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
2968        cause memory overwriting.
2970    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
2971        string has been matched (to stop an infinite loop). It was not recognizing
2972        a conditional subpattern that could match an empty string if that
2973        subpattern was within another subpattern. For example, it looped when
2974        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
2975        condition was not nested. This bug has been fixed.
2977    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
2978        past the start of the subject in the presence of bytes with the top bit
2979        set, for example "\x8aBCD".
2981    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
2982        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
2984    14. Optimized (?!) to (*FAIL).
2986    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
2987        This restricts code points to be within the range 0 to 0x10FFFF, excluding
2988        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
2989        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
2990        does: it's just the validity check that is more restrictive.
2992    16. Inserted checks for integer overflows during escape sequence (backslash)
2993        processing, and also fixed erroneous offset values for syntax errors during
2994        backslash processing.
2996    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
2997        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
2999    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
3000        caused an overrun.
3002    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
3003        something other than just ASCII characters) inside a group that had an
3004        unlimited repeat caused a loop at compile time (while checking to see
3005        whether the group could match an empty string).
3007    20. Debugging a pattern containing \p or \P could cause a crash. For example,
3008        [\P{Any}] did so. (Error in the code for printing property names.)
3010    21. An orphan \E inside a character class could cause a crash.
3012    22. A repeated capturing bracket such as (A)? could cause a wild memory
3013        reference during compilation.
3015    23. There are several functions in pcre_compile() that scan along a compiled
3016        expression for various reasons (e.g. to see if it's fixed length for look
3017        behind). There were bugs in these functions when a repeated \p or \P was
3018        present in the pattern. These operators have additional parameters compared
3019        with \d, etc, and these were not being taken into account when moving along
3020        the compiled data. Specifically:
3022        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
3023            length.
3025        (b) An item such as \pL+ within a repeated group could cause crashes or
3026            loops.
3028        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
3029            "reference to non-existent subpattern" error.
3031        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
3033    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
3034        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
3036    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
3038    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
3039        character were causing crashes (broken optimization).
3041    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
3042        \p or \P) caused a compile-time loop.
3044    28. More problems have arisen in unanchored patterns when CRLF is a valid line
3045        break. For example, the unstudied pattern [\r\n]A does not match the string
3046        "\r\nA" because change 7.0/46 below moves the current point on by two
3047        characters after failing to match at the start. However, the pattern \nA
3048        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
3049        the same is true. There doesn't seem any very clean way out of this, but
3050        what I have chosen to do makes the common cases work: PCRE now takes note
3051        of whether there can be an explicit match for \r or \n anywhere in the
3052        pattern, and if so, 7.0/46 no longer applies. As part of this change,
3053        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
3054        pattern has explicit CR or LF references.
3056    29. Added (*CR) etc for changing newline setting at start of pattern.
3059    Version 7.2 19-Jun-07
3060    ---------------------
3062     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
3063        which is apparently normally available under Windows.
3065     2. Re-jig the pcregrep tests with different newline settings in an attempt
3066        to make them independent of the local environment's newline setting.
3068     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
3070     4. Some of the "internals" tests were previously cut out when the link size
3071        was not 2, because the output contained actual offsets. The recent new
3072        "Z" feature of pcretest means that these can be cut out, making the tests
3073        usable with all link sizes.
3075     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
3076        stack recursion. This gives a massive performance boost under BSD, but just
3077        a small improvement under Linux. However, it saves one field in the frame
3078        in all cases.
3080     6. Added more features from the forthcoming Perl 5.10:
3082        (a) (?-n) (where n is a string of digits) is a relative subroutine or
3083            recursion call. It refers to the nth most recently opened parentheses.
3085        (b) (?+n) is also a relative subroutine call; it refers to the nth next
3086            to be opened parentheses.
3088        (c) Conditions that refer to capturing parentheses can be specified
3089            relatively, for example, (?(-2)... or (?(+3)...
3091        (d) \K resets the start of the current match so that everything before
3092            is not part of it.
3094        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
3096        (f) \g{name} is another synonym - part of Perl 5.10's unification of
3097            reference syntax.
3099        (g) (?| introduces a group in which the numbering of parentheses in each
3100            alternative starts with the same number.
3102        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
3104     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
3107     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
3108        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
3109        for detecting groups that can match an empty string.
3111     9. A pattern with a very large number of alternatives (more than several
3112        hundred) was running out of internal workspace during the pre-compile
3113        phase, where pcre_compile() figures out how much memory will be needed. A
3114        bit of new cunning has reduced the workspace needed for groups with
3115        alternatives. The 1000-alternative test pattern now uses 12 bytes of
3116        workspace instead of running out of the 4096 that are available.
3118    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
3120    11. Applied patch from Google to remove an optimization that didn't quite work.
3121        The report of the bug said:
3123          pcrecpp::RE("a*").FullMatch("aaa") matches, while
3124          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
3125          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
3127    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
3128        it matched the wrong number of bytes.
3131    Version 7.1 24-Apr-07
3132    ---------------------
3134     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
3135        that is more "standard", making use of automake and other Autotools. There
3136        is some re-arrangement of the files and adjustment of comments consequent
3137        on this.
3139     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
3140        for recursive directory scanning broke on some systems because the files
3141        are not scanned in any specific order and on different systems the order
3142        was different. A call to "sort" has been inserted into RunGrepTest for the
3143        approprate test as a short-term fix. In the longer term there may be an
3144        alternative.
3146     3. I had an email from Eric Raymond about problems translating some of PCRE's
3147        man pages to HTML (despite the fact that I distribute HTML pages, some
3148        people do their own conversions for various reasons). The problems
3149        concerned the use of low-level troff macros .br and .in. I have therefore
3150        removed all such uses from the man pages (some were redundant, some could
3151        be replaced by .nf/.fi pairs). The 132html script that I use to generate
3152        HTML has been updated to handle .nf/.fi and to complain if it encounters
3153        .br or .in.
3155     4. Updated comments in configure.ac that get placed in config.h.in and also
3156        arranged for config.h to be included in the distribution, with the name
3157        config.h.generic, for the benefit of those who have to compile without
3158        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
3160     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
3161        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
3162        makevp.bat for the current PCRE, using the additional files
3163        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
3165     6. A Windows user reported a minor discrepancy with test 2, which turned out
3166        to be caused by a trailing space on an input line that had got lost in his
3167        copy. The trailing space was an accident, so I've just removed it.
3169     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
3170        that is needed.
3172     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
3173        as "const" (a) because they are and (b) because it helps the PHP
3174        maintainers who have recently made a script to detect big data structures
3175        in the php code that should be moved to the .rodata section. I remembered
3176        to update Builducptable as well, so it won't revert if ucptable.h is ever
3177        re-created.
3179     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
3180        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
3181        order to be able to cut out the UTF-8 tables in the latter when UTF-8
3182        support is not required. This saves 1.5-2K of code, which is important in
3183        some applications.
3185        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
3186        so as not to refer to the tables, even though these functions will never be
3187        called when UTF-8 support is disabled. Otherwise there are problems with a
3188        shared library.
3190    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
3192        (a) It was defining its arguments as char * instead of void *.
3194        (b) It was assuming that all moves were upwards in memory; this was true
3195            a long time ago when I wrote it, but is no longer the case.
3197        The emulated memove() is provided for those environments that have neither
3198        memmove() nor bcopy(). I didn't think anyone used it these days, but that
3199        is clearly not the case, as these two bugs were recently reported.
3201    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
3202        and Detrail to create the HTML documentation, the .txt form of the man
3203        pages, and it removes trailing spaces from listed files. It also creates
3204        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
3205        case, it wraps all the #defines with #ifndefs. This script should be run
3206        before "make dist".
3208    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
3209        with Unicode property support.
3211        (a) For a maximizing quantifier, if the two different cases of the
3212            character were of different lengths in their UTF-8 codings (there are
3213            some cases like this - I found 11), and the matching function had to
3214            back up over a mixture of the two cases, it incorrectly assumed they
3215            were both the same length.
3217        (b) When PCRE was configured to use the heap rather than the stack for
3218            recursion during matching, it was not correctly preserving the data for
3219            the other case of a UTF-8 character when checking ahead for a match
3220            while processing a minimizing repeat. If the check also involved
3221            matching a wide character, but failed, corruption could cause an
3222            erroneous result when trying to check for a repeat of the original
3223            character.
3225    13. Some tidying changes to the testing mechanism:
3227        (a) The RunTest script now detects the internal link size and whether there
3228            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
3229            values substituted by "configure". (The RunGrepTest script already did
3230            this for UTF-8.) The configure.ac script no longer substitutes the
3231            relevant variables.
3233        (b) The debugging options /B and /D in pcretest show the compiled bytecode
3234            with length and offset values. This means that the output is different
3235            for different internal link sizes. Test 2 is skipped for link sizes
3236            other than 2 because of this, bypassing the problem. Unfortunately,
3237            there was also a test in test 3 (the locale tests) that used /B and
3238            failed for link sizes other than 2. Rather than cut the whole test out,
3239            I have added a new /Z option to pcretest that replaces the length and
3240            offset values with spaces. This is now used to make test 3 independent
3241            of link size. (Test 2 will be tidied up later.)
3243    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
3244        segmentation fault instead of returning the appropriate error message.
3246    15. In multiline mode when the newline sequence was set to "any", the pattern
3247        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
3248        This doesn't seem right; it now treats the CRLF combination as the line
3249        ending, and so does not match in that case. It's only a pattern such as ^$
3250        that would hit this one: something like ^ABC$ would have failed after \r
3251        and then tried again after \r\n.
3253    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
3254        in an attempt to make files that differ only in their line terminators
3255        compare equal. This works on Linux.
3257    17. Under certain error circumstances pcregrep might try to free random memory
3258        as it exited. This is now fixed, thanks to valgrind.
3260    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
3261        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
3262        was because its rules for how to advance for /g after matching an empty
3263        string at the end of a line did not allow for this case. They now check for
3264        it specially.
3266    20. pcretest is supposed to handle patterns and data of any length, by
3267        extending its buffers when necessary. It was getting this wrong when the
3268        buffer for a data line had to be extended.
3270    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
3271        CRLF as a newline sequence.
3273    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
3274        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
3275        I have nevertheless tidied it up.
3277    23. Added some casts to kill warnings from HP-UX ia64 compiler.
3279    24. Added a man page for pcre-config.
3282    Version 7.0 19-Dec-06
3283    ---------------------
3285     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
3286        moving to gcc 4.1.1.
3288     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
3289        sys/time.h, which is documented as needed for this function. It doesn't
3290        seem to matter on Linux, but it showed up on some releases of OS X.
3292     3. It seems that there are systems where bytes whose values are greater than
3293        127 match isprint() in the "C" locale. The "C" locale should be the
3294        default when a C program starts up. In most systems, only ASCII printing
3295        characters match isprint(). This difference caused the output from pcretest
3296        to vary, making some of the tests fail. I have changed pcretest so that:
3298        (a) When it is outputting text in the compiled version of a pattern, bytes
3299            other than 32-126 are always shown as hex escapes.
3301        (b) When it is outputting text that is a matched part of a subject string,
3302            it does the same, unless a different locale has been set for the match
3303            (using the /L modifier). In this case, it uses isprint() to decide.
3305     4. Fixed a major bug that caused incorrect computation of the amount of memory
3306        required for a compiled pattern when options that changed within the
3307        pattern affected the logic of the preliminary scan that determines the
3308        length. The relevant options are -x, and -i in UTF-8 mode. The result was
3309        that the computed length was too small. The symptoms of this bug were
3310        either the PCRE error "internal error: code overflow" from pcre_compile(),
3311        or a glibc crash with a message such as "pcretest: free(): invalid next
3312        size (fast)". Examples of patterns that provoked this bug (shown in
3313        pcretest format) are:
3315          /(?-x: )/x
3316          /(?x)(?-x: \s*#\s*)/
3317          /((?i)[\x{c0}])/8
3318          /(?i:[\x{c0}])/8
3320        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
3321        is now done differently.
3323     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
3324        wrapper classes; (b) implement a new function in the C++ scanner that is
3325        more efficient than the old way of doing things because it avoids levels of
3326        recursion in the regex matching; (c) add a paragraph to the documentation
3327        for the FullMatch() function.
3329     6. The escape sequence \n was being treated as whatever was defined as
3330        "newline". Not only was this contrary to the documentation, which states
3331        that \n is character 10 (hex 0A), but it also went horribly wrong when
3332        "newline" was defined as CRLF. This has been fixed.
3334     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
3335        was being set to -1 for the "end of line" case (supposedly a value that no
3336        character can have). Though this value is never used (the check for end of
3337        line is "zero bytes in current character"), it caused compiler complaints.
3338        I've changed it to 0xffffffff.
3340     8. In pcre_version.c, the version string was being built by a sequence of
3341        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
3342        string (as it is for production releases) called a macro with an empty
3343        argument. The C standard says the result of this is undefined. The gcc
3344        compiler treats it as an empty string (which was what was wanted) but it is
3345        reported that Visual C gives an error. The source has been hacked around to
3346        avoid this problem.
3348     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
3349        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
3350        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
3351        of them did).
3353    10. Originally, pcretest opened its input and output without "b"; then I was
3354        told that "b" was needed in some environments, so it was added for release
3355        5.0 to both the input and output. (It makes no difference on Unix-like
3356        systems.) Later I was told that it is wrong for the input on Windows. I've
3357        now abstracted the modes into two macros, to make it easier to fiddle with
3358        them, and removed "b" from the input mode under Windows.
3360    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
3362    12. Added -help and --help to pcretest as an official way of being reminded
3363        of the options.
3365    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
3366        and pcrecpp.cc because they annoy compilers at high warning levels.
3368    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
3370    15. Fixed an occurrence of == in configure.ac that should have been = (shell
3371        scripts are not C programs :-) and which was not noticed because it works
3372        on Linux.
3374    16. pcretest is supposed to handle any length of pattern and data line (as one
3375        line or as a continued sequence of lines) by extending its input buffer if
3376        necessary. This feature was broken for very long pattern lines, leading to
3377        a string of junk being passed to pcre_compile() if the pattern was longer
3378        than about 50K.
3380    17. I have done a major re-factoring of the way pcre_compile() computes the
3381        amount of memory needed for a compiled pattern. Previously, there was code
3382        that made a preliminary scan of the pattern in order to do this. That was
3383        OK when PCRE was new, but as the facilities have expanded, it has become
3384        harder and harder to keep it in step with the real compile phase, and there
3385        have been a number of bugs (see for example, 4 above). I have now found a
3386        cunning way of running the real compile function in a "fake" mode that
3387        enables it to compute how much memory it would need, while actually only
3388        ever using a few hundred bytes of working memory and without too many
3389        tests of the mode. This should make future maintenance and development
3390        easier. A side effect of this work is that the limit of 200 on the nesting
3391        depth of parentheses has been removed (though this was never a serious
3392        limitation, I suspect). However, there is a downside: pcre_compile() now
3393        runs more slowly than before (30% or more, depending on the pattern). I
3394        hope this isn't a big issue. There is no effect on runtime performance.
3396    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
3397        newline (only possible for the last line of a file) and it was a
3398        pattern that set a locale (followed by /Lsomething), pcretest crashed.
3400    19. Added additional timing features to pcretest. (1) The -tm option now times
3401        matching only, not compiling. (2) Both -t and -tm can be followed, as a
3402        separate command line item, by a number that specifies the number of
3403        repeats to use when timing. The default is 50000; this gives better
3404        precision, but takes uncomfortably long for very large patterns.
3406    20. Extended pcre_study() to be more clever in cases where a branch of a
3407        subpattern has no definite first character. For example, (a*|b*)[cd] would
3408        previously give no result from pcre_study(). Now it recognizes that the
3409        first character must be a, b, c, or d.
3411    21. There was an incorrect error "recursive call could loop indefinitely" if
3412        a subpattern (or the entire pattern) that was being tested for matching an
3413        empty string contained only one non-empty item after a nested subpattern.
3414        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
3415        incorrectly, because the \d was being skipped in the check.
3417    22. The pcretest program now has a new pattern option /B and a command line
3418        option -b, which is equivalent to adding /B to every pattern. This causes
3419        it to show the compiled bytecode, without the additional information that
3420        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
3421        is the same as /B/I).
3423    23. A new optimization is now able automatically to treat some sequences such
3424        as a*b as a*+b. More specifically, if something simple (such as a character
3425        or a simple class like \d) has an unlimited quantifier, and is followed by
3426        something that cannot possibly match the quantified thing, the quantifier
3427        is automatically "possessified".
3429    24. A recursive reference to a subpattern whose number was greater than 39
3430        went wrong under certain circumstances in UTF-8 mode. This bug could also
3431        have affected the operation of pcre_study().
3433    25. Realized that a little bit of performance could be had by replacing
3434        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
3436    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
3438    27. Possessive quantifiers such as a++ were previously implemented by turning
3439        them into atomic groups such as ($>a+). Now they have their own opcodes,
3440        which improves performance. This includes the automatically created ones
3441        from 23 above.
3443    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
3444        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
3445        the first matched character to be a colon. This applied both to named and
3446        numbered groups.
3448    29. The ucpinternal.h header file was missing its idempotency #ifdef.
3450    30. I was sent a "project" file called libpcre.a.dev which I understand makes
3451        building PCRE on Windows easier, so I have included it in the distribution.
3453    31. There is now a check in pcretest against a ridiculously large number being
3454        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
3455        loop, the loop is abandoned.
3457    32. Forward references to subpatterns in conditions such as (?(2)...) where
3458        subpattern 2 is defined later cause pcre_compile() to search forwards in
3459        the pattern for the relevant set of parentheses. This search went wrong
3460        when there were unescaped parentheses in a character class, parentheses
3461        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
3463    33. "Subroutine" calls and backreferences were previously restricted to
3464        referencing subpatterns earlier in the regex. This restriction has now
3465        been removed.
3467    34. Added a number of extra features that are going to be in Perl 5.10. On the
3468        whole, these are just syntactic alternatives for features that PCRE had
3469        previously implemented using the Python syntax or my own invention. The
3470        other formats are all retained for compatibility.
3472        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
3473            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
3474            also .NET compatible.
3476        (b) A recursion or subroutine call to a named group can now be defined as
3477            (?&name) as well as (?P>name).
3479        (c) A backreference to a named group can now be defined as \k<name> or
3480            \k'name' as well as (?P=name). The new forms, as well as being in Perl
3481            5.10, are also .NET compatible.
3483        (d) A conditional reference to a named group can now use the syntax
3484            (?(<name>) or (?('name') as well as (?(name).
3486        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
3487            groups (named and numbered) that are never evaluated inline, but can be
3488            called as "subroutines" from elsewhere. In effect, the DEFINE condition
3489            is always false. There may be only one alternative in such a group.
3491        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
3492            as the simple (?(R). The condition is true only if the most recent
3493            recursion is that of the given number or name. It does not search out
3494            through the entire recursion stack.
3496        (g) The escape \gN or \g{N} has been added, where N is a positive or
3497            negative number, specifying an absolute or relative reference.
3499    35. Tidied to get rid of some further signed/unsigned compiler warnings and
3500        some "unreachable code" warnings.
3502    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
3503        things, this adds five new scripts.
3505    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
3506        There were also incompatibilities regarding the handling of \Q..\E inside
3507        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
3508        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
3510    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
3511        matches an empty string, and forcibly breaks the loop. There were bugs in
3512        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
3513        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
3514        separate and independent bugs (that affected different cases) have been
3515        fixed.
3517    39. Refactored the code to abolish the use of different opcodes for small
3518        capturing bracket numbers. This is a tidy that I avoided doing when I
3519        removed the limit on the number of capturing brackets for 3.5 back in 2001.
3520        The new approach is not only tidier, it makes it possible to reduce the
3521        memory needed to fix the previous bug (38).
3523    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
3524        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
3525        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
3526        mode.
3528    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
3529        report.
3531    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
3532        copy construction and assignment in the C++ wrapper.
3534    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
3535        couple of bugs that could have given wrong results in the "--newline=crlf"
3536        case.
3538    44. Added a number of casts and did some reorganization of signed/unsigned int
3539        variables following suggestions from Dair Grant. Also renamed the variable
3540        "this" as "item" because it is a C++ keyword.
3542    45. Arranged for dftables to add
3544          #include "pcre_internal.h"
3546        to pcre_chartables.c because without it, gcc 4.x may remove the array
3547        definition from the final binary if PCRE is built into a static library and
3548        dead code stripping is activated.
3550    46. For an unanchored pattern, if a match attempt fails at the start of a
3551        newline sequence, and the newline setting is CRLF or ANY, and the next two
3552        characters are CRLF, advance by two characters instead of one.
3555    Version 6.7 04-Jul-06
3556    ---------------------
3558     1. In order to handle tests when input lines are enormously long, pcretest has
3559        been re-factored so that it automatically extends its buffers when
3560        necessary. The code is crude, but this _is_ just a test program. The
3561        default size has been increased from 32K to 50K.
3563     2. The code in pcre_study() was using the value of the re argument before
3564        testing it for NULL. (Of course, in any sensible call of the function, it
3565        won't be NULL.)
3567     3. The memmove() emulation function in pcre_internal.h, which is used on
3568        systems that lack both memmove() and bcopy() - that is, hardly ever -
3569        was missing a "static" storage class specifier.
3571     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
3572        containing an extended class (one that cannot be represented by a bitmap
3573        because it contains high-valued characters or Unicode property items, e.g.
3574        [\pZ]). Almost always one would set UTF-8 mode when processing such a
3575        pattern, but PCRE should not loop if you do not (it no longer does).
3576        [Detail: two cases were found: (a) a repeated subpattern containing an
3577        extended class; (b) a recursive reference to a subpattern that followed a
3578        previous extended class. It wasn't skipping over the extended class
3579        correctly when UTF-8 mode was not set.]
3581     5. A negated single-character class was not being recognized as fixed-length
3582        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
3583        compile error "lookbehind assertion is not fixed length".