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

Diff of /code/trunk/ChangeLog

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

revision 538 by ph10, Wed Jun 9 19:30:57 2010 UTC revision 1597 by ph10, Sat Aug 29 15:23:48 2015 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4  Version 8.10 03-Jun-2010  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    20. A possessively repeated conditional group that could match an empty string,
82        for example, /(?(R))*+/, was incorrectly compiled.
83    
84    21. Fix infinite recursion in the JIT compiler when certain patterns such as
85        /(?:|a|){100}x/ are analysed.
86    
87    22. Some patterns with character classes involving [: and \\ were incorrectly
88        compiled and could cause reading from uninitialized memory or an incorrect
89        error diagnosis.
90    
91    23. Pathological patterns containing many nested occurrences of [: caused
92        pcre_compile() to run for a very long time.
93    
94    24. A conditional group with only one branch has an implicit empty alternative
95        branch and must therefore be treated as potentially matching an empty
96        string.
97    
98    25. If (?R was followed by - or + incorrect behaviour happened instead of a
99        diagnostic.
100    
101    26. Arrange to give up on finding the minimum matching length for overly
102        complex patterns.
103    
104    27. Similar to (4) above: in a pattern with duplicated named groups and an
105        occurrence of (?| it is possible for an apparently non-recursive back
106        reference to become recursive if a later named group with the relevant
107        number is encountered. This could lead to a buffer overflow. Wen Guanxing
108        from Venustech ADLAB discovered this bug.
109    
110    28. If pcregrep was given the -q option with -c or -l, or when handling a
111        binary file, it incorrectly wrote output to stdout.
112    
113    29. The JIT compiler did not restore the control verb head in case of *THEN
114        control verbs. This issue was found by Karl Skomski with a custom LLVM
115        fuzzer.
116    
117    30. Error messages for syntax errors following \g and \k were giving inaccurate
118        offsets in the pattern.
119    
120    31. Added a check for integer overflow in conditions (?(<digits>) and
121        (?(R<digits>). This omission was discovered by Karl Skomski with the LLVM
122        fuzzer.
123    
124    32. Handling recursive references such as (?2) when the reference is to a group
125        later in the pattern uses code that is very hacked about and error-prone.
126        It has been re-written for PCRE2. Here in PCRE1, a check has been added to
127        give an internal error if it is obvious that compiling has gone wrong.
128    
129    33. The JIT compiler should not check repeats after a {0,1} repeat byte code.
130        This issue was found by Karl Skomski with a custom LLVM fuzzer.
131    
132    34. The JIT compiler should restore the control chain for empty possessive
133        repeats. This issue was found by Karl Skomski with a custom LLVM fuzzer.
134    
135    35. Match limit check added to JIT recursion. This issue was found by Karl
136        Skomski with a custom LLVM fuzzer.
137    
138    36. Yet another case similar to 27 above has been circumvented by an
139        unconditional allocation of extra memory. This issue is fixed "properly" in
140        PCRE2 by refactoring the way references are handled. Wen Guanxing
141        from Venustech ADLAB discovered this bug.
142    
143    37. Fix two assertion fails in JIT. These issues were found by Karl Skomski
144        with a custom LLVM fuzzer.
145    
146    38. Fixed a corner case of range optimization in JIT.
147    
148    39. An incorrect error "overran compiling workspace" was given if there were
149        exactly enough group forward references such that the last one extended
150        into the workspace safety margin. The next one would have expanded the
151        workspace. The test for overflow was not including the safety margin.
152    
153    
154    Version 8.37 28-April-2015
155    --------------------------
156    
157    1.  When an (*ACCEPT) is triggered inside capturing parentheses, it arranges
158        for those parentheses to be closed with whatever has been captured so far.
159        However, it was failing to mark any other groups between the hightest
160        capture so far and the currrent group as "unset". Thus, the ovector for
161        those groups contained whatever was previously there. An example is the
162        pattern /(x)|((*ACCEPT))/ when matched against "abcd".
163    
164    2.  If an assertion condition was quantified with a minimum of zero (an odd
165        thing to do, but it happened), SIGSEGV or other misbehaviour could occur.
166    
167    3.  If a pattern in pcretest input had the P (POSIX) modifier followed by an
168        unrecognized modifier, a crash could occur.
169    
170    4.  An attempt to do global matching in pcretest with a zero-length ovector
171        caused a crash.
172    
173    5.  Fixed a memory leak during matching that could occur for a subpattern
174        subroutine call (recursive or otherwise) if the number of captured groups
175        that had to be saved was greater than ten.
176    
177    6.  Catch a bad opcode during auto-possessification after compiling a bad UTF
178        string with NO_UTF_CHECK. This is a tidyup, not a bug fix, as passing bad
179        UTF with NO_UTF_CHECK is documented as having an undefined outcome.
180    
181    7.  A UTF pattern containing a "not" match of a non-ASCII character and a
182        subroutine reference could loop at compile time. Example: /[^\xff]((?1))/.
183    
184    8. When a pattern is compiled, it remembers the highest back reference so that
185       when matching, if the ovector is too small, extra memory can be obtained to
186       use instead. A conditional subpattern whose condition is a check on a
187       capture having happened, such as, for example in the pattern
188       /^(?:(a)|b)(?(1)A|B)/, is another kind of back reference, but it was not
189       setting the highest backreference number. This mattered only if pcre_exec()
190       was called with an ovector that was too small to hold the capture, and there
191       was no other kind of back reference (a situation which is probably quite
192       rare). The effect of the bug was that the condition was always treated as
193       FALSE when the capture could not be consulted, leading to a incorrect
194       behaviour by pcre_exec(). This bug has been fixed.
195    
196    9. A reference to a duplicated named group (either a back reference or a test
197       for being set in a conditional) that occurred in a part of the pattern where
198       PCRE_DUPNAMES was not set caused the amount of memory needed for the pattern
199       to be incorrectly calculated, leading to overwriting.
200    
201    10. A mutually recursive set of back references such as (\2)(\1) caused a
202        segfault at study time (while trying to find the minimum matching length).
203        The infinite loop is now broken (with the minimum length unset, that is,
204        zero).
205    
206    11. If an assertion that was used as a condition was quantified with a minimum
207        of zero, matching went wrong. In particular, if the whole group had
208        unlimited repetition and could match an empty string, a segfault was
209        likely. The pattern (?(?=0)?)+ is an example that caused this. Perl allows
210        assertions to be quantified, but not if they are being used as conditions,
211        so the above pattern is faulted by Perl. PCRE has now been changed so that
212        it also rejects such patterns.
213    
214    12. A possessive capturing group such as (a)*+ with a minimum repeat of zero
215        failed to allow the zero-repeat case if pcre2_exec() was called with an
216        ovector too small to capture the group.
217    
218    13. Fixed two bugs in pcretest that were discovered by fuzzing and reported by
219        Red Hat Product Security:
220    
221        (a) A crash if /K and /F were both set with the option to save the compiled
222        pattern.
223    
224        (b) Another crash if the option to print captured substrings in a callout
225        was combined with setting a null ovector, for example \O\C+ as a subject
226        string.
227    
228    14. A pattern such as "((?2){0,1999}())?", which has a group containing a
229        forward reference repeated a large (but limited) number of times within a
230        repeated outer group that has a zero minimum quantifier, caused incorrect
231        code to be compiled, leading to the error "internal error:
232        previously-checked referenced subpattern not found" when an incorrect
233        memory address was read. This bug was reported as "heap overflow",
234        discovered by Kai Lu of Fortinet's FortiGuard Labs and given the CVE number
235        CVE-2015-2325.
236    
237    23. A pattern such as "((?+1)(\1))/" containing a forward reference subroutine
238        call within a group that also contained a recursive back reference caused
239        incorrect code to be compiled. This bug was reported as "heap overflow",
240        discovered by Kai Lu of Fortinet's FortiGuard Labs, and given the CVE
241        number CVE-2015-2326.
242    
243    24. Computing the size of the JIT read-only data in advance has been a source
244        of various issues, and new ones are still appear unfortunately. To fix
245        existing and future issues, size computation is eliminated from the code,
246        and replaced by on-demand memory allocation.
247    
248    25. A pattern such as /(?i)[A-`]/, where characters in the other case are
249        adjacent to the end of the range, and the range contained characters with
250        more than one other case, caused incorrect behaviour when compiled in UTF
251        mode. In that example, the range a-j was left out of the class.
252    
253    26. Fix JIT compilation of conditional blocks, which assertion
254        is converted to (*FAIL). E.g: /(?(?!))/.
255    
256    27. The pattern /(?(?!)^)/ caused references to random memory. This bug was
257        discovered by the LLVM fuzzer.
258    
259    28. The assertion (?!) is optimized to (*FAIL). This was not handled correctly
260        when this assertion was used as a condition, for example (?(?!)a|b). In
261        pcre2_match() it worked by luck; in pcre2_dfa_match() it gave an incorrect
262        error about an unsupported item.
263    
264    29. For some types of pattern, for example /Z*(|d*){216}/, the auto-
265        possessification code could take exponential time to complete. A recursion
266        depth limit of 1000 has been imposed to limit the resources used by this
267        optimization.
268    
269    30. A pattern such as /(*UTF)[\S\V\H]/, which contains a negated special class
270        such as \S in non-UCP mode, explicit wide characters (> 255) can be ignored
271        because \S ensures they are all in the class. The code for doing this was
272        interacting badly with the code for computing the amount of space needed to
273        compile the pattern, leading to a buffer overflow. This bug was discovered
274        by the LLVM fuzzer.
275    
276    31. A pattern such as /((?2)+)((?1))/ which has mutual recursion nested inside
277        other kinds of group caused stack overflow at compile time. This bug was
278        discovered by the LLVM fuzzer.
279    
280    32. A pattern such as /(?1)(?#?'){8}(a)/ which had a parenthesized comment
281        between a subroutine call and its quantifier was incorrectly compiled,
282        leading to buffer overflow or other errors. This bug was discovered by the
283        LLVM fuzzer.
284    
285    33. The illegal pattern /(?(?<E>.*!.*)?)/ was not being diagnosed as missing an
286        assertion after (?(. The code was failing to check the character after
287        (?(?< for the ! or = that would indicate a lookbehind assertion. This bug
288        was discovered by the LLVM fuzzer.
289    
290    34. A pattern such as /X((?2)()*+){2}+/ which has a possessive quantifier with
291        a fixed maximum following a group that contains a subroutine reference was
292        incorrectly compiled and could trigger buffer overflow. This bug was
293        discovered by the LLVM fuzzer.
294    
295    35. A mutual recursion within a lookbehind assertion such as (?<=((?2))((?1)))
296        caused a stack overflow instead of the diagnosis of a non-fixed length
297        lookbehind assertion. This bug was discovered by the LLVM fuzzer.
298    
299    36. The use of \K in a positive lookbehind assertion in a non-anchored pattern
300        (e.g. /(?<=\Ka)/) could make pcregrep loop.
301    
302    37. There was a similar problem to 36 in pcretest for global matches.
303    
304    38. If a greedy quantified \X was preceded by \C in UTF mode (e.g. \C\X*),
305        and a subsequent item in the pattern caused a non-match, backtracking over
306        the repeated \X did not stop, but carried on past the start of the subject,
307        causing reference to random memory and/or a segfault. There were also some
308        other cases where backtracking after \C could crash. This set of bugs was
309        discovered by the LLVM fuzzer.
310    
311    39. The function for finding the minimum length of a matching string could take
312        a very long time if mutual recursion was present many times in a pattern,
313        for example, /((?2){73}(?2))((?1))/. A better mutual recursion detection
314        method has been implemented. This infelicity was discovered by the LLVM
315        fuzzer.
316    
317    40. Static linking against the PCRE library using the pkg-config module was
318        failing on missing pthread symbols.
319    
320    
321    Version 8.36 26-September-2014
322    ------------------------------
323    
324    1.  Got rid of some compiler warnings in the C++ modules that were shown up by
325        -Wmissing-field-initializers and -Wunused-parameter.
326    
327    2.  The tests for quantifiers being too big (greater than 65535) were being
328        applied after reading the number, and stupidly assuming that integer
329        overflow would give a negative number. The tests are now applied as the
330        numbers are read.
331    
332    3.  Tidy code in pcre_exec.c where two branches that used to be different are
333        now the same.
334    
335    4.  The JIT compiler did not generate match limit checks for certain
336        bracketed expressions with quantifiers. This may lead to exponential
337        backtracking, instead of returning with PCRE_ERROR_MATCHLIMIT. This
338        issue should be resolved now.
339    
340    5.  Fixed an issue, which occures when nested alternatives are optimized
341        with table jumps.
342    
343    6.  Inserted two casts and changed some ints to size_t in the light of some
344        reported 64-bit compiler warnings (Bugzilla 1477).
345    
346    7.  Fixed a bug concerned with zero-minimum possessive groups that could match
347        an empty string, which sometimes were behaving incorrectly in the
348        interpreter (though correctly in the JIT matcher). This pcretest input is
349        an example:
350    
351          '\A(?:[^"]++|"(?:[^"]*+|"")*+")++'
352          NON QUOTED "QUOT""ED" AFTER "NOT MATCHED
353    
354        the interpreter was reporting a match of 'NON QUOTED ' only, whereas the
355        JIT matcher and Perl both matched 'NON QUOTED "QUOT""ED" AFTER '. The test
356        for an empty string was breaking the inner loop and carrying on at a lower
357        level, when possessive repeated groups should always return to a higher
358        level as they have no backtrack points in them. The empty string test now
359        occurs at the outer level.
360    
361    8.  Fixed a bug that was incorrectly auto-possessifying \w+ in the pattern
362        ^\w+(?>\s*)(?<=\w) which caused it not to match "test test".
363    
364    9.  Give a compile-time error for \o{} (as Perl does) and for \x{} (which Perl
365        doesn't).
366    
367    10. Change 8.34/15 introduced a bug that caused the amount of memory needed
368        to hold a pattern to be incorrectly computed (too small) when there were
369        named back references to duplicated names. This could cause "internal
370        error: code overflow" or "double free or corruption" or other memory
371        handling errors.
372    
373    11. When named subpatterns had the same prefixes, back references could be
374        confused. For example, in this pattern:
375    
376          /(?P<Name>a)?(?P<Name2>b)?(?(<Name>)c|d)*l/
377    
378        the reference to 'Name' was incorrectly treated as a reference to a
379        duplicate name.
380    
381    12. A pattern such as /^s?c/mi8 where the optional character has more than
382        one "other case" was incorrectly compiled such that it would only try to
383        match starting at "c".
384    
385    13. When a pattern starting with \s was studied, VT was not included in the
386        list of possible starting characters; this should have been part of the
387        8.34/18 patch.
388    
389    14. If a character class started [\Qx]... where x is any character, the class
390        was incorrectly terminated at the ].
391    
392    15. If a pattern that started with a caseless match for a character with more
393        than one "other case" was studied, PCRE did not set up the starting code
394        unit bit map for the list of possible characters. Now it does. This is an
395        optimization improvement, not a bug fix.
396    
397    16. The Unicode data tables have been updated to Unicode 7.0.0.
398    
399    17. Fixed a number of memory leaks in pcregrep.
400    
401    18. Avoid a compiler warning (from some compilers) for a function call with
402        a cast that removes "const" from an lvalue by using an intermediate
403        variable (to which the compiler does not object).
404    
405    19. Incorrect code was compiled if a group that contained an internal recursive
406        back reference was optional (had quantifier with a minimum of zero). This
407        example compiled incorrect code: /(((a\2)|(a*)\g<-1>))*/ and other examples
408        caused segmentation faults because of stack overflows at compile time.
409    
410    20. A pattern such as /((?(R)a|(?1)))+/, which contains a recursion within a
411        group that is quantified with an indefinite repeat, caused a compile-time
412        loop which used up all the system stack and provoked a segmentation fault.
413        This was not the same bug as 19 above.
414    
415    21. Add PCRECPP_EXP_DECL declaration to operator<< in pcre_stringpiece.h.
416        Patch by Mike Frysinger.
417    
418    
419    Version 8.35 04-April-2014
420    --------------------------
421    
422    1.  A new flag is set, when property checks are present in an XCLASS.
423        When this flag is not set, PCRE can perform certain optimizations
424        such as studying these XCLASS-es.
425    
426    2.  The auto-possessification of character sets were improved: a normal
427        and an extended character set can be compared now. Furthermore
428        the JIT compiler optimizes more character set checks.
429    
430    3.  Got rid of some compiler warnings for potentially uninitialized variables
431        that show up only when compiled with -O2.
432    
433    4.  A pattern such as (?=ab\K) that uses \K in an assertion can set the start
434        of a match later then the end of the match. The pcretest program was not
435        handling the case sensibly - it was outputting from the start to the next
436        binary zero. It now reports this situation in a message, and outputs the
437        text from the end to the start.
438    
439    5.  Fast forward search is improved in JIT. Instead of the first three
440        characters, any three characters with fixed position can be searched.
441        Search order: first, last, middle.
442    
443    6.  Improve character range checks in JIT. Characters are read by an inprecise
444        function now, which returns with an unknown value if the character code is
445        above a certain threshold (e.g: 256). The only limitation is that the value
446        must be bigger than the threshold as well. This function is useful when
447        the characters above the threshold are handled in the same way.
448    
449    7.  The macros whose names start with RAWUCHAR are placeholders for a future
450        mode in which only the bottom 21 bits of 32-bit data items are used. To
451        make this more memorable for those maintaining the code, the names have
452        been changed to start with UCHAR21, and an extensive comment has been added
453        to their definition.
454    
455    8.  Add missing (new) files sljitNativeTILEGX.c and sljitNativeTILEGX-encoder.c
456        to the export list in Makefile.am (they were accidentally omitted from the
457        8.34 tarball).
458    
459    9.  The informational output from pcretest used the phrase "starting byte set"
460        which is inappropriate for the 16-bit and 32-bit libraries. As the output
461        for "first char" and "need char" really means "non-UTF-char", I've changed
462        "byte" to "char", and slightly reworded the output. The documentation about
463        these values has also been (I hope) clarified.
464    
465    10. Another JIT related optimization: use table jumps for selecting the correct
466        backtracking path, when more than four alternatives are present inside a
467        bracket.
468    
469    11. Empty match is not possible, when the minimum length is greater than zero,
470        and there is no \K in the pattern. JIT should avoid empty match checks in
471        such cases.
472    
473    12. In a caseless character class with UCP support, when a character with more
474        than one alternative case was not the first character of a range, not all
475        the alternative cases were added to the class. For example, s and \x{17f}
476        are both alternative cases for S: the class [RST] was handled correctly,
477        but [R-T] was not.
478    
479    13. The configure.ac file always checked for pthread support when JIT was
480        enabled. This is not used in Windows, so I have put this test inside a
481        check for the presence of windows.h (which was already tested for).
482    
483    14. Improve pattern prefix search by a simplified Boyer-Moore algorithm in JIT.
484        The algorithm provides a way to skip certain starting offsets, and usually
485        faster than linear prefix searches.
486    
487    15. Change 13 for 8.20 updated RunTest to check for the 'fr' locale as well
488        as for 'fr_FR' and 'french'. For some reason, however, it then used the
489        Windows-specific input and output files, which have 'french' screwed in.
490        So this could never have worked. One of the problems with locales is that
491        they aren't always the same. I have now updated RunTest so that it checks
492        the output of the locale test (test 3) against three different output
493        files, and it allows the test to pass if any one of them matches. With luck
494        this should make the test pass on some versions of Solaris where it was
495        failing. Because of the uncertainty, the script did not used to stop if
496        test 3 failed; it now does. If further versions of a French locale ever
497        come to light, they can now easily be added.
498    
499    16. If --with-pcregrep-bufsize was given a non-integer value such as "50K",
500        there was a message during ./configure, but it did not stop. This now
501        provokes an error. The invalid example in README has been corrected.
502        If a value less than the minimum is given, the minimum value has always
503        been used, but now a warning is given.
504    
505    17. If --enable-bsr-anycrlf was set, the special 16/32-bit test failed. This
506        was a bug in the test system, which is now fixed. Also, the list of various
507        configurations that are tested for each release did not have one with both
508        16/32 bits and --enable-bar-anycrlf. It now does.
509    
510    18. pcretest was missing "-C bsr" for displaying the \R default setting.
511    
512    19. Little endian PowerPC systems are supported now by the JIT compiler.
513    
514    20. The fast forward newline mechanism could enter to an infinite loop on
515        certain invalid UTF-8 input. Although we don't support these cases
516        this issue can be fixed by a performance optimization.
517    
518    21. Change 33 of 8.34 is not sufficient to ensure stack safety because it does
519        not take account if existing stack usage. There is now a new global
520        variable called pcre_stack_guard that can be set to point to an external
521        function to check stack availability. It is called at the start of
522        processing every parenthesized group.
523    
524    22. A typo in the code meant that in ungreedy mode the max/min qualifier
525        behaved like a min-possessive qualifier, and, for example, /a{1,3}b/U did
526        not match "ab".
527    
528    23. When UTF was disabled, the JIT program reported some incorrect compile
529        errors. These messages are silenced now.
530    
531    24. Experimental support for ARM-64 and MIPS-64 has been added to the JIT
532        compiler.
533    
534    25. Change all the temporary files used in RunGrepTest to be different to those
535        used by RunTest so that the tests can be run simultaneously, for example by
536        "make -j check".
537    
538    
539    Version 8.34 15-December-2013
540    -----------------------------
541    
542    1.  Add pcre[16|32]_jit_free_unused_memory to forcibly free unused JIT
543        executable memory. Patch inspired by Carsten Klein.
544    
545    2.  ./configure --enable-coverage defined SUPPORT_GCOV in config.h, although
546        this macro is never tested and has no effect, because the work to support
547        coverage involves only compiling and linking options and special targets in
548        the Makefile. The comment in config.h implied that defining the macro would
549        enable coverage support, which is totally false. There was also support for
550        setting this macro in the CMake files (my fault, I just copied it from
551        configure). SUPPORT_GCOV has now been removed.
552    
553    3.  Make a small performance improvement in strlen16() and strlen32() in
554        pcretest.
555    
556    4.  Change 36 for 8.33 left some unreachable statements in pcre_exec.c,
557        detected by the Solaris compiler (gcc doesn't seem to be able to diagnose
558        these cases). There was also one in pcretest.c.
559    
560    5.  Cleaned up a "may be uninitialized" compiler warning in pcre_exec.c.
561    
562    6.  In UTF mode, the code for checking whether a group could match an empty
563        string (which is used for indefinitely repeated groups to allow for
564        breaking an infinite loop) was broken when the group contained a repeated
565        negated single-character class with a character that occupied more than one
566        data item and had a minimum repetition of zero (for example, [^\x{100}]* in
567        UTF-8 mode). The effect was undefined: the group might or might not be
568        deemed as matching an empty string, or the program might have crashed.
569    
570    7.  The code for checking whether a group could match an empty string was not
571        recognizing that \h, \H, \v, \V, and \R must match a character.
572    
573    8.  Implemented PCRE_INFO_MATCH_EMPTY, which yields 1 if the pattern can match
574        an empty string. If it can, pcretest shows this in its information output.
575    
576    9.  Fixed two related bugs that applied to Unicode extended grapheme clusters
577        that were repeated with a maximizing qualifier (e.g. \X* or \X{2,5}) when
578        matched by pcre_exec() without using JIT:
579    
580        (a) If the rest of the pattern did not match after a maximal run of
581            grapheme clusters, the code for backing up to try with fewer of them
582            did not always back up over a full grapheme when characters that do not
583            have the modifier quality were involved, e.g. Hangul syllables.
584    
585        (b) If the match point in a subject started with modifier character, and
586            there was no match, the code could incorrectly back up beyond the match
587            point, and potentially beyond the first character in the subject,
588            leading to a segfault or an incorrect match result.
589    
590    10. A conditional group with an assertion condition could lead to PCRE
591        recording an incorrect first data item for a match if no other first data
592        item was recorded. For example, the pattern (?(?=ab)ab) recorded "a" as a
593        first data item, and therefore matched "ca" after "c" instead of at the
594        start.
595    
596    11. Change 40 for 8.33 (allowing pcregrep to find empty strings) showed up a
597        bug that caused the command "echo a | ./pcregrep -M '|a'" to loop.
598    
599    12. The source of pcregrep now includes z/OS-specific code so that it can be
600        compiled for z/OS as part of the special z/OS distribution.
601    
602    13. Added the -T and -TM options to pcretest.
603    
604    14. The code in pcre_compile.c for creating the table of named capturing groups
605        has been refactored. Instead of creating the table dynamically during the
606        actual compiling pass, the information is remembered during the pre-compile
607        pass (on the stack unless there are more than 20 named groups, in which
608        case malloc() is used) and the whole table is created before the actual
609        compile happens. This has simplified the code (it is now nearly 150 lines
610        shorter) and prepared the way for better handling of references to groups
611        with duplicate names.
612    
613    15. A back reference to a named subpattern when there is more than one of the
614        same name now checks them in the order in which they appear in the pattern.
615        The first one that is set is used for the reference. Previously only the
616        first one was inspected. This change makes PCRE more compatible with Perl.
617    
618    16. Unicode character properties were updated from Unicode 6.3.0.
619    
620    17. The compile-time code for auto-possessification has been refactored, based
621        on a patch by Zoltan Herczeg. It now happens after instead of during
622        compilation. The code is cleaner, and more cases are handled. The option
623        PCRE_NO_AUTO_POSSESS is added for testing purposes, and the -O and /O
624        options in pcretest are provided to set it. It can also be set by
625        (*NO_AUTO_POSSESS) at the start of a pattern.
626    
627    18. The character VT has been added to the default ("C" locale) set of
628        characters that match \s and are generally treated as white space,
629        following this same change in Perl 5.18. There is now no difference between
630        "Perl space" and "POSIX space". Whether VT is treated as white space in
631        other locales depends on the locale.
632    
633    19. The code for checking named groups as conditions, either for being set or
634        for being recursed, has been refactored (this is related to 14 and 15
635        above). Processing unduplicated named groups should now be as fast at
636        numerical groups, and processing duplicated groups should be faster than
637        before.
638    
639    20. Two patches to the CMake build system, by Alexander Barkov:
640    
641          (1) Replace the "source" command by "." in CMakeLists.txt because
642              "source" is a bash-ism.
643    
644          (2) Add missing HAVE_STDINT_H and HAVE_INTTYPES_H to config-cmake.h.in;
645              without these the CMake build does not work on Solaris.
646    
647    21. Perl has changed its handling of \8 and \9. If there is no previously
648        encountered capturing group of those numbers, they are treated as the
649        literal characters 8 and 9 instead of a binary zero followed by the
650        literals. PCRE now does the same.
651    
652    22. Following Perl, added \o{} to specify codepoints in octal, making it
653        possible to specify values greater than 0777 and also making them
654        unambiguous.
655    
656    23. Perl now gives an error for missing closing braces after \x{... instead of
657        treating the string as literal. PCRE now does the same.
658    
659    24. RunTest used to grumble if an inappropriate test was selected explicitly,
660        but just skip it when running all tests. This make it awkward to run ranges
661        of tests when one of them was inappropriate. Now it just skips any
662        inappropriate tests, as it always did when running all tests.
663    
664    25. If PCRE_AUTO_CALLOUT and PCRE_UCP were set for a pattern that contained
665        character types such as \d or \w, too many callouts were inserted, and the
666        data that they returned was rubbish.
667    
668    26. In UCP mode, \s was not matching two of the characters that Perl matches,
669        namely NEL (U+0085) and MONGOLIAN VOWEL SEPARATOR (U+180E), though they
670        were matched by \h. The code has now been refactored so that the lists of
671        the horizontal and vertical whitespace characters used for \h and \v (which
672        are defined only in one place) are now also used for \s.
673    
674    27. Add JIT support for the 64 bit TileGX architecture.
675        Patch by Jiong Wang (Tilera Corporation).
676    
677    28. Possessive quantifiers for classes (both explicit and automatically
678        generated) now use special opcodes instead of wrapping in ONCE brackets.
679    
680    29. Whereas an item such as A{4}+ ignored the possessivenes of the quantifier
681        (because it's meaningless), this was not happening when PCRE_CASELESS was
682        set. Not wrong, but inefficient.
683    
684    30. Updated perltest.pl to add /u (force Unicode mode) when /W (use Unicode
685        properties for \w, \d, etc) is present in a test regex. Otherwise if the
686        test contains no characters greater than 255, Perl doesn't realise it
687        should be using Unicode semantics.
688    
689    31. Upgraded the handling of the POSIX classes [:graph:], [:print:], and
690        [:punct:] when PCRE_UCP is set so as to include the same characters as Perl
691        does in Unicode mode.
692    
693    32. Added the "forbid" facility to pcretest so that putting tests into the
694        wrong test files can sometimes be quickly detected.
695    
696    33. There is now a limit (default 250) on the depth of nesting of parentheses.
697        This limit is imposed to control the amount of system stack used at compile
698        time. It can be changed at build time by --with-parens-nest-limit=xxx or
699        the equivalent in CMake.
700    
701    34. Character classes such as [A-\d] or [a-[:digit:]] now cause compile-time
702        errors. Perl warns for these when in warning mode, but PCRE has no facility
703        for giving warnings.
704    
705    35. Change 34 for 8.13 allowed quantifiers on assertions, because Perl does.
706        However, this was not working for (?!) because it is optimized to (*FAIL),
707        for which PCRE does not allow quantifiers. The optimization is now disabled
708        when a quantifier follows (?!). I can't see any use for this, but it makes
709        things uniform.
710    
711    36. Perl no longer allows group names to start with digits, so I have made this
712        change also in PCRE. It simplifies the code a bit.
713    
714    37. In extended mode, Perl ignores spaces before a + that indicates a
715        possessive quantifier. PCRE allowed a space before the quantifier, but not
716        before the possessive +. It now does.
717    
718    38. The use of \K (reset reported match start) within a repeated possessive
719        group such as (a\Kb)*+ was not working.
720    
721    40. Document that the same character tables must be used at compile time and
722        run time, and that the facility to pass tables to pcre_exec() and
723        pcre_dfa_exec() is for use only with saved/restored patterns.
724    
725    41. Applied Jeff Trawick's patch CMakeLists.txt, which "provides two new
726        features for Builds with MSVC:
727    
728        1. Support pcre.rc and/or pcreposix.rc (as is already done for MinGW
729           builds). The .rc files can be used to set FileDescription and many other
730           attributes.
731    
732        2. Add an option (-DINSTALL_MSVC_PDB) to enable installation of .pdb files.
733           This allows higher-level build scripts which want .pdb files to avoid
734           hard-coding the exact files needed."
735    
736    42. Added support for [[:<:]] and [[:>:]] as used in the BSD POSIX library to
737        mean "start of word" and "end of word", respectively, as a transition aid.
738    
739    43. A minimizing repeat of a class containing codepoints greater than 255 in
740        non-UTF 16-bit or 32-bit modes caused an internal error when PCRE was
741        compiled to use the heap for recursion.
742    
743    44. Got rid of some compiler warnings for unused variables when UTF but not UCP
744        is configured.
745    
746    
747    Version 8.33 28-May-2013
748    ------------------------
749    
750    1.  Added 'U' to some constants that are compared to unsigned integers, to
751        avoid compiler signed/unsigned warnings. Added (int) casts to unsigned
752        variables that are added to signed variables, to ensure the result is
753        signed and can be negated.
754    
755    2.  Applied patch by Daniel Richard G for quashing MSVC warnings to the
756        CMake config files.
757    
758    3.  Revise the creation of config.h.generic so that all boolean macros are
759        #undefined, whereas non-boolean macros are #ifndef/#endif-ed. This makes
760        overriding via -D on the command line possible.
761    
762    4.  Changing the definition of the variable "op" in pcre_exec.c from pcre_uchar
763        to unsigned int is reported to make a quite noticeable speed difference in
764        a specific Windows environment. Testing on Linux did also appear to show
765        some benefit (and it is clearly not harmful). Also fixed the definition of
766        Xop which should be unsigned.
767    
768    5.  Related to (4), changing the definition of the intermediate variable cc
769        in repeated character loops from pcre_uchar to pcre_uint32 also gave speed
770        improvements.
771    
772    6.  Fix forward search in JIT when link size is 3 or greater. Also removed some
773        unnecessary spaces.
774    
775    7.  Adjust autogen.sh and configure.ac to lose warnings given by automake 1.12
776        and later.
777    
778    8.  Fix two buffer over read issues in 16 and 32 bit modes. Affects JIT only.
779    
780    9.  Optimizing fast_forward_start_bits in JIT.
781    
782    10. Adding support for callouts in JIT, and fixing some issues revealed
783        during this work. Namely:
784    
785        (a) Unoptimized capturing brackets incorrectly reset on backtrack.
786    
787        (b) Minimum length was not checked before the matching is started.
788    
789    11. The value of capture_last that is passed to callouts was incorrect in some
790        cases when there was a capture on one path that was subsequently abandoned
791        after a backtrack. Also, the capture_last value is now reset after a
792        recursion, since all captures are also reset in this case.
793    
794    12. The interpreter no longer returns the "too many substrings" error in the
795        case when an overflowing capture is in a branch that is subsequently
796        abandoned after a backtrack.
797    
798    13. In the pathological case when an offset vector of size 2 is used, pcretest
799        now prints out the matched string after a yield of 0 or 1.
800    
801    14. Inlining subpatterns in recursions, when certain conditions are fulfilled.
802        Only supported by the JIT compiler at the moment.
803    
804    15. JIT compiler now supports 32 bit Macs thanks to Lawrence Velazquez.
805    
806    16. Partial matches now set offsets[2] to the "bumpalong" value, that is, the
807        offset of the starting point of the matching process, provided the offsets
808        vector is large enough.
809    
810    17. The \A escape now records a lookbehind value of 1, though its execution
811        does not actually inspect the previous character. This is to ensure that,
812        in partial multi-segment matching, at least one character from the old
813        segment is retained when a new segment is processed. Otherwise, if there
814        are no lookbehinds in the pattern, \A might match incorrectly at the start
815        of a new segment.
816    
817    18. Added some #ifdef __VMS code into pcretest.c to help VMS implementations.
818    
819    19. Redefined some pcre_uchar variables in pcre_exec.c as pcre_uint32; this
820        gives some modest performance improvement in 8-bit mode.
821    
822    20. Added the PCRE-specific property \p{Xuc} for matching characters that can
823        be expressed in certain programming languages using Universal Character
824        Names.
825    
826    21. Unicode validation has been updated in the light of Unicode Corrigendum #9,
827        which points out that "non characters" are not "characters that may not
828        appear in Unicode strings" but rather "characters that are reserved for
829        internal use and have only local meaning".
830    
831    22. When a pattern was compiled with automatic callouts (PCRE_AUTO_CALLOUT) and
832        there was a conditional group that depended on an assertion, if the
833        assertion was false, the callout that immediately followed the alternation
834        in the condition was skipped when pcre_exec() was used for matching.
835    
836    23. Allow an explicit callout to be inserted before an assertion that is the
837        condition for a conditional group, for compatibility with automatic
838        callouts, which always insert a callout at this point.
839    
840    24. In 8.31, (*COMMIT) was confined to within a recursive subpattern. Perl also
841        confines (*SKIP) and (*PRUNE) in the same way, and this has now been done.
842    
843    25. (*PRUNE) is now supported by the JIT compiler.
844    
845    26. Fix infinite loop when /(?<=(*SKIP)ac)a/ is matched against aa.
846    
847    27. Fix the case where there are two or more SKIPs with arguments that may be
848        ignored.
849    
850    28. (*SKIP) is now supported by the JIT compiler.
851    
852    29. (*THEN) is now supported by the JIT compiler.
853    
854    30. Update RunTest with additional test selector options.
855    
856    31. The way PCRE handles backtracking verbs has been changed in two ways.
857    
858        (1) Previously, in something like (*COMMIT)(*SKIP), COMMIT would override
859        SKIP. Now, PCRE acts on whichever backtracking verb is reached first by
860        backtracking. In some cases this makes it more Perl-compatible, but Perl's
861        rather obscure rules do not always do the same thing.
862    
863        (2) Previously, backtracking verbs were confined within assertions. This is
864        no longer the case for positive assertions, except for (*ACCEPT). Again,
865        this sometimes improves Perl compatibility, and sometimes does not.
866    
867    32. A number of tests that were in test 2 because Perl did things differently
868        have been moved to test 1, because either Perl or PCRE has changed, and
869        these tests are now compatible.
870    
871    32. Backtracking control verbs are now handled in the same way in JIT and
872        interpreter.
873    
874    33. An opening parenthesis in a MARK/PRUNE/SKIP/THEN name in a pattern that
875        contained a forward subroutine reference caused a compile error.
876    
877    34. Auto-detect and optimize limited repetitions in JIT.
878    
879    35. Implement PCRE_NEVER_UTF to lock out the use of UTF, in particular,
880        blocking (*UTF) etc.
881    
882    36. In the interpreter, maximizing pattern repetitions for characters and
883        character types now use tail recursion, which reduces stack usage.
884    
885    37. The value of the max lookbehind was not correctly preserved if a compiled
886        and saved regex was reloaded on a host of different endianness.
887    
888    38. Implemented (*LIMIT_MATCH) and (*LIMIT_RECURSION). As part of the extension
889        of the compiled pattern block, expand the flags field from 16 to 32 bits
890        because it was almost full.
891    
892    39. Try madvise first before posix_madvise.
893    
894    40. Change 7 for PCRE 7.9 made it impossible for pcregrep to find empty lines
895        with a pattern such as ^$. It has taken 4 years for anybody to notice! The
896        original change locked out all matches of empty strings. This has been
897        changed so that one match of an empty string per line is recognized.
898        Subsequent searches on the same line (for colouring or for --only-matching,
899        for example) do not recognize empty strings.
900    
901    41. Applied a user patch to fix a number of spelling mistakes in comments.
902    
903    42. Data lines longer than 65536 caused pcretest to crash.
904    
905    43. Clarified the data type for length and startoffset arguments for pcre_exec
906        and pcre_dfa_exec in the function-specific man pages, where they were
907        explicitly stated to be in bytes, never having been updated. I also added
908        some clarification to the pcreapi man page.
909    
910    44. A call to pcre_dfa_exec() with an output vector size less than 2 caused
911        a segmentation fault.
912    
913    
914    Version 8.32 30-November-2012
915    -----------------------------
916    
917    1.  Improved JIT compiler optimizations for first character search and single
918        character iterators.
919    
920    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
921        Patch by Daniel Richard G.
922    
923    3.  Single character iterator optimizations in the JIT compiler.
924    
925    4.  Improved JIT compiler optimizations for character ranges.
926    
927    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
928        Reported by Giuseppe D'Angelo.
929    
930    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
931        of a line, was being set incorrectly in cases where .* appeared inside
932        atomic brackets at the start of a pattern, or where there was a subsequent
933        *PRUNE or *SKIP.
934    
935    7.  Improved instruction cache flush for POWER/PowerPC.
936        Patch by Daniel Richard G.
937    
938    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
939        grep:
940    
941        (a) There is now no limit to the number of patterns to be matched.
942    
943        (b) An error is given if a pattern is too long.
944    
945        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
946            are now supported.
947    
948        (d) --exclude-from and --include-from (multiple use) have been added.
949    
950        (e) Exclusions and inclusions now apply to all files and directories, not
951            just to those obtained from scanning a directory recursively.
952    
953        (f) Multiple uses of -f and --file-list are now supported.
954    
955        (g) In a Windows environment, the default for -d has been changed from
956            "read" (the GNU grep default) to "skip", because otherwise the presence
957            of a directory in the file list provokes an error.
958    
959        (h) The documentation has been revised and clarified in places.
960    
961    9.  Improve the matching speed of capturing brackets.
962    
963    10. Changed the meaning of \X so that it now matches a Unicode extended
964        grapheme cluster.
965    
966    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
967        out POSIX threads when JIT support is configured.
968    
969    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
970    
971    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
972        structure could go wrong in environments where size_t is not the same size
973        as int.
974    
975    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
976        set.
977    
978    15. The EBCDIC support had decayed; later updates to the code had included
979        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
980        general tidy up of EBCDIC-related issues, and the documentation was also
981        not quite right. There is now a test that can be run on ASCII systems to
982        check some of the EBCDIC-related things (but is it not a full test).
983    
984    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
985        in a small tidy to the code.
986    
987    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
988    
989    18. If the --only-matching (-o) option in pcregrep is specified multiple
990        times, each one causes appropriate output. For example, -o1 -o2 outputs the
991        substrings matched by the 1st and 2nd capturing parentheses. A separating
992        string can be specified by --om-separator (default empty).
993    
994    19. Improving the first n character searches.
995    
996    20. Turn case lists for horizontal and vertical white space into macros so that
997        they are defined only once.
998    
999    21. This set of changes together give more compatible Unicode case-folding
1000        behaviour for characters that have more than one other case when UCP
1001        support is available.
1002    
1003        (a) The Unicode property table now has offsets into a new table of sets of
1004            three or more characters that are case-equivalent. The MultiStage2.py
1005            script that generates these tables (the pcre_ucd.c file) now scans
1006            CaseFolding.txt instead of UnicodeData.txt for character case
1007            information.
1008    
1009        (b) The code for adding characters or ranges of characters to a character
1010            class has been abstracted into a generalized function that also handles
1011            case-independence. In UTF-mode with UCP support, this uses the new data
1012            to handle characters with more than one other case.
1013    
1014        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
1015            whose other case is greater than 256 are now correctly matched
1016            caselessly. Previously, the high codepoint matched the low one, but not
1017            vice versa.
1018    
1019        (d) The processing of \h, \H, \v, and \ in character classes now makes use
1020            of the new class addition function, using character lists defined as
1021            macros alongside the case definitions of 20 above.
1022    
1023        (e) Caseless back references now work with characters that have more than
1024            one other case.
1025    
1026        (f) General caseless matching of characters with more than one other case
1027            is supported.
1028    
1029    22. Unicode character properties were updated from Unicode 6.2.0
1030    
1031    23. Improved CMake support under Windows. Patch by Daniel Richard G.
1032    
1033    24. Add support for 32-bit character strings, and UTF-32
1034    
1035    25. Major JIT compiler update (code refactoring and bugfixing).
1036        Experimental Sparc 32 support is added.
1037    
1038    26. Applied a modified version of Daniel Richard G's patch to create
1039        pcre.h.generic and config.h.generic by "make" instead of in the
1040        PrepareRelease script.
1041    
1042    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
1043        pcre_compile.c when checking for a zero character.
1044    
1045    28. Introducing a native interface for JIT. Through this interface, the compiled
1046        machine code can be directly executed. The purpose of this interface is to
1047        provide fast pattern matching, so several sanity checks are not performed.
1048        However, feature tests are still performed. The new interface provides
1049        1.4x speedup compared to the old one.
1050    
1051    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
1052        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
1053        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
1054    
1055    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
1056        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
1057        the "old" RFC 2279). Instead, it ended up passing a negative length to
1058        pcre_exec().
1059    
1060    31. Add support for GCC's visibility feature to hide internal functions.
1061    
1062    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
1063        "unknown -C option" after outputting 0 or 1.
1064    
1065    33. There is now support for generating a code coverage report for the test
1066        suite in environments where gcc is the compiler and lcov is installed. This
1067        is mainly for the benefit of the developers.
1068    
1069    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
1070        unaddressable using valgrind annotations, allowing valgrind to detect
1071        invalid memory accesses. This is mainly for the benefit of the developers.
1072    
1073    25. (*UTF) can now be used to start a pattern in any of the three libraries.
1074    
1075    26. Give configure error if --enable-cpp but no C++ compiler found.
1076    
1077    
1078    Version 8.31 06-July-2012
1079    -------------------------
1080    
1081    1.  Fixing a wrong JIT test case and some compiler warnings.
1082    
1083    2.  Removed a bashism from the RunTest script.
1084    
1085    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
1086        to unsigned type, result still unsigned" that was given by an MS compiler
1087        on encountering the code "-sizeof(xxx)".
1088    
1089    4.  Partial matching support is added to the JIT compiler.
1090    
1091    5.  Fixed several bugs concerned with partial matching of items that consist
1092        of more than one character:
1093    
1094        (a) /^(..)\1/ did not partially match "aba" because checking references was
1095            done on an "all or nothing" basis. This also applied to repeated
1096            references.
1097    
1098        (b) \R did not give a hard partial match if \r was found at the end of the
1099            subject.
1100    
1101        (c) \X did not give a hard partial match after matching one or more
1102            characters at the end of the subject.
1103    
1104        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
1105            a partial match for the string "\r".
1106    
1107        (e) When newline was set to CRLF, the metacharacter "." did not recognize
1108            a partial match for a CR character at the end of the subject string.
1109    
1110    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
1111        running pcretest, the text "(JIT)" added to the output whenever JIT is
1112        actually used to run the match.
1113    
1114    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
1115        or /S+[+] with a digit between 1 and 7.
1116    
1117    8.  OP_NOT now supports any UTF character not just single-byte ones.
1118    
1119    9.  (*MARK) control verb is now supported by the JIT compiler.
1120    
1121    10. The command "./RunTest list" lists the available tests without actually
1122        running any of them. (Because I keep forgetting what they all are.)
1123    
1124    11. Add PCRE_INFO_MAXLOOKBEHIND.
1125    
1126    12. Applied a (slightly modified) user-supplied patch that improves performance
1127        when the heap is used for recursion (compiled with --disable-stack-for-
1128        recursion). Instead of malloc and free for each heap frame each time a
1129        logical recursion happens, frames are retained on a chain and re-used where
1130        possible. This sometimes gives as much as 30% improvement.
1131    
1132    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
1133        call.
1134    
1135    14. As documented, (*COMMIT) is now confined to within a positive assertion.
1136    
1137    15. It is now possible to link pcretest with libedit as an alternative to
1138        libreadline.
1139    
1140    16. (*COMMIT) control verb is now supported by the JIT compiler.
1141    
1142    17. The Unicode data tables have been updated to Unicode 6.1.0.
1143    
1144    18. Added --file-list option to pcregrep.
1145    
1146    19. Added binary file support to pcregrep, including the -a, --binary-files,
1147        -I, and --text options.
1148    
1149    20. The madvise function is renamed for posix_madvise for QNX compatibility
1150        reasons. Fixed by Giuseppe D'Angelo.
1151    
1152    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
1153    
1154    22. Changed the option for creating symbolic links for 16-bit man pages from
1155        -s to -sf so that re-installing does not cause issues.
1156    
1157    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
1158    
1159    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
1160        matches in certain environments (the workspace was not being correctly
1161        retained). Also added to pcre_dfa_exec() a simple plausibility check on
1162        some of the workspace data at the beginning of a restart.
1163    
1164    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
1165        was not doing so when it should - probably a typo introduced by SVN 528
1166        (change 8.10/14).
1167    
1168    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
1169        \w+ when the character tables indicated that \x{c4} was a word character.
1170        There were several related cases, all because the tests for doing a table
1171        lookup were testing for characters less than 127 instead of 255.
1172    
1173    27. If a pattern contains capturing parentheses that are not used in a match,
1174        their slots in the ovector are set to -1. For those that are higher than
1175        any matched groups, this happens at the end of processing. In the case when
1176        there were back references that the ovector was too small to contain
1177        (causing temporary malloc'd memory to be used during matching), and the
1178        highest capturing number was not used, memory off the end of the ovector
1179        was incorrectly being set to -1. (It was using the size of the temporary
1180        memory instead of the true size.)
1181    
1182    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
1183        ovector size, it uses memory at the end of the block that it has got.
1184    
1185    29. Check for an overlong MARK name and give an error at compile time. The
1186        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
1187    
1188    30. JIT compiler update.
1189    
1190    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
1191        Rill for the patch.
1192    
1193    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
1194    
1195    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
1196    
1197    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
1198        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
1199        was enabled.
1200    
1201    35. Improve JIT code generation for greedy plus quantifier.
1202    
1203    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
1204        1 to "aa" instead of to an empty string. The bug affected repeated groups
1205        that could potentially match an empty string.
1206    
1207    37. Optimizing single character iterators in JIT.
1208    
1209    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
1210        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
1211        codepoints that are too big for the mode are faulted, and in a UTF mode,
1212        disallowed codepoints are also faulted.
1213    
1214    39. If PCRE was compiled with UTF support, in three places in the DFA
1215        matcher there was code that should only have been obeyed in UTF mode, but
1216        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
1217        processing when bytes with values greater than 127 were present. In 16-bit
1218        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
1219        both cases the values are those that cannot be the first data item in a UTF
1220        character. The three items that might have provoked this were recursions,
1221        possessively repeated groups, and atomic groups.
1222    
1223    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
1224        and pcregrep, because some OS require shared objects to be explicitly
1225        passed to ld, causing the link step to fail if they are not.
1226    
1227    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
1228        mode, patterns that started with \h* or \R* might be incorrectly matched.
1229    
1230    
1231    Version 8.30 04-February-2012
1232    -----------------------------
1233    
1234    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
1235        name is defined in ctype.h.
1236    
1237    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
1238        only in quite long subpatterns.
1239    
1240    3.  Removed the function pcre_info(), which has been obsolete and deprecated
1241        since it was replaced by pcre_fullinfo() in February 2000.
1242    
1243    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
1244        match a (*MARK), and the match failed at the start of the subject, a
1245        reference to memory before the start of the subject could occur. This bug
1246        was introduced by fix 17 of release 8.21.
1247    
1248    5.  A reference to an unset group with zero minimum repetition was giving
1249        totally wrong answers (in non-JavaScript-compatibility mode). For example,
1250        /(another)?(\1?)test/ matched against "hello world test". This bug was
1251        introduced in release 8.13.
1252    
1253    6.  Add support for 16-bit character strings (a large amount of work involving
1254        many changes and refactorings).
1255    
1256    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
1257        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
1258        from a file.
1259    
1260    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
1261        rounding is not applied in this particular case).
1262    
1263    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
1264        if they appear, or are escaped, in patterns.
1265    
1266    10. Get rid of a number of -Wunused-but-set-variable warnings.
1267    
1268    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
1269        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
1270        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
1271        also returns the mark "x". This bug applied to capturing parentheses,
1272        non-capturing parentheses, and atomic parentheses. It also applied to some
1273        assertions.
1274    
1275    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
1276        information out of configure.ac instead of relying on pcre.h.generic, which
1277        is not stored in the repository.
1278    
1279    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
1280        -lreadline.
1281    
1282    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
1283    
1284    15. Applied Graycode's patch to put the top-level frame on the stack rather
1285        than the heap when not using the stack for recursion. This gives a
1286        performance improvement in many cases when recursion is not deep.
1287    
1288    16. Experimental code added to "pcretest -C" to output the stack frame size.
1289    
1290    
1291    Version 8.21 12-Dec-2011
1292    ------------------------
1293    
1294    1.  Updating the JIT compiler.
1295    
1296    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
1297        are added as well.
1298    
1299    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
1300        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
1301        calling _pcre_jit_exec. Some extra comments are added.
1302    
1303    4.  (*MARK) settings inside atomic groups that do not contain any capturing
1304        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
1305        was introduced by change 18 for 8.20.
1306    
1307    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
1308        ECMA-262 standard.
1309    
1310    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
1311        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
1312        This bug was probably introduced by change 9 of 8.13.
1313    
1314    7.  While fixing 6 above, I noticed that a number of other items were being
1315        incorrectly rejected as "not fixed length". This arose partly because newer
1316        opcodes had not been added to the fixed-length checking code. I have (a)
1317        corrected the bug and added tests for these items, and (b) arranged for an
1318        error to occur if an unknown opcode is encountered while checking for fixed
1319        length instead of just assuming "not fixed length". The items that were
1320        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
1321        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
1322        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
1323    
1324    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
1325        being incorrectly compiled and would have given unpredicatble results.
1326    
1327    9.  A possessively repeated subpattern with minimum repeat count greater than
1328        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
1329        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
1330        the first (A) could occur when it should not.
1331    
1332    10. Add a cast and remove a redundant test from the code.
1333    
1334    11. JIT should use pcre_malloc/pcre_free for allocation.
1335    
1336    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
1337        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
1338        is anything other than /usr, -L is still shown).
1339    
1340    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
1341    
1342    14. Perl does not support \N without a following name in a [] class; PCRE now
1343        also gives an error.
1344    
1345    15. If a forward reference was repeated with an upper limit of around 2000,
1346        it caused the error "internal error: overran compiling workspace". The
1347        maximum number of forward references (including repeats) was limited by the
1348        internal workspace, and dependent on the LINK_SIZE. The code has been
1349        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
1350        the default depends on LINK_SIZE. There is a new upper limit (for safety)
1351        of around 200,000 forward references. While doing this, I also speeded up
1352        the filling in of repeated forward references.
1353    
1354    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
1355        incorrectly expecting the subject to contain another "a" after the start.
1356    
1357    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
1358        in the match, the SKIP should be ignored. This was not happening; instead
1359        the SKIP was being treated as NOMATCH. For patterns such as
1360        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
1361        tested.
1362    
1363    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
1364        now much more compatible with Perl, in particular in cases where the result
1365        is a non-match for a non-anchored pattern. For example, if
1366        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
1367        "m", where previously it did not return a name. A side effect of this
1368        change is that for partial matches, the last encountered mark name is
1369        returned, as for non matches. A number of tests that were previously not
1370        Perl-compatible have been moved into the Perl-compatible test files. The
1371        refactoring has had the pleasing side effect of removing one argument from
1372        the match() function, thus reducing its stack requirements.
1373    
1374    19. If the /S+ option was used in pcretest to study a pattern using JIT,
1375        subsequent uses of /S (without +) incorrectly behaved like /S+.
1376    
1377    21. Retrieve executable code size support for the JIT compiler and fixing
1378        some warnings.
1379    
1380    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
1381        not work when the shorter character appeared right at the end of the
1382        subject string.
1383    
1384    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
1385        systems.
1386    
1387    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
1388        output it when the /M option is used in pcretest.
1389    
1390    25. The CheckMan script was not being included in the distribution. Also, added
1391        an explicit "perl" to run Perl scripts from the PrepareRelease script
1392        because this is reportedly needed in Windows.
1393    
1394    26. If study data was being save in a file and studying had not found a set of
1395        "starts with" bytes for the pattern, the data written to the file (though
1396        never used) was taken from uninitialized memory and so caused valgrind to
1397        complain.
1398    
1399    27. Updated RunTest.bat as provided by Sheri Pierce.
1400    
1401    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
1402    
1403    29. Computation of memory usage for the table of capturing group names was
1404        giving an unnecessarily large value.
1405    
1406    
1407    Version 8.20 21-Oct-2011
1408    ------------------------
1409    
1410    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
1411        a POSIX class. After further experiments with Perl, which convinced me that
1412        Perl has bugs and confusions, a closing square bracket is no longer allowed
1413        in a POSIX name. This bug also affected patterns with classes that started
1414        with full stops.
1415    
1416    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
1417        captured substring, but while checking the failing first alternative,
1418        substring 1 is temporarily captured. If the output vector supplied to
1419        pcre_exec() was not big enough for this capture, the yield of the function
1420        was still zero ("insufficient space for captured substrings"). This cannot
1421        be totally fixed without adding another stack variable, which seems a lot
1422        of expense for a edge case. However, I have improved the situation in cases
1423        such as /(a)(b)x|abc/ matched against "abc", where the return code
1424        indicates that fewer than the maximum number of slots in the ovector have
1425        been set.
1426    
1427    3.  Related to (2) above: when there are more back references in a pattern than
1428        slots in the output vector, pcre_exec() uses temporary memory during
1429        matching, and copies in the captures as far as possible afterwards. It was
1430        using the entire output vector, but this conflicts with the specification
1431        that only 2/3 is used for passing back captured substrings. Now it uses
1432        only the first 2/3, for compatibility. This is, of course, another edge
1433        case.
1434    
1435    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
1436        main code base, and can be used by building with --enable-jit. When this is
1437        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
1438        runtime --no-jit option is given.
1439    
1440    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
1441        ovector, the return from the function was zero, implying that there were
1442        other matches that did not fit. The correct "exactly full" value is now
1443        returned.
1444    
1445    6.  If a subpattern that was called recursively or as a subroutine contained
1446        (*PRUNE) or any other control that caused it to give a non-standard return,
1447        invalid errors such as "Error -26 (nested recursion at the same subject
1448        position)" or even infinite loops could occur.
1449    
1450    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
1451        computing the minimum length on reaching *ACCEPT, and so ended up with the
1452        wrong value of 1 rather than 0. Further investigation indicates that
1453        computing a minimum subject length in the presence of *ACCEPT is difficult
1454        (think back references, subroutine calls), and so I have changed the code
1455        so that no minimum is registered for a pattern that contains *ACCEPT.
1456    
1457    8.  If (*THEN) was present in the first (true) branch of a conditional group,
1458        it was not handled as intended. [But see 16 below.]
1459    
1460    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
1461        Sheri Pierce.
1462    
1463    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
1464        the first byte in a match must be "a".
1465    
1466    11. Change 17 for 8.13 increased the recursion depth for patterns like
1467        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
1468        pattern contains any instances of (*THEN). If it does not, the old
1469        optimizations are restored. It would be nice to do this on a per-group
1470        basis, but at the moment that is not feasible.
1471    
1472    12. In some environments, the output of pcretest -C is CRLF terminated. This
1473        broke RunTest's code that checks for the link size. A single white space
1474        character after the value is now allowed for.
1475    
1476    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
1477        For "fr", it uses the Windows-specific input and output files.
1478    
1479    14. If (*THEN) appeared in a group that was called recursively or as a
1480        subroutine, it did not work as intended. [But see next item.]
1481    
1482    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
1483        pattern fragments (but not containing any | characters). If A and B are
1484        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
1485        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
1486        D. In other words, Perl considers parentheses that do not contain any |
1487        characters to be part of a surrounding alternative, whereas PCRE was
1488        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
1489        differently. PCRE now behaves in the same way as Perl, except in the case
1490        of subroutine/recursion calls such as (?1) which have in any case always
1491        been different (but PCRE had them first :-).
1492    
1493    16. Related to 15 above: Perl does not treat the | in a conditional group as
1494        creating alternatives. Such a group is treated in the same way as an
1495        ordinary group without any | characters when processing (*THEN). PCRE has
1496        been changed to match Perl's behaviour.
1497    
1498    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
1499        RunGrepTest script failed.
1500    
1501    18. Change 22 for version 13 caused atomic groups to use more stack. This is
1502        inevitable for groups that contain captures, but it can lead to a lot of
1503        stack use in large patterns. The old behaviour has been restored for atomic
1504        groups that do not contain any capturing parentheses.
1505    
1506    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
1507        suppress the check for a minimum subject length at run time. (If it was
1508        given to pcre_exec() or pcre_dfa_exec() it did work.)
1509    
1510    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
1511        fail to work when decoding hex characters in data strings in EBCDIC
1512        environments.
1513    
1514    21. It appears that in at least one Mac OS environment, the isxdigit() function
1515        is implemented as a macro that evaluates to its argument more than once,
1516        contravening the C 90 Standard (I haven't checked a later standard). There
1517        was an instance in pcretest which caused it to go wrong when processing
1518        \x{...} escapes in subject strings. The has been rewritten to avoid using
1519        things like p++ in the argument of isxdigit().
1520    
1521    
1522    Version 8.13 16-Aug-2011
1523    ------------------------
1524    
1525    1.  The Unicode data tables have been updated to Unicode 6.0.0.
1526    
1527    2.  Two minor typos in pcre_internal.h have been fixed.
1528    
1529    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
1530        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
1531        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
1532    
1533    4.  There were a number of related bugs in the code for matching backrefences
1534        caselessly in UTF-8 mode when codes for the characters concerned were
1535        different numbers of bytes. For example, U+023A and U+2C65 are an upper
1536        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
1537        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
1538        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
1539        2-byte code at the end of the subject (it thought there wasn't enough data
1540        left).
1541    
1542    5.  Comprehensive information about what went wrong is now returned by
1543        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
1544        as the output vector has at least 2 elements. The offset of the start of
1545        the failing character and a reason code are placed in the vector.
1546    
1547    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
1548        now returned is for the first byte of the failing character, instead of the
1549        last byte inspected. This is an incompatible change, but I hope it is small
1550        enough not to be a problem. It makes the returned offset consistent with
1551        pcre_exec() and pcre_dfa_exec().
1552    
1553    7.  pcretest now gives a text phrase as well as the error number when
1554        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
1555        failure, the offset and reason code are output.
1556    
1557    8.  When \R was used with a maximizing quantifier it failed to skip backwards
1558        over a \r\n pair if the subsequent match failed. Instead, it just skipped
1559        back over a single character (\n). This seems wrong (because it treated the
1560        two characters as a single entity when going forwards), conflicts with the
1561        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
1562        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
1563        has been changed.
1564    
1565    9.  Some internal refactoring has changed the processing so that the handling
1566        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
1567        time (the PCRE_DOTALL option was changed this way some time ago: version
1568        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
1569        which was always a bit of a fudge. It also means that there is one less
1570        argument for the match() function, which reduces its stack requirements
1571        slightly. This change also fixes an incompatibility with Perl: the pattern
1572        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
1573    
1574    10. More internal refactoring has drastically reduced the number of recursive
1575        calls to match() for possessively repeated groups such as (abc)++ when
1576        using pcre_exec().
1577    
1578    11. While implementing 10, a number of bugs in the handling of groups were
1579        discovered and fixed:
1580    
1581        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
1582        (a|)*(?1) gave a compile-time internal error.
1583        ((a|)+)+  did not notice that the outer group could match an empty string.
1584        (^a|^)+   was not marked as anchored.
1585        (.*a|.*)+ was not marked as matching at start or after a newline.
1586    
1587    12. Yet more internal refactoring has removed another argument from the match()
1588        function. Special calls to this function are now indicated by setting a
1589        value in a variable in the "match data" data block.
1590    
1591    13. Be more explicit in pcre_study() instead of relying on "default" for
1592        opcodes that mean there is no starting character; this means that when new
1593        ones are added and accidentally left out of pcre_study(), testing should
1594        pick them up.
1595    
1596    14. The -s option of pcretest has been documented for ages as being an old
1597        synonym of -m (show memory usage). I have changed it to mean "force study
1598        for every regex", that is, assume /S for every regex. This is similar to -i
1599        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
1600        using it. It makes it easier to run collections of tests with and without
1601        study enabled, and thereby test pcre_study() more easily. All the standard
1602        tests are now run with and without -s (but some patterns can be marked as
1603        "never study" - see 20 below).
1604    
1605    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
1606        restoration of the capturing data to the outer values was not happening
1607        correctly.
1608    
1609    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
1610        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
1611        pattern had matched an empty string, and so incorrectly returned a no
1612        match.
1613    
1614    17. There was optimizing code for the last branch of non-capturing parentheses,
1615        and also for the obeyed branch of a conditional subexpression, which used
1616        tail recursion to cut down on stack usage. Unfortunately, now that there is
1617        the possibility of (*THEN) occurring in these branches, tail recursion is
1618        no longer possible because the return has to be checked for (*THEN). These
1619        two optimizations have therefore been removed. [But see 8.20/11 above.]
1620    
1621    18. If a pattern containing \R was studied, it was assumed that \R always
1622        matched two bytes, thus causing the minimum subject length to be
1623        incorrectly computed because \R can also match just one byte.
1624    
1625    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
1626        was incorrectly computed.
1627    
1628    20. If /S is present twice on a test pattern in pcretest input, it now
1629        *disables* studying, thereby overriding the use of -s on the command line
1630        (see 14 above). This is necessary for one or two tests to keep the output
1631        identical in both cases.
1632    
1633    21. When (*ACCEPT) was used in an assertion that matched an empty string and
1634        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
1635    
1636    22. When an atomic group that contained a capturing parenthesis was
1637        successfully matched, but the branch in which it appeared failed, the
1638        capturing was not being forgotten if a higher numbered group was later
1639        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
1640        group 1 to "a", when in fact it should be unset. This applied to multi-
1641        branched capturing and non-capturing groups, repeated or not, and also to
1642        positive assertions (capturing in negative assertions does not happen
1643        in PCRE) and also to nested atomic groups.
1644    
1645    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
1646        subject after a captured substring, to make it easier to tell which of a
1647        number of identical substrings has been captured.
1648    
1649    24. The way atomic groups are processed by pcre_exec() has been changed so that
1650        if they are repeated, backtracking one repetition now resets captured
1651        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
1652        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
1653        "aaa". Previously, it would have been "a". As part of this code
1654        refactoring, the way recursive calls are handled has also been changed.
1655    
1656    25. If an assertion condition captured any substrings, they were not passed
1657        back unless some other capturing happened later. For example, if
1658        (?(?=(a))a) was matched against "a", no capturing was returned.
1659    
1660    26. When studying a pattern that contained subroutine calls or assertions,
1661        the code for finding the minimum length of a possible match was handling
1662        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
1663        group 1 called group 2 while simultaneously a separate group 2 called group
1664        1). A stack overflow occurred in this case. I have fixed this by limiting
1665        the recursion depth to 10.
1666    
1667    27. Updated RunTest.bat in the distribution to the version supplied by Tom
1668        Fortmann. This supports explicit test numbers on the command line, and has
1669        argument validation and error reporting.
1670    
1671    28. An instance of \X with an unlimited repeat could fail if at any point the
1672        first character it looked at was a mark character.
1673    
1674    29. Some minor code refactoring concerning Unicode properties and scripts
1675        should reduce the stack requirement of match() slightly.
1676    
1677    30. Added the '=' option to pcretest to check the setting of unused capturing
1678        slots at the end of the pattern, which are documented as being -1, but are
1679        not included in the return count.
1680    
1681    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
1682        compiled something random. Now it gives a compile-time error (as does
1683        Perl).
1684    
1685    32. A *MARK encountered during the processing of a positive assertion is now
1686        recorded and passed back (compatible with Perl).
1687    
1688    33. If --only-matching or --colour was set on a pcregrep call whose pattern
1689        had alternative anchored branches, the search for a second match in a line
1690        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
1691        matched the line "0102" twice. The same bug affected patterns that started
1692        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
1693        twice.
1694    
1695    34. Previously, PCRE did not allow quantification of assertions. However, Perl
1696        does, and because of capturing effects, quantifying parenthesized
1697        assertions may at times be useful. Quantifiers are now allowed for
1698        parenthesized assertions.
1699    
1700    35. A minor code tidy in pcre_compile() when checking options for \R usage.
1701    
1702    36. \g was being checked for fancy things in a character class, when it should
1703        just be a literal "g".
1704    
1705    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
1706        appearance of a nested POSIX class supersedes an apparent external class.
1707        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
1708        unescaped square brackets may also appear as part of class names. For
1709        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
1710        more like Perl. (But see 8.20/1 above.)
1711    
1712    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
1713        was because it thought it was \N{name}, which is not supported).
1714    
1715    39. Add minix to OS list not supporting the -S option in pcretest.
1716    
1717    40. PCRE tries to detect cases of infinite recursion at compile time, but it
1718        cannot analyze patterns in sufficient detail to catch mutual recursions
1719        such as ((?1))((?2)). There is now a runtime test that gives an error if a
1720        subgroup is called recursively as a subpattern for a second time at the
1721        same position in the subject string. In previous releases this might have
1722        been caught by the recursion limit, or it might have run out of stack.
1723    
1724    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
1725        happen only once. PCRE was, however incorrectly giving a compile time error
1726        "recursive call could loop indefinitely" because it cannot analyze the
1727        pattern in sufficient detail. The compile time test no longer happens when
1728        PCRE is compiling a conditional subpattern, but actual runaway loops are
1729        now caught at runtime (see 40 above).
1730    
1731    42. It seems that Perl allows any characters other than a closing parenthesis
1732        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
1733        has been changed to be the same.
1734    
1735    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
1736        as not to get warnings when autogen.sh is called. Also changed
1737        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
1738    
1739    44. To help people who use pcregrep to scan files containing exceedingly long
1740        lines, the following changes have been made:
1741    
1742        (a) The default value of the buffer size parameter has been increased from
1743            8K to 20K. (The actual buffer used is three times this size.)
1744    
1745        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
1746            PCRE is built.
1747    
1748        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
1749            to be set at run time.
1750    
1751        (d) Numerical values in pcregrep options can be followed by K or M, for
1752            example --buffer-size=50K.
1753    
1754        (e) If a line being scanned overflows pcregrep's buffer, an error is now
1755            given and the return code is set to 2.
1756    
1757    45. Add a pointer to the latest mark to the callout data block.
1758    
1759    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
1760        partial match of an empty string instead of no match. This was specific to
1761        the use of ".".
1762    
1763    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
1764        complete match instead of a partial match. This bug was dependent on both
1765        the PCRE_UTF8 and PCRE_DOTALL options being set.
1766    
1767    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
1768        starting byte set, because \b was not being ignored.
1769    
1770    
1771    Version 8.12 15-Jan-2011
1772    ------------------------
1773    
1774    1.  Fixed some typos in the markup of the man pages, and wrote a script that
1775        checks for such things as part of the documentation building process.
1776    
1777    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
1778        --match-limit and --recursion-limit options (added for 8.11). In
1779        particular, this made one of the standard tests fail. (The integer value
1780        went into the wrong half of a long int.)
1781    
1782    3.  If the --colour option was given to pcregrep with -v (invert match), it
1783        did strange things, either producing crazy output, or crashing. It should,
1784        of course, ignore a request for colour when reporting lines that do not
1785        match.
1786    
1787    4.  Another pcregrep bug caused similar problems if --colour was specified with
1788        -M (multiline) and the pattern match finished with a line ending.
1789    
1790    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
1791        matched in multiline mode, the following line was shown as part of the
1792        match. This seems wrong, so I have changed it.
1793    
1794    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
1795        the check for further matches in the same line (so they could be coloured)
1796        to overrun the end of the current line. If another match was found, it was
1797        incorrectly shown (and then shown again when found in the next line).
1798    
1799    7.  If pcregrep was compiled under Windows, there was a reference to the
1800        function pcregrep_exit() before it was defined. I am assuming this was
1801        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
1802        reported by a user. I've moved the definition above the reference.
1803    
1804    
1805    Version 8.11 10-Dec-2010
1806    ------------------------
1807    
1808    1.  (*THEN) was not working properly if there were untried alternatives prior
1809        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
1810        backtracked to try for "b" instead of moving to the next alternative branch
1811        at the same level (in this case, to look for "c"). The Perl documentation
1812        is clear that when (*THEN) is backtracked onto, it goes to the "next
1813        alternative in the innermost enclosing group".
1814    
1815    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
1816        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
1817        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
1818        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
1819        (*THEN).
1820    
1821    3.  If \s appeared in a character class, it removed the VT character from
1822        the class, even if it had been included by some previous item, for example
1823        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
1824        of \s, but is part of the POSIX "space" class.)
1825    
1826    4.  A partial match never returns an empty string (because you can always
1827        match an empty string at the end of the subject); however the checking for
1828        an empty string was starting at the "start of match" point. This has been
1829        changed to the "earliest inspected character" point, because the returned
1830        data for a partial match starts at this character. This means that, for
1831        example, /(?<=abc)def/ gives a partial match for the subject "abc"
1832        (previously it gave "no match").
1833    
1834    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
1835        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
1836        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
1837        has an implication that the given string is incomplete (because a partial
1838        match is preferred over a full match). For this reason, these items now
1839        give a partial match in this situation. [Aside: previously, the one case
1840        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
1841        match rather than a full match, which was wrong by the old rules, but is
1842        now correct.]
1843    
1844    6.  There was a bug in the handling of #-introduced comments, recognized when
1845        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
1846        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
1847        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
1848        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
1849        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
1850        places in pcre_compile().
1851    
1852    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
1853        comments when looking ahead for named forward references to subpatterns,
1854        the only newline sequence it recognized was NL. It now handles newlines
1855        according to the set newline convention.
1856    
1857    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
1858        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
1859        cater for a lack of strerror(). These oversights have been fixed.
1860    
1861    9.  Added --match-limit and --recursion-limit to pcregrep.
1862    
1863    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
1864    
1865    11. When the -o option was used, pcregrep was setting a return code of 1, even
1866        when matches were found, and --line-buffered was not being honoured.
1867    
1868    12. Added an optional parentheses number to the -o and --only-matching options
1869        of pcregrep.
1870    
1871    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
1872        can match an empty string. The code to do it in pcretest and pcredemo
1873        needed fixing:
1874    
1875        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
1876            only one byte after an empty string match just before CRLF (this case
1877            just got forgotten; "any" and "anycrlf" were OK).
1878    
1879        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
1880            mode when an empty string match preceded an ASCII character followed by
1881            a non-ASCII character. (The code for advancing by one character rather
1882            than one byte was nonsense.)
1883    
1884        (c) The pcredemo.c sample program did not have any code at all to handle
1885            the cases when CRLF is a valid newline sequence.
1886    
1887    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
1888        as a starting offset was within the subject string. There is now a new
1889        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
1890        negative or greater than the length of the string. In order to test this,
1891        pcretest is extended to allow the setting of negative starting offsets.
1892    
1893    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
1894        starting offset points to the beginning of a UTF-8 character was
1895        unnecessarily clumsy. I tidied it up.
1896    
1897    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
1898        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
1899    
1900    17. Nobody had reported that the --include_dir option, which was added in
1901        release 7.7 should have been called --include-dir (hyphen, not underscore)
1902        for compatibility with GNU grep. I have changed it to --include-dir, but
1903        left --include_dir as an undocumented synonym, and the same for
1904        --exclude-dir, though that is not available in GNU grep, at least as of
1905        release 2.5.4.
1906    
1907    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
1908        characters from a string of bytes) have been redefined so as not to use
1909        loops, in order to improve performance in some environments. At the same
1910        time, I abstracted some of the common code into auxiliary macros to save
1911        repetition (this should not affect the compiled code).
1912    
1913    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1914        compile-time error is now given if \c is not followed by an ASCII
1915        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1916        different, and any byte value is allowed.)
1917    
1918    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1919        START_OPTIMIZE option, which is now allowed at compile time - but just
1920        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1921        to pcregrep and other applications that have no direct access to PCRE
1922        options. The new /Y option in pcretest sets this option when calling
1923        pcre_compile().
1924    
1925    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1926        back references. Groups containing recursive back references were forced to
1927        be atomic by that change, but in the case of named groups, the amount of
1928        memory required was incorrectly computed, leading to "Failed: internal
1929        error: code overflow". This has been fixed.
1930    
1931    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1932        pcretest.c, to avoid build problems in some Borland environments.
1933    
1934    
1935    Version 8.10 25-Jun-2010
1936  ------------------------  ------------------------
1937    
1938  1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and  1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
# Line 42  Version 8.10 03-Jun-2010 Line 1973  Version 8.10 03-Jun-2010
1973  11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was  11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1974      studied, and the match started with a letter with a code point greater than      studied, and the match started with a letter with a code point greater than
1975      127 whose first byte was different to the first byte of the other case of      127 whose first byte was different to the first byte of the other case of
1976      the letter, the other case of this starting letter was not recognized      the letter, the other case of this starting letter was not recognized
1977      (#976).      (#976).
1978    
1979  12. If a pattern that was studied started with a repeated Unicode property  12. If a pattern that was studied started with a repeated Unicode property
# Line 68  Version 8.10 03-Jun-2010 Line 1999  Version 8.10 03-Jun-2010
1999    
2000  17. Added a check for running out of memory when PCRE is compiled with  17. Added a check for running out of memory when PCRE is compiled with
2001      --disable-stack-for-recursion (#990).      --disable-stack-for-recursion (#990).
2002    
2003  18. If the last data line in a file for pcretest does not have a newline on  18. If the last data line in a file for pcretest does not have a newline on
2004      the end, a newline was missing in the output.      the end, a newline was missing in the output.
2005    
2006  19. The default pcre_chartables.c file recognizes only ASCII characters (values  19. The default pcre_chartables.c file recognizes only ASCII characters (values
2007      less than 128) in its various bitmaps. However, there is a facility for      less than 128) in its various bitmaps. However, there is a facility for
2008      generating tables according to the current locale when PCRE is compiled. It      generating tables according to the current locale when PCRE is compiled. It
2009      turns out that in some environments, 0x85 and 0xa0, which are Unicode space      turns out that in some environments, 0x85 and 0xa0, which are Unicode space
2010      characters, are recognized by isspace() and therefore were getting set in      characters, are recognized by isspace() and therefore were getting set in
2011      these tables. This caused a problem in UTF-8 mode when pcre_study() was      these tables, and indeed these tables seem to approximate to ISO 8859. This
2012      used to create a list of bytes that can start a match. For \s, it was      caused a problem in UTF-8 mode when pcre_study() was used to create a list
2013      including 0x85 and 0xa0, which of course cannot start UTF-8 characters. I      of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
2014      have changed the code so that only real ASCII characters (less than 128)      which of course cannot start UTF-8 characters. I have changed the code so
2015      are set in this case because the \s etc escapes are documented as      that only real ASCII characters (less than 128) and the correct starting
2016      recognizing only ASCII characters. (When PCRE_UCP is set - see 9 above -      bytes for UTF-8 encodings are set for characters greater than 127 when in
2017      the code is different altogether.)      UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
2018        altogether.)
2019    
2020    20. Added the /T option to pcretest so as to be able to run tests with non-
2021        standard character tables, thus making it possible to include the tests
2022        used for 19 above in the standard set of tests.
2023    
2024    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
2025        reference to a subpattern the other side of a comment that contains an
2026        opening parenthesis caused either an internal compiling error, or a
2027        reference to the wrong subpattern.
2028    
2029    
2030  Version 8.02 19-Mar-2010  Version 8.02 19-Mar-2010
# Line 477  Version 7.9 11-Apr-09 Line 2418  Version 7.9 11-Apr-09
2418  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
2419      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
2420      locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this      locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
2421      seems to be how GNU grep behaves.      seems to be how GNU grep behaves. [But see later change 40 for release
2422        8.33.]
2423    
2424  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
2425      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 720  Version 7.7 07-May-08 Line 2662  Version 7.7 07-May-08
2662      containing () gave an internal compiling error instead of "reference to      containing () gave an internal compiling error instead of "reference to
2663      non-existent subpattern". Fortunately, when the pattern did exist, the      non-existent subpattern". Fortunately, when the pattern did exist, the
2664      compiled code was correct. (When scanning forwards to check for the      compiled code was correct. (When scanning forwards to check for the
2665      existencd of the subpattern, it was treating the data ']' as terminating      existence of the subpattern, it was treating the data ']' as terminating
2666      the class, so got the count wrong. When actually compiling, the reference      the class, so got the count wrong. When actually compiling, the reference
2667      was subsequently set up correctly.)      was subsequently set up correctly.)
2668    

Legend:
Removed from v.538  
changed lines
  Added in v.1597

  ViewVC Help
Powered by ViewVC 1.1.5