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

Diff of /code/trunk/ChangeLog

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

revision 740 by zherczeg, Mon Oct 31 06:10:14 2011 UTC revision 1574 by zherczeg, Tue Jun 30 07:13:23 2015 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4  Version 8.21  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.
6    
7    Version 8.38 xx-xxx-xxxx
8    ------------------------
9    
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.
15    
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.
20    
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.
24    
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.
28    
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.
33    
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.
36    
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.)
42    
43    8.  There was no check for integer overflow in subroutine calls such as (?123).
44    
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.
47    
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.
50    
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|)+)+$()/.
54    
55    12. In an EBCDIC environment, pcretest was mishandling the escape sequences
56        \a and \e in test subject lines.
57    
58    13. In an EBCDIC environment, \a in a pattern was converted to the ASCII
59        instead of the EBCDIC value.
60    
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.
63    
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.
67    
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.
71    
72    17. \h within a class was incorrectly compiled in EBCDIC environments.
73    
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))|(?=(.))))/.
77    
78    19. JIT should return with error when the compiled pattern requires more stack
79        space than the maximum.
80    
81    
82    Version 8.37 28-April-2015
83    --------------------------
84    
85    1.  When an (*ACCEPT) is triggered inside capturing parentheses, it arranges
86        for those parentheses to be closed with whatever has been captured so far.
87        However, it was failing to mark any other groups between the hightest
88        capture so far and the currrent group as "unset". Thus, the ovector for
89        those groups contained whatever was previously there. An example is the
90        pattern /(x)|((*ACCEPT))/ when matched against "abcd".
91    
92    2.  If an assertion condition was quantified with a minimum of zero (an odd
93        thing to do, but it happened), SIGSEGV or other misbehaviour could occur.
94    
95    3.  If a pattern in pcretest input had the P (POSIX) modifier followed by an
96        unrecognized modifier, a crash could occur.
97    
98    4.  An attempt to do global matching in pcretest with a zero-length ovector
99        caused a crash.
100    
101    5.  Fixed a memory leak during matching that could occur for a subpattern
102        subroutine call (recursive or otherwise) if the number of captured groups
103        that had to be saved was greater than ten.
104    
105    6.  Catch a bad opcode during auto-possessification after compiling a bad UTF
106        string with NO_UTF_CHECK. This is a tidyup, not a bug fix, as passing bad
107        UTF with NO_UTF_CHECK is documented as having an undefined outcome.
108    
109    7.  A UTF pattern containing a "not" match of a non-ASCII character and a
110        subroutine reference could loop at compile time. Example: /[^\xff]((?1))/.
111    
112    8. When a pattern is compiled, it remembers the highest back reference so that
113       when matching, if the ovector is too small, extra memory can be obtained to
114       use instead. A conditional subpattern whose condition is a check on a
115       capture having happened, such as, for example in the pattern
116       /^(?:(a)|b)(?(1)A|B)/, is another kind of back reference, but it was not
117       setting the highest backreference number. This mattered only if pcre_exec()
118       was called with an ovector that was too small to hold the capture, and there
119       was no other kind of back reference (a situation which is probably quite
120       rare). The effect of the bug was that the condition was always treated as
121       FALSE when the capture could not be consulted, leading to a incorrect
122       behaviour by pcre_exec(). This bug has been fixed.
123    
124    9. A reference to a duplicated named group (either a back reference or a test
125       for being set in a conditional) that occurred in a part of the pattern where
126       PCRE_DUPNAMES was not set caused the amount of memory needed for the pattern
127       to be incorrectly calculated, leading to overwriting.
128    
129    10. A mutually recursive set of back references such as (\2)(\1) caused a
130        segfault at study time (while trying to find the minimum matching length).
131        The infinite loop is now broken (with the minimum length unset, that is,
132        zero).
133    
134    11. If an assertion that was used as a condition was quantified with a minimum
135        of zero, matching went wrong. In particular, if the whole group had
136        unlimited repetition and could match an empty string, a segfault was
137        likely. The pattern (?(?=0)?)+ is an example that caused this. Perl allows
138        assertions to be quantified, but not if they are being used as conditions,
139        so the above pattern is faulted by Perl. PCRE has now been changed so that
140        it also rejects such patterns.
141    
142    12. A possessive capturing group such as (a)*+ with a minimum repeat of zero
143        failed to allow the zero-repeat case if pcre2_exec() was called with an
144        ovector too small to capture the group.
145    
146    13. Fixed two bugs in pcretest that were discovered by fuzzing and reported by
147        Red Hat Product Security:
148    
149        (a) A crash if /K and /F were both set with the option to save the compiled
150        pattern.
151    
152        (b) Another crash if the option to print captured substrings in a callout
153        was combined with setting a null ovector, for example \O\C+ as a subject
154        string.
155    
156    14. A pattern such as "((?2){0,1999}())?", which has a group containing a
157        forward reference repeated a large (but limited) number of times within a
158        repeated outer group that has a zero minimum quantifier, caused incorrect
159        code to be compiled, leading to the error "internal error:
160        previously-checked referenced subpattern not found" when an incorrect
161        memory address was read. This bug was reported as "heap overflow",
162        discovered by Kai Lu of Fortinet's FortiGuard Labs and given the CVE number
163        CVE-2015-2325.
164    
165    23. A pattern such as "((?+1)(\1))/" containing a forward reference subroutine
166        call within a group that also contained a recursive back reference caused
167        incorrect code to be compiled. This bug was reported as "heap overflow",
168        discovered by Kai Lu of Fortinet's FortiGuard Labs, and given the CVE
169        number CVE-2015-2326.
170    
171    24. Computing the size of the JIT read-only data in advance has been a source
172        of various issues, and new ones are still appear unfortunately. To fix
173        existing and future issues, size computation is eliminated from the code,
174        and replaced by on-demand memory allocation.
175    
176    25. A pattern such as /(?i)[A-`]/, where characters in the other case are
177        adjacent to the end of the range, and the range contained characters with
178        more than one other case, caused incorrect behaviour when compiled in UTF
179        mode. In that example, the range a-j was left out of the class.
180    
181    26. Fix JIT compilation of conditional blocks, which assertion
182        is converted to (*FAIL). E.g: /(?(?!))/.
183    
184    27. The pattern /(?(?!)^)/ caused references to random memory. This bug was
185        discovered by the LLVM fuzzer.
186    
187    28. The assertion (?!) is optimized to (*FAIL). This was not handled correctly
188        when this assertion was used as a condition, for example (?(?!)a|b). In
189        pcre2_match() it worked by luck; in pcre2_dfa_match() it gave an incorrect
190        error about an unsupported item.
191    
192    29. For some types of pattern, for example /Z*(|d*){216}/, the auto-
193        possessification code could take exponential time to complete. A recursion
194        depth limit of 1000 has been imposed to limit the resources used by this
195        optimization.
196    
197    30. A pattern such as /(*UTF)[\S\V\H]/, which contains a negated special class
198        such as \S in non-UCP mode, explicit wide characters (> 255) can be ignored
199        because \S ensures they are all in the class. The code for doing this was
200        interacting badly with the code for computing the amount of space needed to
201        compile the pattern, leading to a buffer overflow. This bug was discovered
202        by the LLVM fuzzer.
203    
204    31. A pattern such as /((?2)+)((?1))/ which has mutual recursion nested inside
205        other kinds of group caused stack overflow at compile time. This bug was
206        discovered by the LLVM fuzzer.
207    
208    32. A pattern such as /(?1)(?#?'){8}(a)/ which had a parenthesized comment
209        between a subroutine call and its quantifier was incorrectly compiled,
210        leading to buffer overflow or other errors. This bug was discovered by the
211        LLVM fuzzer.
212    
213    33. The illegal pattern /(?(?<E>.*!.*)?)/ was not being diagnosed as missing an
214        assertion after (?(. The code was failing to check the character after
215        (?(?< for the ! or = that would indicate a lookbehind assertion. This bug
216        was discovered by the LLVM fuzzer.
217    
218    34. A pattern such as /X((?2)()*+){2}+/ which has a possessive quantifier with
219        a fixed maximum following a group that contains a subroutine reference was
220        incorrectly compiled and could trigger buffer overflow. This bug was
221        discovered by the LLVM fuzzer.
222    
223    35. A mutual recursion within a lookbehind assertion such as (?<=((?2))((?1)))
224        caused a stack overflow instead of the diagnosis of a non-fixed length
225        lookbehind assertion. This bug was discovered by the LLVM fuzzer.
226    
227    36. The use of \K in a positive lookbehind assertion in a non-anchored pattern
228        (e.g. /(?<=\Ka)/) could make pcregrep loop.
229    
230    37. There was a similar problem to 36 in pcretest for global matches.
231    
232    38. If a greedy quantified \X was preceded by \C in UTF mode (e.g. \C\X*),
233        and a subsequent item in the pattern caused a non-match, backtracking over
234        the repeated \X did not stop, but carried on past the start of the subject,
235        causing reference to random memory and/or a segfault. There were also some
236        other cases where backtracking after \C could crash. This set of bugs was
237        discovered by the LLVM fuzzer.
238    
239    39. The function for finding the minimum length of a matching string could take
240        a very long time if mutual recursion was present many times in a pattern,
241        for example, /((?2){73}(?2))((?1))/. A better mutual recursion detection
242        method has been implemented. This infelicity was discovered by the LLVM
243        fuzzer.
244    
245    40. Static linking against the PCRE library using the pkg-config module was
246        failing on missing pthread symbols.
247    
248    
249    Version 8.36 26-September-2014
250    ------------------------------
251    
252    1.  Got rid of some compiler warnings in the C++ modules that were shown up by
253        -Wmissing-field-initializers and -Wunused-parameter.
254    
255    2.  The tests for quantifiers being too big (greater than 65535) were being
256        applied after reading the number, and stupidly assuming that integer
257        overflow would give a negative number. The tests are now applied as the
258        numbers are read.
259    
260    3.  Tidy code in pcre_exec.c where two branches that used to be different are
261        now the same.
262    
263    4.  The JIT compiler did not generate match limit checks for certain
264        bracketed expressions with quantifiers. This may lead to exponential
265        backtracking, instead of returning with PCRE_ERROR_MATCHLIMIT. This
266        issue should be resolved now.
267    
268    5.  Fixed an issue, which occures when nested alternatives are optimized
269        with table jumps.
270    
271    6.  Inserted two casts and changed some ints to size_t in the light of some
272        reported 64-bit compiler warnings (Bugzilla 1477).
273    
274    7.  Fixed a bug concerned with zero-minimum possessive groups that could match
275        an empty string, which sometimes were behaving incorrectly in the
276        interpreter (though correctly in the JIT matcher). This pcretest input is
277        an example:
278    
279          '\A(?:[^"]++|"(?:[^"]*+|"")*+")++'
280          NON QUOTED "QUOT""ED" AFTER "NOT MATCHED
281    
282        the interpreter was reporting a match of 'NON QUOTED ' only, whereas the
283        JIT matcher and Perl both matched 'NON QUOTED "QUOT""ED" AFTER '. The test
284        for an empty string was breaking the inner loop and carrying on at a lower
285        level, when possessive repeated groups should always return to a higher
286        level as they have no backtrack points in them. The empty string test now
287        occurs at the outer level.
288    
289    8.  Fixed a bug that was incorrectly auto-possessifying \w+ in the pattern
290        ^\w+(?>\s*)(?<=\w) which caused it not to match "test test".
291    
292    9.  Give a compile-time error for \o{} (as Perl does) and for \x{} (which Perl
293        doesn't).
294    
295    10. Change 8.34/15 introduced a bug that caused the amount of memory needed
296        to hold a pattern to be incorrectly computed (too small) when there were
297        named back references to duplicated names. This could cause "internal
298        error: code overflow" or "double free or corruption" or other memory
299        handling errors.
300    
301    11. When named subpatterns had the same prefixes, back references could be
302        confused. For example, in this pattern:
303    
304          /(?P<Name>a)?(?P<Name2>b)?(?(<Name>)c|d)*l/
305    
306        the reference to 'Name' was incorrectly treated as a reference to a
307        duplicate name.
308    
309    12. A pattern such as /^s?c/mi8 where the optional character has more than
310        one "other case" was incorrectly compiled such that it would only try to
311        match starting at "c".
312    
313    13. When a pattern starting with \s was studied, VT was not included in the
314        list of possible starting characters; this should have been part of the
315        8.34/18 patch.
316    
317    14. If a character class started [\Qx]... where x is any character, the class
318        was incorrectly terminated at the ].
319    
320    15. If a pattern that started with a caseless match for a character with more
321        than one "other case" was studied, PCRE did not set up the starting code
322        unit bit map for the list of possible characters. Now it does. This is an
323        optimization improvement, not a bug fix.
324    
325    16. The Unicode data tables have been updated to Unicode 7.0.0.
326    
327    17. Fixed a number of memory leaks in pcregrep.
328    
329    18. Avoid a compiler warning (from some compilers) for a function call with
330        a cast that removes "const" from an lvalue by using an intermediate
331        variable (to which the compiler does not object).
332    
333    19. Incorrect code was compiled if a group that contained an internal recursive
334        back reference was optional (had quantifier with a minimum of zero). This
335        example compiled incorrect code: /(((a\2)|(a*)\g<-1>))*/ and other examples
336        caused segmentation faults because of stack overflows at compile time.
337    
338    20. A pattern such as /((?(R)a|(?1)))+/, which contains a recursion within a
339        group that is quantified with an indefinite repeat, caused a compile-time
340        loop which used up all the system stack and provoked a segmentation fault.
341        This was not the same bug as 19 above.
342    
343    21. Add PCRECPP_EXP_DECL declaration to operator<< in pcre_stringpiece.h.
344        Patch by Mike Frysinger.
345    
346    
347    Version 8.35 04-April-2014
348    --------------------------
349    
350    1.  A new flag is set, when property checks are present in an XCLASS.
351        When this flag is not set, PCRE can perform certain optimizations
352        such as studying these XCLASS-es.
353    
354    2.  The auto-possessification of character sets were improved: a normal
355        and an extended character set can be compared now. Furthermore
356        the JIT compiler optimizes more character set checks.
357    
358    3.  Got rid of some compiler warnings for potentially uninitialized variables
359        that show up only when compiled with -O2.
360    
361    4.  A pattern such as (?=ab\K) that uses \K in an assertion can set the start
362        of a match later then the end of the match. The pcretest program was not
363        handling the case sensibly - it was outputting from the start to the next
364        binary zero. It now reports this situation in a message, and outputs the
365        text from the end to the start.
366    
367    5.  Fast forward search is improved in JIT. Instead of the first three
368        characters, any three characters with fixed position can be searched.
369        Search order: first, last, middle.
370    
371    6.  Improve character range checks in JIT. Characters are read by an inprecise
372        function now, which returns with an unknown value if the character code is
373        above a certain threshold (e.g: 256). The only limitation is that the value
374        must be bigger than the threshold as well. This function is useful when
375        the characters above the threshold are handled in the same way.
376    
377    7.  The macros whose names start with RAWUCHAR are placeholders for a future
378        mode in which only the bottom 21 bits of 32-bit data items are used. To
379        make this more memorable for those maintaining the code, the names have
380        been changed to start with UCHAR21, and an extensive comment has been added
381        to their definition.
382    
383    8.  Add missing (new) files sljitNativeTILEGX.c and sljitNativeTILEGX-encoder.c
384        to the export list in Makefile.am (they were accidentally omitted from the
385        8.34 tarball).
386    
387    9.  The informational output from pcretest used the phrase "starting byte set"
388        which is inappropriate for the 16-bit and 32-bit libraries. As the output
389        for "first char" and "need char" really means "non-UTF-char", I've changed
390        "byte" to "char", and slightly reworded the output. The documentation about
391        these values has also been (I hope) clarified.
392    
393    10. Another JIT related optimization: use table jumps for selecting the correct
394        backtracking path, when more than four alternatives are present inside a
395        bracket.
396    
397    11. Empty match is not possible, when the minimum length is greater than zero,
398        and there is no \K in the pattern. JIT should avoid empty match checks in
399        such cases.
400    
401    12. In a caseless character class with UCP support, when a character with more
402        than one alternative case was not the first character of a range, not all
403        the alternative cases were added to the class. For example, s and \x{17f}
404        are both alternative cases for S: the class [RST] was handled correctly,
405        but [R-T] was not.
406    
407    13. The configure.ac file always checked for pthread support when JIT was
408        enabled. This is not used in Windows, so I have put this test inside a
409        check for the presence of windows.h (which was already tested for).
410    
411    14. Improve pattern prefix search by a simplified Boyer-Moore algorithm in JIT.
412        The algorithm provides a way to skip certain starting offsets, and usually
413        faster than linear prefix searches.
414    
415    15. Change 13 for 8.20 updated RunTest to check for the 'fr' locale as well
416        as for 'fr_FR' and 'french'. For some reason, however, it then used the
417        Windows-specific input and output files, which have 'french' screwed in.
418        So this could never have worked. One of the problems with locales is that
419        they aren't always the same. I have now updated RunTest so that it checks
420        the output of the locale test (test 3) against three different output
421        files, and it allows the test to pass if any one of them matches. With luck
422        this should make the test pass on some versions of Solaris where it was
423        failing. Because of the uncertainty, the script did not used to stop if
424        test 3 failed; it now does. If further versions of a French locale ever
425        come to light, they can now easily be added.
426    
427    16. If --with-pcregrep-bufsize was given a non-integer value such as "50K",
428        there was a message during ./configure, but it did not stop. This now
429        provokes an error. The invalid example in README has been corrected.
430        If a value less than the minimum is given, the minimum value has always
431        been used, but now a warning is given.
432    
433    17. If --enable-bsr-anycrlf was set, the special 16/32-bit test failed. This
434        was a bug in the test system, which is now fixed. Also, the list of various
435        configurations that are tested for each release did not have one with both
436        16/32 bits and --enable-bar-anycrlf. It now does.
437    
438    18. pcretest was missing "-C bsr" for displaying the \R default setting.
439    
440    19. Little endian PowerPC systems are supported now by the JIT compiler.
441    
442    20. The fast forward newline mechanism could enter to an infinite loop on
443        certain invalid UTF-8 input. Although we don't support these cases
444        this issue can be fixed by a performance optimization.
445    
446    21. Change 33 of 8.34 is not sufficient to ensure stack safety because it does
447        not take account if existing stack usage. There is now a new global
448        variable called pcre_stack_guard that can be set to point to an external
449        function to check stack availability. It is called at the start of
450        processing every parenthesized group.
451    
452    22. A typo in the code meant that in ungreedy mode the max/min qualifier
453        behaved like a min-possessive qualifier, and, for example, /a{1,3}b/U did
454        not match "ab".
455    
456    23. When UTF was disabled, the JIT program reported some incorrect compile
457        errors. These messages are silenced now.
458    
459    24. Experimental support for ARM-64 and MIPS-64 has been added to the JIT
460        compiler.
461    
462    25. Change all the temporary files used in RunGrepTest to be different to those
463        used by RunTest so that the tests can be run simultaneously, for example by
464        "make -j check".
465    
466    
467    Version 8.34 15-December-2013
468    -----------------------------
469    
470    1.  Add pcre[16|32]_jit_free_unused_memory to forcibly free unused JIT
471        executable memory. Patch inspired by Carsten Klein.
472    
473    2.  ./configure --enable-coverage defined SUPPORT_GCOV in config.h, although
474        this macro is never tested and has no effect, because the work to support
475        coverage involves only compiling and linking options and special targets in
476        the Makefile. The comment in config.h implied that defining the macro would
477        enable coverage support, which is totally false. There was also support for
478        setting this macro in the CMake files (my fault, I just copied it from
479        configure). SUPPORT_GCOV has now been removed.
480    
481    3.  Make a small performance improvement in strlen16() and strlen32() in
482        pcretest.
483    
484    4.  Change 36 for 8.33 left some unreachable statements in pcre_exec.c,
485        detected by the Solaris compiler (gcc doesn't seem to be able to diagnose
486        these cases). There was also one in pcretest.c.
487    
488    5.  Cleaned up a "may be uninitialized" compiler warning in pcre_exec.c.
489    
490    6.  In UTF mode, the code for checking whether a group could match an empty
491        string (which is used for indefinitely repeated groups to allow for
492        breaking an infinite loop) was broken when the group contained a repeated
493        negated single-character class with a character that occupied more than one
494        data item and had a minimum repetition of zero (for example, [^\x{100}]* in
495        UTF-8 mode). The effect was undefined: the group might or might not be
496        deemed as matching an empty string, or the program might have crashed.
497    
498    7.  The code for checking whether a group could match an empty string was not
499        recognizing that \h, \H, \v, \V, and \R must match a character.
500    
501    8.  Implemented PCRE_INFO_MATCH_EMPTY, which yields 1 if the pattern can match
502        an empty string. If it can, pcretest shows this in its information output.
503    
504    9.  Fixed two related bugs that applied to Unicode extended grapheme clusters
505        that were repeated with a maximizing qualifier (e.g. \X* or \X{2,5}) when
506        matched by pcre_exec() without using JIT:
507    
508        (a) If the rest of the pattern did not match after a maximal run of
509            grapheme clusters, the code for backing up to try with fewer of them
510            did not always back up over a full grapheme when characters that do not
511            have the modifier quality were involved, e.g. Hangul syllables.
512    
513        (b) If the match point in a subject started with modifier character, and
514            there was no match, the code could incorrectly back up beyond the match
515            point, and potentially beyond the first character in the subject,
516            leading to a segfault or an incorrect match result.
517    
518    10. A conditional group with an assertion condition could lead to PCRE
519        recording an incorrect first data item for a match if no other first data
520        item was recorded. For example, the pattern (?(?=ab)ab) recorded "a" as a
521        first data item, and therefore matched "ca" after "c" instead of at the
522        start.
523    
524    11. Change 40 for 8.33 (allowing pcregrep to find empty strings) showed up a
525        bug that caused the command "echo a | ./pcregrep -M '|a'" to loop.
526    
527    12. The source of pcregrep now includes z/OS-specific code so that it can be
528        compiled for z/OS as part of the special z/OS distribution.
529    
530    13. Added the -T and -TM options to pcretest.
531    
532    14. The code in pcre_compile.c for creating the table of named capturing groups
533        has been refactored. Instead of creating the table dynamically during the
534        actual compiling pass, the information is remembered during the pre-compile
535        pass (on the stack unless there are more than 20 named groups, in which
536        case malloc() is used) and the whole table is created before the actual
537        compile happens. This has simplified the code (it is now nearly 150 lines
538        shorter) and prepared the way for better handling of references to groups
539        with duplicate names.
540    
541    15. A back reference to a named subpattern when there is more than one of the
542        same name now checks them in the order in which they appear in the pattern.
543        The first one that is set is used for the reference. Previously only the
544        first one was inspected. This change makes PCRE more compatible with Perl.
545    
546    16. Unicode character properties were updated from Unicode 6.3.0.
547    
548    17. The compile-time code for auto-possessification has been refactored, based
549        on a patch by Zoltan Herczeg. It now happens after instead of during
550        compilation. The code is cleaner, and more cases are handled. The option
551        PCRE_NO_AUTO_POSSESS is added for testing purposes, and the -O and /O
552        options in pcretest are provided to set it. It can also be set by
553        (*NO_AUTO_POSSESS) at the start of a pattern.
554    
555    18. The character VT has been added to the default ("C" locale) set of
556        characters that match \s and are generally treated as white space,
557        following this same change in Perl 5.18. There is now no difference between
558        "Perl space" and "POSIX space". Whether VT is treated as white space in
559        other locales depends on the locale.
560    
561    19. The code for checking named groups as conditions, either for being set or
562        for being recursed, has been refactored (this is related to 14 and 15
563        above). Processing unduplicated named groups should now be as fast at
564        numerical groups, and processing duplicated groups should be faster than
565        before.
566    
567    20. Two patches to the CMake build system, by Alexander Barkov:
568    
569          (1) Replace the "source" command by "." in CMakeLists.txt because
570              "source" is a bash-ism.
571    
572          (2) Add missing HAVE_STDINT_H and HAVE_INTTYPES_H to config-cmake.h.in;
573              without these the CMake build does not work on Solaris.
574    
575    21. Perl has changed its handling of \8 and \9. If there is no previously
576        encountered capturing group of those numbers, they are treated as the
577        literal characters 8 and 9 instead of a binary zero followed by the
578        literals. PCRE now does the same.
579    
580    22. Following Perl, added \o{} to specify codepoints in octal, making it
581        possible to specify values greater than 0777 and also making them
582        unambiguous.
583    
584    23. Perl now gives an error for missing closing braces after \x{... instead of
585        treating the string as literal. PCRE now does the same.
586    
587    24. RunTest used to grumble if an inappropriate test was selected explicitly,
588        but just skip it when running all tests. This make it awkward to run ranges
589        of tests when one of them was inappropriate. Now it just skips any
590        inappropriate tests, as it always did when running all tests.
591    
592    25. If PCRE_AUTO_CALLOUT and PCRE_UCP were set for a pattern that contained
593        character types such as \d or \w, too many callouts were inserted, and the
594        data that they returned was rubbish.
595    
596    26. In UCP mode, \s was not matching two of the characters that Perl matches,
597        namely NEL (U+0085) and MONGOLIAN VOWEL SEPARATOR (U+180E), though they
598        were matched by \h. The code has now been refactored so that the lists of
599        the horizontal and vertical whitespace characters used for \h and \v (which
600        are defined only in one place) are now also used for \s.
601    
602    27. Add JIT support for the 64 bit TileGX architecture.
603        Patch by Jiong Wang (Tilera Corporation).
604    
605    28. Possessive quantifiers for classes (both explicit and automatically
606        generated) now use special opcodes instead of wrapping in ONCE brackets.
607    
608    29. Whereas an item such as A{4}+ ignored the possessivenes of the quantifier
609        (because it's meaningless), this was not happening when PCRE_CASELESS was
610        set. Not wrong, but inefficient.
611    
612    30. Updated perltest.pl to add /u (force Unicode mode) when /W (use Unicode
613        properties for \w, \d, etc) is present in a test regex. Otherwise if the
614        test contains no characters greater than 255, Perl doesn't realise it
615        should be using Unicode semantics.
616    
617    31. Upgraded the handling of the POSIX classes [:graph:], [:print:], and
618        [:punct:] when PCRE_UCP is set so as to include the same characters as Perl
619        does in Unicode mode.
620    
621    32. Added the "forbid" facility to pcretest so that putting tests into the
622        wrong test files can sometimes be quickly detected.
623    
624    33. There is now a limit (default 250) on the depth of nesting of parentheses.
625        This limit is imposed to control the amount of system stack used at compile
626        time. It can be changed at build time by --with-parens-nest-limit=xxx or
627        the equivalent in CMake.
628    
629    34. Character classes such as [A-\d] or [a-[:digit:]] now cause compile-time
630        errors. Perl warns for these when in warning mode, but PCRE has no facility
631        for giving warnings.
632    
633    35. Change 34 for 8.13 allowed quantifiers on assertions, because Perl does.
634        However, this was not working for (?!) because it is optimized to (*FAIL),
635        for which PCRE does not allow quantifiers. The optimization is now disabled
636        when a quantifier follows (?!). I can't see any use for this, but it makes
637        things uniform.
638    
639    36. Perl no longer allows group names to start with digits, so I have made this
640        change also in PCRE. It simplifies the code a bit.
641    
642    37. In extended mode, Perl ignores spaces before a + that indicates a
643        possessive quantifier. PCRE allowed a space before the quantifier, but not
644        before the possessive +. It now does.
645    
646    38. The use of \K (reset reported match start) within a repeated possessive
647        group such as (a\Kb)*+ was not working.
648    
649    40. Document that the same character tables must be used at compile time and
650        run time, and that the facility to pass tables to pcre_exec() and
651        pcre_dfa_exec() is for use only with saved/restored patterns.
652    
653    41. Applied Jeff Trawick's patch CMakeLists.txt, which "provides two new
654        features for Builds with MSVC:
655    
656        1. Support pcre.rc and/or pcreposix.rc (as is already done for MinGW
657           builds). The .rc files can be used to set FileDescription and many other
658           attributes.
659    
660        2. Add an option (-DINSTALL_MSVC_PDB) to enable installation of .pdb files.
661           This allows higher-level build scripts which want .pdb files to avoid
662           hard-coding the exact files needed."
663    
664    42. Added support for [[:<:]] and [[:>:]] as used in the BSD POSIX library to
665        mean "start of word" and "end of word", respectively, as a transition aid.
666    
667    43. A minimizing repeat of a class containing codepoints greater than 255 in
668        non-UTF 16-bit or 32-bit modes caused an internal error when PCRE was
669        compiled to use the heap for recursion.
670    
671    44. Got rid of some compiler warnings for unused variables when UTF but not UCP
672        is configured.
673    
674    
675    Version 8.33 28-May-2013
676    ------------------------
677    
678    1.  Added 'U' to some constants that are compared to unsigned integers, to
679        avoid compiler signed/unsigned warnings. Added (int) casts to unsigned
680        variables that are added to signed variables, to ensure the result is
681        signed and can be negated.
682    
683    2.  Applied patch by Daniel Richard G for quashing MSVC warnings to the
684        CMake config files.
685    
686    3.  Revise the creation of config.h.generic so that all boolean macros are
687        #undefined, whereas non-boolean macros are #ifndef/#endif-ed. This makes
688        overriding via -D on the command line possible.
689    
690    4.  Changing the definition of the variable "op" in pcre_exec.c from pcre_uchar
691        to unsigned int is reported to make a quite noticeable speed difference in
692        a specific Windows environment. Testing on Linux did also appear to show
693        some benefit (and it is clearly not harmful). Also fixed the definition of
694        Xop which should be unsigned.
695    
696    5.  Related to (4), changing the definition of the intermediate variable cc
697        in repeated character loops from pcre_uchar to pcre_uint32 also gave speed
698        improvements.
699    
700    6.  Fix forward search in JIT when link size is 3 or greater. Also removed some
701        unnecessary spaces.
702    
703    7.  Adjust autogen.sh and configure.ac to lose warnings given by automake 1.12
704        and later.
705    
706    8.  Fix two buffer over read issues in 16 and 32 bit modes. Affects JIT only.
707    
708    9.  Optimizing fast_forward_start_bits in JIT.
709    
710    10. Adding support for callouts in JIT, and fixing some issues revealed
711        during this work. Namely:
712    
713        (a) Unoptimized capturing brackets incorrectly reset on backtrack.
714    
715        (b) Minimum length was not checked before the matching is started.
716    
717    11. The value of capture_last that is passed to callouts was incorrect in some
718        cases when there was a capture on one path that was subsequently abandoned
719        after a backtrack. Also, the capture_last value is now reset after a
720        recursion, since all captures are also reset in this case.
721    
722    12. The interpreter no longer returns the "too many substrings" error in the
723        case when an overflowing capture is in a branch that is subsequently
724        abandoned after a backtrack.
725    
726    13. In the pathological case when an offset vector of size 2 is used, pcretest
727        now prints out the matched string after a yield of 0 or 1.
728    
729    14. Inlining subpatterns in recursions, when certain conditions are fulfilled.
730        Only supported by the JIT compiler at the moment.
731    
732    15. JIT compiler now supports 32 bit Macs thanks to Lawrence Velazquez.
733    
734    16. Partial matches now set offsets[2] to the "bumpalong" value, that is, the
735        offset of the starting point of the matching process, provided the offsets
736        vector is large enough.
737    
738    17. The \A escape now records a lookbehind value of 1, though its execution
739        does not actually inspect the previous character. This is to ensure that,
740        in partial multi-segment matching, at least one character from the old
741        segment is retained when a new segment is processed. Otherwise, if there
742        are no lookbehinds in the pattern, \A might match incorrectly at the start
743        of a new segment.
744    
745    18. Added some #ifdef __VMS code into pcretest.c to help VMS implementations.
746    
747    19. Redefined some pcre_uchar variables in pcre_exec.c as pcre_uint32; this
748        gives some modest performance improvement in 8-bit mode.
749    
750    20. Added the PCRE-specific property \p{Xuc} for matching characters that can
751        be expressed in certain programming languages using Universal Character
752        Names.
753    
754    21. Unicode validation has been updated in the light of Unicode Corrigendum #9,
755        which points out that "non characters" are not "characters that may not
756        appear in Unicode strings" but rather "characters that are reserved for
757        internal use and have only local meaning".
758    
759    22. When a pattern was compiled with automatic callouts (PCRE_AUTO_CALLOUT) and
760        there was a conditional group that depended on an assertion, if the
761        assertion was false, the callout that immediately followed the alternation
762        in the condition was skipped when pcre_exec() was used for matching.
763    
764    23. Allow an explicit callout to be inserted before an assertion that is the
765        condition for a conditional group, for compatibility with automatic
766        callouts, which always insert a callout at this point.
767    
768    24. In 8.31, (*COMMIT) was confined to within a recursive subpattern. Perl also
769        confines (*SKIP) and (*PRUNE) in the same way, and this has now been done.
770    
771    25. (*PRUNE) is now supported by the JIT compiler.
772    
773    26. Fix infinite loop when /(?<=(*SKIP)ac)a/ is matched against aa.
774    
775    27. Fix the case where there are two or more SKIPs with arguments that may be
776        ignored.
777    
778    28. (*SKIP) is now supported by the JIT compiler.
779    
780    29. (*THEN) is now supported by the JIT compiler.
781    
782    30. Update RunTest with additional test selector options.
783    
784    31. The way PCRE handles backtracking verbs has been changed in two ways.
785    
786        (1) Previously, in something like (*COMMIT)(*SKIP), COMMIT would override
787        SKIP. Now, PCRE acts on whichever backtracking verb is reached first by
788        backtracking. In some cases this makes it more Perl-compatible, but Perl's
789        rather obscure rules do not always do the same thing.
790    
791        (2) Previously, backtracking verbs were confined within assertions. This is
792        no longer the case for positive assertions, except for (*ACCEPT). Again,
793        this sometimes improves Perl compatibility, and sometimes does not.
794    
795    32. A number of tests that were in test 2 because Perl did things differently
796        have been moved to test 1, because either Perl or PCRE has changed, and
797        these tests are now compatible.
798    
799    32. Backtracking control verbs are now handled in the same way in JIT and
800        interpreter.
801    
802    33. An opening parenthesis in a MARK/PRUNE/SKIP/THEN name in a pattern that
803        contained a forward subroutine reference caused a compile error.
804    
805    34. Auto-detect and optimize limited repetitions in JIT.
806    
807    35. Implement PCRE_NEVER_UTF to lock out the use of UTF, in particular,
808        blocking (*UTF) etc.
809    
810    36. In the interpreter, maximizing pattern repetitions for characters and
811        character types now use tail recursion, which reduces stack usage.
812    
813    37. The value of the max lookbehind was not correctly preserved if a compiled
814        and saved regex was reloaded on a host of different endianness.
815    
816    38. Implemented (*LIMIT_MATCH) and (*LIMIT_RECURSION). As part of the extension
817        of the compiled pattern block, expand the flags field from 16 to 32 bits
818        because it was almost full.
819    
820    39. Try madvise first before posix_madvise.
821    
822    40. Change 7 for PCRE 7.9 made it impossible for pcregrep to find empty lines
823        with a pattern such as ^$. It has taken 4 years for anybody to notice! The
824        original change locked out all matches of empty strings. This has been
825        changed so that one match of an empty string per line is recognized.
826        Subsequent searches on the same line (for colouring or for --only-matching,
827        for example) do not recognize empty strings.
828    
829    41. Applied a user patch to fix a number of spelling mistakes in comments.
830    
831    42. Data lines longer than 65536 caused pcretest to crash.
832    
833    43. Clarified the data type for length and startoffset arguments for pcre_exec
834        and pcre_dfa_exec in the function-specific man pages, where they were
835        explicitly stated to be in bytes, never having been updated. I also added
836        some clarification to the pcreapi man page.
837    
838    44. A call to pcre_dfa_exec() with an output vector size less than 2 caused
839        a segmentation fault.
840    
841    
842    Version 8.32 30-November-2012
843    -----------------------------
844    
845    1.  Improved JIT compiler optimizations for first character search and single
846        character iterators.
847    
848    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
849        Patch by Daniel Richard G.
850    
851    3.  Single character iterator optimizations in the JIT compiler.
852    
853    4.  Improved JIT compiler optimizations for character ranges.
854    
855    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
856        Reported by Giuseppe D'Angelo.
857    
858    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
859        of a line, was being set incorrectly in cases where .* appeared inside
860        atomic brackets at the start of a pattern, or where there was a subsequent
861        *PRUNE or *SKIP.
862    
863    7.  Improved instruction cache flush for POWER/PowerPC.
864        Patch by Daniel Richard G.
865    
866    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
867        grep:
868    
869        (a) There is now no limit to the number of patterns to be matched.
870    
871        (b) An error is given if a pattern is too long.
872    
873        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
874            are now supported.
875    
876        (d) --exclude-from and --include-from (multiple use) have been added.
877    
878        (e) Exclusions and inclusions now apply to all files and directories, not
879            just to those obtained from scanning a directory recursively.
880    
881        (f) Multiple uses of -f and --file-list are now supported.
882    
883        (g) In a Windows environment, the default for -d has been changed from
884            "read" (the GNU grep default) to "skip", because otherwise the presence
885            of a directory in the file list provokes an error.
886    
887        (h) The documentation has been revised and clarified in places.
888    
889    9.  Improve the matching speed of capturing brackets.
890    
891    10. Changed the meaning of \X so that it now matches a Unicode extended
892        grapheme cluster.
893    
894    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
895        out POSIX threads when JIT support is configured.
896    
897    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
898    
899    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
900        structure could go wrong in environments where size_t is not the same size
901        as int.
902    
903    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
904        set.
905    
906    15. The EBCDIC support had decayed; later updates to the code had included
907        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
908        general tidy up of EBCDIC-related issues, and the documentation was also
909        not quite right. There is now a test that can be run on ASCII systems to
910        check some of the EBCDIC-related things (but is it not a full test).
911    
912    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
913        in a small tidy to the code.
914    
915    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
916    
917    18. If the --only-matching (-o) option in pcregrep is specified multiple
918        times, each one causes appropriate output. For example, -o1 -o2 outputs the
919        substrings matched by the 1st and 2nd capturing parentheses. A separating
920        string can be specified by --om-separator (default empty).
921    
922    19. Improving the first n character searches.
923    
924    20. Turn case lists for horizontal and vertical white space into macros so that
925        they are defined only once.
926    
927    21. This set of changes together give more compatible Unicode case-folding
928        behaviour for characters that have more than one other case when UCP
929        support is available.
930    
931        (a) The Unicode property table now has offsets into a new table of sets of
932            three or more characters that are case-equivalent. The MultiStage2.py
933            script that generates these tables (the pcre_ucd.c file) now scans
934            CaseFolding.txt instead of UnicodeData.txt for character case
935            information.
936    
937        (b) The code for adding characters or ranges of characters to a character
938            class has been abstracted into a generalized function that also handles
939            case-independence. In UTF-mode with UCP support, this uses the new data
940            to handle characters with more than one other case.
941    
942        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
943            whose other case is greater than 256 are now correctly matched
944            caselessly. Previously, the high codepoint matched the low one, but not
945            vice versa.
946    
947        (d) The processing of \h, \H, \v, and \ in character classes now makes use
948            of the new class addition function, using character lists defined as
949            macros alongside the case definitions of 20 above.
950    
951        (e) Caseless back references now work with characters that have more than
952            one other case.
953    
954        (f) General caseless matching of characters with more than one other case
955            is supported.
956    
957    22. Unicode character properties were updated from Unicode 6.2.0
958    
959    23. Improved CMake support under Windows. Patch by Daniel Richard G.
960    
961    24. Add support for 32-bit character strings, and UTF-32
962    
963    25. Major JIT compiler update (code refactoring and bugfixing).
964        Experimental Sparc 32 support is added.
965    
966    26. Applied a modified version of Daniel Richard G's patch to create
967        pcre.h.generic and config.h.generic by "make" instead of in the
968        PrepareRelease script.
969    
970    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
971        pcre_compile.c when checking for a zero character.
972    
973    28. Introducing a native interface for JIT. Through this interface, the compiled
974        machine code can be directly executed. The purpose of this interface is to
975        provide fast pattern matching, so several sanity checks are not performed.
976        However, feature tests are still performed. The new interface provides
977        1.4x speedup compared to the old one.
978    
979    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
980        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
981        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
982    
983    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
984        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
985        the "old" RFC 2279). Instead, it ended up passing a negative length to
986        pcre_exec().
987    
988    31. Add support for GCC's visibility feature to hide internal functions.
989    
990    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
991        "unknown -C option" after outputting 0 or 1.
992    
993    33. There is now support for generating a code coverage report for the test
994        suite in environments where gcc is the compiler and lcov is installed. This
995        is mainly for the benefit of the developers.
996    
997    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
998        unaddressable using valgrind annotations, allowing valgrind to detect
999        invalid memory accesses. This is mainly for the benefit of the developers.
1000    
1001    25. (*UTF) can now be used to start a pattern in any of the three libraries.
1002    
1003    26. Give configure error if --enable-cpp but no C++ compiler found.
1004    
1005    
1006    Version 8.31 06-July-2012
1007    -------------------------
1008    
1009    1.  Fixing a wrong JIT test case and some compiler warnings.
1010    
1011    2.  Removed a bashism from the RunTest script.
1012    
1013    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
1014        to unsigned type, result still unsigned" that was given by an MS compiler
1015        on encountering the code "-sizeof(xxx)".
1016    
1017    4.  Partial matching support is added to the JIT compiler.
1018    
1019    5.  Fixed several bugs concerned with partial matching of items that consist
1020        of more than one character:
1021    
1022        (a) /^(..)\1/ did not partially match "aba" because checking references was
1023            done on an "all or nothing" basis. This also applied to repeated
1024            references.
1025    
1026        (b) \R did not give a hard partial match if \r was found at the end of the
1027            subject.
1028    
1029        (c) \X did not give a hard partial match after matching one or more
1030            characters at the end of the subject.
1031    
1032        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
1033            a partial match for the string "\r".
1034    
1035        (e) When newline was set to CRLF, the metacharacter "." did not recognize
1036            a partial match for a CR character at the end of the subject string.
1037    
1038    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
1039        running pcretest, the text "(JIT)" added to the output whenever JIT is
1040        actually used to run the match.
1041    
1042    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
1043        or /S+[+] with a digit between 1 and 7.
1044    
1045    8.  OP_NOT now supports any UTF character not just single-byte ones.
1046    
1047    9.  (*MARK) control verb is now supported by the JIT compiler.
1048    
1049    10. The command "./RunTest list" lists the available tests without actually
1050        running any of them. (Because I keep forgetting what they all are.)
1051    
1052    11. Add PCRE_INFO_MAXLOOKBEHIND.
1053    
1054    12. Applied a (slightly modified) user-supplied patch that improves performance
1055        when the heap is used for recursion (compiled with --disable-stack-for-
1056        recursion). Instead of malloc and free for each heap frame each time a
1057        logical recursion happens, frames are retained on a chain and re-used where
1058        possible. This sometimes gives as much as 30% improvement.
1059    
1060    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
1061        call.
1062    
1063    14. As documented, (*COMMIT) is now confined to within a positive assertion.
1064    
1065    15. It is now possible to link pcretest with libedit as an alternative to
1066        libreadline.
1067    
1068    16. (*COMMIT) control verb is now supported by the JIT compiler.
1069    
1070    17. The Unicode data tables have been updated to Unicode 6.1.0.
1071    
1072    18. Added --file-list option to pcregrep.
1073    
1074    19. Added binary file support to pcregrep, including the -a, --binary-files,
1075        -I, and --text options.
1076    
1077    20. The madvise function is renamed for posix_madvise for QNX compatibility
1078        reasons. Fixed by Giuseppe D'Angelo.
1079    
1080    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
1081    
1082    22. Changed the option for creating symbolic links for 16-bit man pages from
1083        -s to -sf so that re-installing does not cause issues.
1084    
1085    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
1086    
1087    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
1088        matches in certain environments (the workspace was not being correctly
1089        retained). Also added to pcre_dfa_exec() a simple plausibility check on
1090        some of the workspace data at the beginning of a restart.
1091    
1092    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
1093        was not doing so when it should - probably a typo introduced by SVN 528
1094        (change 8.10/14).
1095    
1096    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
1097        \w+ when the character tables indicated that \x{c4} was a word character.
1098        There were several related cases, all because the tests for doing a table
1099        lookup were testing for characters less than 127 instead of 255.
1100    
1101    27. If a pattern contains capturing parentheses that are not used in a match,
1102        their slots in the ovector are set to -1. For those that are higher than
1103        any matched groups, this happens at the end of processing. In the case when
1104        there were back references that the ovector was too small to contain
1105        (causing temporary malloc'd memory to be used during matching), and the
1106        highest capturing number was not used, memory off the end of the ovector
1107        was incorrectly being set to -1. (It was using the size of the temporary
1108        memory instead of the true size.)
1109    
1110    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
1111        ovector size, it uses memory at the end of the block that it has got.
1112    
1113    29. Check for an overlong MARK name and give an error at compile time. The
1114        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
1115    
1116    30. JIT compiler update.
1117    
1118    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
1119        Rill for the patch.
1120    
1121    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
1122    
1123    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
1124    
1125    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
1126        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
1127        was enabled.
1128    
1129    35. Improve JIT code generation for greedy plus quantifier.
1130    
1131    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
1132        1 to "aa" instead of to an empty string. The bug affected repeated groups
1133        that could potentially match an empty string.
1134    
1135    37. Optimizing single character iterators in JIT.
1136    
1137    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
1138        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
1139        codepoints that are too big for the mode are faulted, and in a UTF mode,
1140        disallowed codepoints are also faulted.
1141    
1142    39. If PCRE was compiled with UTF support, in three places in the DFA
1143        matcher there was code that should only have been obeyed in UTF mode, but
1144        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
1145        processing when bytes with values greater than 127 were present. In 16-bit
1146        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
1147        both cases the values are those that cannot be the first data item in a UTF
1148        character. The three items that might have provoked this were recursions,
1149        possessively repeated groups, and atomic groups.
1150    
1151    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
1152        and pcregrep, because some OS require shared objects to be explicitly
1153        passed to ld, causing the link step to fail if they are not.
1154    
1155    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
1156        mode, patterns that started with \h* or \R* might be incorrectly matched.
1157    
1158    
1159    Version 8.30 04-February-2012
1160    -----------------------------
1161    
1162    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
1163        name is defined in ctype.h.
1164    
1165    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
1166        only in quite long subpatterns.
1167    
1168    3.  Removed the function pcre_info(), which has been obsolete and deprecated
1169        since it was replaced by pcre_fullinfo() in February 2000.
1170    
1171    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
1172        match a (*MARK), and the match failed at the start of the subject, a
1173        reference to memory before the start of the subject could occur. This bug
1174        was introduced by fix 17 of release 8.21.
1175    
1176    5.  A reference to an unset group with zero minimum repetition was giving
1177        totally wrong answers (in non-JavaScript-compatibility mode). For example,
1178        /(another)?(\1?)test/ matched against "hello world test". This bug was
1179        introduced in release 8.13.
1180    
1181    6.  Add support for 16-bit character strings (a large amount of work involving
1182        many changes and refactorings).
1183    
1184    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
1185        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
1186        from a file.
1187    
1188    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
1189        rounding is not applied in this particular case).
1190    
1191    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
1192        if they appear, or are escaped, in patterns.
1193    
1194    10. Get rid of a number of -Wunused-but-set-variable warnings.
1195    
1196    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
1197        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
1198        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
1199        also returns the mark "x". This bug applied to capturing parentheses,
1200        non-capturing parentheses, and atomic parentheses. It also applied to some
1201        assertions.
1202    
1203    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
1204        information out of configure.ac instead of relying on pcre.h.generic, which
1205        is not stored in the repository.
1206    
1207    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
1208        -lreadline.
1209    
1210    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
1211    
1212    15. Applied Graycode's patch to put the top-level frame on the stack rather
1213        than the heap when not using the stack for recursion. This gives a
1214        performance improvement in many cases when recursion is not deep.
1215    
1216    16. Experimental code added to "pcretest -C" to output the stack frame size.
1217    
1218    
1219    Version 8.21 12-Dec-2011
1220    ------------------------
1221    
1222  1.  Updating the JIT compiler.  1.  Updating the JIT compiler.
1223    
1224    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
1225        are added as well.
1226    
1227    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
1228        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
1229        calling _pcre_jit_exec. Some extra comments are added.
1230    
1231    4.  (*MARK) settings inside atomic groups that do not contain any capturing
1232        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
1233        was introduced by change 18 for 8.20.
1234    
1235    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
1236        ECMA-262 standard.
1237    
1238    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
1239        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
1240        This bug was probably introduced by change 9 of 8.13.
1241    
1242    7.  While fixing 6 above, I noticed that a number of other items were being
1243        incorrectly rejected as "not fixed length". This arose partly because newer
1244        opcodes had not been added to the fixed-length checking code. I have (a)
1245        corrected the bug and added tests for these items, and (b) arranged for an
1246        error to occur if an unknown opcode is encountered while checking for fixed
1247        length instead of just assuming "not fixed length". The items that were
1248        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
1249        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
1250        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
1251    
1252    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
1253        being incorrectly compiled and would have given unpredicatble results.
1254    
1255    9.  A possessively repeated subpattern with minimum repeat count greater than
1256        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
1257        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
1258        the first (A) could occur when it should not.
1259    
1260    10. Add a cast and remove a redundant test from the code.
1261    
1262    11. JIT should use pcre_malloc/pcre_free for allocation.
1263    
1264    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
1265        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
1266        is anything other than /usr, -L is still shown).
1267    
1268    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
1269    
1270    14. Perl does not support \N without a following name in a [] class; PCRE now
1271        also gives an error.
1272    
1273    15. If a forward reference was repeated with an upper limit of around 2000,
1274        it caused the error "internal error: overran compiling workspace". The
1275        maximum number of forward references (including repeats) was limited by the
1276        internal workspace, and dependent on the LINK_SIZE. The code has been
1277        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
1278        the default depends on LINK_SIZE. There is a new upper limit (for safety)
1279        of around 200,000 forward references. While doing this, I also speeded up
1280        the filling in of repeated forward references.
1281    
1282    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
1283        incorrectly expecting the subject to contain another "a" after the start.
1284    
1285    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
1286        in the match, the SKIP should be ignored. This was not happening; instead
1287        the SKIP was being treated as NOMATCH. For patterns such as
1288        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
1289        tested.
1290    
1291    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
1292        now much more compatible with Perl, in particular in cases where the result
1293        is a non-match for a non-anchored pattern. For example, if
1294        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
1295        "m", where previously it did not return a name. A side effect of this
1296        change is that for partial matches, the last encountered mark name is
1297        returned, as for non matches. A number of tests that were previously not
1298        Perl-compatible have been moved into the Perl-compatible test files. The
1299        refactoring has had the pleasing side effect of removing one argument from
1300        the match() function, thus reducing its stack requirements.
1301    
1302    19. If the /S+ option was used in pcretest to study a pattern using JIT,
1303        subsequent uses of /S (without +) incorrectly behaved like /S+.
1304    
1305    21. Retrieve executable code size support for the JIT compiler and fixing
1306        some warnings.
1307    
1308    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
1309        not work when the shorter character appeared right at the end of the
1310        subject string.
1311    
1312    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
1313        systems.
1314    
1315    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
1316        output it when the /M option is used in pcretest.
1317    
1318    25. The CheckMan script was not being included in the distribution. Also, added
1319        an explicit "perl" to run Perl scripts from the PrepareRelease script
1320        because this is reportedly needed in Windows.
1321    
1322    26. If study data was being save in a file and studying had not found a set of
1323        "starts with" bytes for the pattern, the data written to the file (though
1324        never used) was taken from uninitialized memory and so caused valgrind to
1325        complain.
1326    
1327    27. Updated RunTest.bat as provided by Sheri Pierce.
1328    
1329    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
1330    
1331    29. Computation of memory usage for the table of capturing group names was
1332        giving an unnecessarily large value.
1333    
1334    
1335  Version 8.20 21-Oct-2011  Version 8.20 21-Oct-2011
1336  ------------------------  ------------------------
# Line 1021  Version 7.9 11-Apr-09 Line 2346  Version 7.9 11-Apr-09
2346  7.  A pattern that could match an empty string could cause pcregrep to loop; it  7.  A pattern that could match an empty string could cause pcregrep to loop; it
2347      doesn't make sense to accept an empty string match in pcregrep, so I have      doesn't make sense to accept an empty string match in pcregrep, so I have
2348      locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this      locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
2349      seems to be how GNU grep behaves.      seems to be how GNU grep behaves. [But see later change 40 for release
2350        8.33.]
2351    
2352  8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at  8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
2353      start or after a newline", because the conditional assertion was not being      start or after a newline", because the conditional assertion was not being
# Line 1264  Version 7.7 07-May-08 Line 2590  Version 7.7 07-May-08
2590      containing () gave an internal compiling error instead of "reference to      containing () gave an internal compiling error instead of "reference to
2591      non-existent subpattern". Fortunately, when the pattern did exist, the      non-existent subpattern". Fortunately, when the pattern did exist, the
2592      compiled code was correct. (When scanning forwards to check for the      compiled code was correct. (When scanning forwards to check for the
2593      existencd of the subpattern, it was treating the data ']' as terminating      existence of the subpattern, it was treating the data ']' as terminating
2594      the class, so got the count wrong. When actually compiling, the reference      the class, so got the count wrong. When actually compiling, the reference
2595      was subsequently set up correctly.)      was subsequently set up correctly.)
2596    

Legend:
Removed from v.740  
changed lines
  Added in v.1574

  ViewVC Help
Powered by ViewVC 1.1.5