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

Diff of /code/trunk/ChangeLog

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

revision 505 by ph10, Tue Mar 9 16:50:47 2010 UTC revision 1603 by ph10, Wed Oct 28 09:43:06 2015 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4  Version 8.02 01-Mar-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 27-October-2015
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    40. A match limit issue is fixed in JIT which was found by Karl Skomski
154        with a custom LLVM fuzzer.
155    
156    41. Remove the use of /dev/null in testdata/testinput2, because it doesn't
157        work under Windows. (Why has it taken so long for anyone to notice?)
158    
159    42. In a character class such as [\W\p{Any}] where both a negative-type escape
160        ("not a word character") and a property escape were present, the property
161        escape was being ignored.
162    
163    43. Fix crash caused by very long (*MARK) or (*THEN) names.
164    
165    
166    Version 8.37 28-April-2015
167    --------------------------
168    
169    1.  When an (*ACCEPT) is triggered inside capturing parentheses, it arranges
170        for those parentheses to be closed with whatever has been captured so far.
171        However, it was failing to mark any other groups between the hightest
172        capture so far and the currrent group as "unset". Thus, the ovector for
173        those groups contained whatever was previously there. An example is the
174        pattern /(x)|((*ACCEPT))/ when matched against "abcd".
175    
176    2.  If an assertion condition was quantified with a minimum of zero (an odd
177        thing to do, but it happened), SIGSEGV or other misbehaviour could occur.
178    
179    3.  If a pattern in pcretest input had the P (POSIX) modifier followed by an
180        unrecognized modifier, a crash could occur.
181    
182    4.  An attempt to do global matching in pcretest with a zero-length ovector
183        caused a crash.
184    
185    5.  Fixed a memory leak during matching that could occur for a subpattern
186        subroutine call (recursive or otherwise) if the number of captured groups
187        that had to be saved was greater than ten.
188    
189    6.  Catch a bad opcode during auto-possessification after compiling a bad UTF
190        string with NO_UTF_CHECK. This is a tidyup, not a bug fix, as passing bad
191        UTF with NO_UTF_CHECK is documented as having an undefined outcome.
192    
193    7.  A UTF pattern containing a "not" match of a non-ASCII character and a
194        subroutine reference could loop at compile time. Example: /[^\xff]((?1))/.
195    
196    8. When a pattern is compiled, it remembers the highest back reference so that
197       when matching, if the ovector is too small, extra memory can be obtained to
198       use instead. A conditional subpattern whose condition is a check on a
199       capture having happened, such as, for example in the pattern
200       /^(?:(a)|b)(?(1)A|B)/, is another kind of back reference, but it was not
201       setting the highest backreference number. This mattered only if pcre_exec()
202       was called with an ovector that was too small to hold the capture, and there
203       was no other kind of back reference (a situation which is probably quite
204       rare). The effect of the bug was that the condition was always treated as
205       FALSE when the capture could not be consulted, leading to a incorrect
206       behaviour by pcre_exec(). This bug has been fixed.
207    
208    9. A reference to a duplicated named group (either a back reference or a test
209       for being set in a conditional) that occurred in a part of the pattern where
210       PCRE_DUPNAMES was not set caused the amount of memory needed for the pattern
211       to be incorrectly calculated, leading to overwriting.
212    
213    10. A mutually recursive set of back references such as (\2)(\1) caused a
214        segfault at study time (while trying to find the minimum matching length).
215        The infinite loop is now broken (with the minimum length unset, that is,
216        zero).
217    
218    11. If an assertion that was used as a condition was quantified with a minimum
219        of zero, matching went wrong. In particular, if the whole group had
220        unlimited repetition and could match an empty string, a segfault was
221        likely. The pattern (?(?=0)?)+ is an example that caused this. Perl allows
222        assertions to be quantified, but not if they are being used as conditions,
223        so the above pattern is faulted by Perl. PCRE has now been changed so that
224        it also rejects such patterns.
225    
226    12. A possessive capturing group such as (a)*+ with a minimum repeat of zero
227        failed to allow the zero-repeat case if pcre2_exec() was called with an
228        ovector too small to capture the group.
229    
230    13. Fixed two bugs in pcretest that were discovered by fuzzing and reported by
231        Red Hat Product Security:
232    
233        (a) A crash if /K and /F were both set with the option to save the compiled
234        pattern.
235    
236        (b) Another crash if the option to print captured substrings in a callout
237        was combined with setting a null ovector, for example \O\C+ as a subject
238        string.
239    
240    14. A pattern such as "((?2){0,1999}())?", which has a group containing a
241        forward reference repeated a large (but limited) number of times within a
242        repeated outer group that has a zero minimum quantifier, caused incorrect
243        code to be compiled, leading to the error "internal error:
244        previously-checked referenced subpattern not found" when an incorrect
245        memory address was read. This bug was reported as "heap overflow",
246        discovered by Kai Lu of Fortinet's FortiGuard Labs and given the CVE number
247        CVE-2015-2325.
248    
249    23. A pattern such as "((?+1)(\1))/" containing a forward reference subroutine
250        call within a group that also contained a recursive back reference caused
251        incorrect code to be compiled. This bug was reported as "heap overflow",
252        discovered by Kai Lu of Fortinet's FortiGuard Labs, and given the CVE
253        number CVE-2015-2326.
254    
255    24. Computing the size of the JIT read-only data in advance has been a source
256        of various issues, and new ones are still appear unfortunately. To fix
257        existing and future issues, size computation is eliminated from the code,
258        and replaced by on-demand memory allocation.
259    
260    25. A pattern such as /(?i)[A-`]/, where characters in the other case are
261        adjacent to the end of the range, and the range contained characters with
262        more than one other case, caused incorrect behaviour when compiled in UTF
263        mode. In that example, the range a-j was left out of the class.
264    
265    26. Fix JIT compilation of conditional blocks, which assertion
266        is converted to (*FAIL). E.g: /(?(?!))/.
267    
268    27. The pattern /(?(?!)^)/ caused references to random memory. This bug was
269        discovered by the LLVM fuzzer.
270    
271    28. The assertion (?!) is optimized to (*FAIL). This was not handled correctly
272        when this assertion was used as a condition, for example (?(?!)a|b). In
273        pcre2_match() it worked by luck; in pcre2_dfa_match() it gave an incorrect
274        error about an unsupported item.
275    
276    29. For some types of pattern, for example /Z*(|d*){216}/, the auto-
277        possessification code could take exponential time to complete. A recursion
278        depth limit of 1000 has been imposed to limit the resources used by this
279        optimization.
280    
281    30. A pattern such as /(*UTF)[\S\V\H]/, which contains a negated special class
282        such as \S in non-UCP mode, explicit wide characters (> 255) can be ignored
283        because \S ensures they are all in the class. The code for doing this was
284        interacting badly with the code for computing the amount of space needed to
285        compile the pattern, leading to a buffer overflow. This bug was discovered
286        by the LLVM fuzzer.
287    
288    31. A pattern such as /((?2)+)((?1))/ which has mutual recursion nested inside
289        other kinds of group caused stack overflow at compile time. This bug was
290        discovered by the LLVM fuzzer.
291    
292    32. A pattern such as /(?1)(?#?'){8}(a)/ which had a parenthesized comment
293        between a subroutine call and its quantifier was incorrectly compiled,
294        leading to buffer overflow or other errors. This bug was discovered by the
295        LLVM fuzzer.
296    
297    33. The illegal pattern /(?(?<E>.*!.*)?)/ was not being diagnosed as missing an
298        assertion after (?(. The code was failing to check the character after
299        (?(?< for the ! or = that would indicate a lookbehind assertion. This bug
300        was discovered by the LLVM fuzzer.
301    
302    34. A pattern such as /X((?2)()*+){2}+/ which has a possessive quantifier with
303        a fixed maximum following a group that contains a subroutine reference was
304        incorrectly compiled and could trigger buffer overflow. This bug was
305        discovered by the LLVM fuzzer.
306    
307    35. A mutual recursion within a lookbehind assertion such as (?<=((?2))((?1)))
308        caused a stack overflow instead of the diagnosis of a non-fixed length
309        lookbehind assertion. This bug was discovered by the LLVM fuzzer.
310    
311    36. The use of \K in a positive lookbehind assertion in a non-anchored pattern
312        (e.g. /(?<=\Ka)/) could make pcregrep loop.
313    
314    37. There was a similar problem to 36 in pcretest for global matches.
315    
316    38. If a greedy quantified \X was preceded by \C in UTF mode (e.g. \C\X*),
317        and a subsequent item in the pattern caused a non-match, backtracking over
318        the repeated \X did not stop, but carried on past the start of the subject,
319        causing reference to random memory and/or a segfault. There were also some
320        other cases where backtracking after \C could crash. This set of bugs was
321        discovered by the LLVM fuzzer.
322    
323    39. The function for finding the minimum length of a matching string could take
324        a very long time if mutual recursion was present many times in a pattern,
325        for example, /((?2){73}(?2))((?1))/. A better mutual recursion detection
326        method has been implemented. This infelicity was discovered by the LLVM
327        fuzzer.
328    
329    40. Static linking against the PCRE library using the pkg-config module was
330        failing on missing pthread symbols.
331    
332    
333    Version 8.36 26-September-2014
334    ------------------------------
335    
336    1.  Got rid of some compiler warnings in the C++ modules that were shown up by
337        -Wmissing-field-initializers and -Wunused-parameter.
338    
339    2.  The tests for quantifiers being too big (greater than 65535) were being
340        applied after reading the number, and stupidly assuming that integer
341        overflow would give a negative number. The tests are now applied as the
342        numbers are read.
343    
344    3.  Tidy code in pcre_exec.c where two branches that used to be different are
345        now the same.
346    
347    4.  The JIT compiler did not generate match limit checks for certain
348        bracketed expressions with quantifiers. This may lead to exponential
349        backtracking, instead of returning with PCRE_ERROR_MATCHLIMIT. This
350        issue should be resolved now.
351    
352    5.  Fixed an issue, which occures when nested alternatives are optimized
353        with table jumps.
354    
355    6.  Inserted two casts and changed some ints to size_t in the light of some
356        reported 64-bit compiler warnings (Bugzilla 1477).
357    
358    7.  Fixed a bug concerned with zero-minimum possessive groups that could match
359        an empty string, which sometimes were behaving incorrectly in the
360        interpreter (though correctly in the JIT matcher). This pcretest input is
361        an example:
362    
363          '\A(?:[^"]++|"(?:[^"]*+|"")*+")++'
364          NON QUOTED "QUOT""ED" AFTER "NOT MATCHED
365    
366        the interpreter was reporting a match of 'NON QUOTED ' only, whereas the
367        JIT matcher and Perl both matched 'NON QUOTED "QUOT""ED" AFTER '. The test
368        for an empty string was breaking the inner loop and carrying on at a lower
369        level, when possessive repeated groups should always return to a higher
370        level as they have no backtrack points in them. The empty string test now
371        occurs at the outer level.
372    
373    8.  Fixed a bug that was incorrectly auto-possessifying \w+ in the pattern
374        ^\w+(?>\s*)(?<=\w) which caused it not to match "test test".
375    
376    9.  Give a compile-time error for \o{} (as Perl does) and for \x{} (which Perl
377        doesn't).
378    
379    10. Change 8.34/15 introduced a bug that caused the amount of memory needed
380        to hold a pattern to be incorrectly computed (too small) when there were
381        named back references to duplicated names. This could cause "internal
382        error: code overflow" or "double free or corruption" or other memory
383        handling errors.
384    
385    11. When named subpatterns had the same prefixes, back references could be
386        confused. For example, in this pattern:
387    
388          /(?P<Name>a)?(?P<Name2>b)?(?(<Name>)c|d)*l/
389    
390        the reference to 'Name' was incorrectly treated as a reference to a
391        duplicate name.
392    
393    12. A pattern such as /^s?c/mi8 where the optional character has more than
394        one "other case" was incorrectly compiled such that it would only try to
395        match starting at "c".
396    
397    13. When a pattern starting with \s was studied, VT was not included in the
398        list of possible starting characters; this should have been part of the
399        8.34/18 patch.
400    
401    14. If a character class started [\Qx]... where x is any character, the class
402        was incorrectly terminated at the ].
403    
404    15. If a pattern that started with a caseless match for a character with more
405        than one "other case" was studied, PCRE did not set up the starting code
406        unit bit map for the list of possible characters. Now it does. This is an
407        optimization improvement, not a bug fix.
408    
409    16. The Unicode data tables have been updated to Unicode 7.0.0.
410    
411    17. Fixed a number of memory leaks in pcregrep.
412    
413    18. Avoid a compiler warning (from some compilers) for a function call with
414        a cast that removes "const" from an lvalue by using an intermediate
415        variable (to which the compiler does not object).
416    
417    19. Incorrect code was compiled if a group that contained an internal recursive
418        back reference was optional (had quantifier with a minimum of zero). This
419        example compiled incorrect code: /(((a\2)|(a*)\g<-1>))*/ and other examples
420        caused segmentation faults because of stack overflows at compile time.
421    
422    20. A pattern such as /((?(R)a|(?1)))+/, which contains a recursion within a
423        group that is quantified with an indefinite repeat, caused a compile-time
424        loop which used up all the system stack and provoked a segmentation fault.
425        This was not the same bug as 19 above.
426    
427    21. Add PCRECPP_EXP_DECL declaration to operator<< in pcre_stringpiece.h.
428        Patch by Mike Frysinger.
429    
430    
431    Version 8.35 04-April-2014
432    --------------------------
433    
434    1.  A new flag is set, when property checks are present in an XCLASS.
435        When this flag is not set, PCRE can perform certain optimizations
436        such as studying these XCLASS-es.
437    
438    2.  The auto-possessification of character sets were improved: a normal
439        and an extended character set can be compared now. Furthermore
440        the JIT compiler optimizes more character set checks.
441    
442    3.  Got rid of some compiler warnings for potentially uninitialized variables
443        that show up only when compiled with -O2.
444    
445    4.  A pattern such as (?=ab\K) that uses \K in an assertion can set the start
446        of a match later then the end of the match. The pcretest program was not
447        handling the case sensibly - it was outputting from the start to the next
448        binary zero. It now reports this situation in a message, and outputs the
449        text from the end to the start.
450    
451    5.  Fast forward search is improved in JIT. Instead of the first three
452        characters, any three characters with fixed position can be searched.
453        Search order: first, last, middle.
454    
455    6.  Improve character range checks in JIT. Characters are read by an inprecise
456        function now, which returns with an unknown value if the character code is
457        above a certain threshold (e.g: 256). The only limitation is that the value
458        must be bigger than the threshold as well. This function is useful when
459        the characters above the threshold are handled in the same way.
460    
461    7.  The macros whose names start with RAWUCHAR are placeholders for a future
462        mode in which only the bottom 21 bits of 32-bit data items are used. To
463        make this more memorable for those maintaining the code, the names have
464        been changed to start with UCHAR21, and an extensive comment has been added
465        to their definition.
466    
467    8.  Add missing (new) files sljitNativeTILEGX.c and sljitNativeTILEGX-encoder.c
468        to the export list in Makefile.am (they were accidentally omitted from the
469        8.34 tarball).
470    
471    9.  The informational output from pcretest used the phrase "starting byte set"
472        which is inappropriate for the 16-bit and 32-bit libraries. As the output
473        for "first char" and "need char" really means "non-UTF-char", I've changed
474        "byte" to "char", and slightly reworded the output. The documentation about
475        these values has also been (I hope) clarified.
476    
477    10. Another JIT related optimization: use table jumps for selecting the correct
478        backtracking path, when more than four alternatives are present inside a
479        bracket.
480    
481    11. Empty match is not possible, when the minimum length is greater than zero,
482        and there is no \K in the pattern. JIT should avoid empty match checks in
483        such cases.
484    
485    12. In a caseless character class with UCP support, when a character with more
486        than one alternative case was not the first character of a range, not all
487        the alternative cases were added to the class. For example, s and \x{17f}
488        are both alternative cases for S: the class [RST] was handled correctly,
489        but [R-T] was not.
490    
491    13. The configure.ac file always checked for pthread support when JIT was
492        enabled. This is not used in Windows, so I have put this test inside a
493        check for the presence of windows.h (which was already tested for).
494    
495    14. Improve pattern prefix search by a simplified Boyer-Moore algorithm in JIT.
496        The algorithm provides a way to skip certain starting offsets, and usually
497        faster than linear prefix searches.
498    
499    15. Change 13 for 8.20 updated RunTest to check for the 'fr' locale as well
500        as for 'fr_FR' and 'french'. For some reason, however, it then used the
501        Windows-specific input and output files, which have 'french' screwed in.
502        So this could never have worked. One of the problems with locales is that
503        they aren't always the same. I have now updated RunTest so that it checks
504        the output of the locale test (test 3) against three different output
505        files, and it allows the test to pass if any one of them matches. With luck
506        this should make the test pass on some versions of Solaris where it was
507        failing. Because of the uncertainty, the script did not used to stop if
508        test 3 failed; it now does. If further versions of a French locale ever
509        come to light, they can now easily be added.
510    
511    16. If --with-pcregrep-bufsize was given a non-integer value such as "50K",
512        there was a message during ./configure, but it did not stop. This now
513        provokes an error. The invalid example in README has been corrected.
514        If a value less than the minimum is given, the minimum value has always
515        been used, but now a warning is given.
516    
517    17. If --enable-bsr-anycrlf was set, the special 16/32-bit test failed. This
518        was a bug in the test system, which is now fixed. Also, the list of various
519        configurations that are tested for each release did not have one with both
520        16/32 bits and --enable-bar-anycrlf. It now does.
521    
522    18. pcretest was missing "-C bsr" for displaying the \R default setting.
523    
524    19. Little endian PowerPC systems are supported now by the JIT compiler.
525    
526    20. The fast forward newline mechanism could enter to an infinite loop on
527        certain invalid UTF-8 input. Although we don't support these cases
528        this issue can be fixed by a performance optimization.
529    
530    21. Change 33 of 8.34 is not sufficient to ensure stack safety because it does
531        not take account if existing stack usage. There is now a new global
532        variable called pcre_stack_guard that can be set to point to an external
533        function to check stack availability. It is called at the start of
534        processing every parenthesized group.
535    
536    22. A typo in the code meant that in ungreedy mode the max/min qualifier
537        behaved like a min-possessive qualifier, and, for example, /a{1,3}b/U did
538        not match "ab".
539    
540    23. When UTF was disabled, the JIT program reported some incorrect compile
541        errors. These messages are silenced now.
542    
543    24. Experimental support for ARM-64 and MIPS-64 has been added to the JIT
544        compiler.
545    
546    25. Change all the temporary files used in RunGrepTest to be different to those
547        used by RunTest so that the tests can be run simultaneously, for example by
548        "make -j check".
549    
550    
551    Version 8.34 15-December-2013
552    -----------------------------
553    
554    1.  Add pcre[16|32]_jit_free_unused_memory to forcibly free unused JIT
555        executable memory. Patch inspired by Carsten Klein.
556    
557    2.  ./configure --enable-coverage defined SUPPORT_GCOV in config.h, although
558        this macro is never tested and has no effect, because the work to support
559        coverage involves only compiling and linking options and special targets in
560        the Makefile. The comment in config.h implied that defining the macro would
561        enable coverage support, which is totally false. There was also support for
562        setting this macro in the CMake files (my fault, I just copied it from
563        configure). SUPPORT_GCOV has now been removed.
564    
565    3.  Make a small performance improvement in strlen16() and strlen32() in
566        pcretest.
567    
568    4.  Change 36 for 8.33 left some unreachable statements in pcre_exec.c,
569        detected by the Solaris compiler (gcc doesn't seem to be able to diagnose
570        these cases). There was also one in pcretest.c.
571    
572    5.  Cleaned up a "may be uninitialized" compiler warning in pcre_exec.c.
573    
574    6.  In UTF mode, the code for checking whether a group could match an empty
575        string (which is used for indefinitely repeated groups to allow for
576        breaking an infinite loop) was broken when the group contained a repeated
577        negated single-character class with a character that occupied more than one
578        data item and had a minimum repetition of zero (for example, [^\x{100}]* in
579        UTF-8 mode). The effect was undefined: the group might or might not be
580        deemed as matching an empty string, or the program might have crashed.
581    
582    7.  The code for checking whether a group could match an empty string was not
583        recognizing that \h, \H, \v, \V, and \R must match a character.
584    
585    8.  Implemented PCRE_INFO_MATCH_EMPTY, which yields 1 if the pattern can match
586        an empty string. If it can, pcretest shows this in its information output.
587    
588    9.  Fixed two related bugs that applied to Unicode extended grapheme clusters
589        that were repeated with a maximizing qualifier (e.g. \X* or \X{2,5}) when
590        matched by pcre_exec() without using JIT:
591    
592        (a) If the rest of the pattern did not match after a maximal run of
593            grapheme clusters, the code for backing up to try with fewer of them
594            did not always back up over a full grapheme when characters that do not
595            have the modifier quality were involved, e.g. Hangul syllables.
596    
597        (b) If the match point in a subject started with modifier character, and
598            there was no match, the code could incorrectly back up beyond the match
599            point, and potentially beyond the first character in the subject,
600            leading to a segfault or an incorrect match result.
601    
602    10. A conditional group with an assertion condition could lead to PCRE
603        recording an incorrect first data item for a match if no other first data
604        item was recorded. For example, the pattern (?(?=ab)ab) recorded "a" as a
605        first data item, and therefore matched "ca" after "c" instead of at the
606        start.
607    
608    11. Change 40 for 8.33 (allowing pcregrep to find empty strings) showed up a
609        bug that caused the command "echo a | ./pcregrep -M '|a'" to loop.
610    
611    12. The source of pcregrep now includes z/OS-specific code so that it can be
612        compiled for z/OS as part of the special z/OS distribution.
613    
614    13. Added the -T and -TM options to pcretest.
615    
616    14. The code in pcre_compile.c for creating the table of named capturing groups
617        has been refactored. Instead of creating the table dynamically during the
618        actual compiling pass, the information is remembered during the pre-compile
619        pass (on the stack unless there are more than 20 named groups, in which
620        case malloc() is used) and the whole table is created before the actual
621        compile happens. This has simplified the code (it is now nearly 150 lines
622        shorter) and prepared the way for better handling of references to groups
623        with duplicate names.
624    
625    15. A back reference to a named subpattern when there is more than one of the
626        same name now checks them in the order in which they appear in the pattern.
627        The first one that is set is used for the reference. Previously only the
628        first one was inspected. This change makes PCRE more compatible with Perl.
629    
630    16. Unicode character properties were updated from Unicode 6.3.0.
631    
632    17. The compile-time code for auto-possessification has been refactored, based
633        on a patch by Zoltan Herczeg. It now happens after instead of during
634        compilation. The code is cleaner, and more cases are handled. The option
635        PCRE_NO_AUTO_POSSESS is added for testing purposes, and the -O and /O
636        options in pcretest are provided to set it. It can also be set by
637        (*NO_AUTO_POSSESS) at the start of a pattern.
638    
639    18. The character VT has been added to the default ("C" locale) set of
640        characters that match \s and are generally treated as white space,
641        following this same change in Perl 5.18. There is now no difference between
642        "Perl space" and "POSIX space". Whether VT is treated as white space in
643        other locales depends on the locale.
644    
645    19. The code for checking named groups as conditions, either for being set or
646        for being recursed, has been refactored (this is related to 14 and 15
647        above). Processing unduplicated named groups should now be as fast at
648        numerical groups, and processing duplicated groups should be faster than
649        before.
650    
651    20. Two patches to the CMake build system, by Alexander Barkov:
652    
653          (1) Replace the "source" command by "." in CMakeLists.txt because
654              "source" is a bash-ism.
655    
656          (2) Add missing HAVE_STDINT_H and HAVE_INTTYPES_H to config-cmake.h.in;
657              without these the CMake build does not work on Solaris.
658    
659    21. Perl has changed its handling of \8 and \9. If there is no previously
660        encountered capturing group of those numbers, they are treated as the
661        literal characters 8 and 9 instead of a binary zero followed by the
662        literals. PCRE now does the same.
663    
664    22. Following Perl, added \o{} to specify codepoints in octal, making it
665        possible to specify values greater than 0777 and also making them
666        unambiguous.
667    
668    23. Perl now gives an error for missing closing braces after \x{... instead of
669        treating the string as literal. PCRE now does the same.
670    
671    24. RunTest used to grumble if an inappropriate test was selected explicitly,
672        but just skip it when running all tests. This make it awkward to run ranges
673        of tests when one of them was inappropriate. Now it just skips any
674        inappropriate tests, as it always did when running all tests.
675    
676    25. If PCRE_AUTO_CALLOUT and PCRE_UCP were set for a pattern that contained
677        character types such as \d or \w, too many callouts were inserted, and the
678        data that they returned was rubbish.
679    
680    26. In UCP mode, \s was not matching two of the characters that Perl matches,
681        namely NEL (U+0085) and MONGOLIAN VOWEL SEPARATOR (U+180E), though they
682        were matched by \h. The code has now been refactored so that the lists of
683        the horizontal and vertical whitespace characters used for \h and \v (which
684        are defined only in one place) are now also used for \s.
685    
686    27. Add JIT support for the 64 bit TileGX architecture.
687        Patch by Jiong Wang (Tilera Corporation).
688    
689    28. Possessive quantifiers for classes (both explicit and automatically
690        generated) now use special opcodes instead of wrapping in ONCE brackets.
691    
692    29. Whereas an item such as A{4}+ ignored the possessivenes of the quantifier
693        (because it's meaningless), this was not happening when PCRE_CASELESS was
694        set. Not wrong, but inefficient.
695    
696    30. Updated perltest.pl to add /u (force Unicode mode) when /W (use Unicode
697        properties for \w, \d, etc) is present in a test regex. Otherwise if the
698        test contains no characters greater than 255, Perl doesn't realise it
699        should be using Unicode semantics.
700    
701    31. Upgraded the handling of the POSIX classes [:graph:], [:print:], and
702        [:punct:] when PCRE_UCP is set so as to include the same characters as Perl
703        does in Unicode mode.
704    
705    32. Added the "forbid" facility to pcretest so that putting tests into the
706        wrong test files can sometimes be quickly detected.
707    
708    33. There is now a limit (default 250) on the depth of nesting of parentheses.
709        This limit is imposed to control the amount of system stack used at compile
710        time. It can be changed at build time by --with-parens-nest-limit=xxx or
711        the equivalent in CMake.
712    
713    34. Character classes such as [A-\d] or [a-[:digit:]] now cause compile-time
714        errors. Perl warns for these when in warning mode, but PCRE has no facility
715        for giving warnings.
716    
717    35. Change 34 for 8.13 allowed quantifiers on assertions, because Perl does.
718        However, this was not working for (?!) because it is optimized to (*FAIL),
719        for which PCRE does not allow quantifiers. The optimization is now disabled
720        when a quantifier follows (?!). I can't see any use for this, but it makes
721        things uniform.
722    
723    36. Perl no longer allows group names to start with digits, so I have made this
724        change also in PCRE. It simplifies the code a bit.
725    
726    37. In extended mode, Perl ignores spaces before a + that indicates a
727        possessive quantifier. PCRE allowed a space before the quantifier, but not
728        before the possessive +. It now does.
729    
730    38. The use of \K (reset reported match start) within a repeated possessive
731        group such as (a\Kb)*+ was not working.
732    
733    40. Document that the same character tables must be used at compile time and
734        run time, and that the facility to pass tables to pcre_exec() and
735        pcre_dfa_exec() is for use only with saved/restored patterns.
736    
737    41. Applied Jeff Trawick's patch CMakeLists.txt, which "provides two new
738        features for Builds with MSVC:
739    
740        1. Support pcre.rc and/or pcreposix.rc (as is already done for MinGW
741           builds). The .rc files can be used to set FileDescription and many other
742           attributes.
743    
744        2. Add an option (-DINSTALL_MSVC_PDB) to enable installation of .pdb files.
745           This allows higher-level build scripts which want .pdb files to avoid
746           hard-coding the exact files needed."
747    
748    42. Added support for [[:<:]] and [[:>:]] as used in the BSD POSIX library to
749        mean "start of word" and "end of word", respectively, as a transition aid.
750    
751    43. A minimizing repeat of a class containing codepoints greater than 255 in
752        non-UTF 16-bit or 32-bit modes caused an internal error when PCRE was
753        compiled to use the heap for recursion.
754    
755    44. Got rid of some compiler warnings for unused variables when UTF but not UCP
756        is configured.
757    
758    
759    Version 8.33 28-May-2013
760    ------------------------
761    
762    1.  Added 'U' to some constants that are compared to unsigned integers, to
763        avoid compiler signed/unsigned warnings. Added (int) casts to unsigned
764        variables that are added to signed variables, to ensure the result is
765        signed and can be negated.
766    
767    2.  Applied patch by Daniel Richard G for quashing MSVC warnings to the
768        CMake config files.
769    
770    3.  Revise the creation of config.h.generic so that all boolean macros are
771        #undefined, whereas non-boolean macros are #ifndef/#endif-ed. This makes
772        overriding via -D on the command line possible.
773    
774    4.  Changing the definition of the variable "op" in pcre_exec.c from pcre_uchar
775        to unsigned int is reported to make a quite noticeable speed difference in
776        a specific Windows environment. Testing on Linux did also appear to show
777        some benefit (and it is clearly not harmful). Also fixed the definition of
778        Xop which should be unsigned.
779    
780    5.  Related to (4), changing the definition of the intermediate variable cc
781        in repeated character loops from pcre_uchar to pcre_uint32 also gave speed
782        improvements.
783    
784    6.  Fix forward search in JIT when link size is 3 or greater. Also removed some
785        unnecessary spaces.
786    
787    7.  Adjust autogen.sh and configure.ac to lose warnings given by automake 1.12
788        and later.
789    
790    8.  Fix two buffer over read issues in 16 and 32 bit modes. Affects JIT only.
791    
792    9.  Optimizing fast_forward_start_bits in JIT.
793    
794    10. Adding support for callouts in JIT, and fixing some issues revealed
795        during this work. Namely:
796    
797        (a) Unoptimized capturing brackets incorrectly reset on backtrack.
798    
799        (b) Minimum length was not checked before the matching is started.
800    
801    11. The value of capture_last that is passed to callouts was incorrect in some
802        cases when there was a capture on one path that was subsequently abandoned
803        after a backtrack. Also, the capture_last value is now reset after a
804        recursion, since all captures are also reset in this case.
805    
806    12. The interpreter no longer returns the "too many substrings" error in the
807        case when an overflowing capture is in a branch that is subsequently
808        abandoned after a backtrack.
809    
810    13. In the pathological case when an offset vector of size 2 is used, pcretest
811        now prints out the matched string after a yield of 0 or 1.
812    
813    14. Inlining subpatterns in recursions, when certain conditions are fulfilled.
814        Only supported by the JIT compiler at the moment.
815    
816    15. JIT compiler now supports 32 bit Macs thanks to Lawrence Velazquez.
817    
818    16. Partial matches now set offsets[2] to the "bumpalong" value, that is, the
819        offset of the starting point of the matching process, provided the offsets
820        vector is large enough.
821    
822    17. The \A escape now records a lookbehind value of 1, though its execution
823        does not actually inspect the previous character. This is to ensure that,
824        in partial multi-segment matching, at least one character from the old
825        segment is retained when a new segment is processed. Otherwise, if there
826        are no lookbehinds in the pattern, \A might match incorrectly at the start
827        of a new segment.
828    
829    18. Added some #ifdef __VMS code into pcretest.c to help VMS implementations.
830    
831    19. Redefined some pcre_uchar variables in pcre_exec.c as pcre_uint32; this
832        gives some modest performance improvement in 8-bit mode.
833    
834    20. Added the PCRE-specific property \p{Xuc} for matching characters that can
835        be expressed in certain programming languages using Universal Character
836        Names.
837    
838    21. Unicode validation has been updated in the light of Unicode Corrigendum #9,
839        which points out that "non characters" are not "characters that may not
840        appear in Unicode strings" but rather "characters that are reserved for
841        internal use and have only local meaning".
842    
843    22. When a pattern was compiled with automatic callouts (PCRE_AUTO_CALLOUT) and
844        there was a conditional group that depended on an assertion, if the
845        assertion was false, the callout that immediately followed the alternation
846        in the condition was skipped when pcre_exec() was used for matching.
847    
848    23. Allow an explicit callout to be inserted before an assertion that is the
849        condition for a conditional group, for compatibility with automatic
850        callouts, which always insert a callout at this point.
851    
852    24. In 8.31, (*COMMIT) was confined to within a recursive subpattern. Perl also
853        confines (*SKIP) and (*PRUNE) in the same way, and this has now been done.
854    
855    25. (*PRUNE) is now supported by the JIT compiler.
856    
857    26. Fix infinite loop when /(?<=(*SKIP)ac)a/ is matched against aa.
858    
859    27. Fix the case where there are two or more SKIPs with arguments that may be
860        ignored.
861    
862    28. (*SKIP) is now supported by the JIT compiler.
863    
864    29. (*THEN) is now supported by the JIT compiler.
865    
866    30. Update RunTest with additional test selector options.
867    
868    31. The way PCRE handles backtracking verbs has been changed in two ways.
869    
870        (1) Previously, in something like (*COMMIT)(*SKIP), COMMIT would override
871        SKIP. Now, PCRE acts on whichever backtracking verb is reached first by
872        backtracking. In some cases this makes it more Perl-compatible, but Perl's
873        rather obscure rules do not always do the same thing.
874    
875        (2) Previously, backtracking verbs were confined within assertions. This is
876        no longer the case for positive assertions, except for (*ACCEPT). Again,
877        this sometimes improves Perl compatibility, and sometimes does not.
878    
879    32. A number of tests that were in test 2 because Perl did things differently
880        have been moved to test 1, because either Perl or PCRE has changed, and
881        these tests are now compatible.
882    
883    32. Backtracking control verbs are now handled in the same way in JIT and
884        interpreter.
885    
886    33. An opening parenthesis in a MARK/PRUNE/SKIP/THEN name in a pattern that
887        contained a forward subroutine reference caused a compile error.
888    
889    34. Auto-detect and optimize limited repetitions in JIT.
890    
891    35. Implement PCRE_NEVER_UTF to lock out the use of UTF, in particular,
892        blocking (*UTF) etc.
893    
894    36. In the interpreter, maximizing pattern repetitions for characters and
895        character types now use tail recursion, which reduces stack usage.
896    
897    37. The value of the max lookbehind was not correctly preserved if a compiled
898        and saved regex was reloaded on a host of different endianness.
899    
900    38. Implemented (*LIMIT_MATCH) and (*LIMIT_RECURSION). As part of the extension
901        of the compiled pattern block, expand the flags field from 16 to 32 bits
902        because it was almost full.
903    
904    39. Try madvise first before posix_madvise.
905    
906    40. Change 7 for PCRE 7.9 made it impossible for pcregrep to find empty lines
907        with a pattern such as ^$. It has taken 4 years for anybody to notice! The
908        original change locked out all matches of empty strings. This has been
909        changed so that one match of an empty string per line is recognized.
910        Subsequent searches on the same line (for colouring or for --only-matching,
911        for example) do not recognize empty strings.
912    
913    41. Applied a user patch to fix a number of spelling mistakes in comments.
914    
915    42. Data lines longer than 65536 caused pcretest to crash.
916    
917    43. Clarified the data type for length and startoffset arguments for pcre_exec
918        and pcre_dfa_exec in the function-specific man pages, where they were
919        explicitly stated to be in bytes, never having been updated. I also added
920        some clarification to the pcreapi man page.
921    
922    44. A call to pcre_dfa_exec() with an output vector size less than 2 caused
923        a segmentation fault.
924    
925    
926    Version 8.32 30-November-2012
927    -----------------------------
928    
929    1.  Improved JIT compiler optimizations for first character search and single
930        character iterators.
931    
932    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
933        Patch by Daniel Richard G.
934    
935    3.  Single character iterator optimizations in the JIT compiler.
936    
937    4.  Improved JIT compiler optimizations for character ranges.
938    
939    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
940        Reported by Giuseppe D'Angelo.
941    
942    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
943        of a line, was being set incorrectly in cases where .* appeared inside
944        atomic brackets at the start of a pattern, or where there was a subsequent
945        *PRUNE or *SKIP.
946    
947    7.  Improved instruction cache flush for POWER/PowerPC.
948        Patch by Daniel Richard G.
949    
950    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
951        grep:
952    
953        (a) There is now no limit to the number of patterns to be matched.
954    
955        (b) An error is given if a pattern is too long.
956    
957        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
958            are now supported.
959    
960        (d) --exclude-from and --include-from (multiple use) have been added.
961    
962        (e) Exclusions and inclusions now apply to all files and directories, not
963            just to those obtained from scanning a directory recursively.
964    
965        (f) Multiple uses of -f and --file-list are now supported.
966    
967        (g) In a Windows environment, the default for -d has been changed from
968            "read" (the GNU grep default) to "skip", because otherwise the presence
969            of a directory in the file list provokes an error.
970    
971        (h) The documentation has been revised and clarified in places.
972    
973    9.  Improve the matching speed of capturing brackets.
974    
975    10. Changed the meaning of \X so that it now matches a Unicode extended
976        grapheme cluster.
977    
978    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
979        out POSIX threads when JIT support is configured.
980    
981    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
982    
983    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
984        structure could go wrong in environments where size_t is not the same size
985        as int.
986    
987    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
988        set.
989    
990    15. The EBCDIC support had decayed; later updates to the code had included
991        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
992        general tidy up of EBCDIC-related issues, and the documentation was also
993        not quite right. There is now a test that can be run on ASCII systems to
994        check some of the EBCDIC-related things (but is it not a full test).
995    
996    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
997        in a small tidy to the code.
998    
999    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
1000    
1001    18. If the --only-matching (-o) option in pcregrep is specified multiple
1002        times, each one causes appropriate output. For example, -o1 -o2 outputs the
1003        substrings matched by the 1st and 2nd capturing parentheses. A separating
1004        string can be specified by --om-separator (default empty).
1005    
1006    19. Improving the first n character searches.
1007    
1008    20. Turn case lists for horizontal and vertical white space into macros so that
1009        they are defined only once.
1010    
1011    21. This set of changes together give more compatible Unicode case-folding
1012        behaviour for characters that have more than one other case when UCP
1013        support is available.
1014    
1015        (a) The Unicode property table now has offsets into a new table of sets of
1016            three or more characters that are case-equivalent. The MultiStage2.py
1017            script that generates these tables (the pcre_ucd.c file) now scans
1018            CaseFolding.txt instead of UnicodeData.txt for character case
1019            information.
1020    
1021        (b) The code for adding characters or ranges of characters to a character
1022            class has been abstracted into a generalized function that also handles
1023            case-independence. In UTF-mode with UCP support, this uses the new data
1024            to handle characters with more than one other case.
1025    
1026        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
1027            whose other case is greater than 256 are now correctly matched
1028            caselessly. Previously, the high codepoint matched the low one, but not
1029            vice versa.
1030    
1031        (d) The processing of \h, \H, \v, and \ in character classes now makes use
1032            of the new class addition function, using character lists defined as
1033            macros alongside the case definitions of 20 above.
1034    
1035        (e) Caseless back references now work with characters that have more than
1036            one other case.
1037    
1038        (f) General caseless matching of characters with more than one other case
1039            is supported.
1040    
1041    22. Unicode character properties were updated from Unicode 6.2.0
1042    
1043    23. Improved CMake support under Windows. Patch by Daniel Richard G.
1044    
1045    24. Add support for 32-bit character strings, and UTF-32
1046    
1047    25. Major JIT compiler update (code refactoring and bugfixing).
1048        Experimental Sparc 32 support is added.
1049    
1050    26. Applied a modified version of Daniel Richard G's patch to create
1051        pcre.h.generic and config.h.generic by "make" instead of in the
1052        PrepareRelease script.
1053    
1054    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
1055        pcre_compile.c when checking for a zero character.
1056    
1057    28. Introducing a native interface for JIT. Through this interface, the compiled
1058        machine code can be directly executed. The purpose of this interface is to
1059        provide fast pattern matching, so several sanity checks are not performed.
1060        However, feature tests are still performed. The new interface provides
1061        1.4x speedup compared to the old one.
1062    
1063    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
1064        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
1065        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
1066    
1067    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
1068        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
1069        the "old" RFC 2279). Instead, it ended up passing a negative length to
1070        pcre_exec().
1071    
1072    31. Add support for GCC's visibility feature to hide internal functions.
1073    
1074    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
1075        "unknown -C option" after outputting 0 or 1.
1076    
1077    33. There is now support for generating a code coverage report for the test
1078        suite in environments where gcc is the compiler and lcov is installed. This
1079        is mainly for the benefit of the developers.
1080    
1081    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
1082        unaddressable using valgrind annotations, allowing valgrind to detect
1083        invalid memory accesses. This is mainly for the benefit of the developers.
1084    
1085    25. (*UTF) can now be used to start a pattern in any of the three libraries.
1086    
1087    26. Give configure error if --enable-cpp but no C++ compiler found.
1088    
1089    
1090    Version 8.31 06-July-2012
1091    -------------------------
1092    
1093    1.  Fixing a wrong JIT test case and some compiler warnings.
1094    
1095    2.  Removed a bashism from the RunTest script.
1096    
1097    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
1098        to unsigned type, result still unsigned" that was given by an MS compiler
1099        on encountering the code "-sizeof(xxx)".
1100    
1101    4.  Partial matching support is added to the JIT compiler.
1102    
1103    5.  Fixed several bugs concerned with partial matching of items that consist
1104        of more than one character:
1105    
1106        (a) /^(..)\1/ did not partially match "aba" because checking references was
1107            done on an "all or nothing" basis. This also applied to repeated
1108            references.
1109    
1110        (b) \R did not give a hard partial match if \r was found at the end of the
1111            subject.
1112    
1113        (c) \X did not give a hard partial match after matching one or more
1114            characters at the end of the subject.
1115    
1116        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
1117            a partial match for the string "\r".
1118    
1119        (e) When newline was set to CRLF, the metacharacter "." did not recognize
1120            a partial match for a CR character at the end of the subject string.
1121    
1122    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
1123        running pcretest, the text "(JIT)" added to the output whenever JIT is
1124        actually used to run the match.
1125    
1126    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
1127        or /S+[+] with a digit between 1 and 7.
1128    
1129    8.  OP_NOT now supports any UTF character not just single-byte ones.
1130    
1131    9.  (*MARK) control verb is now supported by the JIT compiler.
1132    
1133    10. The command "./RunTest list" lists the available tests without actually
1134        running any of them. (Because I keep forgetting what they all are.)
1135    
1136    11. Add PCRE_INFO_MAXLOOKBEHIND.
1137    
1138    12. Applied a (slightly modified) user-supplied patch that improves performance
1139        when the heap is used for recursion (compiled with --disable-stack-for-
1140        recursion). Instead of malloc and free for each heap frame each time a
1141        logical recursion happens, frames are retained on a chain and re-used where
1142        possible. This sometimes gives as much as 30% improvement.
1143    
1144    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
1145        call.
1146    
1147    14. As documented, (*COMMIT) is now confined to within a positive assertion.
1148    
1149    15. It is now possible to link pcretest with libedit as an alternative to
1150        libreadline.
1151    
1152    16. (*COMMIT) control verb is now supported by the JIT compiler.
1153    
1154    17. The Unicode data tables have been updated to Unicode 6.1.0.
1155    
1156    18. Added --file-list option to pcregrep.
1157    
1158    19. Added binary file support to pcregrep, including the -a, --binary-files,
1159        -I, and --text options.
1160    
1161    20. The madvise function is renamed for posix_madvise for QNX compatibility
1162        reasons. Fixed by Giuseppe D'Angelo.
1163    
1164    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
1165    
1166    22. Changed the option for creating symbolic links for 16-bit man pages from
1167        -s to -sf so that re-installing does not cause issues.
1168    
1169    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
1170    
1171    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
1172        matches in certain environments (the workspace was not being correctly
1173        retained). Also added to pcre_dfa_exec() a simple plausibility check on
1174        some of the workspace data at the beginning of a restart.
1175    
1176    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
1177        was not doing so when it should - probably a typo introduced by SVN 528
1178        (change 8.10/14).
1179    
1180    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
1181        \w+ when the character tables indicated that \x{c4} was a word character.
1182        There were several related cases, all because the tests for doing a table
1183        lookup were testing for characters less than 127 instead of 255.
1184    
1185    27. If a pattern contains capturing parentheses that are not used in a match,
1186        their slots in the ovector are set to -1. For those that are higher than
1187        any matched groups, this happens at the end of processing. In the case when
1188        there were back references that the ovector was too small to contain
1189        (causing temporary malloc'd memory to be used during matching), and the
1190        highest capturing number was not used, memory off the end of the ovector
1191        was incorrectly being set to -1. (It was using the size of the temporary
1192        memory instead of the true size.)
1193    
1194    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
1195        ovector size, it uses memory at the end of the block that it has got.
1196    
1197    29. Check for an overlong MARK name and give an error at compile time. The
1198        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
1199    
1200    30. JIT compiler update.
1201    
1202    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
1203        Rill for the patch.
1204    
1205    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
1206    
1207    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
1208    
1209    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
1210        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
1211        was enabled.
1212    
1213    35. Improve JIT code generation for greedy plus quantifier.
1214    
1215    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
1216        1 to "aa" instead of to an empty string. The bug affected repeated groups
1217        that could potentially match an empty string.
1218    
1219    37. Optimizing single character iterators in JIT.
1220    
1221    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
1222        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
1223        codepoints that are too big for the mode are faulted, and in a UTF mode,
1224        disallowed codepoints are also faulted.
1225    
1226    39. If PCRE was compiled with UTF support, in three places in the DFA
1227        matcher there was code that should only have been obeyed in UTF mode, but
1228        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
1229        processing when bytes with values greater than 127 were present. In 16-bit
1230        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
1231        both cases the values are those that cannot be the first data item in a UTF
1232        character. The three items that might have provoked this were recursions,
1233        possessively repeated groups, and atomic groups.
1234    
1235    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
1236        and pcregrep, because some OS require shared objects to be explicitly
1237        passed to ld, causing the link step to fail if they are not.
1238    
1239    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
1240        mode, patterns that started with \h* or \R* might be incorrectly matched.
1241    
1242    
1243    Version 8.30 04-February-2012
1244    -----------------------------
1245    
1246    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
1247        name is defined in ctype.h.
1248    
1249    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
1250        only in quite long subpatterns.
1251    
1252    3.  Removed the function pcre_info(), which has been obsolete and deprecated
1253        since it was replaced by pcre_fullinfo() in February 2000.
1254    
1255    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
1256        match a (*MARK), and the match failed at the start of the subject, a
1257        reference to memory before the start of the subject could occur. This bug
1258        was introduced by fix 17 of release 8.21.
1259    
1260    5.  A reference to an unset group with zero minimum repetition was giving
1261        totally wrong answers (in non-JavaScript-compatibility mode). For example,
1262        /(another)?(\1?)test/ matched against "hello world test". This bug was
1263        introduced in release 8.13.
1264    
1265    6.  Add support for 16-bit character strings (a large amount of work involving
1266        many changes and refactorings).
1267    
1268    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
1269        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
1270        from a file.
1271    
1272    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
1273        rounding is not applied in this particular case).
1274    
1275    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
1276        if they appear, or are escaped, in patterns.
1277    
1278    10. Get rid of a number of -Wunused-but-set-variable warnings.
1279    
1280    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
1281        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
1282        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
1283        also returns the mark "x". This bug applied to capturing parentheses,
1284        non-capturing parentheses, and atomic parentheses. It also applied to some
1285        assertions.
1286    
1287    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
1288        information out of configure.ac instead of relying on pcre.h.generic, which
1289        is not stored in the repository.
1290    
1291    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
1292        -lreadline.
1293    
1294    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
1295    
1296    15. Applied Graycode's patch to put the top-level frame on the stack rather
1297        than the heap when not using the stack for recursion. This gives a
1298        performance improvement in many cases when recursion is not deep.
1299    
1300    16. Experimental code added to "pcretest -C" to output the stack frame size.
1301    
1302    
1303    Version 8.21 12-Dec-2011
1304    ------------------------
1305    
1306    1.  Updating the JIT compiler.
1307    
1308    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
1309        are added as well.
1310    
1311    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
1312        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
1313        calling _pcre_jit_exec. Some extra comments are added.
1314    
1315    4.  (*MARK) settings inside atomic groups that do not contain any capturing
1316        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
1317        was introduced by change 18 for 8.20.
1318    
1319    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
1320        ECMA-262 standard.
1321    
1322    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
1323        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
1324        This bug was probably introduced by change 9 of 8.13.
1325    
1326    7.  While fixing 6 above, I noticed that a number of other items were being
1327        incorrectly rejected as "not fixed length". This arose partly because newer
1328        opcodes had not been added to the fixed-length checking code. I have (a)
1329        corrected the bug and added tests for these items, and (b) arranged for an
1330        error to occur if an unknown opcode is encountered while checking for fixed
1331        length instead of just assuming "not fixed length". The items that were
1332        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
1333        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
1334        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
1335    
1336    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
1337        being incorrectly compiled and would have given unpredicatble results.
1338    
1339    9.  A possessively repeated subpattern with minimum repeat count greater than
1340        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
1341        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
1342        the first (A) could occur when it should not.
1343    
1344    10. Add a cast and remove a redundant test from the code.
1345    
1346    11. JIT should use pcre_malloc/pcre_free for allocation.
1347    
1348    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
1349        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
1350        is anything other than /usr, -L is still shown).
1351    
1352    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
1353    
1354    14. Perl does not support \N without a following name in a [] class; PCRE now
1355        also gives an error.
1356    
1357    15. If a forward reference was repeated with an upper limit of around 2000,
1358        it caused the error "internal error: overran compiling workspace". The
1359        maximum number of forward references (including repeats) was limited by the
1360        internal workspace, and dependent on the LINK_SIZE. The code has been
1361        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
1362        the default depends on LINK_SIZE. There is a new upper limit (for safety)
1363        of around 200,000 forward references. While doing this, I also speeded up
1364        the filling in of repeated forward references.
1365    
1366    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
1367        incorrectly expecting the subject to contain another "a" after the start.
1368    
1369    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
1370        in the match, the SKIP should be ignored. This was not happening; instead
1371        the SKIP was being treated as NOMATCH. For patterns such as
1372        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
1373        tested.
1374    
1375    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
1376        now much more compatible with Perl, in particular in cases where the result
1377        is a non-match for a non-anchored pattern. For example, if
1378        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
1379        "m", where previously it did not return a name. A side effect of this
1380        change is that for partial matches, the last encountered mark name is
1381        returned, as for non matches. A number of tests that were previously not
1382        Perl-compatible have been moved into the Perl-compatible test files. The
1383        refactoring has had the pleasing side effect of removing one argument from
1384        the match() function, thus reducing its stack requirements.
1385    
1386    19. If the /S+ option was used in pcretest to study a pattern using JIT,
1387        subsequent uses of /S (without +) incorrectly behaved like /S+.
1388    
1389    21. Retrieve executable code size support for the JIT compiler and fixing
1390        some warnings.
1391    
1392    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
1393        not work when the shorter character appeared right at the end of the
1394        subject string.
1395    
1396    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
1397        systems.
1398    
1399    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
1400        output it when the /M option is used in pcretest.
1401    
1402    25. The CheckMan script was not being included in the distribution. Also, added
1403        an explicit "perl" to run Perl scripts from the PrepareRelease script
1404        because this is reportedly needed in Windows.
1405    
1406    26. If study data was being save in a file and studying had not found a set of
1407        "starts with" bytes for the pattern, the data written to the file (though
1408        never used) was taken from uninitialized memory and so caused valgrind to
1409        complain.
1410    
1411    27. Updated RunTest.bat as provided by Sheri Pierce.
1412    
1413    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
1414    
1415    29. Computation of memory usage for the table of capturing group names was
1416        giving an unnecessarily large value.
1417    
1418    
1419    Version 8.20 21-Oct-2011
1420    ------------------------
1421    
1422    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
1423        a POSIX class. After further experiments with Perl, which convinced me that
1424        Perl has bugs and confusions, a closing square bracket is no longer allowed
1425        in a POSIX name. This bug also affected patterns with classes that started
1426        with full stops.
1427    
1428    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
1429        captured substring, but while checking the failing first alternative,
1430        substring 1 is temporarily captured. If the output vector supplied to
1431        pcre_exec() was not big enough for this capture, the yield of the function
1432        was still zero ("insufficient space for captured substrings"). This cannot
1433        be totally fixed without adding another stack variable, which seems a lot
1434        of expense for a edge case. However, I have improved the situation in cases
1435        such as /(a)(b)x|abc/ matched against "abc", where the return code
1436        indicates that fewer than the maximum number of slots in the ovector have
1437        been set.
1438    
1439    3.  Related to (2) above: when there are more back references in a pattern than
1440        slots in the output vector, pcre_exec() uses temporary memory during
1441        matching, and copies in the captures as far as possible afterwards. It was
1442        using the entire output vector, but this conflicts with the specification
1443        that only 2/3 is used for passing back captured substrings. Now it uses
1444        only the first 2/3, for compatibility. This is, of course, another edge
1445        case.
1446    
1447    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
1448        main code base, and can be used by building with --enable-jit. When this is
1449        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
1450        runtime --no-jit option is given.
1451    
1452    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
1453        ovector, the return from the function was zero, implying that there were
1454        other matches that did not fit. The correct "exactly full" value is now
1455        returned.
1456    
1457    6.  If a subpattern that was called recursively or as a subroutine contained
1458        (*PRUNE) or any other control that caused it to give a non-standard return,
1459        invalid errors such as "Error -26 (nested recursion at the same subject
1460        position)" or even infinite loops could occur.
1461    
1462    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
1463        computing the minimum length on reaching *ACCEPT, and so ended up with the
1464        wrong value of 1 rather than 0. Further investigation indicates that
1465        computing a minimum subject length in the presence of *ACCEPT is difficult
1466        (think back references, subroutine calls), and so I have changed the code
1467        so that no minimum is registered for a pattern that contains *ACCEPT.
1468    
1469    8.  If (*THEN) was present in the first (true) branch of a conditional group,
1470        it was not handled as intended. [But see 16 below.]
1471    
1472    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
1473        Sheri Pierce.
1474    
1475    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
1476        the first byte in a match must be "a".
1477    
1478    11. Change 17 for 8.13 increased the recursion depth for patterns like
1479        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
1480        pattern contains any instances of (*THEN). If it does not, the old
1481        optimizations are restored. It would be nice to do this on a per-group
1482        basis, but at the moment that is not feasible.
1483    
1484    12. In some environments, the output of pcretest -C is CRLF terminated. This
1485        broke RunTest's code that checks for the link size. A single white space
1486        character after the value is now allowed for.
1487    
1488    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
1489        For "fr", it uses the Windows-specific input and output files.
1490    
1491    14. If (*THEN) appeared in a group that was called recursively or as a
1492        subroutine, it did not work as intended. [But see next item.]
1493    
1494    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
1495        pattern fragments (but not containing any | characters). If A and B are
1496        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
1497        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
1498        D. In other words, Perl considers parentheses that do not contain any |
1499        characters to be part of a surrounding alternative, whereas PCRE was
1500        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
1501        differently. PCRE now behaves in the same way as Perl, except in the case
1502        of subroutine/recursion calls such as (?1) which have in any case always
1503        been different (but PCRE had them first :-).
1504    
1505    16. Related to 15 above: Perl does not treat the | in a conditional group as
1506        creating alternatives. Such a group is treated in the same way as an
1507        ordinary group without any | characters when processing (*THEN). PCRE has
1508        been changed to match Perl's behaviour.
1509    
1510    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
1511        RunGrepTest script failed.
1512    
1513    18. Change 22 for version 13 caused atomic groups to use more stack. This is
1514        inevitable for groups that contain captures, but it can lead to a lot of
1515        stack use in large patterns. The old behaviour has been restored for atomic
1516        groups that do not contain any capturing parentheses.
1517    
1518    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
1519        suppress the check for a minimum subject length at run time. (If it was
1520        given to pcre_exec() or pcre_dfa_exec() it did work.)
1521    
1522    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
1523        fail to work when decoding hex characters in data strings in EBCDIC
1524        environments.
1525    
1526    21. It appears that in at least one Mac OS environment, the isxdigit() function
1527        is implemented as a macro that evaluates to its argument more than once,
1528        contravening the C 90 Standard (I haven't checked a later standard). There
1529        was an instance in pcretest which caused it to go wrong when processing
1530        \x{...} escapes in subject strings. The has been rewritten to avoid using
1531        things like p++ in the argument of isxdigit().
1532    
1533    
1534    Version 8.13 16-Aug-2011
1535    ------------------------
1536    
1537    1.  The Unicode data tables have been updated to Unicode 6.0.0.
1538    
1539    2.  Two minor typos in pcre_internal.h have been fixed.
1540    
1541    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
1542        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
1543        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
1544    
1545    4.  There were a number of related bugs in the code for matching backrefences
1546        caselessly in UTF-8 mode when codes for the characters concerned were
1547        different numbers of bytes. For example, U+023A and U+2C65 are an upper
1548        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
1549        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
1550        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
1551        2-byte code at the end of the subject (it thought there wasn't enough data
1552        left).
1553    
1554    5.  Comprehensive information about what went wrong is now returned by
1555        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
1556        as the output vector has at least 2 elements. The offset of the start of
1557        the failing character and a reason code are placed in the vector.
1558    
1559    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
1560        now returned is for the first byte of the failing character, instead of the
1561        last byte inspected. This is an incompatible change, but I hope it is small
1562        enough not to be a problem. It makes the returned offset consistent with
1563        pcre_exec() and pcre_dfa_exec().
1564    
1565    7.  pcretest now gives a text phrase as well as the error number when
1566        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
1567        failure, the offset and reason code are output.
1568    
1569    8.  When \R was used with a maximizing quantifier it failed to skip backwards
1570        over a \r\n pair if the subsequent match failed. Instead, it just skipped
1571        back over a single character (\n). This seems wrong (because it treated the
1572        two characters as a single entity when going forwards), conflicts with the
1573        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
1574        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
1575        has been changed.
1576    
1577    9.  Some internal refactoring has changed the processing so that the handling
1578        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
1579        time (the PCRE_DOTALL option was changed this way some time ago: version
1580        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
1581        which was always a bit of a fudge. It also means that there is one less
1582        argument for the match() function, which reduces its stack requirements
1583        slightly. This change also fixes an incompatibility with Perl: the pattern
1584        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
1585    
1586    10. More internal refactoring has drastically reduced the number of recursive
1587        calls to match() for possessively repeated groups such as (abc)++ when
1588        using pcre_exec().
1589    
1590    11. While implementing 10, a number of bugs in the handling of groups were
1591        discovered and fixed:
1592    
1593        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
1594        (a|)*(?1) gave a compile-time internal error.
1595        ((a|)+)+  did not notice that the outer group could match an empty string.
1596        (^a|^)+   was not marked as anchored.
1597        (.*a|.*)+ was not marked as matching at start or after a newline.
1598    
1599    12. Yet more internal refactoring has removed another argument from the match()
1600        function. Special calls to this function are now indicated by setting a
1601        value in a variable in the "match data" data block.
1602    
1603    13. Be more explicit in pcre_study() instead of relying on "default" for
1604        opcodes that mean there is no starting character; this means that when new
1605        ones are added and accidentally left out of pcre_study(), testing should
1606        pick them up.
1607    
1608    14. The -s option of pcretest has been documented for ages as being an old
1609        synonym of -m (show memory usage). I have changed it to mean "force study
1610        for every regex", that is, assume /S for every regex. This is similar to -i
1611        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
1612        using it. It makes it easier to run collections of tests with and without
1613        study enabled, and thereby test pcre_study() more easily. All the standard
1614        tests are now run with and without -s (but some patterns can be marked as
1615        "never study" - see 20 below).
1616    
1617    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
1618        restoration of the capturing data to the outer values was not happening
1619        correctly.
1620    
1621    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
1622        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
1623        pattern had matched an empty string, and so incorrectly returned a no
1624        match.
1625    
1626    17. There was optimizing code for the last branch of non-capturing parentheses,
1627        and also for the obeyed branch of a conditional subexpression, which used
1628        tail recursion to cut down on stack usage. Unfortunately, now that there is
1629        the possibility of (*THEN) occurring in these branches, tail recursion is
1630        no longer possible because the return has to be checked for (*THEN). These
1631        two optimizations have therefore been removed. [But see 8.20/11 above.]
1632    
1633    18. If a pattern containing \R was studied, it was assumed that \R always
1634        matched two bytes, thus causing the minimum subject length to be
1635        incorrectly computed because \R can also match just one byte.
1636    
1637    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
1638        was incorrectly computed.
1639    
1640    20. If /S is present twice on a test pattern in pcretest input, it now
1641        *disables* studying, thereby overriding the use of -s on the command line
1642        (see 14 above). This is necessary for one or two tests to keep the output
1643        identical in both cases.
1644    
1645    21. When (*ACCEPT) was used in an assertion that matched an empty string and
1646        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
1647    
1648    22. When an atomic group that contained a capturing parenthesis was
1649        successfully matched, but the branch in which it appeared failed, the
1650        capturing was not being forgotten if a higher numbered group was later
1651        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
1652        group 1 to "a", when in fact it should be unset. This applied to multi-
1653        branched capturing and non-capturing groups, repeated or not, and also to
1654        positive assertions (capturing in negative assertions does not happen
1655        in PCRE) and also to nested atomic groups.
1656    
1657    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
1658        subject after a captured substring, to make it easier to tell which of a
1659        number of identical substrings has been captured.
1660    
1661    24. The way atomic groups are processed by pcre_exec() has been changed so that
1662        if they are repeated, backtracking one repetition now resets captured
1663        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
1664        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
1665        "aaa". Previously, it would have been "a". As part of this code
1666        refactoring, the way recursive calls are handled has also been changed.
1667    
1668    25. If an assertion condition captured any substrings, they were not passed
1669        back unless some other capturing happened later. For example, if
1670        (?(?=(a))a) was matched against "a", no capturing was returned.
1671    
1672    26. When studying a pattern that contained subroutine calls or assertions,
1673        the code for finding the minimum length of a possible match was handling
1674        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
1675        group 1 called group 2 while simultaneously a separate group 2 called group
1676        1). A stack overflow occurred in this case. I have fixed this by limiting
1677        the recursion depth to 10.
1678    
1679    27. Updated RunTest.bat in the distribution to the version supplied by Tom
1680        Fortmann. This supports explicit test numbers on the command line, and has
1681        argument validation and error reporting.
1682    
1683    28. An instance of \X with an unlimited repeat could fail if at any point the
1684        first character it looked at was a mark character.
1685    
1686    29. Some minor code refactoring concerning Unicode properties and scripts
1687        should reduce the stack requirement of match() slightly.
1688    
1689    30. Added the '=' option to pcretest to check the setting of unused capturing
1690        slots at the end of the pattern, which are documented as being -1, but are
1691        not included in the return count.
1692    
1693    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
1694        compiled something random. Now it gives a compile-time error (as does
1695        Perl).
1696    
1697    32. A *MARK encountered during the processing of a positive assertion is now
1698        recorded and passed back (compatible with Perl).
1699    
1700    33. If --only-matching or --colour was set on a pcregrep call whose pattern
1701        had alternative anchored branches, the search for a second match in a line
1702        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
1703        matched the line "0102" twice. The same bug affected patterns that started
1704        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
1705        twice.
1706    
1707    34. Previously, PCRE did not allow quantification of assertions. However, Perl
1708        does, and because of capturing effects, quantifying parenthesized
1709        assertions may at times be useful. Quantifiers are now allowed for
1710        parenthesized assertions.
1711    
1712    35. A minor code tidy in pcre_compile() when checking options for \R usage.
1713    
1714    36. \g was being checked for fancy things in a character class, when it should
1715        just be a literal "g".
1716    
1717    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
1718        appearance of a nested POSIX class supersedes an apparent external class.
1719        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
1720        unescaped square brackets may also appear as part of class names. For
1721        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
1722        more like Perl. (But see 8.20/1 above.)
1723    
1724    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
1725        was because it thought it was \N{name}, which is not supported).
1726    
1727    39. Add minix to OS list not supporting the -S option in pcretest.
1728    
1729    40. PCRE tries to detect cases of infinite recursion at compile time, but it
1730        cannot analyze patterns in sufficient detail to catch mutual recursions
1731        such as ((?1))((?2)). There is now a runtime test that gives an error if a
1732        subgroup is called recursively as a subpattern for a second time at the
1733        same position in the subject string. In previous releases this might have
1734        been caught by the recursion limit, or it might have run out of stack.
1735    
1736    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
1737        happen only once. PCRE was, however incorrectly giving a compile time error
1738        "recursive call could loop indefinitely" because it cannot analyze the
1739        pattern in sufficient detail. The compile time test no longer happens when
1740        PCRE is compiling a conditional subpattern, but actual runaway loops are
1741        now caught at runtime (see 40 above).
1742    
1743    42. It seems that Perl allows any characters other than a closing parenthesis
1744        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
1745        has been changed to be the same.
1746    
1747    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
1748        as not to get warnings when autogen.sh is called. Also changed
1749        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
1750    
1751    44. To help people who use pcregrep to scan files containing exceedingly long
1752        lines, the following changes have been made:
1753    
1754        (a) The default value of the buffer size parameter has been increased from
1755            8K to 20K. (The actual buffer used is three times this size.)
1756    
1757        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
1758            PCRE is built.
1759    
1760        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
1761            to be set at run time.
1762    
1763        (d) Numerical values in pcregrep options can be followed by K or M, for
1764            example --buffer-size=50K.
1765    
1766        (e) If a line being scanned overflows pcregrep's buffer, an error is now
1767            given and the return code is set to 2.
1768    
1769    45. Add a pointer to the latest mark to the callout data block.
1770    
1771    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
1772        partial match of an empty string instead of no match. This was specific to
1773        the use of ".".
1774    
1775    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
1776        complete match instead of a partial match. This bug was dependent on both
1777        the PCRE_UTF8 and PCRE_DOTALL options being set.
1778    
1779    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
1780        starting byte set, because \b was not being ignored.
1781    
1782    
1783    Version 8.12 15-Jan-2011
1784    ------------------------
1785    
1786    1.  Fixed some typos in the markup of the man pages, and wrote a script that
1787        checks for such things as part of the documentation building process.
1788    
1789    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
1790        --match-limit and --recursion-limit options (added for 8.11). In
1791        particular, this made one of the standard tests fail. (The integer value
1792        went into the wrong half of a long int.)
1793    
1794    3.  If the --colour option was given to pcregrep with -v (invert match), it
1795        did strange things, either producing crazy output, or crashing. It should,
1796        of course, ignore a request for colour when reporting lines that do not
1797        match.
1798    
1799    4.  Another pcregrep bug caused similar problems if --colour was specified with
1800        -M (multiline) and the pattern match finished with a line ending.
1801    
1802    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
1803        matched in multiline mode, the following line was shown as part of the
1804        match. This seems wrong, so I have changed it.
1805    
1806    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
1807        the check for further matches in the same line (so they could be coloured)
1808        to overrun the end of the current line. If another match was found, it was
1809        incorrectly shown (and then shown again when found in the next line).
1810    
1811    7.  If pcregrep was compiled under Windows, there was a reference to the
1812        function pcregrep_exit() before it was defined. I am assuming this was
1813        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
1814        reported by a user. I've moved the definition above the reference.
1815    
1816    
1817    Version 8.11 10-Dec-2010
1818    ------------------------
1819    
1820    1.  (*THEN) was not working properly if there were untried alternatives prior
1821        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
1822        backtracked to try for "b" instead of moving to the next alternative branch
1823        at the same level (in this case, to look for "c"). The Perl documentation
1824        is clear that when (*THEN) is backtracked onto, it goes to the "next
1825        alternative in the innermost enclosing group".
1826    
1827    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
1828        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
1829        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
1830        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
1831        (*THEN).
1832    
1833    3.  If \s appeared in a character class, it removed the VT character from
1834        the class, even if it had been included by some previous item, for example
1835        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
1836        of \s, but is part of the POSIX "space" class.)
1837    
1838    4.  A partial match never returns an empty string (because you can always
1839        match an empty string at the end of the subject); however the checking for
1840        an empty string was starting at the "start of match" point. This has been
1841        changed to the "earliest inspected character" point, because the returned
1842        data for a partial match starts at this character. This means that, for
1843        example, /(?<=abc)def/ gives a partial match for the subject "abc"
1844        (previously it gave "no match").
1845    
1846    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
1847        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
1848        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
1849        has an implication that the given string is incomplete (because a partial
1850        match is preferred over a full match). For this reason, these items now
1851        give a partial match in this situation. [Aside: previously, the one case
1852        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
1853        match rather than a full match, which was wrong by the old rules, but is
1854        now correct.]
1855    
1856    6.  There was a bug in the handling of #-introduced comments, recognized when
1857        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
1858        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
1859        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
1860        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
1861        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
1862        places in pcre_compile().
1863    
1864    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
1865        comments when looking ahead for named forward references to subpatterns,
1866        the only newline sequence it recognized was NL. It now handles newlines
1867        according to the set newline convention.
1868    
1869    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
1870        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
1871        cater for a lack of strerror(). These oversights have been fixed.
1872    
1873    9.  Added --match-limit and --recursion-limit to pcregrep.
1874    
1875    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
1876    
1877    11. When the -o option was used, pcregrep was setting a return code of 1, even
1878        when matches were found, and --line-buffered was not being honoured.
1879    
1880    12. Added an optional parentheses number to the -o and --only-matching options
1881        of pcregrep.
1882    
1883    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
1884        can match an empty string. The code to do it in pcretest and pcredemo
1885        needed fixing:
1886    
1887        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
1888            only one byte after an empty string match just before CRLF (this case
1889            just got forgotten; "any" and "anycrlf" were OK).
1890    
1891        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
1892            mode when an empty string match preceded an ASCII character followed by
1893            a non-ASCII character. (The code for advancing by one character rather
1894            than one byte was nonsense.)
1895    
1896        (c) The pcredemo.c sample program did not have any code at all to handle
1897            the cases when CRLF is a valid newline sequence.
1898    
1899    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
1900        as a starting offset was within the subject string. There is now a new
1901        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
1902        negative or greater than the length of the string. In order to test this,
1903        pcretest is extended to allow the setting of negative starting offsets.
1904    
1905    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
1906        starting offset points to the beginning of a UTF-8 character was
1907        unnecessarily clumsy. I tidied it up.
1908    
1909    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
1910        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
1911    
1912    17. Nobody had reported that the --include_dir option, which was added in
1913        release 7.7 should have been called --include-dir (hyphen, not underscore)
1914        for compatibility with GNU grep. I have changed it to --include-dir, but
1915        left --include_dir as an undocumented synonym, and the same for
1916        --exclude-dir, though that is not available in GNU grep, at least as of
1917        release 2.5.4.
1918    
1919    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
1920        characters from a string of bytes) have been redefined so as not to use
1921        loops, in order to improve performance in some environments. At the same
1922        time, I abstracted some of the common code into auxiliary macros to save
1923        repetition (this should not affect the compiled code).
1924    
1925    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1926        compile-time error is now given if \c is not followed by an ASCII
1927        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1928        different, and any byte value is allowed.)
1929    
1930    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1931        START_OPTIMIZE option, which is now allowed at compile time - but just
1932        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1933        to pcregrep and other applications that have no direct access to PCRE
1934        options. The new /Y option in pcretest sets this option when calling
1935        pcre_compile().
1936    
1937    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1938        back references. Groups containing recursive back references were forced to
1939        be atomic by that change, but in the case of named groups, the amount of
1940        memory required was incorrectly computed, leading to "Failed: internal
1941        error: code overflow". This has been fixed.
1942    
1943    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1944        pcretest.c, to avoid build problems in some Borland environments.
1945    
1946    
1947    Version 8.10 25-Jun-2010
1948    ------------------------
1949    
1950    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1951        THEN.
1952    
1953    2.  (*ACCEPT) was not working when inside an atomic group.
1954    
1955    3.  Inside a character class, \B is treated as a literal by default, but
1956        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1957        causes the error). The code is unchanged, but I tidied the documentation.
1958    
1959    4.  Inside a character class, PCRE always treated \R and \X as literals,
1960        whereas Perl faults them if its -w option is set. I have changed PCRE so
1961        that it faults them when PCRE_EXTRA is set.
1962    
1963    5.  Added support for \N, which always matches any character other than
1964        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1965    
1966    6.  When compiling pcregrep with newer versions of gcc which may have
1967        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1968        declared with attribute warn_unused_result" were given. Just casting the
1969        result to (void) does not stop the warnings; a more elaborate fudge is
1970        needed. I've used a macro to implement this.
1971    
1972    7.  Minor change to pcretest.c to avoid a compiler warning.
1973    
1974    8.  Added four artifical Unicode properties to help with an option to make
1975        \s etc use properties (see next item). The new properties are: Xan
1976        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1977    
1978    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1979        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1980        this option. Modified pcretest to add /W to test this facility. Added
1981        REG_UCP to make it available via the POSIX interface.
1982    
1983    10. Added --line-buffered to pcregrep.
1984    
1985    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1986        studied, and the match started with a letter with a code point greater than
1987        127 whose first byte was different to the first byte of the other case of
1988        the letter, the other case of this starting letter was not recognized
1989        (#976).
1990    
1991    12. If a pattern that was studied started with a repeated Unicode property
1992        test, for example, \p{Nd}+, there was the theoretical possibility of
1993        setting up an incorrect bitmap of starting bytes, but fortunately it could
1994        not have actually happened in practice until change 8 above was made (it
1995        added property types that matched character-matching opcodes).
1996    
1997    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1998        possible starting bytes for non-anchored patterns.
1999    
2000    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
2001        \R, and also a number of cases that involve Unicode properties, both
2002        explicit and implicit when PCRE_UCP is set.
2003    
2004    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
2005        input, it could crash or give wrong results if characters with values
2006        greater than 0xc0 were present in the subject string. (Detail: it assumed
2007        UTF-8 input when processing these items.)
2008    
2009    16. Added a lot of (int) casts to avoid compiler warnings in systems where
2010        size_t is 64-bit (#991).
2011    
2012    17. Added a check for running out of memory when PCRE is compiled with
2013        --disable-stack-for-recursion (#990).
2014    
2015    18. If the last data line in a file for pcretest does not have a newline on
2016        the end, a newline was missing in the output.
2017    
2018    19. The default pcre_chartables.c file recognizes only ASCII characters (values
2019        less than 128) in its various bitmaps. However, there is a facility for
2020        generating tables according to the current locale when PCRE is compiled. It
2021        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
2022        characters, are recognized by isspace() and therefore were getting set in
2023        these tables, and indeed these tables seem to approximate to ISO 8859. This
2024        caused a problem in UTF-8 mode when pcre_study() was used to create a list
2025        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
2026        which of course cannot start UTF-8 characters. I have changed the code so
2027        that only real ASCII characters (less than 128) and the correct starting
2028        bytes for UTF-8 encodings are set for characters greater than 127 when in
2029        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
2030        altogether.)
2031    
2032    20. Added the /T option to pcretest so as to be able to run tests with non-
2033        standard character tables, thus making it possible to include the tests
2034        used for 19 above in the standard set of tests.
2035    
2036    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
2037        reference to a subpattern the other side of a comment that contains an
2038        opening parenthesis caused either an internal compiling error, or a
2039        reference to the wrong subpattern.
2040    
2041    
2042    Version 8.02 19-Mar-2010
2043  ------------------------  ------------------------
2044    
2045  1.  The Unicode data tables have been updated to Unicode 5.2.0.  1.  The Unicode data tables have been updated to Unicode 5.2.0.
2046    
2047  2.  Added the option --libs-cpp to pcre-config, but only when C++ support is  2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
2048      configured.      configured.
2049    
2050  3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the  3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
2051      original author of that file, following a query about its status.      original author of that file, following a query about its status.
2052    
2053  4.  On systems that do not have stdint.h (e.g. Solaris), check for and include  4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
2054      inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.      inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
2055    
2056  5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive  5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
2057      quantifier applied to a forward-referencing subroutine call, could compile      quantifier applied to a forward-referencing subroutine call, could compile
2058      incorrect code or give the error "internal error: previously-checked      incorrect code or give the error "internal error: previously-checked
2059      referenced subpattern not found".      referenced subpattern not found".
2060    
2061  6.  Both MS Visual Studio and Symbian OS have problems with initializing  6.  Both MS Visual Studio and Symbian OS have problems with initializing
2062      variables to point to external functions. For these systems, therefore,      variables to point to external functions. For these systems, therefore,
2063      pcre_malloc etc. are now initialized to local functions that call the      pcre_malloc etc. are now initialized to local functions that call the
2064      relevant global functions.      relevant global functions.
2065    
2066  7.  There were two entries missing in the vectors called coptable and poptable  7.  There were two entries missing in the vectors called coptable and poptable
2067      in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.      in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
2068      I've fixed the data, and added a kludgy way of testing at compile time that      I've fixed the data, and added a kludgy way of testing at compile time that
2069      the lengths are correct (equal to the number of opcodes).      the lengths are correct (equal to the number of opcodes).
2070    
2071  8.  Following on from 7, I added a similar kludge to check the length of the  8.  Following on from 7, I added a similar kludge to check the length of the
2072      eint vector in pcreposix.c.      eint vector in pcreposix.c.
2073    
2074  9.  Error texts for pcre_compile() are held as one long string to avoid too  9.  Error texts for pcre_compile() are held as one long string to avoid too
2075      much relocation at load time. To find a text, the string is searched,      much relocation at load time. To find a text, the string is searched,
2076      counting zeros. There was no check for running off the end of the string,      counting zeros. There was no check for running off the end of the string,
2077      which could happen if a new error number was added without updating the      which could happen if a new error number was added without updating the
2078      string.      string.
2079    
2080  10. \K gave a compile-time error if it appeared in a lookbehind assersion.  10. \K gave a compile-time error if it appeared in a lookbehind assersion.
2081    
2082  11. \K was not working if it appeared in an atomic group or in a group that  11. \K was not working if it appeared in an atomic group or in a group that
2083      was called as a "subroutine", or in an assertion. Perl 5.11 documents that      was called as a "subroutine", or in an assertion. Perl 5.11 documents that
2084      \K is "not well defined" if used in an assertion. PCRE now accepts it if      \K is "not well defined" if used in an assertion. PCRE now accepts it if
2085      the assertion is positive, but not if it is negative.      the assertion is positive, but not if it is negative.
2086    
2087  12. Change 11 fortuitously reduced the size of the stack frame used in the  12. Change 11 fortuitously reduced the size of the stack frame used in the
2088      "match()" function of pcre_exec.c by one pointer. Forthcoming      "match()" function of pcre_exec.c by one pointer. Forthcoming
2089      implementation of support for (*MARK) will need an extra pointer on the      implementation of support for (*MARK) will need an extra pointer on the
2090      stack; I have reserved it now, so that the stack frame size does not      stack; I have reserved it now, so that the stack frame size does not
2091      decrease.      decrease.
2092    
2093  13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other  13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
2094      item in branch that calls a recursion is a subroutine call - as in the      item in branch that calls a recursion is a subroutine call - as in the
2095      second branch in the above example - was incorrectly given the compile-      second branch in the above example - was incorrectly given the compile-
2096      time error "recursive call could loop indefinitely" because pcre_compile()      time error "recursive call could loop indefinitely" because pcre_compile()
2097      was not correctly checking the subroutine for matching a non-empty string.      was not correctly checking the subroutine for matching a non-empty string.
2098    
2099  14. The checks for overrunning compiling workspace could trigger after an  14. The checks for overrunning compiling workspace could trigger after an
2100      overrun had occurred. This is a "should never occur" error, but it can be      overrun had occurred. This is a "should never occur" error, but it can be
2101      triggered by pathological patterns such as hundreds of nested parentheses.      triggered by pathological patterns such as hundreds of nested parentheses.
2102      The checks now trigger 100 bytes before the end of the workspace.      The checks now trigger 100 bytes before the end of the workspace.
2103    
2104    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
2105    
2106    
2107  Version 8.01 19-Jan-2010  Version 8.01 19-Jan-2010
# Line 390  Version 7.9 11-Apr-09 Line 2430  Version 7.9 11-Apr-09
2430  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
2431      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
2432      locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this      locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
2433      seems to be how GNU grep behaves.      seems to be how GNU grep behaves. [But see later change 40 for release
2434        8.33.]
2435    
2436  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
2437      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 633  Version 7.7 07-May-08 Line 2674  Version 7.7 07-May-08
2674      containing () gave an internal compiling error instead of "reference to      containing () gave an internal compiling error instead of "reference to
2675      non-existent subpattern". Fortunately, when the pattern did exist, the      non-existent subpattern". Fortunately, when the pattern did exist, the
2676      compiled code was correct. (When scanning forwards to check for the      compiled code was correct. (When scanning forwards to check for the
2677      existencd of the subpattern, it was treating the data ']' as terminating      existence of the subpattern, it was treating the data ']' as terminating
2678      the class, so got the count wrong. When actually compiling, the reference      the class, so got the count wrong. When actually compiling, the reference
2679      was subsequently set up correctly.)      was subsequently set up correctly.)
2680    

Legend:
Removed from v.505  
changed lines
  Added in v.1603

  ViewVC Help
Powered by ViewVC 1.1.5