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

Diff of /code/trunk/ChangeLog

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

revision 1033 by ph10, Mon Sep 10 11:02:48 2012 UTC revision 1617 by ph10, Thu Dec 3 17:05:40 2015 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4  Version 8.32  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 8.39 xx-xxxxxx-201x
8    ---------------------------
9    
10    1.  If PCRE_AUTO_CALLOUT was set on a pattern that had a (?# comment between
11        an item and its qualifier (for example, A(?#comment)?B) pcre_compile()
12        misbehaved. This bug was found by the LLVM fuzzer.
13    
14    2.  Similar to the above, if an isolated \E was present between an item and its
15        qualifier when PCRE_AUTO_CALLOUT was set, pcre_compile() misbehaved. This
16        bug was found by the LLVM fuzzer.
17    
18    3.  Further to 8.38/46, negated classes such as [^[:^ascii:]\d] were also not
19        working correctly in UCP mode.
20    
21    4.  The POSIX wrapper function regexec() crashed if the option REG_STARTEND
22        was set when the pmatch argument was NULL. It now returns REG_INVARG.
23    
24    5.  Allow for up to 32-bit numbers in the ordin() function in pcregrep.
25    
26    6 . An empty \Q\E sequence between an item and its qualifier caused
27        pcre_compile() to misbehave when auto callouts were enabled. This bug was
28        found by the LLVM fuzzer.
29    
30    7 . If a pattern that was compiled with PCRE_EXTENDED started with white
31        space or a #-type comment that was followed by (?-x), which turns off
32        PCRE_EXTENDED, and there was no subsequent (?x) to turn it on again,
33        pcre_compile() assumed that (?-x) applied to the whole pattern and
34        consequently mis-compiled it. This bug was found by the LLVM fuzzer.
35    
36    
37    Version 8.38 23-November-2015
38    -----------------------------
39    
40    1.  If a group that contained a recursive back reference also contained a
41        forward reference subroutine call followed by a non-forward-reference
42        subroutine call, for example /.((?2)(?R)\1)()/, pcre2_compile() failed to
43        compile correct code, leading to undefined behaviour or an internally
44        detected error. This bug was discovered by the LLVM fuzzer.
45    
46    2.  Quantification of certain items (e.g. atomic back references) could cause
47        incorrect code to be compiled when recursive forward references were
48        involved. For example, in this pattern: /(?1)()((((((\1++))\x85)+)|))/.
49        This bug was discovered by the LLVM fuzzer.
50    
51    3.  A repeated conditional group whose condition was a reference by name caused
52        a buffer overflow if there was more than one group with the given name.
53        This bug was discovered by the LLVM fuzzer.
54    
55    4.  A recursive back reference by name within a group that had the same name as
56        another group caused a buffer overflow. For example:
57        /(?J)(?'d'(?'d'\g{d}))/. This bug was discovered by the LLVM fuzzer.
58    
59    5.  A forward reference by name to a group whose number is the same as the
60        current group, for example in this pattern: /(?|(\k'Pm')|(?'Pm'))/, caused
61        a buffer overflow at compile time. This bug was discovered by the LLVM
62        fuzzer.
63    
64    6.  A lookbehind assertion within a set of mutually recursive subpatterns could
65        provoke a buffer overflow. This bug was discovered by the LLVM fuzzer.
66    
67    7.  Another buffer overflow bug involved duplicate named groups with a
68        reference between their definition, with a group that reset capture
69        numbers, for example: /(?J:(?|(?'R')(\k'R')|((?'R'))))/. This has been
70        fixed by always allowing for more memory, even if not needed. (A proper fix
71        is implemented in PCRE2, but it involves more refactoring.)
72    
73    8.  There was no check for integer overflow in subroutine calls such as (?123).
74    
75    9.  The table entry for \l in EBCDIC environments was incorrect, leading to its
76        being treated as a literal 'l' instead of causing an error.
77    
78    10. There was a buffer overflow if pcre_exec() was called with an ovector of
79        size 1. This bug was found by american fuzzy lop.
80    
81    11. If a non-capturing group containing a conditional group that could match
82        an empty string was repeated, it was not identified as matching an empty
83        string itself. For example: /^(?:(?(1)x|)+)+$()/.
84    
85    12. In an EBCDIC environment, pcretest was mishandling the escape sequences
86        \a and \e in test subject lines.
87    
88    13. In an EBCDIC environment, \a in a pattern was converted to the ASCII
89        instead of the EBCDIC value.
90    
91    14. The handling of \c in an EBCDIC environment has been revised so that it is
92        now compatible with the specification in Perl's perlebcdic page.
93    
94    15. The EBCDIC character 0x41 is a non-breaking space, equivalent to 0xa0 in
95        ASCII/Unicode. This has now been added to the list of characters that are
96        recognized as white space in EBCDIC.
97    
98    16. When PCRE was compiled without UCP support, the use of \p and \P gave an
99        error (correctly) when used outside a class, but did not give an error
100        within a class.
101    
102    17. \h within a class was incorrectly compiled in EBCDIC environments.
103    
104    18. A pattern with an unmatched closing parenthesis that contained a backward
105        assertion which itself contained a forward reference caused buffer
106        overflow. And example pattern is: /(?=di(?<=(?1))|(?=(.))))/.
107    
108    19. JIT should return with error when the compiled pattern requires more stack
109        space than the maximum.
110    
111    20. A possessively repeated conditional group that could match an empty string,
112        for example, /(?(R))*+/, was incorrectly compiled.
113    
114    21. Fix infinite recursion in the JIT compiler when certain patterns such as
115        /(?:|a|){100}x/ are analysed.
116    
117    22. Some patterns with character classes involving [: and \\ were incorrectly
118        compiled and could cause reading from uninitialized memory or an incorrect
119        error diagnosis.
120    
121    23. Pathological patterns containing many nested occurrences of [: caused
122        pcre_compile() to run for a very long time.
123    
124    24. A conditional group with only one branch has an implicit empty alternative
125        branch and must therefore be treated as potentially matching an empty
126        string.
127    
128    25. If (?R was followed by - or + incorrect behaviour happened instead of a
129        diagnostic.
130    
131    26. Arrange to give up on finding the minimum matching length for overly
132        complex patterns.
133    
134    27. Similar to (4) above: in a pattern with duplicated named groups and an
135        occurrence of (?| it is possible for an apparently non-recursive back
136        reference to become recursive if a later named group with the relevant
137        number is encountered. This could lead to a buffer overflow. Wen Guanxing
138        from Venustech ADLAB discovered this bug.
139    
140    28. If pcregrep was given the -q option with -c or -l, or when handling a
141        binary file, it incorrectly wrote output to stdout.
142    
143    29. The JIT compiler did not restore the control verb head in case of *THEN
144        control verbs. This issue was found by Karl Skomski with a custom LLVM
145        fuzzer.
146    
147    30. Error messages for syntax errors following \g and \k were giving inaccurate
148        offsets in the pattern.
149    
150    31. Added a check for integer overflow in conditions (?(<digits>) and
151        (?(R<digits>). This omission was discovered by Karl Skomski with the LLVM
152        fuzzer.
153    
154    32. Handling recursive references such as (?2) when the reference is to a group
155        later in the pattern uses code that is very hacked about and error-prone.
156        It has been re-written for PCRE2. Here in PCRE1, a check has been added to
157        give an internal error if it is obvious that compiling has gone wrong.
158    
159    33. The JIT compiler should not check repeats after a {0,1} repeat byte code.
160        This issue was found by Karl Skomski with a custom LLVM fuzzer.
161    
162    34. The JIT compiler should restore the control chain for empty possessive
163        repeats. This issue was found by Karl Skomski with a custom LLVM fuzzer.
164    
165    35. Match limit check added to JIT recursion. This issue was found by Karl
166        Skomski with a custom LLVM fuzzer.
167    
168    36. Yet another case similar to 27 above has been circumvented by an
169        unconditional allocation of extra memory. This issue is fixed "properly" in
170        PCRE2 by refactoring the way references are handled. Wen Guanxing
171        from Venustech ADLAB discovered this bug.
172    
173    37. Fix two assertion fails in JIT. These issues were found by Karl Skomski
174        with a custom LLVM fuzzer.
175    
176    38. Fixed a corner case of range optimization in JIT.
177    
178    39. An incorrect error "overran compiling workspace" was given if there were
179        exactly enough group forward references such that the last one extended
180        into the workspace safety margin. The next one would have expanded the
181        workspace. The test for overflow was not including the safety margin.
182    
183    40. A match limit issue is fixed in JIT which was found by Karl Skomski
184        with a custom LLVM fuzzer.
185    
186    41. Remove the use of /dev/null in testdata/testinput2, because it doesn't
187        work under Windows. (Why has it taken so long for anyone to notice?)
188    
189    42. In a character class such as [\W\p{Any}] where both a negative-type escape
190        ("not a word character") and a property escape were present, the property
191        escape was being ignored.
192    
193    43. Fix crash caused by very long (*MARK) or (*THEN) names.
194    
195    44. A sequence such as [[:punct:]b] that is, a POSIX character class followed
196        by a single ASCII character in a class item, was incorrectly compiled in
197        UCP mode. The POSIX class got lost, but only if the single character
198        followed it.
199    
200    45. [:punct:] in UCP mode was matching some characters in the range 128-255
201        that should not have been matched.
202    
203    46. If [:^ascii:] or [:^xdigit:] or [:^cntrl:] are present in a non-negated
204        class, all characters with code points greater than 255 are in the class.
205        When a Unicode property was also in the class (if PCRE_UCP is set, escapes
206        such as \w are turned into Unicode properties), wide characters were not
207        correctly handled, and could fail to match.
208    
209    
210    Version 8.37 28-April-2015
211    --------------------------
212    
213    1.  When an (*ACCEPT) is triggered inside capturing parentheses, it arranges
214        for those parentheses to be closed with whatever has been captured so far.
215        However, it was failing to mark any other groups between the hightest
216        capture so far and the currrent group as "unset". Thus, the ovector for
217        those groups contained whatever was previously there. An example is the
218        pattern /(x)|((*ACCEPT))/ when matched against "abcd".
219    
220    2.  If an assertion condition was quantified with a minimum of zero (an odd
221        thing to do, but it happened), SIGSEGV or other misbehaviour could occur.
222    
223    3.  If a pattern in pcretest input had the P (POSIX) modifier followed by an
224        unrecognized modifier, a crash could occur.
225    
226    4.  An attempt to do global matching in pcretest with a zero-length ovector
227        caused a crash.
228    
229    5.  Fixed a memory leak during matching that could occur for a subpattern
230        subroutine call (recursive or otherwise) if the number of captured groups
231        that had to be saved was greater than ten.
232    
233    6.  Catch a bad opcode during auto-possessification after compiling a bad UTF
234        string with NO_UTF_CHECK. This is a tidyup, not a bug fix, as passing bad
235        UTF with NO_UTF_CHECK is documented as having an undefined outcome.
236    
237    7.  A UTF pattern containing a "not" match of a non-ASCII character and a
238        subroutine reference could loop at compile time. Example: /[^\xff]((?1))/.
239    
240    8. When a pattern is compiled, it remembers the highest back reference so that
241       when matching, if the ovector is too small, extra memory can be obtained to
242       use instead. A conditional subpattern whose condition is a check on a
243       capture having happened, such as, for example in the pattern
244       /^(?:(a)|b)(?(1)A|B)/, is another kind of back reference, but it was not
245       setting the highest backreference number. This mattered only if pcre_exec()
246       was called with an ovector that was too small to hold the capture, and there
247       was no other kind of back reference (a situation which is probably quite
248       rare). The effect of the bug was that the condition was always treated as
249       FALSE when the capture could not be consulted, leading to a incorrect
250       behaviour by pcre_exec(). This bug has been fixed.
251    
252    9. A reference to a duplicated named group (either a back reference or a test
253       for being set in a conditional) that occurred in a part of the pattern where
254       PCRE_DUPNAMES was not set caused the amount of memory needed for the pattern
255       to be incorrectly calculated, leading to overwriting.
256    
257    10. A mutually recursive set of back references such as (\2)(\1) caused a
258        segfault at study time (while trying to find the minimum matching length).
259        The infinite loop is now broken (with the minimum length unset, that is,
260        zero).
261    
262    11. If an assertion that was used as a condition was quantified with a minimum
263        of zero, matching went wrong. In particular, if the whole group had
264        unlimited repetition and could match an empty string, a segfault was
265        likely. The pattern (?(?=0)?)+ is an example that caused this. Perl allows
266        assertions to be quantified, but not if they are being used as conditions,
267        so the above pattern is faulted by Perl. PCRE has now been changed so that
268        it also rejects such patterns.
269    
270    12. A possessive capturing group such as (a)*+ with a minimum repeat of zero
271        failed to allow the zero-repeat case if pcre2_exec() was called with an
272        ovector too small to capture the group.
273    
274    13. Fixed two bugs in pcretest that were discovered by fuzzing and reported by
275        Red Hat Product Security:
276    
277        (a) A crash if /K and /F were both set with the option to save the compiled
278        pattern.
279    
280        (b) Another crash if the option to print captured substrings in a callout
281        was combined with setting a null ovector, for example \O\C+ as a subject
282        string.
283    
284    14. A pattern such as "((?2){0,1999}())?", which has a group containing a
285        forward reference repeated a large (but limited) number of times within a
286        repeated outer group that has a zero minimum quantifier, caused incorrect
287        code to be compiled, leading to the error "internal error:
288        previously-checked referenced subpattern not found" when an incorrect
289        memory address was read. This bug was reported as "heap overflow",
290        discovered by Kai Lu of Fortinet's FortiGuard Labs and given the CVE number
291        CVE-2015-2325.
292    
293    23. A pattern such as "((?+1)(\1))/" containing a forward reference subroutine
294        call within a group that also contained a recursive back reference caused
295        incorrect code to be compiled. This bug was reported as "heap overflow",
296        discovered by Kai Lu of Fortinet's FortiGuard Labs, and given the CVE
297        number CVE-2015-2326.
298    
299    24. Computing the size of the JIT read-only data in advance has been a source
300        of various issues, and new ones are still appear unfortunately. To fix
301        existing and future issues, size computation is eliminated from the code,
302        and replaced by on-demand memory allocation.
303    
304    25. A pattern such as /(?i)[A-`]/, where characters in the other case are
305        adjacent to the end of the range, and the range contained characters with
306        more than one other case, caused incorrect behaviour when compiled in UTF
307        mode. In that example, the range a-j was left out of the class.
308    
309    26. Fix JIT compilation of conditional blocks, which assertion
310        is converted to (*FAIL). E.g: /(?(?!))/.
311    
312    27. The pattern /(?(?!)^)/ caused references to random memory. This bug was
313        discovered by the LLVM fuzzer.
314    
315    28. The assertion (?!) is optimized to (*FAIL). This was not handled correctly
316        when this assertion was used as a condition, for example (?(?!)a|b). In
317        pcre2_match() it worked by luck; in pcre2_dfa_match() it gave an incorrect
318        error about an unsupported item.
319    
320    29. For some types of pattern, for example /Z*(|d*){216}/, the auto-
321        possessification code could take exponential time to complete. A recursion
322        depth limit of 1000 has been imposed to limit the resources used by this
323        optimization.
324    
325    30. A pattern such as /(*UTF)[\S\V\H]/, which contains a negated special class
326        such as \S in non-UCP mode, explicit wide characters (> 255) can be ignored
327        because \S ensures they are all in the class. The code for doing this was
328        interacting badly with the code for computing the amount of space needed to
329        compile the pattern, leading to a buffer overflow. This bug was discovered
330        by the LLVM fuzzer.
331    
332    31. A pattern such as /((?2)+)((?1))/ which has mutual recursion nested inside
333        other kinds of group caused stack overflow at compile time. This bug was
334        discovered by the LLVM fuzzer.
335    
336    32. A pattern such as /(?1)(?#?'){8}(a)/ which had a parenthesized comment
337        between a subroutine call and its quantifier was incorrectly compiled,
338        leading to buffer overflow or other errors. This bug was discovered by the
339        LLVM fuzzer.
340    
341    33. The illegal pattern /(?(?<E>.*!.*)?)/ was not being diagnosed as missing an
342        assertion after (?(. The code was failing to check the character after
343        (?(?< for the ! or = that would indicate a lookbehind assertion. This bug
344        was discovered by the LLVM fuzzer.
345    
346    34. A pattern such as /X((?2)()*+){2}+/ which has a possessive quantifier with
347        a fixed maximum following a group that contains a subroutine reference was
348        incorrectly compiled and could trigger buffer overflow. This bug was
349        discovered by the LLVM fuzzer.
350    
351    35. A mutual recursion within a lookbehind assertion such as (?<=((?2))((?1)))
352        caused a stack overflow instead of the diagnosis of a non-fixed length
353        lookbehind assertion. This bug was discovered by the LLVM fuzzer.
354    
355    36. The use of \K in a positive lookbehind assertion in a non-anchored pattern
356        (e.g. /(?<=\Ka)/) could make pcregrep loop.
357    
358    37. There was a similar problem to 36 in pcretest for global matches.
359    
360    38. If a greedy quantified \X was preceded by \C in UTF mode (e.g. \C\X*),
361        and a subsequent item in the pattern caused a non-match, backtracking over
362        the repeated \X did not stop, but carried on past the start of the subject,
363        causing reference to random memory and/or a segfault. There were also some
364        other cases where backtracking after \C could crash. This set of bugs was
365        discovered by the LLVM fuzzer.
366    
367    39. The function for finding the minimum length of a matching string could take
368        a very long time if mutual recursion was present many times in a pattern,
369        for example, /((?2){73}(?2))((?1))/. A better mutual recursion detection
370        method has been implemented. This infelicity was discovered by the LLVM
371        fuzzer.
372    
373    40. Static linking against the PCRE library using the pkg-config module was
374        failing on missing pthread symbols.
375    
376    
377    Version 8.36 26-September-2014
378    ------------------------------
379    
380    1.  Got rid of some compiler warnings in the C++ modules that were shown up by
381        -Wmissing-field-initializers and -Wunused-parameter.
382    
383    2.  The tests for quantifiers being too big (greater than 65535) were being
384        applied after reading the number, and stupidly assuming that integer
385        overflow would give a negative number. The tests are now applied as the
386        numbers are read.
387    
388    3.  Tidy code in pcre_exec.c where two branches that used to be different are
389        now the same.
390    
391    4.  The JIT compiler did not generate match limit checks for certain
392        bracketed expressions with quantifiers. This may lead to exponential
393        backtracking, instead of returning with PCRE_ERROR_MATCHLIMIT. This
394        issue should be resolved now.
395    
396    5.  Fixed an issue, which occures when nested alternatives are optimized
397        with table jumps.
398    
399    6.  Inserted two casts and changed some ints to size_t in the light of some
400        reported 64-bit compiler warnings (Bugzilla 1477).
401    
402    7.  Fixed a bug concerned with zero-minimum possessive groups that could match
403        an empty string, which sometimes were behaving incorrectly in the
404        interpreter (though correctly in the JIT matcher). This pcretest input is
405        an example:
406    
407          '\A(?:[^"]++|"(?:[^"]*+|"")*+")++'
408          NON QUOTED "QUOT""ED" AFTER "NOT MATCHED
409    
410        the interpreter was reporting a match of 'NON QUOTED ' only, whereas the
411        JIT matcher and Perl both matched 'NON QUOTED "QUOT""ED" AFTER '. The test
412        for an empty string was breaking the inner loop and carrying on at a lower
413        level, when possessive repeated groups should always return to a higher
414        level as they have no backtrack points in them. The empty string test now
415        occurs at the outer level.
416    
417    8.  Fixed a bug that was incorrectly auto-possessifying \w+ in the pattern
418        ^\w+(?>\s*)(?<=\w) which caused it not to match "test test".
419    
420    9.  Give a compile-time error for \o{} (as Perl does) and for \x{} (which Perl
421        doesn't).
422    
423    10. Change 8.34/15 introduced a bug that caused the amount of memory needed
424        to hold a pattern to be incorrectly computed (too small) when there were
425        named back references to duplicated names. This could cause "internal
426        error: code overflow" or "double free or corruption" or other memory
427        handling errors.
428    
429    11. When named subpatterns had the same prefixes, back references could be
430        confused. For example, in this pattern:
431    
432          /(?P<Name>a)?(?P<Name2>b)?(?(<Name>)c|d)*l/
433    
434        the reference to 'Name' was incorrectly treated as a reference to a
435        duplicate name.
436    
437    12. A pattern such as /^s?c/mi8 where the optional character has more than
438        one "other case" was incorrectly compiled such that it would only try to
439        match starting at "c".
440    
441    13. When a pattern starting with \s was studied, VT was not included in the
442        list of possible starting characters; this should have been part of the
443        8.34/18 patch.
444    
445    14. If a character class started [\Qx]... where x is any character, the class
446        was incorrectly terminated at the ].
447    
448    15. If a pattern that started with a caseless match for a character with more
449        than one "other case" was studied, PCRE did not set up the starting code
450        unit bit map for the list of possible characters. Now it does. This is an
451        optimization improvement, not a bug fix.
452    
453    16. The Unicode data tables have been updated to Unicode 7.0.0.
454    
455    17. Fixed a number of memory leaks in pcregrep.
456    
457    18. Avoid a compiler warning (from some compilers) for a function call with
458        a cast that removes "const" from an lvalue by using an intermediate
459        variable (to which the compiler does not object).
460    
461    19. Incorrect code was compiled if a group that contained an internal recursive
462        back reference was optional (had quantifier with a minimum of zero). This
463        example compiled incorrect code: /(((a\2)|(a*)\g<-1>))*/ and other examples
464        caused segmentation faults because of stack overflows at compile time.
465    
466    20. A pattern such as /((?(R)a|(?1)))+/, which contains a recursion within a
467        group that is quantified with an indefinite repeat, caused a compile-time
468        loop which used up all the system stack and provoked a segmentation fault.
469        This was not the same bug as 19 above.
470    
471    21. Add PCRECPP_EXP_DECL declaration to operator<< in pcre_stringpiece.h.
472        Patch by Mike Frysinger.
473    
474    
475    Version 8.35 04-April-2014
476    --------------------------
477    
478    1.  A new flag is set, when property checks are present in an XCLASS.
479        When this flag is not set, PCRE can perform certain optimizations
480        such as studying these XCLASS-es.
481    
482    2.  The auto-possessification of character sets were improved: a normal
483        and an extended character set can be compared now. Furthermore
484        the JIT compiler optimizes more character set checks.
485    
486    3.  Got rid of some compiler warnings for potentially uninitialized variables
487        that show up only when compiled with -O2.
488    
489    4.  A pattern such as (?=ab\K) that uses \K in an assertion can set the start
490        of a match later then the end of the match. The pcretest program was not
491        handling the case sensibly - it was outputting from the start to the next
492        binary zero. It now reports this situation in a message, and outputs the
493        text from the end to the start.
494    
495    5.  Fast forward search is improved in JIT. Instead of the first three
496        characters, any three characters with fixed position can be searched.
497        Search order: first, last, middle.
498    
499    6.  Improve character range checks in JIT. Characters are read by an inprecise
500        function now, which returns with an unknown value if the character code is
501        above a certain threshold (e.g: 256). The only limitation is that the value
502        must be bigger than the threshold as well. This function is useful when
503        the characters above the threshold are handled in the same way.
504    
505    7.  The macros whose names start with RAWUCHAR are placeholders for a future
506        mode in which only the bottom 21 bits of 32-bit data items are used. To
507        make this more memorable for those maintaining the code, the names have
508        been changed to start with UCHAR21, and an extensive comment has been added
509        to their definition.
510    
511    8.  Add missing (new) files sljitNativeTILEGX.c and sljitNativeTILEGX-encoder.c
512        to the export list in Makefile.am (they were accidentally omitted from the
513        8.34 tarball).
514    
515    9.  The informational output from pcretest used the phrase "starting byte set"
516        which is inappropriate for the 16-bit and 32-bit libraries. As the output
517        for "first char" and "need char" really means "non-UTF-char", I've changed
518        "byte" to "char", and slightly reworded the output. The documentation about
519        these values has also been (I hope) clarified.
520    
521    10. Another JIT related optimization: use table jumps for selecting the correct
522        backtracking path, when more than four alternatives are present inside a
523        bracket.
524    
525    11. Empty match is not possible, when the minimum length is greater than zero,
526        and there is no \K in the pattern. JIT should avoid empty match checks in
527        such cases.
528    
529    12. In a caseless character class with UCP support, when a character with more
530        than one alternative case was not the first character of a range, not all
531        the alternative cases were added to the class. For example, s and \x{17f}
532        are both alternative cases for S: the class [RST] was handled correctly,
533        but [R-T] was not.
534    
535    13. The configure.ac file always checked for pthread support when JIT was
536        enabled. This is not used in Windows, so I have put this test inside a
537        check for the presence of windows.h (which was already tested for).
538    
539    14. Improve pattern prefix search by a simplified Boyer-Moore algorithm in JIT.
540        The algorithm provides a way to skip certain starting offsets, and usually
541        faster than linear prefix searches.
542    
543    15. Change 13 for 8.20 updated RunTest to check for the 'fr' locale as well
544        as for 'fr_FR' and 'french'. For some reason, however, it then used the
545        Windows-specific input and output files, which have 'french' screwed in.
546        So this could never have worked. One of the problems with locales is that
547        they aren't always the same. I have now updated RunTest so that it checks
548        the output of the locale test (test 3) against three different output
549        files, and it allows the test to pass if any one of them matches. With luck
550        this should make the test pass on some versions of Solaris where it was
551        failing. Because of the uncertainty, the script did not used to stop if
552        test 3 failed; it now does. If further versions of a French locale ever
553        come to light, they can now easily be added.
554    
555    16. If --with-pcregrep-bufsize was given a non-integer value such as "50K",
556        there was a message during ./configure, but it did not stop. This now
557        provokes an error. The invalid example in README has been corrected.
558        If a value less than the minimum is given, the minimum value has always
559        been used, but now a warning is given.
560    
561    17. If --enable-bsr-anycrlf was set, the special 16/32-bit test failed. This
562        was a bug in the test system, which is now fixed. Also, the list of various
563        configurations that are tested for each release did not have one with both
564        16/32 bits and --enable-bar-anycrlf. It now does.
565    
566    18. pcretest was missing "-C bsr" for displaying the \R default setting.
567    
568    19. Little endian PowerPC systems are supported now by the JIT compiler.
569    
570    20. The fast forward newline mechanism could enter to an infinite loop on
571        certain invalid UTF-8 input. Although we don't support these cases
572        this issue can be fixed by a performance optimization.
573    
574    21. Change 33 of 8.34 is not sufficient to ensure stack safety because it does
575        not take account if existing stack usage. There is now a new global
576        variable called pcre_stack_guard that can be set to point to an external
577        function to check stack availability. It is called at the start of
578        processing every parenthesized group.
579    
580    22. A typo in the code meant that in ungreedy mode the max/min qualifier
581        behaved like a min-possessive qualifier, and, for example, /a{1,3}b/U did
582        not match "ab".
583    
584    23. When UTF was disabled, the JIT program reported some incorrect compile
585        errors. These messages are silenced now.
586    
587    24. Experimental support for ARM-64 and MIPS-64 has been added to the JIT
588        compiler.
589    
590    25. Change all the temporary files used in RunGrepTest to be different to those
591        used by RunTest so that the tests can be run simultaneously, for example by
592        "make -j check".
593    
594    
595    Version 8.34 15-December-2013
596    -----------------------------
597    
598    1.  Add pcre[16|32]_jit_free_unused_memory to forcibly free unused JIT
599        executable memory. Patch inspired by Carsten Klein.
600    
601    2.  ./configure --enable-coverage defined SUPPORT_GCOV in config.h, although
602        this macro is never tested and has no effect, because the work to support
603        coverage involves only compiling and linking options and special targets in
604        the Makefile. The comment in config.h implied that defining the macro would
605        enable coverage support, which is totally false. There was also support for
606        setting this macro in the CMake files (my fault, I just copied it from
607        configure). SUPPORT_GCOV has now been removed.
608    
609    3.  Make a small performance improvement in strlen16() and strlen32() in
610        pcretest.
611    
612    4.  Change 36 for 8.33 left some unreachable statements in pcre_exec.c,
613        detected by the Solaris compiler (gcc doesn't seem to be able to diagnose
614        these cases). There was also one in pcretest.c.
615    
616    5.  Cleaned up a "may be uninitialized" compiler warning in pcre_exec.c.
617    
618    6.  In UTF mode, the code for checking whether a group could match an empty
619        string (which is used for indefinitely repeated groups to allow for
620        breaking an infinite loop) was broken when the group contained a repeated
621        negated single-character class with a character that occupied more than one
622        data item and had a minimum repetition of zero (for example, [^\x{100}]* in
623        UTF-8 mode). The effect was undefined: the group might or might not be
624        deemed as matching an empty string, or the program might have crashed.
625    
626    7.  The code for checking whether a group could match an empty string was not
627        recognizing that \h, \H, \v, \V, and \R must match a character.
628    
629    8.  Implemented PCRE_INFO_MATCH_EMPTY, which yields 1 if the pattern can match
630        an empty string. If it can, pcretest shows this in its information output.
631    
632    9.  Fixed two related bugs that applied to Unicode extended grapheme clusters
633        that were repeated with a maximizing qualifier (e.g. \X* or \X{2,5}) when
634        matched by pcre_exec() without using JIT:
635    
636        (a) If the rest of the pattern did not match after a maximal run of
637            grapheme clusters, the code for backing up to try with fewer of them
638            did not always back up over a full grapheme when characters that do not
639            have the modifier quality were involved, e.g. Hangul syllables.
640    
641        (b) If the match point in a subject started with modifier character, and
642            there was no match, the code could incorrectly back up beyond the match
643            point, and potentially beyond the first character in the subject,
644            leading to a segfault or an incorrect match result.
645    
646    10. A conditional group with an assertion condition could lead to PCRE
647        recording an incorrect first data item for a match if no other first data
648        item was recorded. For example, the pattern (?(?=ab)ab) recorded "a" as a
649        first data item, and therefore matched "ca" after "c" instead of at the
650        start.
651    
652    11. Change 40 for 8.33 (allowing pcregrep to find empty strings) showed up a
653        bug that caused the command "echo a | ./pcregrep -M '|a'" to loop.
654    
655    12. The source of pcregrep now includes z/OS-specific code so that it can be
656        compiled for z/OS as part of the special z/OS distribution.
657    
658    13. Added the -T and -TM options to pcretest.
659    
660    14. The code in pcre_compile.c for creating the table of named capturing groups
661        has been refactored. Instead of creating the table dynamically during the
662        actual compiling pass, the information is remembered during the pre-compile
663        pass (on the stack unless there are more than 20 named groups, in which
664        case malloc() is used) and the whole table is created before the actual
665        compile happens. This has simplified the code (it is now nearly 150 lines
666        shorter) and prepared the way for better handling of references to groups
667        with duplicate names.
668    
669    15. A back reference to a named subpattern when there is more than one of the
670        same name now checks them in the order in which they appear in the pattern.
671        The first one that is set is used for the reference. Previously only the
672        first one was inspected. This change makes PCRE more compatible with Perl.
673    
674    16. Unicode character properties were updated from Unicode 6.3.0.
675    
676    17. The compile-time code for auto-possessification has been refactored, based
677        on a patch by Zoltan Herczeg. It now happens after instead of during
678        compilation. The code is cleaner, and more cases are handled. The option
679        PCRE_NO_AUTO_POSSESS is added for testing purposes, and the -O and /O
680        options in pcretest are provided to set it. It can also be set by
681        (*NO_AUTO_POSSESS) at the start of a pattern.
682    
683    18. The character VT has been added to the default ("C" locale) set of
684        characters that match \s and are generally treated as white space,
685        following this same change in Perl 5.18. There is now no difference between
686        "Perl space" and "POSIX space". Whether VT is treated as white space in
687        other locales depends on the locale.
688    
689    19. The code for checking named groups as conditions, either for being set or
690        for being recursed, has been refactored (this is related to 14 and 15
691        above). Processing unduplicated named groups should now be as fast at
692        numerical groups, and processing duplicated groups should be faster than
693        before.
694    
695    20. Two patches to the CMake build system, by Alexander Barkov:
696    
697          (1) Replace the "source" command by "." in CMakeLists.txt because
698              "source" is a bash-ism.
699    
700          (2) Add missing HAVE_STDINT_H and HAVE_INTTYPES_H to config-cmake.h.in;
701              without these the CMake build does not work on Solaris.
702    
703    21. Perl has changed its handling of \8 and \9. If there is no previously
704        encountered capturing group of those numbers, they are treated as the
705        literal characters 8 and 9 instead of a binary zero followed by the
706        literals. PCRE now does the same.
707    
708    22. Following Perl, added \o{} to specify codepoints in octal, making it
709        possible to specify values greater than 0777 and also making them
710        unambiguous.
711    
712    23. Perl now gives an error for missing closing braces after \x{... instead of
713        treating the string as literal. PCRE now does the same.
714    
715    24. RunTest used to grumble if an inappropriate test was selected explicitly,
716        but just skip it when running all tests. This make it awkward to run ranges
717        of tests when one of them was inappropriate. Now it just skips any
718        inappropriate tests, as it always did when running all tests.
719    
720    25. If PCRE_AUTO_CALLOUT and PCRE_UCP were set for a pattern that contained
721        character types such as \d or \w, too many callouts were inserted, and the
722        data that they returned was rubbish.
723    
724    26. In UCP mode, \s was not matching two of the characters that Perl matches,
725        namely NEL (U+0085) and MONGOLIAN VOWEL SEPARATOR (U+180E), though they
726        were matched by \h. The code has now been refactored so that the lists of
727        the horizontal and vertical whitespace characters used for \h and \v (which
728        are defined only in one place) are now also used for \s.
729    
730    27. Add JIT support for the 64 bit TileGX architecture.
731        Patch by Jiong Wang (Tilera Corporation).
732    
733    28. Possessive quantifiers for classes (both explicit and automatically
734        generated) now use special opcodes instead of wrapping in ONCE brackets.
735    
736    29. Whereas an item such as A{4}+ ignored the possessivenes of the quantifier
737        (because it's meaningless), this was not happening when PCRE_CASELESS was
738        set. Not wrong, but inefficient.
739    
740    30. Updated perltest.pl to add /u (force Unicode mode) when /W (use Unicode
741        properties for \w, \d, etc) is present in a test regex. Otherwise if the
742        test contains no characters greater than 255, Perl doesn't realise it
743        should be using Unicode semantics.
744    
745    31. Upgraded the handling of the POSIX classes [:graph:], [:print:], and
746        [:punct:] when PCRE_UCP is set so as to include the same characters as Perl
747        does in Unicode mode.
748    
749    32. Added the "forbid" facility to pcretest so that putting tests into the
750        wrong test files can sometimes be quickly detected.
751    
752    33. There is now a limit (default 250) on the depth of nesting of parentheses.
753        This limit is imposed to control the amount of system stack used at compile
754        time. It can be changed at build time by --with-parens-nest-limit=xxx or
755        the equivalent in CMake.
756    
757    34. Character classes such as [A-\d] or [a-[:digit:]] now cause compile-time
758        errors. Perl warns for these when in warning mode, but PCRE has no facility
759        for giving warnings.
760    
761    35. Change 34 for 8.13 allowed quantifiers on assertions, because Perl does.
762        However, this was not working for (?!) because it is optimized to (*FAIL),
763        for which PCRE does not allow quantifiers. The optimization is now disabled
764        when a quantifier follows (?!). I can't see any use for this, but it makes
765        things uniform.
766    
767    36. Perl no longer allows group names to start with digits, so I have made this
768        change also in PCRE. It simplifies the code a bit.
769    
770    37. In extended mode, Perl ignores spaces before a + that indicates a
771        possessive quantifier. PCRE allowed a space before the quantifier, but not
772        before the possessive +. It now does.
773    
774    38. The use of \K (reset reported match start) within a repeated possessive
775        group such as (a\Kb)*+ was not working.
776    
777    40. Document that the same character tables must be used at compile time and
778        run time, and that the facility to pass tables to pcre_exec() and
779        pcre_dfa_exec() is for use only with saved/restored patterns.
780    
781    41. Applied Jeff Trawick's patch CMakeLists.txt, which "provides two new
782        features for Builds with MSVC:
783    
784        1. Support pcre.rc and/or pcreposix.rc (as is already done for MinGW
785           builds). The .rc files can be used to set FileDescription and many other
786           attributes.
787    
788        2. Add an option (-DINSTALL_MSVC_PDB) to enable installation of .pdb files.
789           This allows higher-level build scripts which want .pdb files to avoid
790           hard-coding the exact files needed."
791    
792    42. Added support for [[:<:]] and [[:>:]] as used in the BSD POSIX library to
793        mean "start of word" and "end of word", respectively, as a transition aid.
794    
795    43. A minimizing repeat of a class containing codepoints greater than 255 in
796        non-UTF 16-bit or 32-bit modes caused an internal error when PCRE was
797        compiled to use the heap for recursion.
798    
799    44. Got rid of some compiler warnings for unused variables when UTF but not UCP
800        is configured.
801    
802    
803    Version 8.33 28-May-2013
804    ------------------------
805    
806    1.  Added 'U' to some constants that are compared to unsigned integers, to
807        avoid compiler signed/unsigned warnings. Added (int) casts to unsigned
808        variables that are added to signed variables, to ensure the result is
809        signed and can be negated.
810    
811    2.  Applied patch by Daniel Richard G for quashing MSVC warnings to the
812        CMake config files.
813    
814    3.  Revise the creation of config.h.generic so that all boolean macros are
815        #undefined, whereas non-boolean macros are #ifndef/#endif-ed. This makes
816        overriding via -D on the command line possible.
817    
818    4.  Changing the definition of the variable "op" in pcre_exec.c from pcre_uchar
819        to unsigned int is reported to make a quite noticeable speed difference in
820        a specific Windows environment. Testing on Linux did also appear to show
821        some benefit (and it is clearly not harmful). Also fixed the definition of
822        Xop which should be unsigned.
823    
824    5.  Related to (4), changing the definition of the intermediate variable cc
825        in repeated character loops from pcre_uchar to pcre_uint32 also gave speed
826        improvements.
827    
828    6.  Fix forward search in JIT when link size is 3 or greater. Also removed some
829        unnecessary spaces.
830    
831    7.  Adjust autogen.sh and configure.ac to lose warnings given by automake 1.12
832        and later.
833    
834    8.  Fix two buffer over read issues in 16 and 32 bit modes. Affects JIT only.
835    
836    9.  Optimizing fast_forward_start_bits in JIT.
837    
838    10. Adding support for callouts in JIT, and fixing some issues revealed
839        during this work. Namely:
840    
841        (a) Unoptimized capturing brackets incorrectly reset on backtrack.
842    
843        (b) Minimum length was not checked before the matching is started.
844    
845    11. The value of capture_last that is passed to callouts was incorrect in some
846        cases when there was a capture on one path that was subsequently abandoned
847        after a backtrack. Also, the capture_last value is now reset after a
848        recursion, since all captures are also reset in this case.
849    
850    12. The interpreter no longer returns the "too many substrings" error in the
851        case when an overflowing capture is in a branch that is subsequently
852        abandoned after a backtrack.
853    
854    13. In the pathological case when an offset vector of size 2 is used, pcretest
855        now prints out the matched string after a yield of 0 or 1.
856    
857    14. Inlining subpatterns in recursions, when certain conditions are fulfilled.
858        Only supported by the JIT compiler at the moment.
859    
860    15. JIT compiler now supports 32 bit Macs thanks to Lawrence Velazquez.
861    
862    16. Partial matches now set offsets[2] to the "bumpalong" value, that is, the
863        offset of the starting point of the matching process, provided the offsets
864        vector is large enough.
865    
866    17. The \A escape now records a lookbehind value of 1, though its execution
867        does not actually inspect the previous character. This is to ensure that,
868        in partial multi-segment matching, at least one character from the old
869        segment is retained when a new segment is processed. Otherwise, if there
870        are no lookbehinds in the pattern, \A might match incorrectly at the start
871        of a new segment.
872    
873    18. Added some #ifdef __VMS code into pcretest.c to help VMS implementations.
874    
875    19. Redefined some pcre_uchar variables in pcre_exec.c as pcre_uint32; this
876        gives some modest performance improvement in 8-bit mode.
877    
878    20. Added the PCRE-specific property \p{Xuc} for matching characters that can
879        be expressed in certain programming languages using Universal Character
880        Names.
881    
882    21. Unicode validation has been updated in the light of Unicode Corrigendum #9,
883        which points out that "non characters" are not "characters that may not
884        appear in Unicode strings" but rather "characters that are reserved for
885        internal use and have only local meaning".
886    
887    22. When a pattern was compiled with automatic callouts (PCRE_AUTO_CALLOUT) and
888        there was a conditional group that depended on an assertion, if the
889        assertion was false, the callout that immediately followed the alternation
890        in the condition was skipped when pcre_exec() was used for matching.
891    
892    23. Allow an explicit callout to be inserted before an assertion that is the
893        condition for a conditional group, for compatibility with automatic
894        callouts, which always insert a callout at this point.
895    
896    24. In 8.31, (*COMMIT) was confined to within a recursive subpattern. Perl also
897        confines (*SKIP) and (*PRUNE) in the same way, and this has now been done.
898    
899    25. (*PRUNE) is now supported by the JIT compiler.
900    
901    26. Fix infinite loop when /(?<=(*SKIP)ac)a/ is matched against aa.
902    
903    27. Fix the case where there are two or more SKIPs with arguments that may be
904        ignored.
905    
906    28. (*SKIP) is now supported by the JIT compiler.
907    
908    29. (*THEN) is now supported by the JIT compiler.
909    
910    30. Update RunTest with additional test selector options.
911    
912    31. The way PCRE handles backtracking verbs has been changed in two ways.
913    
914        (1) Previously, in something like (*COMMIT)(*SKIP), COMMIT would override
915        SKIP. Now, PCRE acts on whichever backtracking verb is reached first by
916        backtracking. In some cases this makes it more Perl-compatible, but Perl's
917        rather obscure rules do not always do the same thing.
918    
919        (2) Previously, backtracking verbs were confined within assertions. This is
920        no longer the case for positive assertions, except for (*ACCEPT). Again,
921        this sometimes improves Perl compatibility, and sometimes does not.
922    
923    32. A number of tests that were in test 2 because Perl did things differently
924        have been moved to test 1, because either Perl or PCRE has changed, and
925        these tests are now compatible.
926    
927    32. Backtracking control verbs are now handled in the same way in JIT and
928        interpreter.
929    
930    33. An opening parenthesis in a MARK/PRUNE/SKIP/THEN name in a pattern that
931        contained a forward subroutine reference caused a compile error.
932    
933    34. Auto-detect and optimize limited repetitions in JIT.
934    
935    35. Implement PCRE_NEVER_UTF to lock out the use of UTF, in particular,
936        blocking (*UTF) etc.
937    
938    36. In the interpreter, maximizing pattern repetitions for characters and
939        character types now use tail recursion, which reduces stack usage.
940    
941    37. The value of the max lookbehind was not correctly preserved if a compiled
942        and saved regex was reloaded on a host of different endianness.
943    
944    38. Implemented (*LIMIT_MATCH) and (*LIMIT_RECURSION). As part of the extension
945        of the compiled pattern block, expand the flags field from 16 to 32 bits
946        because it was almost full.
947    
948    39. Try madvise first before posix_madvise.
949    
950    40. Change 7 for PCRE 7.9 made it impossible for pcregrep to find empty lines
951        with a pattern such as ^$. It has taken 4 years for anybody to notice! The
952        original change locked out all matches of empty strings. This has been
953        changed so that one match of an empty string per line is recognized.
954        Subsequent searches on the same line (for colouring or for --only-matching,
955        for example) do not recognize empty strings.
956    
957    41. Applied a user patch to fix a number of spelling mistakes in comments.
958    
959    42. Data lines longer than 65536 caused pcretest to crash.
960    
961    43. Clarified the data type for length and startoffset arguments for pcre_exec
962        and pcre_dfa_exec in the function-specific man pages, where they were
963        explicitly stated to be in bytes, never having been updated. I also added
964        some clarification to the pcreapi man page.
965    
966    44. A call to pcre_dfa_exec() with an output vector size less than 2 caused
967        a segmentation fault.
968    
969    
970    Version 8.32 30-November-2012
971    -----------------------------
972    
973  1.  Improved JIT compiler optimizations for first character search and single  1.  Improved JIT compiler optimizations for first character search and single
974      character iterators.      character iterators.
# Line 16  Version 8.32 Line 982  Version 8.32
982    
983  5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.  5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
984      Reported by Giuseppe D'Angelo.      Reported by Giuseppe D'Angelo.
985    
986  6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start  6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
987      of a line, was being set incorrectly in cases where .* appeared inside      of a line, was being set incorrectly in cases where .* appeared inside
988      atomic brackets at the start of a pattern, or where there was a subsequent      atomic brackets at the start of a pattern, or where there was a subsequent
989      *PRUNE or *SKIP.      *PRUNE or *SKIP.
990    
991  7.  Improved instruction cache flush for POWER/PowerPC.  7.  Improved instruction cache flush for POWER/PowerPC.
992      Patch by Daniel Richard G.      Patch by Daniel Richard G.
993    
994  8.  Fixed a number of issues in pcregrep, making it more compatible with GNU  8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
995      grep:      grep:
996    
997      (a) There is now no limit to the number of patterns to be matched.      (a) There is now no limit to the number of patterns to be matched.
998    
999      (b) An error is given if a pattern is too long.      (b) An error is given if a pattern is too long.
1000    
1001      (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir      (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
1002          are now supported.          are now supported.
1003    
1004      (d) --exclude-from and --include-from (multiple use) have been added.      (d) --exclude-from and --include-from (multiple use) have been added.
1005    
1006      (e) Exclusions and inclusions now apply to all files and directories, not      (e) Exclusions and inclusions now apply to all files and directories, not
1007          just to those obtained from scanning a directory recursively.          just to those obtained from scanning a directory recursively.
1008    
1009      (f) Multiple uses of -f and --file-list are now supported.      (f) Multiple uses of -f and --file-list are now supported.
1010    
1011      (g) In a Windows environment, the default for -d has been changed from      (g) In a Windows environment, the default for -d has been changed from
1012          "read" (the GNU grep default) to "skip", because otherwise the presence          "read" (the GNU grep default) to "skip", because otherwise the presence
1013          of a directory in the file list provokes an error.          of a directory in the file list provokes an error.
1014    
1015      (h) The documentation has been revised and clarified in places.      (h) The documentation has been revised and clarified in places.
1016    
1017  9.  Improve the matching speed of capturing brackets.  9.  Improve the matching speed of capturing brackets.
1018    
1019  10. Changed the meaning of \X so that it now matches a Unicode extended  10. Changed the meaning of \X so that it now matches a Unicode extended
1020      grapheme cluster.      grapheme cluster.
1021    
1022  11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting  11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
1023      out POSIX threads when JIT support is configured.      out POSIX threads when JIT support is configured.
   
 12. Added support for PCRE_STUDY_EXTRA_NEEDED.  
1024    
1025  13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg  12. Added support for PCRE_STUDY_EXTRA_NEEDED.
1026      structure could go wrong in environments where size_t is not the same size  
1027    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
1028        structure could go wrong in environments where size_t is not the same size
1029      as int.      as int.
1030    
1031  14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be  14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
1032      set.      set.
1033    
1034  15. The EBCDIC support had decayed; later updates to the code had included  15. The EBCDIC support had decayed; later updates to the code had included
1035      explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a      explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
1036      general tidy up of EBCDIC-related issues, and the documentation was also      general tidy up of EBCDIC-related issues, and the documentation was also
1037      not quite right. There is now a test that can be run on ASCII systems to      not quite right. There is now a test that can be run on ASCII systems to
1038      check some of the EBCDIC-related things (but is it not a full test).      check some of the EBCDIC-related things (but is it not a full test).
1039    
1040    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
1041        in a small tidy to the code.
1042    
1043    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
1044    
1045    18. If the --only-matching (-o) option in pcregrep is specified multiple
1046        times, each one causes appropriate output. For example, -o1 -o2 outputs the
1047        substrings matched by the 1st and 2nd capturing parentheses. A separating
1048        string can be specified by --om-separator (default empty).
1049    
1050    19. Improving the first n character searches.
1051    
1052    20. Turn case lists for horizontal and vertical white space into macros so that
1053        they are defined only once.
1054    
1055    21. This set of changes together give more compatible Unicode case-folding
1056        behaviour for characters that have more than one other case when UCP
1057        support is available.
1058    
1059        (a) The Unicode property table now has offsets into a new table of sets of
1060            three or more characters that are case-equivalent. The MultiStage2.py
1061            script that generates these tables (the pcre_ucd.c file) now scans
1062            CaseFolding.txt instead of UnicodeData.txt for character case
1063            information.
1064    
1065        (b) The code for adding characters or ranges of characters to a character
1066            class has been abstracted into a generalized function that also handles
1067            case-independence. In UTF-mode with UCP support, this uses the new data
1068            to handle characters with more than one other case.
1069    
1070        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
1071            whose other case is greater than 256 are now correctly matched
1072            caselessly. Previously, the high codepoint matched the low one, but not
1073            vice versa.
1074    
1075        (d) The processing of \h, \H, \v, and \ in character classes now makes use
1076            of the new class addition function, using character lists defined as
1077            macros alongside the case definitions of 20 above.
1078    
1079        (e) Caseless back references now work with characters that have more than
1080            one other case.
1081    
1082        (f) General caseless matching of characters with more than one other case
1083            is supported.
1084    
1085    22. Unicode character properties were updated from Unicode 6.2.0
1086    
1087    23. Improved CMake support under Windows. Patch by Daniel Richard G.
1088    
1089    24. Add support for 32-bit character strings, and UTF-32
1090    
1091    25. Major JIT compiler update (code refactoring and bugfixing).
1092        Experimental Sparc 32 support is added.
1093    
1094    26. Applied a modified version of Daniel Richard G's patch to create
1095        pcre.h.generic and config.h.generic by "make" instead of in the
1096        PrepareRelease script.
1097    
1098    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
1099        pcre_compile.c when checking for a zero character.
1100    
1101    28. Introducing a native interface for JIT. Through this interface, the compiled
1102        machine code can be directly executed. The purpose of this interface is to
1103        provide fast pattern matching, so several sanity checks are not performed.
1104        However, feature tests are still performed. The new interface provides
1105        1.4x speedup compared to the old one.
1106    
1107    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
1108        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
1109        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
1110    
1111    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
1112        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
1113        the "old" RFC 2279). Instead, it ended up passing a negative length to
1114        pcre_exec().
1115    
1116    31. Add support for GCC's visibility feature to hide internal functions.
1117    
1118    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
1119        "unknown -C option" after outputting 0 or 1.
1120    
1121    33. There is now support for generating a code coverage report for the test
1122        suite in environments where gcc is the compiler and lcov is installed. This
1123        is mainly for the benefit of the developers.
1124    
1125    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
1126        unaddressable using valgrind annotations, allowing valgrind to detect
1127        invalid memory accesses. This is mainly for the benefit of the developers.
1128    
1129    25. (*UTF) can now be used to start a pattern in any of the three libraries.
1130    
1131    26. Give configure error if --enable-cpp but no C++ compiler found.
1132    
1133    
1134  Version 8.31 06-July-2012  Version 8.31 06-July-2012
# Line 1416  Version 7.9 11-Apr-09 Line 2474  Version 7.9 11-Apr-09
2474  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
2475      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
2476      locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this      locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
2477      seems to be how GNU grep behaves.      seems to be how GNU grep behaves. [But see later change 40 for release
2478        8.33.]
2479    
2480  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
2481      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 1659  Version 7.7 07-May-08 Line 2718  Version 7.7 07-May-08
2718      containing () gave an internal compiling error instead of "reference to      containing () gave an internal compiling error instead of "reference to
2719      non-existent subpattern". Fortunately, when the pattern did exist, the      non-existent subpattern". Fortunately, when the pattern did exist, the
2720      compiled code was correct. (When scanning forwards to check for the      compiled code was correct. (When scanning forwards to check for the
2721      existencd of the subpattern, it was treating the data ']' as terminating      existence of the subpattern, it was treating the data ']' as terminating
2722      the class, so got the count wrong. When actually compiling, the reference      the class, so got the count wrong. When actually compiling, the reference
2723      was subsequently set up correctly.)      was subsequently set up correctly.)
2724    

Legend:
Removed from v.1033  
changed lines
  Added in v.1617

  ViewVC Help
Powered by ViewVC 1.1.5