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

Diff of /code/trunk/ChangeLog

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

revision 65 by nigel, Sat Feb 24 21:40:08 2007 UTC revision 593 by ph10, Sun May 1 08:14:24 2011 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 8.13 30-Apr-2011
5    ------------------------
6    
7    1.  The Unicode data tables have been updated to Unicode 6.0.0.
8    
9    2.  Two minor typos in pcre_internal.h have been fixed.
10    
11    
12    Version 8.12 15-Jan-2011
13    ------------------------
14    
15    1.  Fixed some typos in the markup of the man pages, and wrote a script that
16        checks for such things as part of the documentation building process.
17    
18    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
19        --match-limit and --recursion-limit options (added for 8.11). In
20        particular, this made one of the standard tests fail. (The integer value
21        went into the wrong half of a long int.)
22    
23    3.  If the --colour option was given to pcregrep with -v (invert match), it
24        did strange things, either producing crazy output, or crashing. It should,
25        of course, ignore a request for colour when reporting lines that do not
26        match.
27    
28    4.  Another pcregrep bug caused similar problems if --colour was specified with
29        -M (multiline) and the pattern match finished with a line ending.
30    
31    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
32        matched in multiline mode, the following line was shown as part of the
33        match. This seems wrong, so I have changed it.
34    
35    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
36        the check for further matches in the same line (so they could be coloured)
37        to overrun the end of the current line. If another match was found, it was
38        incorrectly shown (and then shown again when found in the next line).
39    
40    7.  If pcregrep was compiled under Windows, there was a reference to the
41        function pcregrep_exit() before it was defined. I am assuming this was
42        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
43        reported by a user. I've moved the definition above the reference.
44    
45    
46    Version 8.11 10-Dec-2010
47    ------------------------
48    
49    1.  (*THEN) was not working properly if there were untried alternatives prior
50        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
51        backtracked to try for "b" instead of moving to the next alternative branch
52        at the same level (in this case, to look for "c"). The Perl documentation
53        is clear that when (*THEN) is backtracked onto, it goes to the "next
54        alternative in the innermost enclosing group".
55    
56    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
57        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
58        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
59        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
60        (*THEN).
61    
62    3.  If \s appeared in a character class, it removed the VT character from
63        the class, even if it had been included by some previous item, for example
64        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
65        of \s, but is part of the POSIX "space" class.)
66    
67    4.  A partial match never returns an empty string (because you can always
68        match an empty string at the end of the subject); however the checking for
69        an empty string was starting at the "start of match" point. This has been
70        changed to the "earliest inspected character" point, because the returned
71        data for a partial match starts at this character. This means that, for
72        example, /(?<=abc)def/ gives a partial match for the subject "abc"
73        (previously it gave "no match").
74    
75    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
76        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
77        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
78        has an implication that the given string is incomplete (because a partial
79        match is preferred over a full match). For this reason, these items now
80        give a partial match in this situation. [Aside: previously, the one case
81        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
82        match rather than a full match, which was wrong by the old rules, but is
83        now correct.]
84    
85    6.  There was a bug in the handling of #-introduced comments, recognized when
86        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
87        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
88        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
89        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
90        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
91        places in pcre_compile().
92    
93    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
94        comments when looking ahead for named forward references to subpatterns,
95        the only newline sequence it recognized was NL. It now handles newlines
96        according to the set newline convention.
97    
98    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
99        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
100        cater for a lack of strerror(). These oversights have been fixed.
101    
102    9.  Added --match-limit and --recursion-limit to pcregrep.
103    
104    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
105    
106    11. When the -o option was used, pcregrep was setting a return code of 1, even
107        when matches were found, and --line-buffered was not being honoured.
108    
109    12. Added an optional parentheses number to the -o and --only-matching options
110        of pcregrep.
111    
112    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
113        can match an empty string. The code to do it in pcretest and pcredemo
114        needed fixing:
115    
116        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
117            only one byte after an empty string match just before CRLF (this case
118            just got forgotten; "any" and "anycrlf" were OK).
119    
120        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
121            mode when an empty string match preceded an ASCII character followed by
122            a non-ASCII character. (The code for advancing by one character rather
123            than one byte was nonsense.)
124    
125        (c) The pcredemo.c sample program did not have any code at all to handle
126            the cases when CRLF is a valid newline sequence.
127    
128    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
129        as a starting offset was within the subject string. There is now a new
130        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
131        negative or greater than the length of the string. In order to test this,
132        pcretest is extended to allow the setting of negative starting offsets.
133    
134    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
135        starting offset points to the beginning of a UTF-8 character was
136        unnecessarily clumsy. I tidied it up.
137    
138    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
139        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
140    
141    17. Nobody had reported that the --include_dir option, which was added in
142        release 7.7 should have been called --include-dir (hyphen, not underscore)
143        for compatibility with GNU grep. I have changed it to --include-dir, but
144        left --include_dir as an undocumented synonym, and the same for
145        --exclude-dir, though that is not available in GNU grep, at least as of
146        release 2.5.4.
147    
148    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
149        characters from a string of bytes) have been redefined so as not to use
150        loops, in order to improve performance in some environments. At the same
151        time, I abstracted some of the common code into auxiliary macros to save
152        repetition (this should not affect the compiled code).
153    
154    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
155        compile-time error is now given if \c is not followed by an ASCII
156        character, that is, a byte less than 128. (In EBCDIC mode, the code is
157        different, and any byte value is allowed.)
158    
159    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
160        START_OPTIMIZE option, which is now allowed at compile time - but just
161        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
162        to pcregrep and other applications that have no direct access to PCRE
163        options. The new /Y option in pcretest sets this option when calling
164        pcre_compile().
165    
166    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
167        back references. Groups containing recursive back references were forced to
168        be atomic by that change, but in the case of named groups, the amount of
169        memory required was incorrectly computed, leading to "Failed: internal
170        error: code overflow". This has been fixed.
171    
172    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
173        pcretest.c, to avoid build problems in some Borland environments.
174    
175    
176    Version 8.10 25-Jun-2010
177    ------------------------
178    
179    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
180        THEN.
181    
182    2.  (*ACCEPT) was not working when inside an atomic group.
183    
184    3.  Inside a character class, \B is treated as a literal by default, but
185        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
186        causes the error). The code is unchanged, but I tidied the documentation.
187    
188    4.  Inside a character class, PCRE always treated \R and \X as literals,
189        whereas Perl faults them if its -w option is set. I have changed PCRE so
190        that it faults them when PCRE_EXTRA is set.
191    
192    5.  Added support for \N, which always matches any character other than
193        newline. (It is the same as "." when PCRE_DOTALL is not set.)
194    
195    6.  When compiling pcregrep with newer versions of gcc which may have
196        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
197        declared with attribute warn_unused_result" were given. Just casting the
198        result to (void) does not stop the warnings; a more elaborate fudge is
199        needed. I've used a macro to implement this.
200    
201    7.  Minor change to pcretest.c to avoid a compiler warning.
202    
203    8.  Added four artifical Unicode properties to help with an option to make
204        \s etc use properties (see next item). The new properties are: Xan
205        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
206    
207    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
208        use Unicode properties. (*UCP) at the start of a pattern can be used to set
209        this option. Modified pcretest to add /W to test this facility. Added
210        REG_UCP to make it available via the POSIX interface.
211    
212    10. Added --line-buffered to pcregrep.
213    
214    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
215        studied, and the match started with a letter with a code point greater than
216        127 whose first byte was different to the first byte of the other case of
217        the letter, the other case of this starting letter was not recognized
218        (#976).
219    
220    12. If a pattern that was studied started with a repeated Unicode property
221        test, for example, \p{Nd}+, there was the theoretical possibility of
222        setting up an incorrect bitmap of starting bytes, but fortunately it could
223        not have actually happened in practice until change 8 above was made (it
224        added property types that matched character-matching opcodes).
225    
226    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
227        possible starting bytes for non-anchored patterns.
228    
229    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
230        \R, and also a number of cases that involve Unicode properties, both
231        explicit and implicit when PCRE_UCP is set.
232    
233    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
234        input, it could crash or give wrong results if characters with values
235        greater than 0xc0 were present in the subject string. (Detail: it assumed
236        UTF-8 input when processing these items.)
237    
238    16. Added a lot of (int) casts to avoid compiler warnings in systems where
239        size_t is 64-bit (#991).
240    
241    17. Added a check for running out of memory when PCRE is compiled with
242        --disable-stack-for-recursion (#990).
243    
244    18. If the last data line in a file for pcretest does not have a newline on
245        the end, a newline was missing in the output.
246    
247    19. The default pcre_chartables.c file recognizes only ASCII characters (values
248        less than 128) in its various bitmaps. However, there is a facility for
249        generating tables according to the current locale when PCRE is compiled. It
250        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
251        characters, are recognized by isspace() and therefore were getting set in
252        these tables, and indeed these tables seem to approximate to ISO 8859. This
253        caused a problem in UTF-8 mode when pcre_study() was used to create a list
254        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
255        which of course cannot start UTF-8 characters. I have changed the code so
256        that only real ASCII characters (less than 128) and the correct starting
257        bytes for UTF-8 encodings are set for characters greater than 127 when in
258        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
259        altogether.)
260    
261    20. Added the /T option to pcretest so as to be able to run tests with non-
262        standard character tables, thus making it possible to include the tests
263        used for 19 above in the standard set of tests.
264    
265    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
266        reference to a subpattern the other side of a comment that contains an
267        opening parenthesis caused either an internal compiling error, or a
268        reference to the wrong subpattern.
269    
270    
271    Version 8.02 19-Mar-2010
272    ------------------------
273    
274    1.  The Unicode data tables have been updated to Unicode 5.2.0.
275    
276    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
277        configured.
278    
279    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
280        original author of that file, following a query about its status.
281    
282    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
283        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
284    
285    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
286        quantifier applied to a forward-referencing subroutine call, could compile
287        incorrect code or give the error "internal error: previously-checked
288        referenced subpattern not found".
289    
290    6.  Both MS Visual Studio and Symbian OS have problems with initializing
291        variables to point to external functions. For these systems, therefore,
292        pcre_malloc etc. are now initialized to local functions that call the
293        relevant global functions.
294    
295    7.  There were two entries missing in the vectors called coptable and poptable
296        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
297        I've fixed the data, and added a kludgy way of testing at compile time that
298        the lengths are correct (equal to the number of opcodes).
299    
300    8.  Following on from 7, I added a similar kludge to check the length of the
301        eint vector in pcreposix.c.
302    
303    9.  Error texts for pcre_compile() are held as one long string to avoid too
304        much relocation at load time. To find a text, the string is searched,
305        counting zeros. There was no check for running off the end of the string,
306        which could happen if a new error number was added without updating the
307        string.
308    
309    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
310    
311    11. \K was not working if it appeared in an atomic group or in a group that
312        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
313        \K is "not well defined" if used in an assertion. PCRE now accepts it if
314        the assertion is positive, but not if it is negative.
315    
316    12. Change 11 fortuitously reduced the size of the stack frame used in the
317        "match()" function of pcre_exec.c by one pointer. Forthcoming
318        implementation of support for (*MARK) will need an extra pointer on the
319        stack; I have reserved it now, so that the stack frame size does not
320        decrease.
321    
322    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
323        item in branch that calls a recursion is a subroutine call - as in the
324        second branch in the above example - was incorrectly given the compile-
325        time error "recursive call could loop indefinitely" because pcre_compile()
326        was not correctly checking the subroutine for matching a non-empty string.
327    
328    14. The checks for overrunning compiling workspace could trigger after an
329        overrun had occurred. This is a "should never occur" error, but it can be
330        triggered by pathological patterns such as hundreds of nested parentheses.
331        The checks now trigger 100 bytes before the end of the workspace.
332    
333    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
334    
335    
336    Version 8.01 19-Jan-2010
337    ------------------------
338    
339    1.  If a pattern contained a conditional subpattern with only one branch (in
340        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
341        computed the wrong minimum data length (which is of course zero for such
342        subpatterns). This could cause incorrect "no match" results.
343    
344    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
345        the pattern is reset in the first branch, pcre_compile() failed with
346        "internal error: code overflow at offset...". This happened only when
347        the reset was to the original external option setting. (An optimization
348        abstracts leading options settings into an external setting, which was the
349        cause of this.)
350    
351    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
352        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
353        assertion pattern did not match (meaning that the assertion was true), it
354        was incorrectly treated as false if the SKIP had been reached during the
355        matching. This also applied to assertions used as conditions.
356    
357    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
358        assertion subpattern, including such a pattern used as a condition,
359        unpredictable results occurred, instead of the error return
360        PCRE_ERROR_DFA_UITEM.
361    
362    5.  The C++ GlobalReplace function was not working like Perl for the special
363        situation when an empty string is matched. It now does the fancy magic
364        stuff that is necessary.
365    
366    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
367        removed. (These were left over from very, very early versions of PCRE.)
368    
369    7.  Some cosmetic changes to the code to make life easier when compiling it
370        as part of something else:
371    
372        (a) Change DEBUG to PCRE_DEBUG.
373    
374        (b) In pcre_compile(), rename the member of the "branch_chain" structure
375            called "current" as "current_branch", to prevent a collision with the
376            Linux macro when compiled as a kernel module.
377    
378        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
379            prevent a collision with the Linux macro when compiled as a kernel
380            module.
381    
382    8.  In pcre_compile() there are some checks for integer overflows that used to
383        cast potentially large values to (double). This has been changed to that
384        when building, a check for int64_t is made, and if it is found, it is used
385        instead, thus avoiding the use of floating point arithmetic. (There is no
386        other use of FP in PCRE.) If int64_t is not found, the fallback is to
387        double.
388    
389    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
390        2005 (difference between two addresses compared to an unsigned value).
391    
392    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
393        custom one, because of the following reported problem in Windows:
394    
395          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
396              under Win32.
397          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
398              therefore missing the function definition.
399          - The compiler thus generates a "C" signature for the test function.
400          - The linker fails to find the "C" function.
401          - PCRE fails to configure if asked to do so against libbz2.
402    
403    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
404        messages were output:
405    
406          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
407          rerunning libtoolize, to keep the correct libtool macros in-tree.
408          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
409    
410        I have done both of these things.
411    
412    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
413        most of the time, it *can* run out if it is given a pattern that contains a
414        runaway infinite recursion. I updated the discussion in the pcrestack man
415        page.
416    
417    13. Now that we have gone to the x.xx style of version numbers, the minor
418        version may start with zero. Using 08 or 09 is a bad idea because users
419        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
420        interpreted as invalid octal numbers. I've updated the previous comment in
421        configure.ac, and also added a check that gives an error if 08 or 09 are
422        used.
423    
424    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
425        causing partial matching to fail when the end of the subject matched \W
426        in a UTF-8 pattern where \W was quantified with a minimum of 3.
427    
428    15. There were some discrepancies between the declarations in pcre_internal.h
429        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
430        their definitions. The declarations used "const uschar *" and the
431        definitions used USPTR. Even though USPTR is normally defined as "const
432        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
433        reported that: "This difference in casting confuses some C++ compilers, for
434        example, SunCC recognizes above declarations as different functions and
435        generates broken code for hbpcre." I have changed the declarations to use
436        USPTR.
437    
438    16. GNU libtool is named differently on some systems. The autogen.sh script now
439        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
440        (FreeBSD).
441    
442    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
443        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
444        comment: "Figure out how to create a longlong from a string: strtoll and
445        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
446        instance, but it only takes 2 args instead of 3!"
447    
448    18. A subtle bug concerned with back references has been fixed by a change of
449        specification, with a corresponding code fix. A pattern such as
450        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
451        refers, was giving matches when it shouldn't. For example, xa=xaaa would
452        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
453        same bug. Such groups have to be quantified to be useful, or contained
454        inside another quantified group. (If there's no repetition, the reference
455        can never match.) The problem arises because, having left the group and
456        moved on to the rest of the pattern, a later failure that backtracks into
457        the group uses the captured value from the final iteration of the group
458        rather than the correct earlier one. I have fixed this in PCRE by forcing
459        any group that contains a reference to itself to be an atomic group; that
460        is, there cannot be any backtracking into it once it has completed. This is
461        similar to recursive and subroutine calls.
462    
463    
464    Version 8.00 19-Oct-09
465    ----------------------
466    
467    1.  The table for translating pcre_compile() error codes into POSIX error codes
468        was out-of-date, and there was no check on the pcre_compile() error code
469        being within the table. This could lead to an OK return being given in
470        error.
471    
472    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
473        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
474        in a Windows environment.
475    
476    3.  The pcregrep --count option prints the count for each file even when it is
477        zero, as does GNU grep. However, pcregrep was also printing all files when
478        --files-with-matches was added. Now, when both options are given, it prints
479        counts only for those files that have at least one match. (GNU grep just
480        prints the file name in this circumstance, but including the count seems
481        more useful - otherwise, why use --count?) Also ensured that the
482        combination -clh just lists non-zero counts, with no names.
483    
484    4.  The long form of the pcregrep -F option was incorrectly implemented as
485        --fixed_strings instead of --fixed-strings. This is an incompatible change,
486        but it seems right to fix it, and I didn't think it was worth preserving
487        the old behaviour.
488    
489    5.  The command line items --regex=pattern and --regexp=pattern were not
490        recognized by pcregrep, which required --regex pattern or --regexp pattern
491        (with a space rather than an '='). The man page documented the '=' forms,
492        which are compatible with GNU grep; these now work.
493    
494    6.  No libpcreposix.pc file was created for pkg-config; there was just
495        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
496    
497    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
498        when UCP support is not needed, by modifying the Python script that
499        generates it from Unicode data files. This should not matter if the module
500        is correctly used as a library, but I received one complaint about 50K of
501        unwanted data. My guess is that the person linked everything into his
502        program rather than using a library. Anyway, it does no harm.
503    
504    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
505        was a minimum greater than 1 for a wide character in a possessive
506        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
507        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
508        character. Chaos in the form of incorrect output or a compiling loop could
509        result.
510    
511    9.  The restrictions on what a pattern can contain when partial matching is
512        requested for pcre_exec() have been removed. All patterns can now be
513        partially matched by this function. In addition, if there are at least two
514        slots in the offset vector, the offset of the earliest inspected character
515        for the match and the offset of the end of the subject are set in them when
516        PCRE_ERROR_PARTIAL is returned.
517    
518    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
519        synonymous with PCRE_PARTIAL, for backwards compatibility, and
520        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
521        and may be more useful for multi-segment matching.
522    
523    11. Partial matching with pcre_exec() is now more intuitive. A partial match
524        used to be given if ever the end of the subject was reached; now it is
525        given only if matching could not proceed because another character was
526        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
527        string "Z", which now yields "no match" instead of "partial match". In the
528        case of pcre_dfa_exec(), "no match" is given if every matching path for the
529        final character ended with (*FAIL).
530    
531    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
532        if the pattern had a "must contain" character that was already found in the
533        earlier partial match, unless partial matching was again requested. For
534        example, with the pattern /dog.(body)?/, the "must contain" character is
535        "g". If the first part-match was for the string "dog", restarting with
536        "sbody" failed. This bug has been fixed.
537    
538    13. The string returned by pcre_dfa_exec() after a partial match has been
539        changed so that it starts at the first inspected character rather than the
540        first character of the match. This makes a difference only if the pattern
541        starts with a lookbehind assertion or \b or \B (\K is not supported by
542        pcre_dfa_exec()). It's an incompatible change, but it makes the two
543        matching functions compatible, and I think it's the right thing to do.
544    
545    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
546        so that the demonstration program is easily available in environments where
547        PCRE has not been installed from source.
548    
549    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
550        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
551        library.
552    
553    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
554        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
555        is not the first non-POSIX option to be added. Clearly some people find
556        these options useful.
557    
558    17. If a caller to the POSIX matching function regexec() passes a non-zero
559        value for nmatch with a NULL value for pmatch, the value of
560        nmatch is forced to zero.
561    
562    18. RunGrepTest did not have a test for the availability of the -u option of
563        the diff command, as RunTest does. It now checks in the same way as
564        RunTest, and also checks for the -b option.
565    
566    19. If an odd number of negated classes containing just a single character
567        interposed, within parentheses, between a forward reference to a named
568        subpattern and the definition of the subpattern, compilation crashed with
569        an internal error, complaining that it could not find the referenced
570        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
571        [The bug was that it was starting one character too far in when skipping
572        over the character class, thus treating the ] as data rather than
573        terminating the class. This meant it could skip too much.]
574    
575    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
576        /g option in pcretest when the pattern contains \K, which makes it possible
577        to have an empty string match not at the start, even when the pattern is
578        anchored. Updated pcretest and pcredemo to use this option.
579    
580    21. If the maximum number of capturing subpatterns in a recursion was greater
581        than the maximum at the outer level, the higher number was returned, but
582        with unset values at the outer level. The correct (outer level) value is
583        now given.
584    
585    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
586        PCRE did not set those parentheses (unlike Perl). I have now found a way to
587        make it do so. The string so far is captured, making this feature
588        compatible with Perl.
589    
590    23. The tests have been re-organized, adding tests 11 and 12, to make it
591        possible to check the Perl 5.10 features against Perl 5.10.
592    
593    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
594        pattern matches a fixed length string. PCRE did not allow this; now it
595        does. Neither allows recursion.
596    
597    25. I finally figured out how to implement a request to provide the minimum
598        length of subject string that was needed in order to match a given pattern.
599        (It was back references and recursion that I had previously got hung up
600        on.) This code has now been added to pcre_study(); it finds a lower bound
601        to the length of subject needed. It is not necessarily the greatest lower
602        bound, but using it to avoid searching strings that are too short does give
603        some useful speed-ups. The value is available to calling programs via
604        pcre_fullinfo().
605    
606    26. While implementing 25, I discovered to my embarrassment that pcretest had
607        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
608        study optimizations had never been tested with that matching function.
609        Oops. What is worse, even when it was passed study data, there was a bug in
610        pcre_dfa_exec() that meant it never actually used it. Double oops. There
611        were also very few tests of studied patterns with pcre_dfa_exec().
612    
613    27. If (?| is used to create subpatterns with duplicate numbers, they are now
614        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
615        on the other side of the coin, they are no longer allowed to have different
616        names, because these cannot be distinguished in PCRE, and this has caused
617        confusion. (This is a difference from Perl.)
618    
619    28. When duplicate subpattern names are present (necessarily with different
620        numbers, as required by 27 above), and a test is made by name in a
621        conditional pattern, either for a subpattern having been matched, or for
622        recursion in such a pattern, all the associated numbered subpatterns are
623        tested, and the overall condition is true if the condition is true for any
624        one of them. This is the way Perl works, and is also more like the way
625        testing by number works.
626    
627    
628    Version 7.9 11-Apr-09
629    ---------------------
630    
631    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
632        (pcretest), all targets were linked against these libraries. This included
633        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
634        libraries. This caused unwanted dependencies to be created. This problem
635        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
636        pcretest is linked with readline.
637    
638    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
639        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
640        moved outside it again, because FALSE and TRUE are already defined in AIX,
641        but BOOL is not.
642    
643    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
644        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
645    
646    4.  The pcregrep documentation said spaces were inserted as well as colons (or
647        hyphens) following file names and line numbers when outputting matching
648        lines. This is not true; no spaces are inserted. I have also clarified the
649        wording for the --colour (or --color) option.
650    
651    5.  In pcregrep, when --colour was used with -o, the list of matching strings
652        was not coloured; this is different to GNU grep, so I have changed it to be
653        the same.
654    
655    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
656        each matching line was coloured. Now it goes on to look for further matches
657        of any of the test patterns, which is the same behaviour as GNU grep.
658    
659    7.  A pattern that could match an empty string could cause pcregrep to loop; it
660        doesn't make sense to accept an empty string match in pcregrep, so I have
661        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
662        seems to be how GNU grep behaves.
663    
664    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
665        start or after a newline", because the conditional assertion was not being
666        correctly handled. The rule now is that both the assertion and what follows
667        in the first alternative must satisfy the test.
668    
669    9.  If auto-callout was enabled in a pattern with a conditional group whose
670        condition was an assertion, PCRE could crash during matching, both with
671        pcre_exec() and pcre_dfa_exec().
672    
673    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
674        used for matching.
675    
676    11. Unicode property support in character classes was not working for
677        characters (bytes) greater than 127 when not in UTF-8 mode.
678    
679    12. Added the -M command line option to pcretest.
680    
681    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
682    
683    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
684    
685    16. Added comments and documentation about mis-use of no_arg in the C++
686        wrapper.
687    
688    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
689        from Martin Jerabek that uses macro names for all relevant character and
690        string constants.
691    
692    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
693        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
694        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
695        these, but not everybody uses configure.
696    
697    19. A conditional group that had only one branch was not being correctly
698        recognized as an item that could match an empty string. This meant that an
699        enclosing group might also not be so recognized, causing infinite looping
700        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
701        with the subject "ab", where knowledge that the repeated group can match
702        nothing is needed in order to break the loop.
703    
704    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
705        exec(), but without supplying a callout function, matching went wrong.
706    
707    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
708        leak if the size of the offset vector was greater than 30. When the vector
709        is smaller, the saved offsets during recursion go onto a local stack
710        vector, but for larger vectors malloc() is used. It was failing to free
711        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
712        error, in fact).
713    
714    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
715        heapframe that is used only when UTF-8 support is enabled. This caused no
716        problem, but was untidy.
717    
718    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
719        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
720        included within another project.
721    
722    24. Steven Van Ingelgem's patches to add more options to the CMake support,
723        slightly modified by me:
724    
725          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
726              not building pcregrep.
727    
728          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
729              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
730    
731    25. Forward references, both numeric and by name, in patterns that made use of
732        duplicate group numbers, could behave incorrectly or give incorrect errors,
733        because when scanning forward to find the reference group, PCRE was not
734        taking into account the duplicate group numbers. A pattern such as
735        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
736    
737    26. Changed a few more instances of "const unsigned char *" to USPTR, making
738        the feature of a custom pointer more persuasive (as requested by a user).
739    
740    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
741        pcretest.c, inside #ifndefs, because it seems they are sometimes already
742        pre-defined.
743    
744    28. Added support for (*UTF8) at the start of a pattern.
745    
746    29. Arrange for flags added by the "release type" setting in CMake to be shown
747        in the configuration summary.
748    
749    
750    Version 7.8 05-Sep-08
751    ---------------------
752    
753    1.  Replaced UCP searching code with optimized version as implemented for Ad
754        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
755        stage table and inline lookup instead of a function, giving speed ups of 2
756        to 5 times on some simple patterns that I tested. Permission was given to
757        distribute the MultiStage2.py script that generates the tables (it's not in
758        the tarball, but is in the Subversion repository).
759    
760    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
761        scripts.
762    
763    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
764        a group with a zero qualifier. The result of the study could be incorrect,
765        or the function might crash, depending on the pattern.
766    
767    4.  Caseless matching was not working for non-ASCII characters in back
768        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
769        It now works when Unicode Property Support is available.
770    
771    5.  In pcretest, an escape such as \x{de} in the data was always generating
772        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
773        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
774        truncation.
775    
776    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
777    
778    7.  Added two (int) casts to pcregrep when printing the difference of two
779        pointers, in case they are 64-bit values.
780    
781    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
782        test 2 if it fails.
783    
784    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
785        and a #define of that name to empty if it is not externally set. This is to
786        allow users of MSVC to set it if necessary.
787    
788    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
789        the convenience functions in the pcre_get.c source file.
790    
791    11. An option change at the start of a pattern that had top-level alternatives
792        could cause overwriting and/or a crash. This command provoked a crash in
793        some environments:
794    
795          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
796    
797        This potential security problem was recorded as CVE-2008-2371.
798    
799    12. For a pattern where the match had to start at the beginning or immediately
800        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
801        pcre_dfa_exec() could read past the end of the passed subject if there was
802        no match. To help with detecting such bugs (e.g. with valgrind), I modified
803        pcretest so that it places the subject at the end of its malloc-ed buffer.
804    
805    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
806        exec() might read past the end of the data buffer in UTF-8 mode.
807    
808    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
809        the data contained the byte 0x85 as part of a UTF-8 character within its
810        first line. This applied both to normal and DFA matching.
811    
812    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
813        /^[^d]*?$/8 failed to match "abc".
814    
815    16. Added a missing copyright notice to pcrecpp_internal.h.
816    
817    17. Make it more clear in the documentation that values returned from
818        pcre_exec() in ovector are byte offsets, not character counts.
819    
820    18. Tidied a few places to stop certain compilers from issuing warnings.
821    
822    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
823        supplied by Stefan Weber. I made a further small update for 7.8 because
824        there is a change of source arrangements: the pcre_searchfuncs.c module is
825        replaced by pcre_ucd.c.
826    
827    
828    Version 7.7 07-May-08
829    ---------------------
830    
831    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
832        a string to a long long, pretend we don't even have a long long." This is
833        done by checking for the strtoq, strtoll, and _strtoi64 functions.
834    
835    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
836        pre-7.6 versions, which defined a global no_arg variable instead of putting
837        it in the RE class. (See also #8 below.)
838    
839    3.  Remove a line of dead code, identified by coverity and reported by Nuno
840        Lopes.
841    
842    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
843    
844        (1) The include/exclude patterns were being applied to the whole pathnames
845            of files, instead of just to the final components.
846    
847        (2) If there was more than one level of directory, the subdirectories were
848            skipped unless they satisfied the include/exclude conditions. This is
849            inconsistent with GNU grep (and could even be seen as contrary to the
850            pcregrep specification - which I improved to make it absolutely clear).
851            The action now is always to scan all levels of directory, and just
852            apply the include/exclude patterns to regular files.
853    
854    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
855        --exclude_dir in the tests to avoid scanning .svn directories.
856    
857    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
858        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
859        doesn't support NULs in patterns.
860    
861    7.  Added some missing "const"s to declarations of static tables in
862        pcre_compile.c and pcre_dfa_exec.c.
863    
864    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
865        caused by fix #2  above. (Subsequently also a second patch to fix the
866        first patch. And a third patch - this was a messy problem.)
867    
868    9.  Applied Craig's patch to remove the use of push_back().
869    
870    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
871        matching function regexec().
872    
873    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
874        which, however, unlike Perl's \g{...}, are subroutine calls, not back
875        references. PCRE supports relative numbers with this syntax (I don't think
876        Oniguruma does).
877    
878    12. Previously, a group with a zero repeat such as (...){0} was completely
879        omitted from the compiled regex. However, this means that if the group
880        was called as a subroutine from elsewhere in the pattern, things went wrong
881        (an internal error was given). Such groups are now left in the compiled
882        pattern, with a new opcode that causes them to be skipped at execution
883        time.
884    
885    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
886        to the way PCRE behaves:
887    
888        (a) A lone ] character is dis-allowed (Perl treats it as data).
889    
890        (b) A back reference to an unmatched subpattern matches an empty string
891            (Perl fails the current match path).
892    
893        (c) A data ] in a character class must be notated as \] because if the
894            first data character in a class is ], it defines an empty class. (In
895            Perl it is not possible to have an empty class.) The empty class []
896            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
897            The negative empty class [^] matches any one character, independently
898            of the DOTALL setting.
899    
900    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
901        non-existent subpattern following a character class starting with ']' and
902        containing () gave an internal compiling error instead of "reference to
903        non-existent subpattern". Fortunately, when the pattern did exist, the
904        compiled code was correct. (When scanning forwards to check for the
905        existencd of the subpattern, it was treating the data ']' as terminating
906        the class, so got the count wrong. When actually compiling, the reference
907        was subsequently set up correctly.)
908    
909    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
910        it was being rejected as not supported by pcre_dfa_exec(), even though
911        other assertions are supported. I have made pcre_dfa_exec() support
912        (*FAIL).
913    
914    16. The implementation of 13c above involved the invention of a new opcode,
915        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
916        cannot be changed at match time, I realized I could make a small
917        improvement to matching performance by compiling OP_ALLANY instead of
918        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
919        on the OP_ANY path.
920    
921    17. Compiling pcretest on Windows with readline support failed without the
922        following two fixes: (1) Make the unistd.h include conditional on
923        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
924    
925    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
926        ncurses library to be included for pcretest when ReadLine support is
927        requested, but also to allow for it to be overridden. This patch came from
928        Daniel Bergström.
929    
930    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
931        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
932        any errors with the current Unicode tables. Thanks to Peter Kankowski for
933        spotting this.
934    
935    
936    Version 7.6 28-Jan-08
937    ---------------------
938    
939    1.  A character class containing a very large number of characters with
940        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
941        overflow.
942    
943    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
944        HAVE_LONG_LONG is not defined.
945    
946    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
947        bring them up to date and include new features. This patch includes:
948    
949        - Fixed PH's badly added libz and libbz2 support.
950        - Fixed a problem with static linking.
951        - Added pcredemo. [But later removed - see 7 below.]
952        - Fixed dftables problem and added an option.
953        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
954            HAVE_LONG_LONG.
955        - Added readline support for pcretest.
956        - Added an listing of the option settings after cmake has run.
957    
958    4.  A user submitted a patch to Makefile that makes it easy to create
959        "pcre.dll" under mingw when using Configure/Make. I added stuff to
960        Makefile.am that cause it to include this special target, without
961        affecting anything else. Note that the same mingw target plus all
962        the other distribution libraries and programs are now supported
963        when configuring with CMake (see 6 below) instead of with
964        Configure/Make.
965    
966    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
967        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
968        exported in the Windows port". It has not yet been confirmed that the patch
969        solves the problem, but it does no harm.
970    
971    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
972        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
973        with CMake, and also correct the comment about stack recursion.
974    
975    7.  Remove the automatic building of pcredemo from the ./configure system and
976        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
977        of a program that users should build themselves after PCRE is installed, so
978        building it automatically is not really right. What is more, it gave
979        trouble in some build environments.
980    
981    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
982    
983    
984    Version 7.5 10-Jan-08
985    ---------------------
986    
987    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
988        values in parens when parsing an RE using the C++ wrapper."
989    
990    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
991        Characters greater than 255 were excluded from the class instead of being
992        included.
993    
994    3.  The same bug as (2) above applied to negated POSIX classes such as
995        [:^space:].
996    
997    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
998        defined or documented. It seems to have been a typo for PCRE_STATIC, so
999        I have changed it.
1000    
1001    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1002        first named subpattern) and a construct such as (?&a) would reference the
1003        first named subpattern whose name started with "a" (in other words, the
1004        length check was missing). Both these problems are fixed. "Subpattern name
1005        expected" is now given for (?&) (a zero-length name), and this patch also
1006        makes it give the same error for \k'' (previously it complained that that
1007        was a reference to a non-existent subpattern).
1008    
1009    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1010        this is right because (?- can be followed by option settings as well as by
1011        digits. I have, however, made the messages clearer.
1012    
1013    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1014        than the number used in the conditional) now cause a compile-time error.
1015        This is actually not compatible with Perl, which accepts such patterns, but
1016        treats the conditional as always being FALSE (as PCRE used to), but it
1017        seems to me that giving a diagnostic is better.
1018    
1019    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1020        and messages.
1021    
1022    9.  Fix two occurrences of "backslash" in comments that should have been
1023        "backspace".
1024    
1025    10. Remove two redundant lines of code that can never be obeyed (their function
1026        was moved elsewhere).
1027    
1028    11. The program that makes PCRE's Unicode character property table had a bug
1029        which caused it to generate incorrect table entries for sequences of
1030        characters that have the same character type, but are in different scripts.
1031        It amalgamated them into a single range, with the script of the first of
1032        them. In other words, some characters were in the wrong script. There were
1033        thirteen such cases, affecting characters in the following ranges:
1034    
1035          U+002b0 - U+002c1
1036          U+0060c - U+0060d
1037          U+0061e - U+00612
1038          U+0064b - U+0065e
1039          U+0074d - U+0076d
1040          U+01800 - U+01805
1041          U+01d00 - U+01d77
1042          U+01d9b - U+01dbf
1043          U+0200b - U+0200f
1044          U+030fc - U+030fe
1045          U+03260 - U+0327f
1046          U+0fb46 - U+0fbb1
1047          U+10450 - U+1049d
1048    
1049    12. The -o option (show only the matching part of a line) for pcregrep was not
1050        compatible with GNU grep in that, if there was more than one match in a
1051        line, it showed only the first of them. It now behaves in the same way as
1052        GNU grep.
1053    
1054    13. If the -o and -v options were combined for pcregrep, it printed a blank
1055        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1056        does the same. The return code can be used to tell if there were any
1057        non-matching lines.
1058    
1059    14. Added --file-offsets and --line-offsets to pcregrep.
1060    
1061    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1062        infinitely looping recursion. The bug was that positive lookaheads were not
1063        being skipped when checking for a possible empty match (negative lookaheads
1064        and both kinds of lookbehind were skipped).
1065    
1066    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1067        inclusion of <windows.h> to before rather than after the definition of
1068        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1069    
1070    17. Specifying a possessive quantifier with a specific limit for a Unicode
1071        character property caused pcre_compile() to compile bad code, which led at
1072        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1073        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1074        caused the error; without that there was no problem.
1075    
1076    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1077    
1078    19. Added --enable-pcretest-libreadline.
1079    
1080    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1081        RE::GlobalReplace(). As a result, the number of replacements returned was
1082        double what it should be. I removed one of the increments, but Craig sent a
1083        later patch that removed the other one (the right fix) and added unit tests
1084        that check the return values (which was not done before).
1085    
1086    21. Several CMake things:
1087    
1088        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1089            the names libpcre and libpcreposix, not just pcre and pcreposix.
1090    
1091        (2) The above change means that pcretest and pcregrep are now correctly
1092            linked with the newly-built libraries, not previously installed ones.
1093    
1094        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
1095    
1096    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1097        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1098        UTF-8 newline character). The key issue is that the pattern starts .*;
1099        this means that the match must be either at the beginning, or after a
1100        newline. The bug was in the code for advancing after a failed match and
1101        checking that the new position followed a newline. It was not taking
1102        account of UTF-8 characters correctly.
1103    
1104    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1105        character classes. PCRE was not treating the sequence [:...:] as a
1106        character class unless the ... were all letters. Perl, however, seems to
1107        allow any characters between [: and :], though of course it rejects as
1108        unknown any "names" that contain non-letters, because all the known class
1109        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1110        for example, whereas PCRE did not - it did not recognize a POSIX character
1111        class. This seemed a bit dangerous, so the code has been changed to be
1112        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1113        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1114        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1115        Perl does, and where it didn't before.
1116    
1117    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1118        Windows environments %n is disabled by default.
1119    
1120    
1121    Version 7.4 21-Sep-07
1122    ---------------------
1123    
1124    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1125        means that a class such as [\s] counted as "explicit reference to CR or
1126        LF". That isn't really right - the whole point of the change was to try to
1127        help when there was an actual mention of one of the two characters. So now
1128        the change happens only if \r or \n (or a literal CR or LF) character is
1129        encountered.
1130    
1131    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1132        of both had grown to the point where there were only 3 bits left.
1133        Fortunately, there was spare space in the data structure, and so I have
1134        moved the internal flags into a new 16-bit field to free up more option
1135        bits.
1136    
1137    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1138        but did not set the internal JCHANGED flag - either of these is enough to
1139        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1140        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1141        start sets both bits.
1142    
1143    4.  Added options (at build time, compile time, exec time) to change \R from
1144        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1145    
1146    5.  doc/pcresyntax.html was missing from the distribution.
1147    
1148    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1149        compatibility, even though it is no longer used.
1150    
1151    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1152        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1153        windows.h file is present (where different names are used). [This was
1154        reversed later after testing - see 16 below.]
1155    
1156    8.  Changed all #include <config.h> to #include "config.h". There were also
1157        some further <pcre.h> cases that I changed to "pcre.h".
1158    
1159    9.  When pcregrep was used with the --colour option, it missed the line ending
1160        sequence off the lines that it output.
1161    
1162    10. It was pointed out to me that arrays of string pointers cause lots of
1163        relocations when a shared library is dynamically loaded. A technique of
1164        using a single long string with a table of offsets can drastically reduce
1165        these. I have refactored PCRE in four places to do this. The result is
1166        dramatic:
1167    
1168          Originally:                          290
1169          After changing UCP table:            187
1170          After changing error message table:   43
1171          After changing table of "verbs"       36
1172          After changing table of Posix names   22
1173    
1174        Thanks to the folks working on Gregex for glib for this insight.
1175    
1176    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1177        unicode-properties was also set.
1178    
1179    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1180    
1181    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1182        checked only for CRLF.
1183    
1184    14. Added casts to pcretest.c to avoid compiler warnings.
1185    
1186    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1187    
1188    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1189        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1190        entirely. This removes changes made in 7 above.
1191    
1192    17. The CMake files have been updated, and there is now more information about
1193        building with CMake in the NON-UNIX-USE document.
1194    
1195    
1196    Version 7.3 28-Aug-07
1197    ---------------------
1198    
1199     1. In the rejigging of the build system that eventually resulted in 7.1, the
1200        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1201        brackets there is not right, since it causes compilers to look for an
1202        installed pcre.h, not the version that is in the source that is being
1203        compiled (which of course may be different). I have changed it back to:
1204    
1205          #include "pcre.h"
1206    
1207        I have a vague recollection that the change was concerned with compiling in
1208        different directories, but in the new build system, that is taken care of
1209        by the VPATH setting the Makefile.
1210    
1211     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1212        when the subject happened to end in the byte 0x85 (e.g. if the last
1213        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1214        characters but of course it shouldn't be taken as a newline when it is part
1215        of another character. The bug was that, for an unlimited repeat of . in
1216        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1217        characters when looking for a newline.
1218    
1219     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1220    
1221     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1222        in debug output.
1223    
1224     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1225        long printing in the pcrecpp unittest when running under MinGW.
1226    
1227     6. ESC_K was left out of the EBCDIC table.
1228    
1229     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1230        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1231        limit also applies to "virtual nesting" when a pattern is recursive, and in
1232        this case 1000 isn't so big. I have been able to remove this limit at the
1233        expense of backing off one optimization in certain circumstances. Normally,
1234        when pcre_exec() would call its internal match() function recursively and
1235        immediately return the result unconditionally, it uses a "tail recursion"
1236        feature to save stack. However, when a subpattern that can match an empty
1237        string has an unlimited repetition quantifier, it no longer makes this
1238        optimization. That gives it a stack frame in which to save the data for
1239        checking that an empty string has been matched. Previously this was taken
1240        from the 1000-entry workspace that had been reserved. So now there is no
1241        explicit limit, but more stack is used.
1242    
1243     8. Applied Daniel's patches to solve problems with the import/export magic
1244        syntax that is required for Windows, and which was going wrong for the
1245        pcreposix and pcrecpp parts of the library. These were overlooked when this
1246        problem was solved for the main library.
1247    
1248     9. There were some crude static tests to avoid integer overflow when computing
1249        the size of patterns that contain repeated groups with explicit upper
1250        limits. As the maximum quantifier is 65535, the maximum group length was
1251        set at 30,000 so that the product of these two numbers did not overflow a
1252        32-bit integer. However, it turns out that people want to use groups that
1253        are longer than 30,000 bytes (though not repeat them that many times).
1254        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1255        made it possible to implement the integer overflow checks in a much more
1256        dynamic way, which I have now done. The artificial limitation on group
1257        length has been removed - we now have only the limit on the total length of
1258        the compiled pattern, which depends on the LINK_SIZE setting.
1259    
1260    10. Fixed a bug in the documentation for get/copy named substring when
1261        duplicate names are permitted. If none of the named substrings are set, the
1262        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1263        empty string.
1264    
1265    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1266        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1267        because the ] is interpreted as the first data character and the
1268        terminating ] is not found. PCRE has been made compatible with Perl in this
1269        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1270        cause memory overwriting.
1271    
1272    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1273        string has been matched (to stop an infinite loop). It was not recognizing
1274        a conditional subpattern that could match an empty string if that
1275        subpattern was within another subpattern. For example, it looped when
1276        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1277        condition was not nested. This bug has been fixed.
1278    
1279    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1280        past the start of the subject in the presence of bytes with the top bit
1281        set, for example "\x8aBCD".
1282    
1283    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1284        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1285    
1286    14. Optimized (?!) to (*FAIL).
1287    
1288    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1289        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1290        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1291        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1292        does: it's just the validity check that is more restrictive.
1293    
1294    16. Inserted checks for integer overflows during escape sequence (backslash)
1295        processing, and also fixed erroneous offset values for syntax errors during
1296        backslash processing.
1297    
1298    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1299        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1300    
1301    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1302        caused an overrun.
1303    
1304    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1305        something other than just ASCII characters) inside a group that had an
1306        unlimited repeat caused a loop at compile time (while checking to see
1307        whether the group could match an empty string).
1308    
1309    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1310        [\P{Any}] did so. (Error in the code for printing property names.)
1311    
1312    21. An orphan \E inside a character class could cause a crash.
1313    
1314    22. A repeated capturing bracket such as (A)? could cause a wild memory
1315        reference during compilation.
1316    
1317    23. There are several functions in pcre_compile() that scan along a compiled
1318        expression for various reasons (e.g. to see if it's fixed length for look
1319        behind). There were bugs in these functions when a repeated \p or \P was
1320        present in the pattern. These operators have additional parameters compared
1321        with \d, etc, and these were not being taken into account when moving along
1322        the compiled data. Specifically:
1323    
1324        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1325            length.
1326    
1327        (b) An item such as \pL+ within a repeated group could cause crashes or
1328            loops.
1329    
1330        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1331            "reference to non-existent subpattern" error.
1332    
1333        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1334    
1335    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1336        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1337    
1338    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1339    
1340    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1341        character were causing crashes (broken optimization).
1342    
1343    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1344        \p or \P) caused a compile-time loop.
1345    
1346    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1347        break. For example, the unstudied pattern [\r\n]A does not match the string
1348        "\r\nA" because change 7.0/46 below moves the current point on by two
1349        characters after failing to match at the start. However, the pattern \nA
1350        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1351        the same is true. There doesn't seem any very clean way out of this, but
1352        what I have chosen to do makes the common cases work: PCRE now takes note
1353        of whether there can be an explicit match for \r or \n anywhere in the
1354        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1355        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1356        pattern has explicit CR or LF references.
1357    
1358    29. Added (*CR) etc for changing newline setting at start of pattern.
1359    
1360    
1361    Version 7.2 19-Jun-07
1362    ---------------------
1363    
1364     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1365        which is apparently normally available under Windows.
1366    
1367     2. Re-jig the pcregrep tests with different newline settings in an attempt
1368        to make them independent of the local environment's newline setting.
1369    
1370     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1371    
1372     4. Some of the "internals" tests were previously cut out when the link size
1373        was not 2, because the output contained actual offsets. The recent new
1374        "Z" feature of pcretest means that these can be cut out, making the tests
1375        usable with all link sizes.
1376    
1377     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1378        stack recursion. This gives a massive performance boost under BSD, but just
1379        a small improvement under Linux. However, it saves one field in the frame
1380        in all cases.
1381    
1382     6. Added more features from the forthcoming Perl 5.10:
1383    
1384        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1385            recursion call. It refers to the nth most recently opened parentheses.
1386    
1387        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1388            to be opened parentheses.
1389    
1390        (c) Conditions that refer to capturing parentheses can be specified
1391            relatively, for example, (?(-2)... or (?(+3)...
1392    
1393        (d) \K resets the start of the current match so that everything before
1394            is not part of it.
1395    
1396        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1397    
1398        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1399            reference syntax.
1400    
1401        (g) (?| introduces a group in which the numbering of parentheses in each
1402            alternative starts with the same number.
1403    
1404        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1405    
1406     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1407        PCRE_INFO_JCHANGED.
1408    
1409     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1410        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1411        for detecting groups that can match an empty string.
1412    
1413     9. A pattern with a very large number of alternatives (more than several
1414        hundred) was running out of internal workspace during the pre-compile
1415        phase, where pcre_compile() figures out how much memory will be needed. A
1416        bit of new cunning has reduced the workspace needed for groups with
1417        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1418        workspace instead of running out of the 4096 that are available.
1419    
1420    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1421    
1422    11. Applied patch from Google to remove an optimization that didn't quite work.
1423        The report of the bug said:
1424    
1425          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1426          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1427          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1428    
1429    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1430        it matched the wrong number of bytes.
1431    
1432    
1433    Version 7.1 24-Apr-07
1434    ---------------------
1435    
1436     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1437        that is more "standard", making use of automake and other Autotools. There
1438        is some re-arrangement of the files and adjustment of comments consequent
1439        on this.
1440    
1441     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1442        for recursive directory scanning broke on some systems because the files
1443        are not scanned in any specific order and on different systems the order
1444        was different. A call to "sort" has been inserted into RunGrepTest for the
1445        approprate test as a short-term fix. In the longer term there may be an
1446        alternative.
1447    
1448     3. I had an email from Eric Raymond about problems translating some of PCRE's
1449        man pages to HTML (despite the fact that I distribute HTML pages, some
1450        people do their own conversions for various reasons). The problems
1451        concerned the use of low-level troff macros .br and .in. I have therefore
1452        removed all such uses from the man pages (some were redundant, some could
1453        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1454        HTML has been updated to handle .nf/.fi and to complain if it encounters
1455        .br or .in.
1456    
1457     4. Updated comments in configure.ac that get placed in config.h.in and also
1458        arranged for config.h to be included in the distribution, with the name
1459        config.h.generic, for the benefit of those who have to compile without
1460        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1461    
1462     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1463        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1464        makevp.bat for the current PCRE, using the additional files
1465        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1466    
1467     6. A Windows user reported a minor discrepancy with test 2, which turned out
1468        to be caused by a trailing space on an input line that had got lost in his
1469        copy. The trailing space was an accident, so I've just removed it.
1470    
1471     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1472        that is needed.
1473    
1474     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1475        as "const" (a) because they are and (b) because it helps the PHP
1476        maintainers who have recently made a script to detect big data structures
1477        in the php code that should be moved to the .rodata section. I remembered
1478        to update Builducptable as well, so it won't revert if ucptable.h is ever
1479        re-created.
1480    
1481     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1482        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1483        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1484        support is not required. This saves 1.5-2K of code, which is important in
1485        some applications.
1486    
1487        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1488        so as not to refer to the tables, even though these functions will never be
1489        called when UTF-8 support is disabled. Otherwise there are problems with a
1490        shared library.
1491    
1492    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1493    
1494        (a) It was defining its arguments as char * instead of void *.
1495    
1496        (b) It was assuming that all moves were upwards in memory; this was true
1497            a long time ago when I wrote it, but is no longer the case.
1498    
1499        The emulated memove() is provided for those environments that have neither
1500        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1501        is clearly not the case, as these two bugs were recently reported.
1502    
1503    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1504        and Detrail to create the HTML documentation, the .txt form of the man
1505        pages, and it removes trailing spaces from listed files. It also creates
1506        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1507        case, it wraps all the #defines with #ifndefs. This script should be run
1508        before "make dist".
1509    
1510    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1511        with Unicode property support.
1512    
1513        (a) For a maximizing quantifier, if the two different cases of the
1514            character were of different lengths in their UTF-8 codings (there are
1515            some cases like this - I found 11), and the matching function had to
1516            back up over a mixture of the two cases, it incorrectly assumed they
1517            were both the same length.
1518    
1519        (b) When PCRE was configured to use the heap rather than the stack for
1520            recursion during matching, it was not correctly preserving the data for
1521            the other case of a UTF-8 character when checking ahead for a match
1522            while processing a minimizing repeat. If the check also involved
1523            matching a wide character, but failed, corruption could cause an
1524            erroneous result when trying to check for a repeat of the original
1525            character.
1526    
1527    13. Some tidying changes to the testing mechanism:
1528    
1529        (a) The RunTest script now detects the internal link size and whether there
1530            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1531            values substituted by "configure". (The RunGrepTest script already did
1532            this for UTF-8.) The configure.ac script no longer substitutes the
1533            relevant variables.
1534    
1535        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1536            with length and offset values. This means that the output is different
1537            for different internal link sizes. Test 2 is skipped for link sizes
1538            other than 2 because of this, bypassing the problem. Unfortunately,
1539            there was also a test in test 3 (the locale tests) that used /B and
1540            failed for link sizes other than 2. Rather than cut the whole test out,
1541            I have added a new /Z option to pcretest that replaces the length and
1542            offset values with spaces. This is now used to make test 3 independent
1543            of link size. (Test 2 will be tidied up later.)
1544    
1545    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1546        segmentation fault instead of returning the appropriate error message.
1547    
1548    15. In multiline mode when the newline sequence was set to "any", the pattern
1549        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1550        This doesn't seem right; it now treats the CRLF combination as the line
1551        ending, and so does not match in that case. It's only a pattern such as ^$
1552        that would hit this one: something like ^ABC$ would have failed after \r
1553        and then tried again after \r\n.
1554    
1555    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1556        in an attempt to make files that differ only in their line terminators
1557        compare equal. This works on Linux.
1558    
1559    17. Under certain error circumstances pcregrep might try to free random memory
1560        as it exited. This is now fixed, thanks to valgrind.
1561    
1562    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1563        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1564        was because its rules for how to advance for /g after matching an empty
1565        string at the end of a line did not allow for this case. They now check for
1566        it specially.
1567    
1568    20. pcretest is supposed to handle patterns and data of any length, by
1569        extending its buffers when necessary. It was getting this wrong when the
1570        buffer for a data line had to be extended.
1571    
1572    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1573        CRLF as a newline sequence.
1574    
1575    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1576        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1577        I have nevertheless tidied it up.
1578    
1579    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1580    
1581    24. Added a man page for pcre-config.
1582    
1583    
1584    Version 7.0 19-Dec-06
1585    ---------------------
1586    
1587     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1588        moving to gcc 4.1.1.
1589    
1590     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1591        sys/time.h, which is documented as needed for this function. It doesn't
1592        seem to matter on Linux, but it showed up on some releases of OS X.
1593    
1594     3. It seems that there are systems where bytes whose values are greater than
1595        127 match isprint() in the "C" locale. The "C" locale should be the
1596        default when a C program starts up. In most systems, only ASCII printing
1597        characters match isprint(). This difference caused the output from pcretest
1598        to vary, making some of the tests fail. I have changed pcretest so that:
1599    
1600        (a) When it is outputting text in the compiled version of a pattern, bytes
1601            other than 32-126 are always shown as hex escapes.
1602    
1603        (b) When it is outputting text that is a matched part of a subject string,
1604            it does the same, unless a different locale has been set for the match
1605            (using the /L modifier). In this case, it uses isprint() to decide.
1606    
1607     4. Fixed a major bug that caused incorrect computation of the amount of memory
1608        required for a compiled pattern when options that changed within the
1609        pattern affected the logic of the preliminary scan that determines the
1610        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1611        that the computed length was too small. The symptoms of this bug were
1612        either the PCRE error "internal error: code overflow" from pcre_compile(),
1613        or a glibc crash with a message such as "pcretest: free(): invalid next
1614        size (fast)". Examples of patterns that provoked this bug (shown in
1615        pcretest format) are:
1616    
1617          /(?-x: )/x
1618          /(?x)(?-x: \s*#\s*)/
1619          /((?i)[\x{c0}])/8
1620          /(?i:[\x{c0}])/8
1621    
1622        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1623        is now done differently.
1624    
1625     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1626        wrapper classes; (b) implement a new function in the C++ scanner that is
1627        more efficient than the old way of doing things because it avoids levels of
1628        recursion in the regex matching; (c) add a paragraph to the documentation
1629        for the FullMatch() function.
1630    
1631     6. The escape sequence \n was being treated as whatever was defined as
1632        "newline". Not only was this contrary to the documentation, which states
1633        that \n is character 10 (hex 0A), but it also went horribly wrong when
1634        "newline" was defined as CRLF. This has been fixed.
1635    
1636     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1637        was being set to -1 for the "end of line" case (supposedly a value that no
1638        character can have). Though this value is never used (the check for end of
1639        line is "zero bytes in current character"), it caused compiler complaints.
1640        I've changed it to 0xffffffff.
1641    
1642     8. In pcre_version.c, the version string was being built by a sequence of
1643        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1644        string (as it is for production releases) called a macro with an empty
1645        argument. The C standard says the result of this is undefined. The gcc
1646        compiler treats it as an empty string (which was what was wanted) but it is
1647        reported that Visual C gives an error. The source has been hacked around to
1648        avoid this problem.
1649    
1650     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1651        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1652        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1653        of them did).
1654    
1655    10. Originally, pcretest opened its input and output without "b"; then I was
1656        told that "b" was needed in some environments, so it was added for release
1657        5.0 to both the input and output. (It makes no difference on Unix-like
1658        systems.) Later I was told that it is wrong for the input on Windows. I've
1659        now abstracted the modes into two macros, to make it easier to fiddle with
1660        them, and removed "b" from the input mode under Windows.
1661    
1662    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1663    
1664    12. Added -help and --help to pcretest as an official way of being reminded
1665        of the options.
1666    
1667    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1668        and pcrecpp.cc because they annoy compilers at high warning levels.
1669    
1670    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1671    
1672    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1673        scripts are not C programs :-) and which was not noticed because it works
1674        on Linux.
1675    
1676    16. pcretest is supposed to handle any length of pattern and data line (as one
1677        line or as a continued sequence of lines) by extending its input buffer if
1678        necessary. This feature was broken for very long pattern lines, leading to
1679        a string of junk being passed to pcre_compile() if the pattern was longer
1680        than about 50K.
1681    
1682    17. I have done a major re-factoring of the way pcre_compile() computes the
1683        amount of memory needed for a compiled pattern. Previously, there was code
1684        that made a preliminary scan of the pattern in order to do this. That was
1685        OK when PCRE was new, but as the facilities have expanded, it has become
1686        harder and harder to keep it in step with the real compile phase, and there
1687        have been a number of bugs (see for example, 4 above). I have now found a
1688        cunning way of running the real compile function in a "fake" mode that
1689        enables it to compute how much memory it would need, while actually only
1690        ever using a few hundred bytes of working memory and without too many
1691        tests of the mode. This should make future maintenance and development
1692        easier. A side effect of this work is that the limit of 200 on the nesting
1693        depth of parentheses has been removed (though this was never a serious
1694        limitation, I suspect). However, there is a downside: pcre_compile() now
1695        runs more slowly than before (30% or more, depending on the pattern). I
1696        hope this isn't a big issue. There is no effect on runtime performance.
1697    
1698    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1699        newline (only possible for the last line of a file) and it was a
1700        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1701    
1702    19. Added additional timing features to pcretest. (1) The -tm option now times
1703        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1704        separate command line item, by a number that specifies the number of
1705        repeats to use when timing. The default is 50000; this gives better
1706        precision, but takes uncomfortably long for very large patterns.
1707    
1708    20. Extended pcre_study() to be more clever in cases where a branch of a
1709        subpattern has no definite first character. For example, (a*|b*)[cd] would
1710        previously give no result from pcre_study(). Now it recognizes that the
1711        first character must be a, b, c, or d.
1712    
1713    21. There was an incorrect error "recursive call could loop indefinitely" if
1714        a subpattern (or the entire pattern) that was being tested for matching an
1715        empty string contained only one non-empty item after a nested subpattern.
1716        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1717        incorrectly, because the \d was being skipped in the check.
1718    
1719    22. The pcretest program now has a new pattern option /B and a command line
1720        option -b, which is equivalent to adding /B to every pattern. This causes
1721        it to show the compiled bytecode, without the additional information that
1722        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1723        is the same as /B/I).
1724    
1725    23. A new optimization is now able automatically to treat some sequences such
1726        as a*b as a*+b. More specifically, if something simple (such as a character
1727        or a simple class like \d) has an unlimited quantifier, and is followed by
1728        something that cannot possibly match the quantified thing, the quantifier
1729        is automatically "possessified".
1730    
1731    24. A recursive reference to a subpattern whose number was greater than 39
1732        went wrong under certain circumstances in UTF-8 mode. This bug could also
1733        have affected the operation of pcre_study().
1734    
1735    25. Realized that a little bit of performance could be had by replacing
1736        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1737    
1738    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1739    
1740    27. Possessive quantifiers such as a++ were previously implemented by turning
1741        them into atomic groups such as ($>a+). Now they have their own opcodes,
1742        which improves performance. This includes the automatically created ones
1743        from 23 above.
1744    
1745    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1746        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1747        the first matched character to be a colon. This applied both to named and
1748        numbered groups.
1749    
1750    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1751    
1752    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1753        building PCRE on Windows easier, so I have included it in the distribution.
1754    
1755    31. There is now a check in pcretest against a ridiculously large number being
1756        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1757        loop, the loop is abandoned.
1758    
1759    32. Forward references to subpatterns in conditions such as (?(2)...) where
1760        subpattern 2 is defined later cause pcre_compile() to search forwards in
1761        the pattern for the relevant set of parentheses. This search went wrong
1762        when there were unescaped parentheses in a character class, parentheses
1763        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1764    
1765    33. "Subroutine" calls and backreferences were previously restricted to
1766        referencing subpatterns earlier in the regex. This restriction has now
1767        been removed.
1768    
1769    34. Added a number of extra features that are going to be in Perl 5.10. On the
1770        whole, these are just syntactic alternatives for features that PCRE had
1771        previously implemented using the Python syntax or my own invention. The
1772        other formats are all retained for compatibility.
1773    
1774        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1775            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1776            also .NET compatible.
1777    
1778        (b) A recursion or subroutine call to a named group can now be defined as
1779            (?&name) as well as (?P>name).
1780    
1781        (c) A backreference to a named group can now be defined as \k<name> or
1782            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1783            5.10, are also .NET compatible.
1784    
1785        (d) A conditional reference to a named group can now use the syntax
1786            (?(<name>) or (?('name') as well as (?(name).
1787    
1788        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1789            groups (named and numbered) that are never evaluated inline, but can be
1790            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1791            is always false. There may be only one alternative in such a group.
1792    
1793        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1794            as the simple (?(R). The condition is true only if the most recent
1795            recursion is that of the given number or name. It does not search out
1796            through the entire recursion stack.
1797    
1798        (g) The escape \gN or \g{N} has been added, where N is a positive or
1799            negative number, specifying an absolute or relative reference.
1800    
1801    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1802        some "unreachable code" warnings.
1803    
1804    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1805        things, this adds five new scripts.
1806    
1807    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1808        There were also incompatibilities regarding the handling of \Q..\E inside
1809        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1810        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1811    
1812    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1813        matches an empty string, and forcibly breaks the loop. There were bugs in
1814        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1815        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1816        separate and independent bugs (that affected different cases) have been
1817        fixed.
1818    
1819    39. Refactored the code to abolish the use of different opcodes for small
1820        capturing bracket numbers. This is a tidy that I avoided doing when I
1821        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1822        The new approach is not only tidier, it makes it possible to reduce the
1823        memory needed to fix the previous bug (38).
1824    
1825    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1826        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1827        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1828        mode.
1829    
1830    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1831        report.
1832    
1833    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1834        copy construction and assignment in the C++ wrapper.
1835    
1836    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1837        couple of bugs that could have given wrong results in the "--newline=crlf"
1838        case.
1839    
1840    44. Added a number of casts and did some reorganization of signed/unsigned int
1841        variables following suggestions from Dair Grant. Also renamed the variable
1842        "this" as "item" because it is a C++ keyword.
1843    
1844    45. Arranged for dftables to add
1845    
1846          #include "pcre_internal.h"
1847    
1848        to pcre_chartables.c because without it, gcc 4.x may remove the array
1849        definition from the final binary if PCRE is built into a static library and
1850        dead code stripping is activated.
1851    
1852    46. For an unanchored pattern, if a match attempt fails at the start of a
1853        newline sequence, and the newline setting is CRLF or ANY, and the next two
1854        characters are CRLF, advance by two characters instead of one.
1855    
1856    
1857    Version 6.7 04-Jul-06
1858    ---------------------
1859    
1860     1. In order to handle tests when input lines are enormously long, pcretest has
1861        been re-factored so that it automatically extends its buffers when
1862        necessary. The code is crude, but this _is_ just a test program. The
1863        default size has been increased from 32K to 50K.
1864    
1865     2. The code in pcre_study() was using the value of the re argument before
1866        testing it for NULL. (Of course, in any sensible call of the function, it
1867        won't be NULL.)
1868    
1869     3. The memmove() emulation function in pcre_internal.h, which is used on
1870        systems that lack both memmove() and bcopy() - that is, hardly ever -
1871        was missing a "static" storage class specifier.
1872    
1873     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1874        containing an extended class (one that cannot be represented by a bitmap
1875        because it contains high-valued characters or Unicode property items, e.g.
1876        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1877        pattern, but PCRE should not loop if you do not (it no longer does).
1878        [Detail: two cases were found: (a) a repeated subpattern containing an
1879        extended class; (b) a recursive reference to a subpattern that followed a
1880        previous extended class. It wasn't skipping over the extended class
1881        correctly when UTF-8 mode was not set.]
1882    
1883     5. A negated single-character class was not being recognized as fixed-length
1884        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1885        compile error "lookbehind assertion is not fixed length".
1886    
1887     6. The RunPerlTest auxiliary script was showing an unexpected difference
1888        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1889        write a Perl script that can interpret lines of an input file either as
1890        byte characters or as UTF-8, which is what "perltest" was being required to
1891        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1892        can't do is switch easily at run time between having the "use utf8;" pragma
1893        or not. In the end, I fudged it by using the RunPerlTest script to insert
1894        "use utf8;" explicitly for the UTF-8 tests.
1895    
1896     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1897        the end of the subject string, contrary to the documentation and to what
1898        Perl does. This was true of both matching functions. Now it matches only at
1899        the start of the subject and immediately after *internal* newlines.
1900    
1901     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1902        a pointer to an int instead of a pointer to an unsigned long int. This
1903        caused problems on 64-bit systems.
1904    
1905     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1906        instance of the 'standard' template library not being so standard".
1907    
1908    10. There was no check on the number of named subpatterns nor the maximum
1909        length of a subpattern name. The product of these values is used to compute
1910        the size of the memory block for a compiled pattern. By supplying a very
1911        long subpattern name and a large number of named subpatterns, the size
1912        computation could be caused to overflow. This is now prevented by limiting
1913        the length of names to 32 characters, and the number of named subpatterns
1914        to 10,000.
1915    
1916    11. Subpatterns that are repeated with specific counts have to be replicated in
1917        the compiled pattern. The size of memory for this was computed from the
1918        length of the subpattern and the repeat count. The latter is limited to
1919        65535, but there was no limit on the former, meaning that integer overflow
1920        could in principle occur. The compiled length of a repeated subpattern is
1921        now limited to 30,000 bytes in order to prevent this.
1922    
1923    12. Added the optional facility to have named substrings with the same name.
1924    
1925    13. Added the ability to use a named substring as a condition, using the
1926        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1927        are numbers (not recommended). Forward references are permitted.
1928    
1929    14. Added forward references in named backreferences (if you see what I mean).
1930    
1931    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1932        pattern could run off the end of the subject. For example, the pattern
1933        "(?s)(.{1,5})"8 did this with the subject "ab".
1934    
1935    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1936        PCRE_CASELESS was set when matching characters that were quantified with ?
1937        or *.
1938    
1939    17. A character class other than a single negated character that had a minimum
1940        but no maximum quantifier - for example [ab]{6,} - was not handled
1941        correctly by pce_dfa_exec(). It would match only one character.
1942    
1943    18. A valid (though odd) pattern that looked like a POSIX character
1944        class but used an invalid character after [ (for example [[,abc,]]) caused
1945        pcre_compile() to give the error "Failed: internal error: code overflow" or
1946        in some cases to crash with a glibc free() error. This could even happen if
1947        the pattern terminated after [[ but there just happened to be a sequence of
1948        letters, a binary zero, and a closing ] in the memory that followed.
1949    
1950    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1951        over the years. Originally (before any Unicode support), just the bottom 8
1952        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1953        output from "man perlunicode" includes this:
1954    
1955          The regular expression compiler produces polymorphic opcodes.  That
1956          is, the pattern adapts to the data and automatically switches to
1957          the Unicode character scheme when presented with Unicode data--or
1958          instead uses a traditional byte scheme when presented with byte
1959          data.
1960    
1961        Sadly, a wide octal escape does not cause a switch, and in a string with
1962        no other multibyte characters, these octal escapes are treated as before.
1963        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1964        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1965        Unicode string.
1966    
1967        I have not perpetrated such confusion in PCRE. Up till now, it took just
1968        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1969        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1970        translate to the appropriate multibyte character.
1971    
1972    29. Applied some refactoring to reduce the number of warnings from Microsoft
1973        and Borland compilers. This has included removing the fudge introduced
1974        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1975        a warning about an unused variable.
1976    
1977    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1978        characters since release 4.0, because Perl (from release 5.004) does not.
1979        [Or at least, is documented not to: some releases seem to be in conflict
1980        with the documentation.] However, when a pattern was studied with
1981        pcre_study() and all its branches started with \s, PCRE still included VT
1982        as a possible starting character. Of course, this did no harm; it just
1983        caused an unnecessary match attempt.
1984    
1985    22. Removed a now-redundant internal flag bit that recorded the fact that case
1986        dependency changed within the pattern. This was once needed for "required
1987        byte" processing, but is no longer used. This recovers a now-scarce options
1988        bit. Also moved the least significant internal flag bit to the most-
1989        significant bit of the word, which was not previously used (hangover from
1990        the days when it was an int rather than a uint) to free up another bit for
1991        the future.
1992    
1993    23. Added support for CRLF line endings as well as CR and LF. As well as the
1994        default being selectable at build time, it can now be changed at runtime
1995        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1996        specify that it is scanning data with non-default line endings.
1997    
1998    24. Changed the definition of CXXLINK to make it agree with the definition of
1999        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2000    
2001    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2002        recursions. This makes a big different to stack usage for some patterns.
2003    
2004    26. If a subpattern containing a named recursion or subroutine reference such
2005        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2006        the space required for the compiled pattern went wrong and gave too small a
2007        value. Depending on the environment, this could lead to "Failed: internal
2008        error: code overflow at offset 49" or "glibc detected double free or
2009        corruption" errors.
2010    
2011    27. Applied patches from Google (a) to support the new newline modes and (b) to
2012        advance over multibyte UTF-8 characters in GlobalReplace.
2013    
2014    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2015        difference for some implementation of PCRE in some Windows version.
2016    
2017    29. Added some extra testing facilities to pcretest:
2018    
2019        \q<number>   in a data line sets the "match limit" value
2020        \Q<number>   in a data line sets the "match recursion limt" value
2021        -S <number>  sets the stack size, where <number> is in megabytes
2022    
2023        The -S option isn't available for Windows.
2024    
2025    
2026    Version 6.6 06-Feb-06
2027    ---------------------
2028    
2029     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2030        in pcreposix.h. I have copied the definition from pcre.h.
2031    
2032     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2033        because pcre.h is no longer a built file.
2034    
2035     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2036        not normally included in the compiled code.
2037    
2038    
2039    Version 6.5 01-Feb-06
2040    ---------------------
2041    
2042     1. When using the partial match feature with pcre_dfa_exec(), it was not
2043        anchoring the second and subsequent partial matches at the new starting
2044        point. This could lead to incorrect results. For example, with the pattern
2045        /1234/, partially matching against "123" and then "a4" gave a match.
2046    
2047     2. Changes to pcregrep:
2048    
2049        (a) All non-match returns from pcre_exec() were being treated as failures
2050            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2051            error message is output. Some extra information is given for the
2052            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2053            probably the only errors that are likely to be caused by users (by
2054            specifying a regex that has nested indefinite repeats, for instance).
2055            If there are more than 20 of these errors, pcregrep is abandoned.
2056    
2057        (b) A binary zero was treated as data while matching, but terminated the
2058            output line if it was written out. This has been fixed: binary zeroes
2059            are now no different to any other data bytes.
2060    
2061        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2062            used to set a locale for matching. The --locale=xxxx long option has
2063            been added (no short equivalent) to specify a locale explicitly on the
2064            pcregrep command, overriding the environment variables.
2065    
2066        (d) When -B was used with -n, some line numbers in the output were one less
2067            than they should have been.
2068    
2069        (e) Added the -o (--only-matching) option.
2070    
2071        (f) If -A or -C was used with -c (count only), some lines of context were
2072            accidentally printed for the final match.
2073    
2074        (g) Added the -H (--with-filename) option.
2075    
2076        (h) The combination of options -rh failed to suppress file names for files
2077            that were found from directory arguments.
2078    
2079        (i) Added the -D (--devices) and -d (--directories) options.
2080    
2081        (j) Added the -F (--fixed-strings) option.
2082    
2083        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2084    
2085        (l) Added the --colo(u)r option.
2086    
2087        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2088            is not present by default.
2089    
2090     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2091        items such as (?R) or (?1), when the recursion could match a number of
2092        alternatives. If it matched one of the alternatives, but subsequently,
2093        outside the recursion, there was a failure, the code tried to back up into
2094        the recursion. However, because of the way PCRE is implemented, this is not
2095        possible, and the result was an incorrect result from the match.
2096    
2097        In order to prevent this happening, the specification of recursion has
2098        been changed so that all such subpatterns are automatically treated as
2099        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2100    
2101     4. I had overlooked the fact that, in some locales, there are characters for
2102        which isalpha() is true but neither isupper() nor islower() are true. In
2103        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2104        and ordfeminine) are like this. This affected the treatment of \w and \W
2105        when they appeared in character classes, but not when they appeared outside
2106        a character class. The bit map for "word" characters is now created
2107        separately from the results of isalnum() instead of just taking it from the
2108        upper, lower, and digit maps. (Plus the underscore character, of course.)
2109    
2110     5. The above bug also affected the handling of POSIX character classes such as
2111        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2112        permanent tables. Instead, the bit maps for such a class were previously
2113        created as the appropriate unions of the upper, lower, and digit bitmaps.
2114        Now they are created by subtraction from the [[:word:]] class, which has
2115        its own bitmap.
2116    
2117     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2118        It is created by subtracting the vertical space characters (\x09, \x0a,
2119        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2120        subtraction was done in the overall bitmap for a character class, meaning
2121        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2122        be recognized. This bug has been fixed.
2123    
2124     7. Patches from the folks at Google:
2125    
2126          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2127          real life, but is still worth protecting against".
2128    
2129          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2130          regular expressions".
2131    
2132          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2133          have it.
2134    
2135          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2136          "configure" and the latter not, in order to fix a problem somebody had
2137          with compiling the Arg class on HP-UX.
2138    
2139          (e) Improve the error-handling of the C++ wrapper a little bit.
2140    
2141          (f) New tests for checking recursion limiting.
2142    
2143     8. The pcre_memmove() function, which is used only if the environment does not
2144        have a standard memmove() function (and is therefore rarely compiled),
2145        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2146        returning a result (though PCRE never actually uses the result).
2147    
2148     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2149        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2150        returned instead of calling malloc() with an overflowing number that would
2151        most likely cause subsequent chaos.
2152    
2153    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2154    
2155    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2156        with this option is matched, the nmatch and pmatch options of regexec() are
2157        ignored.
2158    
2159    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2160        provided in case anyone wants to the the POSIX interface with UTF-8
2161        strings.
2162    
2163    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2164        C++ linking (needed for some HP-UX environments).
2165    
2166    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2167        (unused parameter) and in the pcre_printint() function (omitted "default"
2168        switch label when the default is to do nothing).
2169    
2170    15. Added some code to make it possible, when PCRE is compiled as a C++
2171        library, to replace subject pointers for pcre_exec() with a smart pointer
2172        class, thus making it possible to process discontinuous strings.
2173    
2174    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2175        much the same function. They were added by different people who were trying
2176        to make PCRE easy to compile on non-Unix systems. It has been suggested
2177        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2178        for compiling on Windows systems. I have therefore replaced it with
2179        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2180        defaults to "extern" for C or "extern C" for C++, which works fine on
2181        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2182        SCOPE with something explicit in config.h. In addition:
2183    
2184        (a) pcreposix.h still had just "extern" instead of either of these macros;
2185            I have replaced it with PCRE_DATA_SCOPE.
2186    
2187        (b) Functions such as _pcre_xclass(), which are internal to the library,
2188            but external in the C sense, all had PCRE_EXPORT in their definitions.
2189            This is apparently wrong for the Windows case, so I have removed it.
2190            (It makes no difference on Unix-like systems.)
2191    
2192    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2193        of recursive calls to match(). This is different to MATCH_LIMIT because
2194        that limits the total number of calls to match(), not all of which increase
2195        the depth of recursion. Limiting the recursion depth limits the amount of
2196        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2197        when PCRE is compiled, and changed at run time. A patch from Google adds
2198        this functionality to the C++ interface.
2199    
2200    18. Changes to the handling of Unicode character properties:
2201    
2202        (a) Updated the table to Unicode 4.1.0.
2203    
2204        (b) Recognize characters that are not in the table as "Cn" (undefined).
2205    
2206        (c) I revised the way the table is implemented to a much improved format
2207            which includes recognition of ranges. It now supports the ranges that
2208            are defined in UnicodeData.txt, and it also amalgamates other
2209            characters into ranges. This has reduced the number of entries in the
2210            table from around 16,000 to around 3,000, thus reducing its size
2211            considerably. I realized I did not need to use a tree structure after
2212            all - a binary chop search is just as efficient. Having reduced the
2213            number of entries, I extended their size from 6 bytes to 8 bytes to
2214            allow for more data.
2215    
2216        (d) Added support for Unicode script names via properties such as \p{Han}.
2217    
2218    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2219        matching that character.
2220    
2221    20. When matching a repeated Unicode property with a minimum greater than zero,
2222        (for example \pL{2,}), PCRE could look past the end of the subject if it
2223        reached it while seeking the minimum number of characters. This could
2224        happen only if some of the characters were more than one byte long, because
2225        there is a check for at least the minimum number of bytes.
2226    
2227    21. Refactored the implementation of \p and \P so as to be more general, to
2228        allow for more different types of property in future. This has changed the
2229        compiled form incompatibly. Anybody with saved compiled patterns that use
2230        \p or \P will have to recompile them.
2231    
2232    22. Added "Any" and "L&" to the supported property types.
2233    
2234    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2235        but give a compile time error if the value is greater than 0xff.
2236    
2237    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2238        accidentally not being installed or uninstalled.
2239    
2240    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2241        made were to insert the current release number. This seemed silly, because
2242        it made things harder for people building PCRE on systems that don't run
2243        "configure". I have turned pcre.h into a distributed file, no longer built
2244        by "configure", with the version identification directly included. There is
2245        no longer a pcre.h.in file.
2246    
2247        However, this change necessitated a change to the pcre-config script as
2248        well. It is built from pcre-config.in, and one of the substitutions was the
2249        release number. I have updated configure.ac so that ./configure now finds
2250        the release number by grepping pcre.h.
2251    
2252    26. Added the ability to run the tests under valgrind.
2253    
2254    
2255    Version 6.4 05-Sep-05
2256    ---------------------
2257    
2258     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2259        "--" to be printed when multiple files were scanned, even when none of the
2260        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2261        consider it to be a bug, and have restored the previous behaviour.
2262    
2263     2. A couple of code tidies to get rid of compiler warnings.
2264    
2265     3. The pcretest program used to cheat by referring to symbols in the library
2266        whose names begin with _pcre_. These are internal symbols that are not
2267        really supposed to be visible externally, and in some environments it is
2268        possible to suppress them. The cheating is now confined to including
2269        certain files from the library's source, which is a bit cleaner.
2270    
2271     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2272        file's purpose clearer.
2273    
2274     5. Reorganized pcre_ucp_findchar().
2275    
2276    
2277    Version 6.3 15-Aug-05
2278    ---------------------
2279    
2280     1. The file libpcre.pc.in did not have general read permission in the tarball.
2281    
2282     2. There were some problems when building without C++ support:
2283    
2284        (a) If C++ support was not built, "make install" and "make test" still
2285            tried to test it.
2286    
2287        (b) There were problems when the value of CXX was explicitly set. Some
2288            changes have been made to try to fix these, and ...
2289    
2290        (c) --disable-cpp can now be used to explicitly disable C++ support.
2291    
2292        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2293            backslash in a target when C++ was disabled. This confuses some
2294            versions of "make", apparently. Using an intermediate variable solves
2295            this. (Same for CPP_LOBJ.)
2296    
2297     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2298        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2299        necessary on certain architectures.
2300    
2301     4. Added a setting of -export-symbols-regex to the link command to remove
2302        those symbols that are exported in the C sense, but actually are local
2303        within the library, and not documented. Their names all begin with
2304        "_pcre_". This is not a perfect job, because (a) we have to except some
2305        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2306        available (and never for static libraries). I have made a note to try to
2307        find a way round (a) in the future.
2308    
2309    
2310    Version 6.2 01-Aug-05
2311    ---------------------
2312    
2313     1. There was no test for integer overflow of quantifier values. A construction
2314        such as {1111111111111111} would give undefined results. What is worse, if
2315        a minimum quantifier for a parenthesized subpattern overflowed and became
2316        negative, the calculation of the memory size went wrong. This could have
2317        led to memory overwriting.
2318    
2319     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2320    
2321     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2322        operating environments where this matters.
2323    
2324     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2325        PCRE options from within the C++ wrapper.
2326    
2327     5. Named capturing subpatterns were not being correctly counted when a pattern
2328        was compiled. This caused two problems: (a) If there were more than 100
2329        such subpatterns, the calculation of the memory needed for the whole
2330        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2331        back references of the form \12, where the number was greater than 9, were
2332        not recognized as back references, even though there were sufficient
2333        previous subpatterns.
2334    
2335     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2336        versions of gcc, e.g. 2.95.4.
2337    
2338    
2339    Version 6.1 21-Jun-05
2340    ---------------------
2341    
2342     1. There was one reference to the variable "posix" in pcretest.c that was not
2343        surrounded by "#if !defined NOPOSIX".
2344    
2345     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2346        the cross-check on the old pcre_info() function, for the benefit of the
2347        cut-down version of PCRE that is currently imported into Exim.
2348    
2349     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2350        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2351        patterns that start (?i) but I don't think that matters. The use of (?i) is
2352        just an example; this all applies to the other options as well.
2353    
2354     4. Since libtool seems to echo the compile commands it is issuing, the output
2355        from "make" can be reduced a bit by putting "@" in front of each libtool
2356        compile command.
2357    
2358     5. Patch from the folks at Google for configure.in to be a bit more thorough
2359        in checking for a suitable C++ installation before trying to compile the
2360        C++ stuff. This should fix a reported problem when a compiler was present,
2361        but no suitable headers.
2362    
2363     6. The man pages all had just "PCRE" as their title. I have changed them to
2364        be the relevant file name. I have also arranged that these names are
2365        retained in the file doc/pcre.txt, which is a concatenation in text format
2366        of all the man pages except the little individual ones for each function.
2367    
2368     7. The NON-UNIX-USE file had not been updated for the different set of source
2369        files that come with release 6. I also added a few comments about the C++
2370        wrapper.
2371    
2372    
2373    Version 6.0 07-Jun-05
2374    ---------------------
2375    
2376     1. Some minor internal re-organization to help with my DFA experiments.
2377    
2378     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2379        didn't matter for the library itself when fully configured, but did matter
2380        when compiling without UCP support, or within Exim, where the ucp files are
2381        not imported.
2382    
2383     3. Refactoring of the library code to split up the various functions into
2384        different source modules. The addition of the new DFA matching code (see
2385        below) to a single monolithic source would have made it really too
2386        unwieldy, quite apart from causing all the code to be include in a
2387        statically linked application, when only some functions are used. This is
2388        relevant even without the DFA addition now that patterns can be compiled in
2389        one application and matched in another.
2390    
2391        The downside of splitting up is that there have to be some external
2392        functions and data tables that are used internally in different modules of
2393        the library but which are not part of the API. These have all had their
2394        names changed to start with "_pcre_" so that they are unlikely to clash
2395        with other external names.
2396    
2397     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2398        a different (DFA) algorithm. Although it is slower than the original
2399        function, it does have some advantages for certain types of matching
2400        problem.
2401    
2402     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2403        including restarting after a partial match.
2404    
2405     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2406        defined when compiling for Windows was sent to me. I have put it into the
2407        code, though I have no means of testing or verifying it.
2408    
2409     7. Added the pcre_refcount() auxiliary function.
2410    
2411     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2412        match before or at the first newline in the subject string. In pcretest,
2413        the /f option on a pattern can be used to set this.
2414    
2415     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2416        would behave wrongly. This has been present in PCRE since release 4.0.
2417    
2418    10. A number of changes to the pcregrep command:
2419    
2420        (a) Refactored how -x works; insert ^(...)$ instead of setting
2421            PCRE_ANCHORED and checking the length, in preparation for adding
2422            something similar for -w.
2423    
2424        (b) Added the -w (match as a word) option.
2425    
2426        (c) Refactored the way lines are read and buffered so as to have more
2427            than one at a time available.
2428    
2429        (d) Implemented a pcregrep test script.
2430    
2431        (e) Added the -M (multiline match) option. This allows patterns to match
2432            over several lines of the subject. The buffering ensures that at least
2433            8K, or the rest of the document (whichever is the shorter) is available
2434            for matching (and similarly the previous 8K for lookbehind assertions).
2435    
2436        (f) Changed the --help output so that it now says
2437    
2438              -w, --word-regex(p)
2439    
2440            instead of two lines, one with "regex" and the other with "regexp"
2441            because that confused at least one person since the short forms are the
2442            same. (This required a bit of code, as the output is generated
2443            automatically from a table. It wasn't just a text change.)
2444    
2445        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2446            option but starts with a hyphen. Could be a pattern or a path name
2447            starting with a hyphen, for instance.
2448    
2449        (h) "-" can be given as a file name to represent stdin.
2450    
2451        (i) When file names are being printed, "(standard input)" is used for
2452            the standard input, for compatibility with GNU grep. Previously
2453            "<stdin>" was used.
2454    
2455        (j) The option --label=xxx can be used to supply a name to be used for
2456            stdin when file names are being printed. There is no short form.
2457    
2458        (k) Re-factored the options decoding logic because we are going to add
2459            two more options that take data. Such options can now be given in four
2460            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2461    
2462        (l) Added the -A, -B, and -C options for requesting that lines of context
2463            around matches be printed.
2464    
2465        (m) Added the -L option to print the names of files that do not contain
2466            any matching lines, that is, the complement of -l.
2467    
2468        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2469            continue to scan other files.
2470    
2471        (o) The -s option was incorrectly implemented. For compatibility with other
2472            greps, it now suppresses the error message for a non-existent or non-
2473            accessible file (but not the return code). There is a new option called
2474            -q that suppresses the output of matching lines, which was what -s was
2475            previously doing.
2476    
2477        (p) Added --include and --exclude options to specify files for inclusion
2478            and exclusion when recursing.
2479    
2480    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2481        Hopefully, it now does.
2482    
2483    12. Missing cast in pcre_study().
2484    
2485    13. Added an "uninstall" target to the makefile.
2486    
2487    14. Replaced "extern" in the function prototypes in Makefile.in with
2488        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2489        world, but is set differently for Windows.
2490    
2491    15. Added a second compiling function called pcre_compile2(). The only
2492        difference is that it has an extra argument, which is a pointer to an
2493        integer error code. When there is a compile-time failure, this is set
2494        non-zero, in addition to the error test pointer being set to point to an
2495        error message. The new argument may be NULL if no error number is required
2496        (but then you may as well call pcre_compile(), which is now just a
2497        wrapper). This facility is provided because some applications need a
2498        numeric error indication, but it has also enabled me to tidy up the way
2499        compile-time errors are handled in the POSIX wrapper.
2500    
2501    16. Added VPATH=.libs to the makefile; this should help when building with one
2502        prefix path and installing with another. (Or so I'm told by someone who
2503        knows more about this stuff than I do.)
2504    
2505    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2506        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2507        match everything, including newlines. This is not POSIX-compatible, but
2508        somebody wanted the feature. From pcretest it can be activated by using
2509        both the P and the s flags.
2510    
2511    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2512    
2513    19. libpcre.pc was being incorrectly installed as executable.
2514    
2515    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2516        it now also looks for '\r' so that it will work unmodified on Windows.
2517    
2518    21. Added Google's contributed C++ wrapper to the distribution.
2519    
2520    22. Added some untidy missing memory free() calls in pcretest, to keep
2521        Electric Fence happy when testing.
2522    
2523    
2524    
2525    Version 5.0 13-Sep-04
2526    ---------------------
2527    
2528     1. Internal change: literal characters are no longer packed up into items
2529        containing multiple characters in a single byte-string. Each character
2530        is now matched using a separate opcode. However, there may be more than one
2531        byte in the character in UTF-8 mode.
2532    
2533     2. The pcre_callout_block structure has two new fields: pattern_position and
2534        next_item_length. These contain the offset in the pattern to the next match
2535        item, and its length, respectively.
2536    
2537     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2538        insertion of callouts before each pattern item. Added the /C option to
2539        pcretest to make use of this.
2540    
2541     4. On the advice of a Windows user, the lines
2542    
2543          #if defined(_WIN32) || defined(WIN32)
2544          _setmode( _fileno( stdout ), 0x8000 );
2545          #endif  /* defined(_WIN32) || defined(WIN32) */
2546    
2547        have been added to the source of pcretest. This apparently does useful
2548        magic in relation to line terminators.
2549    
2550     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2551        for the benefit of those environments where the "b" makes a difference.
2552    
2553     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2554        to know about it. I have put a hack into configure.in that adds in code
2555        to set GCC=yes if CC=icc. This seems to end up at a point in the
2556        generated configure script that is early enough to affect the setting of
2557        compiler options, which is what is needed, but I have no means of testing
2558        whether it really works. (The user who reported this had patched the
2559        generated configure script, which of course I cannot do.)
2560    
2561        LATER: After change 22 below (new libtool files), the configure script
2562        seems to know about icc (and also ecc). Therefore, I have commented out
2563        this hack in configure.in.
2564    
2565     7. Added support for pkg-config (2 patches were sent in).
2566    
2567     8. Negated POSIX character classes that used a combination of internal tables
2568        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2569        [[:^ascii]]. Typically, they would match almost any characters. The other
2570        POSIX classes were not broken in this way.
2571    
2572     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2573        to find the match, as PCRE was deluded into thinking that the match had to
2574        start at the start point or following a newline. The same bug applied to
2575        patterns with negative forward assertions or any backward assertions
2576        preceding ".*" at the start, unless the pattern required a fixed first
2577        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2578    
2579    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2580        starting at the last subject character, bytes beyond the end of the subject
2581        string were read.
2582    
2583    11. Renamed the variable "class" as "classbits" to make life easier for C++
2584        users. (Previously there was a macro definition, but it apparently wasn't
2585        enough.)
2586    
2587    12. Added the new field "tables" to the extra data so that tables can be passed
2588        in at exec time, or the internal tables can be re-selected. This allows
2589        a compiled regex to be saved and re-used at a later time by a different
2590        program that might have everything at different addresses.
2591    
2592    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2593        -R library as well as a -L library.
2594    
2595    14. The debugging options of pcretest (-d on the command line or D on a
2596        pattern) showed incorrect output for anything following an extended class
2597        that contained multibyte characters and which was followed by a quantifier.
2598    
2599    15. Added optional support for general category Unicode character properties
2600        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2601        support. It adds about 90K to the size of the library. The meanings of the
2602        inbuilt class escapes such as \d and \s have NOT been changed.
2603    
2604    16. Updated pcredemo.c to include calls to free() to release the memory for the
2605        compiled pattern.
2606    
2607    17. The generated file chartables.c was being created in the source directory
2608        instead of in the building directory. This caused the build to fail if the
2609        source directory was different from the building directory, and was
2610        read-only.
2611    
2612    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2613        file. No doubt somebody will tell me if they don't make sense... Also added
2614        Dan Mooney's comments about building on OpenVMS.
2615    
2616    19. Added support for partial matching via the PCRE_PARTIAL option for
2617        pcre_exec() and the \P data escape in pcretest.
2618    
2619    20. Extended pcretest with 3 new pattern features:
2620    
2621        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2622              write the compiled pattern to the file whose name is "rest-of-line".
2623              This is a straight binary dump of the data, with the saved pointer to
2624              the character tables forced to be NULL. The study data, if any, is
2625              written too. After writing, pcretest reads a new pattern.
2626    
2627        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2628              compiled pattern from the given file. There must not be any
2629              occurrences of "<" in the file name (pretty unlikely); if there are,
2630              pcretest will instead treat the initial "<" as a pattern delimiter.
2631              After reading in the pattern, pcretest goes on to read data lines as
2632              usual.
2633    
2634        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2635              and 16-bit fields in a compiled pattern, to simulate a pattern that
2636              was compiled on a host of opposite endianness.
2637    
2638    21. The pcre-exec() function can now cope with patterns that were compiled on
2639        hosts of opposite endianness, with this restriction:
2640    
2641          As for any compiled expression that is saved and used later, the tables
2642          pointer field cannot be preserved; the extra_data field in the arguments
2643          to pcre_exec() should be used to pass in a tables address if a value
2644          other than the default internal tables were used at compile time.
2645    
2646    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2647        now diagnosed as an error. Previously, most of the time, a negative number
2648        would have been treated as zero, but if in addition "ovector" was passed as
2649        NULL, a crash could occur.
2650    
2651    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2652        new versions from the libtool 1.5 distribution (the last one is a copy of
2653        a file called libtool.m4). This seems to have fixed the need to patch
2654        "configure" to support Darwin 1.3 (which I used to do). However, I still
2655        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2656        workstation).
2657    
2658    24. Changed the PCRE licence to be the more standard "BSD" licence.
2659    
2660    
2661    Version 4.5 01-Dec-03
2662    ---------------------
2663    
2664     1. There has been some re-arrangement of the code for the match() function so
2665        that it can be compiled in a version that does not call itself recursively.
2666        Instead, it keeps those local variables that need separate instances for
2667        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2668        needs to "recurse". Keeping track of where control must go is done by means
2669        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2670        hide most of the details from the main code, and operates only if
2671        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2672        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2673        operating.
2674    
2675        To make it easier for callers to provide specially tailored get/free
2676        functions for this usage, two new functions, pcre_stack_malloc, and
2677        pcre_stack_free, are used. They are always called in strict stacking order,
2678        and the size of block requested is always the same.
2679    
2680        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2681        PCRE has been compiled to use the stack or the heap for recursion. The
2682        -C option of pcretest uses this to show which version is compiled.
2683    
2684        A new data escape \S, is added to pcretest; it causes the amounts of store
2685        obtained and freed by both kinds of malloc/free at match time to be added
2686        to the output.
2687    
2688     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2689        what's available on my current Linux desktop machine.
2690    
2691     3. When matching a UTF-8 string, the test for a valid string at the start has
2692        been extended. If start_offset is not zero, PCRE now checks that it points
2693        to a byte that is the start of a UTF-8 character. If not, it returns
2694        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2695        this is necessary because there may be backward assertions in the pattern.
2696        When matching the same subject several times, it may save resources to use
2697        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2698    
2699     4. The code for checking the validity of UTF-8 strings has been tightened so
2700        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2701        containing "overlong sequences".
2702    
2703     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2704        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2705        should have been "&", but it just so happened that all the cases this let
2706        through by mistake were picked up later in the function.
2707    
2708     6. I had used a variable called "isblank" - this is a C99 function, causing
2709        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2710    
2711     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2712        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2713        so the version is shown; (c) stop "make test" echoing "./RunTest".
2714    
2715     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2716    
2717     9. The prototype for memmove() for systems that don't have it was using
2718        size_t, but the inclusion of the header that defines size_t was later. I've
2719        moved the #includes for the C headers earlier to avoid this.
2720    
2721    10. Added some adjustments to the code to make it easier to compiler on certain
2722        special systems:
2723    
2724          (a) Some "const" qualifiers were missing.
2725          (b) Added the macro EXPORT before all exported functions; by default this
2726              is defined to be empty.
2727          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2728              that it reads its output file name as an argument instead of writing
2729              to the standard output and assuming this can be redirected.
2730    
2731    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2732        class containing characters with values greater than 255, PCRE compilation
2733        went into a loop.
2734    
2735    12. A recursive reference to a subpattern that was within another subpattern
2736        that had a minimum quantifier of zero caused PCRE to crash. For example,
2737        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2738        recursion. If the recursively-called subpattern itself had a zero repeat,
2739        that was OK.
2740    
2741    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2742        buffer into which it was copied (for escape processing) was still set at
2743        1024, so long lines caused crashes.
2744    
2745    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2746        "internal error: code overflow...". This applied to any character class
2747        that was followed by a possessive quantifier.
2748    
2749    15. Modified the Makefile to add libpcre.la as a prerequisite for
2750        libpcreposix.la because I was told this is needed for a parallel build to
2751        work.
2752    
2753    16. If a pattern that contained .* following optional items at the start was
2754        studied, the wrong optimizing data was generated, leading to matching
2755        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2756        matching string must start with a or b or c. The correct conclusion for
2757        this pattern is that a match can start with any character.
2758    
2759    
2760    Version 4.4 13-Aug-03
2761    ---------------------
2762    
2763     1. In UTF-8 mode, a character class containing characters with values between
2764        127 and 255 was not handled correctly if the compiled pattern was studied.
2765        In fixing this, I have also improved the studying algorithm for such
2766        classes (slightly).
2767    
2768     2. Three internal functions had redundant arguments passed to them. Removal
2769        might give a very teeny performance improvement.
2770    
2771     3. Documentation bug: the value of the capture_top field in a callout is *one
2772        more than* the number of the hightest numbered captured substring.
2773    
2774     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2775        in incorrectly linking with a previously installed version. They now link
2776        explicitly with libpcre.la.
2777    
2778     5. configure.in no longer needs to recognize Cygwin specially.
2779    
2780     6. A problem in pcre.in for Windows platforms is fixed.
2781    
2782     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2783        pcretest, it used to include the size of the study block as part of its
2784        output. Unfortunately, the structure contains a field that has a different
2785        size on different hardware architectures. This meant that the tests that
2786        showed this size failed. As the block is currently always of a fixed size,
2787        this information isn't actually particularly useful in pcretest output, so
2788        I have just removed it.
2789    
2790     8. Three pre-processor statements accidentally did not start in column 1.
2791        Sadly, there are *still* compilers around that complain, even though
2792        standard C has not required this for well over a decade. Sigh.
2793    
2794     9. In pcretest, the code for checking callouts passed small integers in the
2795        callout_data field, which is a void * field. However, some picky compilers
2796        complained about the casts involved for this on 64-bit systems. Now
2797        pcretest passes the address of the small integer instead, which should get
2798        rid of the warnings.
2799    
2800    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2801        both compile and run time, and gives an error if an invalid UTF-8 sequence
2802        is found. There is a option for disabling this check in cases where the
2803        string is known to be correct and/or the maximum performance is wanted.
2804    
2805    11. In response to a bug report, I changed one line in Makefile.in from
2806    
2807            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2808        to
2809            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2810    
2811        to look similar to other lines, but I have no way of telling whether this
2812        is the right thing to do, as I do not use Windows. No doubt I'll get told
2813        if it's wrong...
2814    
2815    
2816    Version 4.3 21-May-03
2817    ---------------------
2818    
2819    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2820       Makefile.
2821    
2822    2. Some refactoring to improve the quality of the code:
2823    
2824       (i)   The utf8_table... variables are now declared "const".
2825    
2826       (ii)  The code for \cx, which used the "case flipping" table to upper case
2827             lower case letters, now just substracts 32. This is ASCII-specific,
2828             but the whole concept of \cx is ASCII-specific, so it seems
2829             reasonable.
2830    
2831       (iii) PCRE was using its character types table to recognize decimal and
2832             hexadecimal digits in the pattern. This is silly, because it handles
2833             only 0-9, a-f, and A-F, but the character types table is locale-
2834             specific, which means strange things might happen. A private
2835             table is now used for this - though it costs 256 bytes, a table is
2836             much faster than multiple explicit tests. Of course, the standard
2837             character types table is still used for matching digits in subject
2838             strings against \d.
2839    
2840       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2841             ending in _t are). So I've renamed it as ESC_tee.
2842    
2843    3. The first argument for regexec() in the POSIX wrapper should have been
2844       defined as "const".
2845    
2846    4. Changed pcretest to use malloc() for its buffers so that they can be
2847       Electric Fenced for debugging.
2848    
2849    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2850       to read one or more bytes before the start of the subject string. Often this
2851       had no effect on PCRE's behaviour, but in some circumstances it could
2852       provoke a segmentation fault.
2853    
2854    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2855       to try to read one or more bytes before the start of the subject string.
2856    
2857    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2858       UTF-8 support could misbehave in various ways if the subject string
2859       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2860       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2861       back over UTF-8 characters.)
2862    
2863    
2864    Version 4.2 14-Apr-03
2865    ---------------------
2866    
2867    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2868    
2869    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2870         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2871         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2872         [WIN_PREFIX]: new variable, "cyg" for Cygwin
2873         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2874           and BUILD_EXEEXT
2875         Note: automatic setting of the BUILD variables is not yet working
2876         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2877           compile-time but not at link-time
2878         [LINK]: use for linking executables only
2879         make different versions for Windows and non-Windows
2880         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2881           libraries
2882         [LINK_FOR_BUILD]: new variable
2883         [OBJEXT]: use throughout
2884         [EXEEXT]: use throughout
2885         <winshared>: new target
2886         <wininstall>: new target
2887         <dftables.o>: use native compiler
2888         <dftables>: use native linker
2889         <install>: handle Windows platform correctly
2890         <clean>: ditto
2891         <check>: ditto
2892         copy DLL to top builddir before testing
2893    
2894       As part of these changes, -no-undefined was removed again. This was reported
2895       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2896       in any case.
2897    
2898    3. Some tidies to get rid of compiler warnings:
2899    
2900       . In the match_data structure, match_limit was an unsigned long int, whereas
2901         match_call_count was an int. I've made them both unsigned long ints.
2902    
2903       . In pcretest the fact that a const uschar * doesn't automatically cast to
2904         a void * provoked a warning.
2905    
2906       . Turning on some more compiler warnings threw up some "shadow" variables
2907         and a few more missing casts.
2908    
2909    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2910       option, a class that contained a single character with a value between 128
2911       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2912    
2913    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2914       option, a class that contained several characters, but with at least one
2915       whose value was between 128 and 255 caused PCRE to crash.
2916    
2917    
2918  Version 4.1 12-Mar-03  Version 4.1 12-Mar-03
2919  ---------------------  ---------------------
2920    

Legend:
Removed from v.65  
changed lines
  Added in v.593

  ViewVC Help
Powered by ViewVC 1.1.5