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

Diff of /code/trunk/ChangeLog

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

revision 71 by nigel, Sat Feb 24 21:40:24 2007 UTC revision 581 by ph10, Fri Dec 10 11:40:00 2010 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 8.11 10-Dec-2010
5    ------------------------
6    
7    1.  (*THEN) was not working properly if there were untried alternatives prior
8        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
9        backtracked to try for "b" instead of moving to the next alternative branch
10        at the same level (in this case, to look for "c"). The Perl documentation
11        is clear that when (*THEN) is backtracked onto, it goes to the "next
12        alternative in the innermost enclosing group".
13    
14    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
15        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
16        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
17        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
18        (*THEN).
19    
20    3.  If \s appeared in a character class, it removed the VT character from
21        the class, even if it had been included by some previous item, for example
22        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
23        of \s, but is part of the POSIX "space" class.)
24    
25    4.  A partial match never returns an empty string (because you can always
26        match an empty string at the end of the subject); however the checking for
27        an empty string was starting at the "start of match" point. This has been
28        changed to the "earliest inspected character" point, because the returned
29        data for a partial match starts at this character. This means that, for
30        example, /(?<=abc)def/ gives a partial match for the subject "abc"
31        (previously it gave "no match").
32    
33    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
34        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
35        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
36        has an implication that the given string is incomplete (because a partial
37        match is preferred over a full match). For this reason, these items now
38        give a partial match in this situation. [Aside: previously, the one case
39        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
40        match rather than a full match, which was wrong by the old rules, but is
41        now correct.]
42    
43    6.  There was a bug in the handling of #-introduced comments, recognized when
44        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
45        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
46        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
47        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
48        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
49        places in pcre_compile().
50    
51    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
52        comments when looking ahead for named forward references to subpatterns,
53        the only newline sequence it recognized was NL. It now handles newlines
54        according to the set newline convention.
55    
56    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
57        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
58        cater for a lack of strerror(). These oversights have been fixed.
59    
60    9.  Added --match-limit and --recursion-limit to pcregrep.
61    
62    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
63    
64    11. When the -o option was used, pcregrep was setting a return code of 1, even
65        when matches were found, and --line-buffered was not being honoured.
66    
67    12. Added an optional parentheses number to the -o and --only-matching options
68        of pcregrep.
69    
70    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
71        can match an empty string. The code to do it in pcretest and pcredemo
72        needed fixing:
73    
74        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
75            only one byte after an empty string match just before CRLF (this case
76            just got forgotten; "any" and "anycrlf" were OK).
77    
78        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
79            mode when an empty string match preceded an ASCII character followed by
80            a non-ASCII character. (The code for advancing by one character rather
81            than one byte was nonsense.)
82    
83        (c) The pcredemo.c sample program did not have any code at all to handle
84            the cases when CRLF is a valid newline sequence.
85    
86    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
87        as a starting offset was within the subject string. There is now a new
88        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
89        negative or greater than the length of the string. In order to test this,
90        pcretest is extended to allow the setting of negative starting offsets.
91    
92    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
93        starting offset points to the beginning of a UTF-8 character was
94        unnecessarily clumsy. I tidied it up.
95    
96    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
97        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
98    
99    17. Nobody had reported that the --include_dir option, which was added in
100        release 7.7 should have been called --include-dir (hyphen, not underscore)
101        for compatibility with GNU grep. I have changed it to --include-dir, but
102        left --include_dir as an undocumented synonym, and the same for
103        --exclude-dir, though that is not available in GNU grep, at least as of
104        release 2.5.4.
105    
106    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
107        characters from a string of bytes) have been redefined so as not to use
108        loops, in order to improve performance in some environments. At the same
109        time, I abstracted some of the common code into auxiliary macros to save
110        repetition (this should not affect the compiled code).
111    
112    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
113        compile-time error is now given if \c is not followed by an ASCII
114        character, that is, a byte less than 128. (In EBCDIC mode, the code is
115        different, and any byte value is allowed.)
116    
117    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
118        START_OPTIMIZE option, which is now allowed at compile time - but just
119        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
120        to pcregrep and other applications that have no direct access to PCRE
121        options. The new /Y option in pcretest sets this option when calling
122        pcre_compile().
123    
124    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
125        back references. Groups containing recursive back references were forced to
126        be atomic by that change, but in the case of named groups, the amount of
127        memory required was incorrectly computed, leading to "Failed: internal
128        error: code overflow". This has been fixed.
129    
130    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
131        pcretest.c, to avoid build problems in some Borland environments.
132    
133    
134    Version 8.10 25-Jun-2010
135    ------------------------
136    
137    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
138        THEN.
139    
140    2.  (*ACCEPT) was not working when inside an atomic group.
141    
142    3.  Inside a character class, \B is treated as a literal by default, but
143        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
144        causes the error). The code is unchanged, but I tidied the documentation.
145    
146    4.  Inside a character class, PCRE always treated \R and \X as literals,
147        whereas Perl faults them if its -w option is set. I have changed PCRE so
148        that it faults them when PCRE_EXTRA is set.
149    
150    5.  Added support for \N, which always matches any character other than
151        newline. (It is the same as "." when PCRE_DOTALL is not set.)
152    
153    6.  When compiling pcregrep with newer versions of gcc which may have
154        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
155        declared with attribute warn_unused_result" were given. Just casting the
156        result to (void) does not stop the warnings; a more elaborate fudge is
157        needed. I've used a macro to implement this.
158    
159    7.  Minor change to pcretest.c to avoid a compiler warning.
160    
161    8.  Added four artifical Unicode properties to help with an option to make
162        \s etc use properties (see next item). The new properties are: Xan
163        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
164    
165    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
166        use Unicode properties. (*UCP) at the start of a pattern can be used to set
167        this option. Modified pcretest to add /W to test this facility. Added
168        REG_UCP to make it available via the POSIX interface.
169    
170    10. Added --line-buffered to pcregrep.
171    
172    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
173        studied, and the match started with a letter with a code point greater than
174        127 whose first byte was different to the first byte of the other case of
175        the letter, the other case of this starting letter was not recognized
176        (#976).
177    
178    12. If a pattern that was studied started with a repeated Unicode property
179        test, for example, \p{Nd}+, there was the theoretical possibility of
180        setting up an incorrect bitmap of starting bytes, but fortunately it could
181        not have actually happened in practice until change 8 above was made (it
182        added property types that matched character-matching opcodes).
183    
184    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
185        possible starting bytes for non-anchored patterns.
186    
187    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
188        \R, and also a number of cases that involve Unicode properties, both
189        explicit and implicit when PCRE_UCP is set.
190    
191    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
192        input, it could crash or give wrong results if characters with values
193        greater than 0xc0 were present in the subject string. (Detail: it assumed
194        UTF-8 input when processing these items.)
195    
196    16. Added a lot of (int) casts to avoid compiler warnings in systems where
197        size_t is 64-bit (#991).
198    
199    17. Added a check for running out of memory when PCRE is compiled with
200        --disable-stack-for-recursion (#990).
201    
202    18. If the last data line in a file for pcretest does not have a newline on
203        the end, a newline was missing in the output.
204    
205    19. The default pcre_chartables.c file recognizes only ASCII characters (values
206        less than 128) in its various bitmaps. However, there is a facility for
207        generating tables according to the current locale when PCRE is compiled. It
208        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
209        characters, are recognized by isspace() and therefore were getting set in
210        these tables, and indeed these tables seem to approximate to ISO 8859. This
211        caused a problem in UTF-8 mode when pcre_study() was used to create a list
212        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
213        which of course cannot start UTF-8 characters. I have changed the code so
214        that only real ASCII characters (less than 128) and the correct starting
215        bytes for UTF-8 encodings are set for characters greater than 127 when in
216        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
217        altogether.)
218    
219    20. Added the /T option to pcretest so as to be able to run tests with non-
220        standard character tables, thus making it possible to include the tests
221        used for 19 above in the standard set of tests.
222    
223    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
224        reference to a subpattern the other side of a comment that contains an
225        opening parenthesis caused either an internal compiling error, or a
226        reference to the wrong subpattern.
227    
228    
229    Version 8.02 19-Mar-2010
230    ------------------------
231    
232    1.  The Unicode data tables have been updated to Unicode 5.2.0.
233    
234    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
235        configured.
236    
237    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
238        original author of that file, following a query about its status.
239    
240    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
241        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
242    
243    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
244        quantifier applied to a forward-referencing subroutine call, could compile
245        incorrect code or give the error "internal error: previously-checked
246        referenced subpattern not found".
247    
248    6.  Both MS Visual Studio and Symbian OS have problems with initializing
249        variables to point to external functions. For these systems, therefore,
250        pcre_malloc etc. are now initialized to local functions that call the
251        relevant global functions.
252    
253    7.  There were two entries missing in the vectors called coptable and poptable
254        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
255        I've fixed the data, and added a kludgy way of testing at compile time that
256        the lengths are correct (equal to the number of opcodes).
257    
258    8.  Following on from 7, I added a similar kludge to check the length of the
259        eint vector in pcreposix.c.
260    
261    9.  Error texts for pcre_compile() are held as one long string to avoid too
262        much relocation at load time. To find a text, the string is searched,
263        counting zeros. There was no check for running off the end of the string,
264        which could happen if a new error number was added without updating the
265        string.
266    
267    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
268    
269    11. \K was not working if it appeared in an atomic group or in a group that
270        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
271        \K is "not well defined" if used in an assertion. PCRE now accepts it if
272        the assertion is positive, but not if it is negative.
273    
274    12. Change 11 fortuitously reduced the size of the stack frame used in the
275        "match()" function of pcre_exec.c by one pointer. Forthcoming
276        implementation of support for (*MARK) will need an extra pointer on the
277        stack; I have reserved it now, so that the stack frame size does not
278        decrease.
279    
280    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
281        item in branch that calls a recursion is a subroutine call - as in the
282        second branch in the above example - was incorrectly given the compile-
283        time error "recursive call could loop indefinitely" because pcre_compile()
284        was not correctly checking the subroutine for matching a non-empty string.
285    
286    14. The checks for overrunning compiling workspace could trigger after an
287        overrun had occurred. This is a "should never occur" error, but it can be
288        triggered by pathological patterns such as hundreds of nested parentheses.
289        The checks now trigger 100 bytes before the end of the workspace.
290    
291    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
292    
293    
294    Version 8.01 19-Jan-2010
295    ------------------------
296    
297    1.  If a pattern contained a conditional subpattern with only one branch (in
298        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
299        computed the wrong minimum data length (which is of course zero for such
300        subpatterns). This could cause incorrect "no match" results.
301    
302    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
303        the pattern is reset in the first branch, pcre_compile() failed with
304        "internal error: code overflow at offset...". This happened only when
305        the reset was to the original external option setting. (An optimization
306        abstracts leading options settings into an external setting, which was the
307        cause of this.)
308    
309    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
310        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
311        assertion pattern did not match (meaning that the assertion was true), it
312        was incorrectly treated as false if the SKIP had been reached during the
313        matching. This also applied to assertions used as conditions.
314    
315    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
316        assertion subpattern, including such a pattern used as a condition,
317        unpredictable results occurred, instead of the error return
318        PCRE_ERROR_DFA_UITEM.
319    
320    5.  The C++ GlobalReplace function was not working like Perl for the special
321        situation when an empty string is matched. It now does the fancy magic
322        stuff that is necessary.
323    
324    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
325        removed. (These were left over from very, very early versions of PCRE.)
326    
327    7.  Some cosmetic changes to the code to make life easier when compiling it
328        as part of something else:
329    
330        (a) Change DEBUG to PCRE_DEBUG.
331    
332        (b) In pcre_compile(), rename the member of the "branch_chain" structure
333            called "current" as "current_branch", to prevent a collision with the
334            Linux macro when compiled as a kernel module.
335    
336        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
337            prevent a collision with the Linux macro when compiled as a kernel
338            module.
339    
340    8.  In pcre_compile() there are some checks for integer overflows that used to
341        cast potentially large values to (double). This has been changed to that
342        when building, a check for int64_t is made, and if it is found, it is used
343        instead, thus avoiding the use of floating point arithmetic. (There is no
344        other use of FP in PCRE.) If int64_t is not found, the fallback is to
345        double.
346    
347    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
348        2005 (difference between two addresses compared to an unsigned value).
349    
350    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
351        custom one, because of the following reported problem in Windows:
352    
353          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
354              under Win32.
355          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
356              therefore missing the function definition.
357          - The compiler thus generates a "C" signature for the test function.
358          - The linker fails to find the "C" function.
359          - PCRE fails to configure if asked to do so against libbz2.
360    
361    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
362        messages were output:
363    
364          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
365          rerunning libtoolize, to keep the correct libtool macros in-tree.
366          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
367    
368        I have done both of these things.
369    
370    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
371        most of the time, it *can* run out if it is given a pattern that contains a
372        runaway infinite recursion. I updated the discussion in the pcrestack man
373        page.
374    
375    13. Now that we have gone to the x.xx style of version numbers, the minor
376        version may start with zero. Using 08 or 09 is a bad idea because users
377        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
378        interpreted as invalid octal numbers. I've updated the previous comment in
379        configure.ac, and also added a check that gives an error if 08 or 09 are
380        used.
381    
382    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
383        causing partial matching to fail when the end of the subject matched \W
384        in a UTF-8 pattern where \W was quantified with a minimum of 3.
385    
386    15. There were some discrepancies between the declarations in pcre_internal.h
387        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
388        their definitions. The declarations used "const uschar *" and the
389        definitions used USPTR. Even though USPTR is normally defined as "const
390        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
391        reported that: "This difference in casting confuses some C++ compilers, for
392        example, SunCC recognizes above declarations as different functions and
393        generates broken code for hbpcre." I have changed the declarations to use
394        USPTR.
395    
396    16. GNU libtool is named differently on some systems. The autogen.sh script now
397        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
398        (FreeBSD).
399    
400    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
401        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
402        comment: "Figure out how to create a longlong from a string: strtoll and
403        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
404        instance, but it only takes 2 args instead of 3!"
405    
406    18. A subtle bug concerned with back references has been fixed by a change of
407        specification, with a corresponding code fix. A pattern such as
408        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
409        refers, was giving matches when it shouldn't. For example, xa=xaaa would
410        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
411        same bug. Such groups have to be quantified to be useful, or contained
412        inside another quantified group. (If there's no repetition, the reference
413        can never match.) The problem arises because, having left the group and
414        moved on to the rest of the pattern, a later failure that backtracks into
415        the group uses the captured value from the final iteration of the group
416        rather than the correct earlier one. I have fixed this in PCRE by forcing
417        any group that contains a reference to itself to be an atomic group; that
418        is, there cannot be any backtracking into it once it has completed. This is
419        similar to recursive and subroutine calls.
420    
421    
422    Version 8.00 19-Oct-09
423    ----------------------
424    
425    1.  The table for translating pcre_compile() error codes into POSIX error codes
426        was out-of-date, and there was no check on the pcre_compile() error code
427        being within the table. This could lead to an OK return being given in
428        error.
429    
430    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
431        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
432        in a Windows environment.
433    
434    3.  The pcregrep --count option prints the count for each file even when it is
435        zero, as does GNU grep. However, pcregrep was also printing all files when
436        --files-with-matches was added. Now, when both options are given, it prints
437        counts only for those files that have at least one match. (GNU grep just
438        prints the file name in this circumstance, but including the count seems
439        more useful - otherwise, why use --count?) Also ensured that the
440        combination -clh just lists non-zero counts, with no names.
441    
442    4.  The long form of the pcregrep -F option was incorrectly implemented as
443        --fixed_strings instead of --fixed-strings. This is an incompatible change,
444        but it seems right to fix it, and I didn't think it was worth preserving
445        the old behaviour.
446    
447    5.  The command line items --regex=pattern and --regexp=pattern were not
448        recognized by pcregrep, which required --regex pattern or --regexp pattern
449        (with a space rather than an '='). The man page documented the '=' forms,
450        which are compatible with GNU grep; these now work.
451    
452    6.  No libpcreposix.pc file was created for pkg-config; there was just
453        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
454    
455    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
456        when UCP support is not needed, by modifying the Python script that
457        generates it from Unicode data files. This should not matter if the module
458        is correctly used as a library, but I received one complaint about 50K of
459        unwanted data. My guess is that the person linked everything into his
460        program rather than using a library. Anyway, it does no harm.
461    
462    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
463        was a minimum greater than 1 for a wide character in a possessive
464        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
465        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
466        character. Chaos in the form of incorrect output or a compiling loop could
467        result.
468    
469    9.  The restrictions on what a pattern can contain when partial matching is
470        requested for pcre_exec() have been removed. All patterns can now be
471        partially matched by this function. In addition, if there are at least two
472        slots in the offset vector, the offset of the earliest inspected character
473        for the match and the offset of the end of the subject are set in them when
474        PCRE_ERROR_PARTIAL is returned.
475    
476    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
477        synonymous with PCRE_PARTIAL, for backwards compatibility, and
478        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
479        and may be more useful for multi-segment matching.
480    
481    11. Partial matching with pcre_exec() is now more intuitive. A partial match
482        used to be given if ever the end of the subject was reached; now it is
483        given only if matching could not proceed because another character was
484        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
485        string "Z", which now yields "no match" instead of "partial match". In the
486        case of pcre_dfa_exec(), "no match" is given if every matching path for the
487        final character ended with (*FAIL).
488    
489    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
490        if the pattern had a "must contain" character that was already found in the
491        earlier partial match, unless partial matching was again requested. For
492        example, with the pattern /dog.(body)?/, the "must contain" character is
493        "g". If the first part-match was for the string "dog", restarting with
494        "sbody" failed. This bug has been fixed.
495    
496    13. The string returned by pcre_dfa_exec() after a partial match has been
497        changed so that it starts at the first inspected character rather than the
498        first character of the match. This makes a difference only if the pattern
499        starts with a lookbehind assertion or \b or \B (\K is not supported by
500        pcre_dfa_exec()). It's an incompatible change, but it makes the two
501        matching functions compatible, and I think it's the right thing to do.
502    
503    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
504        so that the demonstration program is easily available in environments where
505        PCRE has not been installed from source.
506    
507    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
508        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
509        library.
510    
511    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
512        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
513        is not the first non-POSIX option to be added. Clearly some people find
514        these options useful.
515    
516    17. If a caller to the POSIX matching function regexec() passes a non-zero
517        value for nmatch with a NULL value for pmatch, the value of
518        nmatch is forced to zero.
519    
520    18. RunGrepTest did not have a test for the availability of the -u option of
521        the diff command, as RunTest does. It now checks in the same way as
522        RunTest, and also checks for the -b option.
523    
524    19. If an odd number of negated classes containing just a single character
525        interposed, within parentheses, between a forward reference to a named
526        subpattern and the definition of the subpattern, compilation crashed with
527        an internal error, complaining that it could not find the referenced
528        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
529        [The bug was that it was starting one character too far in when skipping
530        over the character class, thus treating the ] as data rather than
531        terminating the class. This meant it could skip too much.]
532    
533    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
534        /g option in pcretest when the pattern contains \K, which makes it possible
535        to have an empty string match not at the start, even when the pattern is
536        anchored. Updated pcretest and pcredemo to use this option.
537    
538    21. If the maximum number of capturing subpatterns in a recursion was greater
539        than the maximum at the outer level, the higher number was returned, but
540        with unset values at the outer level. The correct (outer level) value is
541        now given.
542    
543    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
544        PCRE did not set those parentheses (unlike Perl). I have now found a way to
545        make it do so. The string so far is captured, making this feature
546        compatible with Perl.
547    
548    23. The tests have been re-organized, adding tests 11 and 12, to make it
549        possible to check the Perl 5.10 features against Perl 5.10.
550    
551    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
552        pattern matches a fixed length string. PCRE did not allow this; now it
553        does. Neither allows recursion.
554    
555    25. I finally figured out how to implement a request to provide the minimum
556        length of subject string that was needed in order to match a given pattern.
557        (It was back references and recursion that I had previously got hung up
558        on.) This code has now been added to pcre_study(); it finds a lower bound
559        to the length of subject needed. It is not necessarily the greatest lower
560        bound, but using it to avoid searching strings that are too short does give
561        some useful speed-ups. The value is available to calling programs via
562        pcre_fullinfo().
563    
564    26. While implementing 25, I discovered to my embarrassment that pcretest had
565        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
566        study optimizations had never been tested with that matching function.
567        Oops. What is worse, even when it was passed study data, there was a bug in
568        pcre_dfa_exec() that meant it never actually used it. Double oops. There
569        were also very few tests of studied patterns with pcre_dfa_exec().
570    
571    27. If (?| is used to create subpatterns with duplicate numbers, they are now
572        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
573        on the other side of the coin, they are no longer allowed to have different
574        names, because these cannot be distinguished in PCRE, and this has caused
575        confusion. (This is a difference from Perl.)
576    
577    28. When duplicate subpattern names are present (necessarily with different
578        numbers, as required by 27 above), and a test is made by name in a
579        conditional pattern, either for a subpattern having been matched, or for
580        recursion in such a pattern, all the associated numbered subpatterns are
581        tested, and the overall condition is true if the condition is true for any
582        one of them. This is the way Perl works, and is also more like the way
583        testing by number works.
584    
585    
586    Version 7.9 11-Apr-09
587    ---------------------
588    
589    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
590        (pcretest), all targets were linked against these libraries. This included
591        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
592        libraries. This caused unwanted dependencies to be created. This problem
593        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
594        pcretest is linked with readline.
595    
596    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
597        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
598        moved outside it again, because FALSE and TRUE are already defined in AIX,
599        but BOOL is not.
600    
601    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
602        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
603    
604    4.  The pcregrep documentation said spaces were inserted as well as colons (or
605        hyphens) following file names and line numbers when outputting matching
606        lines. This is not true; no spaces are inserted. I have also clarified the
607        wording for the --colour (or --color) option.
608    
609    5.  In pcregrep, when --colour was used with -o, the list of matching strings
610        was not coloured; this is different to GNU grep, so I have changed it to be
611        the same.
612    
613    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
614        each matching line was coloured. Now it goes on to look for further matches
615        of any of the test patterns, which is the same behaviour as GNU grep.
616    
617    7.  A pattern that could match an empty string could cause pcregrep to loop; it
618        doesn't make sense to accept an empty string match in pcregrep, so I have
619        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
620        seems to be how GNU grep behaves.
621    
622    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
623        start or after a newline", because the conditional assertion was not being
624        correctly handled. The rule now is that both the assertion and what follows
625        in the first alternative must satisfy the test.
626    
627    9.  If auto-callout was enabled in a pattern with a conditional group whose
628        condition was an assertion, PCRE could crash during matching, both with
629        pcre_exec() and pcre_dfa_exec().
630    
631    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
632        used for matching.
633    
634    11. Unicode property support in character classes was not working for
635        characters (bytes) greater than 127 when not in UTF-8 mode.
636    
637    12. Added the -M command line option to pcretest.
638    
639    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
640    
641    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
642    
643    16. Added comments and documentation about mis-use of no_arg in the C++
644        wrapper.
645    
646    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
647        from Martin Jerabek that uses macro names for all relevant character and
648        string constants.
649    
650    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
651        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
652        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
653        these, but not everybody uses configure.
654    
655    19. A conditional group that had only one branch was not being correctly
656        recognized as an item that could match an empty string. This meant that an
657        enclosing group might also not be so recognized, causing infinite looping
658        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
659        with the subject "ab", where knowledge that the repeated group can match
660        nothing is needed in order to break the loop.
661    
662    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
663        exec(), but without supplying a callout function, matching went wrong.
664    
665    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
666        leak if the size of the offset vector was greater than 30. When the vector
667        is smaller, the saved offsets during recursion go onto a local stack
668        vector, but for larger vectors malloc() is used. It was failing to free
669        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
670        error, in fact).
671    
672    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
673        heapframe that is used only when UTF-8 support is enabled. This caused no
674        problem, but was untidy.
675    
676    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
677        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
678        included within another project.
679    
680    24. Steven Van Ingelgem's patches to add more options to the CMake support,
681        slightly modified by me:
682    
683          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
684              not building pcregrep.
685    
686          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
687              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
688    
689    25. Forward references, both numeric and by name, in patterns that made use of
690        duplicate group numbers, could behave incorrectly or give incorrect errors,
691        because when scanning forward to find the reference group, PCRE was not
692        taking into account the duplicate group numbers. A pattern such as
693        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
694    
695    26. Changed a few more instances of "const unsigned char *" to USPTR, making
696        the feature of a custom pointer more persuasive (as requested by a user).
697    
698    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
699        pcretest.c, inside #ifndefs, because it seems they are sometimes already
700        pre-defined.
701    
702    28. Added support for (*UTF8) at the start of a pattern.
703    
704    29. Arrange for flags added by the "release type" setting in CMake to be shown
705        in the configuration summary.
706    
707    
708    Version 7.8 05-Sep-08
709    ---------------------
710    
711    1.  Replaced UCP searching code with optimized version as implemented for Ad
712        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
713        stage table and inline lookup instead of a function, giving speed ups of 2
714        to 5 times on some simple patterns that I tested. Permission was given to
715        distribute the MultiStage2.py script that generates the tables (it's not in
716        the tarball, but is in the Subversion repository).
717    
718    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
719        scripts.
720    
721    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
722        a group with a zero qualifier. The result of the study could be incorrect,
723        or the function might crash, depending on the pattern.
724    
725    4.  Caseless matching was not working for non-ASCII characters in back
726        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
727        It now works when Unicode Property Support is available.
728    
729    5.  In pcretest, an escape such as \x{de} in the data was always generating
730        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
731        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
732        truncation.
733    
734    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
735    
736    7.  Added two (int) casts to pcregrep when printing the difference of two
737        pointers, in case they are 64-bit values.
738    
739    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
740        test 2 if it fails.
741    
742    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
743        and a #define of that name to empty if it is not externally set. This is to
744        allow users of MSVC to set it if necessary.
745    
746    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
747        the convenience functions in the pcre_get.c source file.
748    
749    11. An option change at the start of a pattern that had top-level alternatives
750        could cause overwriting and/or a crash. This command provoked a crash in
751        some environments:
752    
753          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
754    
755        This potential security problem was recorded as CVE-2008-2371.
756    
757    12. For a pattern where the match had to start at the beginning or immediately
758        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
759        pcre_dfa_exec() could read past the end of the passed subject if there was
760        no match. To help with detecting such bugs (e.g. with valgrind), I modified
761        pcretest so that it places the subject at the end of its malloc-ed buffer.
762    
763    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
764        exec() might read past the end of the data buffer in UTF-8 mode.
765    
766    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
767        the data contained the byte 0x85 as part of a UTF-8 character within its
768        first line. This applied both to normal and DFA matching.
769    
770    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
771        /^[^d]*?$/8 failed to match "abc".
772    
773    16. Added a missing copyright notice to pcrecpp_internal.h.
774    
775    17. Make it more clear in the documentation that values returned from
776        pcre_exec() in ovector are byte offsets, not character counts.
777    
778    18. Tidied a few places to stop certain compilers from issuing warnings.
779    
780    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
781        supplied by Stefan Weber. I made a further small update for 7.8 because
782        there is a change of source arrangements: the pcre_searchfuncs.c module is
783        replaced by pcre_ucd.c.
784    
785    
786    Version 7.7 07-May-08
787    ---------------------
788    
789    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
790        a string to a long long, pretend we don't even have a long long." This is
791        done by checking for the strtoq, strtoll, and _strtoi64 functions.
792    
793    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
794        pre-7.6 versions, which defined a global no_arg variable instead of putting
795        it in the RE class. (See also #8 below.)
796    
797    3.  Remove a line of dead code, identified by coverity and reported by Nuno
798        Lopes.
799    
800    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
801    
802        (1) The include/exclude patterns were being applied to the whole pathnames
803            of files, instead of just to the final components.
804    
805        (2) If there was more than one level of directory, the subdirectories were
806            skipped unless they satisfied the include/exclude conditions. This is
807            inconsistent with GNU grep (and could even be seen as contrary to the
808            pcregrep specification - which I improved to make it absolutely clear).
809            The action now is always to scan all levels of directory, and just
810            apply the include/exclude patterns to regular files.
811    
812    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
813        --exclude_dir in the tests to avoid scanning .svn directories.
814    
815    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
816        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
817        doesn't support NULs in patterns.
818    
819    7.  Added some missing "const"s to declarations of static tables in
820        pcre_compile.c and pcre_dfa_exec.c.
821    
822    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
823        caused by fix #2  above. (Subsequently also a second patch to fix the
824        first patch. And a third patch - this was a messy problem.)
825    
826    9.  Applied Craig's patch to remove the use of push_back().
827    
828    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
829        matching function regexec().
830    
831    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
832        which, however, unlike Perl's \g{...}, are subroutine calls, not back
833        references. PCRE supports relative numbers with this syntax (I don't think
834        Oniguruma does).
835    
836    12. Previously, a group with a zero repeat such as (...){0} was completely
837        omitted from the compiled regex. However, this means that if the group
838        was called as a subroutine from elsewhere in the pattern, things went wrong
839        (an internal error was given). Such groups are now left in the compiled
840        pattern, with a new opcode that causes them to be skipped at execution
841        time.
842    
843    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
844        to the way PCRE behaves:
845    
846        (a) A lone ] character is dis-allowed (Perl treats it as data).
847    
848        (b) A back reference to an unmatched subpattern matches an empty string
849            (Perl fails the current match path).
850    
851        (c) A data ] in a character class must be notated as \] because if the
852            first data character in a class is ], it defines an empty class. (In
853            Perl it is not possible to have an empty class.) The empty class []
854            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
855            The negative empty class [^] matches any one character, independently
856            of the DOTALL setting.
857    
858    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
859        non-existent subpattern following a character class starting with ']' and
860        containing () gave an internal compiling error instead of "reference to
861        non-existent subpattern". Fortunately, when the pattern did exist, the
862        compiled code was correct. (When scanning forwards to check for the
863        existencd of the subpattern, it was treating the data ']' as terminating
864        the class, so got the count wrong. When actually compiling, the reference
865        was subsequently set up correctly.)
866    
867    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
868        it was being rejected as not supported by pcre_dfa_exec(), even though
869        other assertions are supported. I have made pcre_dfa_exec() support
870        (*FAIL).
871    
872    16. The implementation of 13c above involved the invention of a new opcode,
873        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
874        cannot be changed at match time, I realized I could make a small
875        improvement to matching performance by compiling OP_ALLANY instead of
876        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
877        on the OP_ANY path.
878    
879    17. Compiling pcretest on Windows with readline support failed without the
880        following two fixes: (1) Make the unistd.h include conditional on
881        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
882    
883    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
884        ncurses library to be included for pcretest when ReadLine support is
885        requested, but also to allow for it to be overridden. This patch came from
886        Daniel Bergström.
887    
888    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
889        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
890        any errors with the current Unicode tables. Thanks to Peter Kankowski for
891        spotting this.
892    
893    
894    Version 7.6 28-Jan-08
895    ---------------------
896    
897    1.  A character class containing a very large number of characters with
898        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
899        overflow.
900    
901    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
902        HAVE_LONG_LONG is not defined.
903    
904    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
905        bring them up to date and include new features. This patch includes:
906    
907        - Fixed PH's badly added libz and libbz2 support.
908        - Fixed a problem with static linking.
909        - Added pcredemo. [But later removed - see 7 below.]
910        - Fixed dftables problem and added an option.
911        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
912            HAVE_LONG_LONG.
913        - Added readline support for pcretest.
914        - Added an listing of the option settings after cmake has run.
915    
916    4.  A user submitted a patch to Makefile that makes it easy to create
917        "pcre.dll" under mingw when using Configure/Make. I added stuff to
918        Makefile.am that cause it to include this special target, without
919        affecting anything else. Note that the same mingw target plus all
920        the other distribution libraries and programs are now supported
921        when configuring with CMake (see 6 below) instead of with
922        Configure/Make.
923    
924    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
925        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
926        exported in the Windows port". It has not yet been confirmed that the patch
927        solves the problem, but it does no harm.
928    
929    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
930        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
931        with CMake, and also correct the comment about stack recursion.
932    
933    7.  Remove the automatic building of pcredemo from the ./configure system and
934        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
935        of a program that users should build themselves after PCRE is installed, so
936        building it automatically is not really right. What is more, it gave
937        trouble in some build environments.
938    
939    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
940    
941    
942    Version 7.5 10-Jan-08
943    ---------------------
944    
945    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
946        values in parens when parsing an RE using the C++ wrapper."
947    
948    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
949        Characters greater than 255 were excluded from the class instead of being
950        included.
951    
952    3.  The same bug as (2) above applied to negated POSIX classes such as
953        [:^space:].
954    
955    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
956        defined or documented. It seems to have been a typo for PCRE_STATIC, so
957        I have changed it.
958    
959    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
960        first named subpattern) and a construct such as (?&a) would reference the
961        first named subpattern whose name started with "a" (in other words, the
962        length check was missing). Both these problems are fixed. "Subpattern name
963        expected" is now given for (?&) (a zero-length name), and this patch also
964        makes it give the same error for \k'' (previously it complained that that
965        was a reference to a non-existent subpattern).
966    
967    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
968        this is right because (?- can be followed by option settings as well as by
969        digits. I have, however, made the messages clearer.
970    
971    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
972        than the number used in the conditional) now cause a compile-time error.
973        This is actually not compatible with Perl, which accepts such patterns, but
974        treats the conditional as always being FALSE (as PCRE used to), but it
975        seems to me that giving a diagnostic is better.
976    
977    8.  Change "alphameric" to the more common word "alphanumeric" in comments
978        and messages.
979    
980    9.  Fix two occurrences of "backslash" in comments that should have been
981        "backspace".
982    
983    10. Remove two redundant lines of code that can never be obeyed (their function
984        was moved elsewhere).
985    
986    11. The program that makes PCRE's Unicode character property table had a bug
987        which caused it to generate incorrect table entries for sequences of
988        characters that have the same character type, but are in different scripts.
989        It amalgamated them into a single range, with the script of the first of
990        them. In other words, some characters were in the wrong script. There were
991        thirteen such cases, affecting characters in the following ranges:
992    
993          U+002b0 - U+002c1
994          U+0060c - U+0060d
995          U+0061e - U+00612
996          U+0064b - U+0065e
997          U+0074d - U+0076d
998          U+01800 - U+01805
999          U+01d00 - U+01d77
1000          U+01d9b - U+01dbf
1001          U+0200b - U+0200f
1002          U+030fc - U+030fe
1003          U+03260 - U+0327f
1004          U+0fb46 - U+0fbb1
1005          U+10450 - U+1049d
1006    
1007    12. The -o option (show only the matching part of a line) for pcregrep was not
1008        compatible with GNU grep in that, if there was more than one match in a
1009        line, it showed only the first of them. It now behaves in the same way as
1010        GNU grep.
1011    
1012    13. If the -o and -v options were combined for pcregrep, it printed a blank
1013        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1014        does the same. The return code can be used to tell if there were any
1015        non-matching lines.
1016    
1017    14. Added --file-offsets and --line-offsets to pcregrep.
1018    
1019    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1020        infinitely looping recursion. The bug was that positive lookaheads were not
1021        being skipped when checking for a possible empty match (negative lookaheads
1022        and both kinds of lookbehind were skipped).
1023    
1024    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1025        inclusion of <windows.h> to before rather than after the definition of
1026        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1027    
1028    17. Specifying a possessive quantifier with a specific limit for a Unicode
1029        character property caused pcre_compile() to compile bad code, which led at
1030        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1031        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1032        caused the error; without that there was no problem.
1033    
1034    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1035    
1036    19. Added --enable-pcretest-libreadline.
1037    
1038    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1039        RE::GlobalReplace(). As a result, the number of replacements returned was
1040        double what it should be. I removed one of the increments, but Craig sent a
1041        later patch that removed the other one (the right fix) and added unit tests
1042        that check the return values (which was not done before).
1043    
1044    21. Several CMake things:
1045    
1046        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1047            the names libpcre and libpcreposix, not just pcre and pcreposix.
1048    
1049        (2) The above change means that pcretest and pcregrep are now correctly
1050            linked with the newly-built libraries, not previously installed ones.
1051    
1052        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
1053    
1054    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1055        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1056        UTF-8 newline character). The key issue is that the pattern starts .*;
1057        this means that the match must be either at the beginning, or after a
1058        newline. The bug was in the code for advancing after a failed match and
1059        checking that the new position followed a newline. It was not taking
1060        account of UTF-8 characters correctly.
1061    
1062    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1063        character classes. PCRE was not treating the sequence [:...:] as a
1064        character class unless the ... were all letters. Perl, however, seems to
1065        allow any characters between [: and :], though of course it rejects as
1066        unknown any "names" that contain non-letters, because all the known class
1067        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1068        for example, whereas PCRE did not - it did not recognize a POSIX character
1069        class. This seemed a bit dangerous, so the code has been changed to be
1070        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1071        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1072        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1073        Perl does, and where it didn't before.
1074    
1075    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1076        Windows environments %n is disabled by default.
1077    
1078    
1079    Version 7.4 21-Sep-07
1080    ---------------------
1081    
1082    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1083        means that a class such as [\s] counted as "explicit reference to CR or
1084        LF". That isn't really right - the whole point of the change was to try to
1085        help when there was an actual mention of one of the two characters. So now
1086        the change happens only if \r or \n (or a literal CR or LF) character is
1087        encountered.
1088    
1089    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1090        of both had grown to the point where there were only 3 bits left.
1091        Fortunately, there was spare space in the data structure, and so I have
1092        moved the internal flags into a new 16-bit field to free up more option
1093        bits.
1094    
1095    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1096        but did not set the internal JCHANGED flag - either of these is enough to
1097        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1098        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1099        start sets both bits.
1100    
1101    4.  Added options (at build time, compile time, exec time) to change \R from
1102        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1103    
1104    5.  doc/pcresyntax.html was missing from the distribution.
1105    
1106    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1107        compatibility, even though it is no longer used.
1108    
1109    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1110        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1111        windows.h file is present (where different names are used). [This was
1112        reversed later after testing - see 16 below.]
1113    
1114    8.  Changed all #include <config.h> to #include "config.h". There were also
1115        some further <pcre.h> cases that I changed to "pcre.h".
1116    
1117    9.  When pcregrep was used with the --colour option, it missed the line ending
1118        sequence off the lines that it output.
1119    
1120    10. It was pointed out to me that arrays of string pointers cause lots of
1121        relocations when a shared library is dynamically loaded. A technique of
1122        using a single long string with a table of offsets can drastically reduce
1123        these. I have refactored PCRE in four places to do this. The result is
1124        dramatic:
1125    
1126          Originally:                          290
1127          After changing UCP table:            187
1128          After changing error message table:   43
1129          After changing table of "verbs"       36
1130          After changing table of Posix names   22
1131    
1132        Thanks to the folks working on Gregex for glib for this insight.
1133    
1134    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1135        unicode-properties was also set.
1136    
1137    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1138    
1139    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1140        checked only for CRLF.
1141    
1142    14. Added casts to pcretest.c to avoid compiler warnings.
1143    
1144    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1145    
1146    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1147        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1148        entirely. This removes changes made in 7 above.
1149    
1150    17. The CMake files have been updated, and there is now more information about
1151        building with CMake in the NON-UNIX-USE document.
1152    
1153    
1154    Version 7.3 28-Aug-07
1155    ---------------------
1156    
1157     1. In the rejigging of the build system that eventually resulted in 7.1, the
1158        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1159        brackets there is not right, since it causes compilers to look for an
1160        installed pcre.h, not the version that is in the source that is being
1161        compiled (which of course may be different). I have changed it back to:
1162    
1163          #include "pcre.h"
1164    
1165        I have a vague recollection that the change was concerned with compiling in
1166        different directories, but in the new build system, that is taken care of
1167        by the VPATH setting the Makefile.
1168    
1169     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1170        when the subject happened to end in the byte 0x85 (e.g. if the last
1171        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1172        characters but of course it shouldn't be taken as a newline when it is part
1173        of another character. The bug was that, for an unlimited repeat of . in
1174        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1175        characters when looking for a newline.
1176    
1177     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1178    
1179     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1180        in debug output.
1181    
1182     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1183        long printing in the pcrecpp unittest when running under MinGW.
1184    
1185     6. ESC_K was left out of the EBCDIC table.
1186    
1187     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1188        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1189        limit also applies to "virtual nesting" when a pattern is recursive, and in
1190        this case 1000 isn't so big. I have been able to remove this limit at the
1191        expense of backing off one optimization in certain circumstances. Normally,
1192        when pcre_exec() would call its internal match() function recursively and
1193        immediately return the result unconditionally, it uses a "tail recursion"
1194        feature to save stack. However, when a subpattern that can match an empty
1195        string has an unlimited repetition quantifier, it no longer makes this
1196        optimization. That gives it a stack frame in which to save the data for
1197        checking that an empty string has been matched. Previously this was taken
1198        from the 1000-entry workspace that had been reserved. So now there is no
1199        explicit limit, but more stack is used.
1200    
1201     8. Applied Daniel's patches to solve problems with the import/export magic
1202        syntax that is required for Windows, and which was going wrong for the
1203        pcreposix and pcrecpp parts of the library. These were overlooked when this
1204        problem was solved for the main library.
1205    
1206     9. There were some crude static tests to avoid integer overflow when computing
1207        the size of patterns that contain repeated groups with explicit upper
1208        limits. As the maximum quantifier is 65535, the maximum group length was
1209        set at 30,000 so that the product of these two numbers did not overflow a
1210        32-bit integer. However, it turns out that people want to use groups that
1211        are longer than 30,000 bytes (though not repeat them that many times).
1212        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1213        made it possible to implement the integer overflow checks in a much more
1214        dynamic way, which I have now done. The artificial limitation on group
1215        length has been removed - we now have only the limit on the total length of
1216        the compiled pattern, which depends on the LINK_SIZE setting.
1217    
1218    10. Fixed a bug in the documentation for get/copy named substring when
1219        duplicate names are permitted. If none of the named substrings are set, the
1220        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1221        empty string.
1222    
1223    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1224        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1225        because the ] is interpreted as the first data character and the
1226        terminating ] is not found. PCRE has been made compatible with Perl in this
1227        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1228        cause memory overwriting.
1229    
1230    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1231        string has been matched (to stop an infinite loop). It was not recognizing
1232        a conditional subpattern that could match an empty string if that
1233        subpattern was within another subpattern. For example, it looped when
1234        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1235        condition was not nested. This bug has been fixed.
1236    
1237    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1238        past the start of the subject in the presence of bytes with the top bit
1239        set, for example "\x8aBCD".
1240    
1241    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1242        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1243    
1244    14. Optimized (?!) to (*FAIL).
1245    
1246    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1247        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1248        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1249        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1250        does: it's just the validity check that is more restrictive.
1251    
1252    16. Inserted checks for integer overflows during escape sequence (backslash)
1253        processing, and also fixed erroneous offset values for syntax errors during
1254        backslash processing.
1255    
1256    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1257        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1258    
1259    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1260        caused an overrun.
1261    
1262    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1263        something other than just ASCII characters) inside a group that had an
1264        unlimited repeat caused a loop at compile time (while checking to see
1265        whether the group could match an empty string).
1266    
1267    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1268        [\P{Any}] did so. (Error in the code for printing property names.)
1269    
1270    21. An orphan \E inside a character class could cause a crash.
1271    
1272    22. A repeated capturing bracket such as (A)? could cause a wild memory
1273        reference during compilation.
1274    
1275    23. There are several functions in pcre_compile() that scan along a compiled
1276        expression for various reasons (e.g. to see if it's fixed length for look
1277        behind). There were bugs in these functions when a repeated \p or \P was
1278        present in the pattern. These operators have additional parameters compared
1279        with \d, etc, and these were not being taken into account when moving along
1280        the compiled data. Specifically:
1281    
1282        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1283            length.
1284    
1285        (b) An item such as \pL+ within a repeated group could cause crashes or
1286            loops.
1287    
1288        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1289            "reference to non-existent subpattern" error.
1290    
1291        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1292    
1293    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1294        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1295    
1296    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1297    
1298    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1299        character were causing crashes (broken optimization).
1300    
1301    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1302        \p or \P) caused a compile-time loop.
1303    
1304    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1305        break. For example, the unstudied pattern [\r\n]A does not match the string
1306        "\r\nA" because change 7.0/46 below moves the current point on by two
1307        characters after failing to match at the start. However, the pattern \nA
1308        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1309        the same is true. There doesn't seem any very clean way out of this, but
1310        what I have chosen to do makes the common cases work: PCRE now takes note
1311        of whether there can be an explicit match for \r or \n anywhere in the
1312        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1313        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1314        pattern has explicit CR or LF references.
1315    
1316    29. Added (*CR) etc for changing newline setting at start of pattern.
1317    
1318    
1319    Version 7.2 19-Jun-07
1320    ---------------------
1321    
1322     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1323        which is apparently normally available under Windows.
1324    
1325     2. Re-jig the pcregrep tests with different newline settings in an attempt
1326        to make them independent of the local environment's newline setting.
1327    
1328     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1329    
1330     4. Some of the "internals" tests were previously cut out when the link size
1331        was not 2, because the output contained actual offsets. The recent new
1332        "Z" feature of pcretest means that these can be cut out, making the tests
1333        usable with all link sizes.
1334    
1335     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1336        stack recursion. This gives a massive performance boost under BSD, but just
1337        a small improvement under Linux. However, it saves one field in the frame
1338        in all cases.
1339    
1340     6. Added more features from the forthcoming Perl 5.10:
1341    
1342        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1343            recursion call. It refers to the nth most recently opened parentheses.
1344    
1345        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1346            to be opened parentheses.
1347    
1348        (c) Conditions that refer to capturing parentheses can be specified
1349            relatively, for example, (?(-2)... or (?(+3)...
1350    
1351        (d) \K resets the start of the current match so that everything before
1352            is not part of it.
1353    
1354        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1355    
1356        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1357            reference syntax.
1358    
1359        (g) (?| introduces a group in which the numbering of parentheses in each
1360            alternative starts with the same number.
1361    
1362        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1363    
1364     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1365        PCRE_INFO_JCHANGED.
1366    
1367     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1368        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1369        for detecting groups that can match an empty string.
1370    
1371     9. A pattern with a very large number of alternatives (more than several
1372        hundred) was running out of internal workspace during the pre-compile
1373        phase, where pcre_compile() figures out how much memory will be needed. A
1374        bit of new cunning has reduced the workspace needed for groups with
1375        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1376        workspace instead of running out of the 4096 that are available.
1377    
1378    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1379    
1380    11. Applied patch from Google to remove an optimization that didn't quite work.
1381        The report of the bug said:
1382    
1383          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1384          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1385          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1386    
1387    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1388        it matched the wrong number of bytes.
1389    
1390    
1391    Version 7.1 24-Apr-07
1392    ---------------------
1393    
1394     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1395        that is more "standard", making use of automake and other Autotools. There
1396        is some re-arrangement of the files and adjustment of comments consequent
1397        on this.
1398    
1399     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1400        for recursive directory scanning broke on some systems because the files
1401        are not scanned in any specific order and on different systems the order
1402        was different. A call to "sort" has been inserted into RunGrepTest for the
1403        approprate test as a short-term fix. In the longer term there may be an
1404        alternative.
1405    
1406     3. I had an email from Eric Raymond about problems translating some of PCRE's
1407        man pages to HTML (despite the fact that I distribute HTML pages, some
1408        people do their own conversions for various reasons). The problems
1409        concerned the use of low-level troff macros .br and .in. I have therefore
1410        removed all such uses from the man pages (some were redundant, some could
1411        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1412        HTML has been updated to handle .nf/.fi and to complain if it encounters
1413        .br or .in.
1414    
1415     4. Updated comments in configure.ac that get placed in config.h.in and also
1416        arranged for config.h to be included in the distribution, with the name
1417        config.h.generic, for the benefit of those who have to compile without
1418        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1419    
1420     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1421        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1422        makevp.bat for the current PCRE, using the additional files
1423        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1424    
1425     6. A Windows user reported a minor discrepancy with test 2, which turned out
1426        to be caused by a trailing space on an input line that had got lost in his
1427        copy. The trailing space was an accident, so I've just removed it.
1428    
1429     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1430        that is needed.
1431    
1432     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1433        as "const" (a) because they are and (b) because it helps the PHP
1434        maintainers who have recently made a script to detect big data structures
1435        in the php code that should be moved to the .rodata section. I remembered
1436        to update Builducptable as well, so it won't revert if ucptable.h is ever
1437        re-created.
1438    
1439     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1440        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1441        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1442        support is not required. This saves 1.5-2K of code, which is important in
1443        some applications.
1444    
1445        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1446        so as not to refer to the tables, even though these functions will never be
1447        called when UTF-8 support is disabled. Otherwise there are problems with a
1448        shared library.
1449    
1450    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1451    
1452        (a) It was defining its arguments as char * instead of void *.
1453    
1454        (b) It was assuming that all moves were upwards in memory; this was true
1455            a long time ago when I wrote it, but is no longer the case.
1456    
1457        The emulated memove() is provided for those environments that have neither
1458        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1459        is clearly not the case, as these two bugs were recently reported.
1460    
1461    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1462        and Detrail to create the HTML documentation, the .txt form of the man
1463        pages, and it removes trailing spaces from listed files. It also creates
1464        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1465        case, it wraps all the #defines with #ifndefs. This script should be run
1466        before "make dist".
1467    
1468    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1469        with Unicode property support.
1470    
1471        (a) For a maximizing quantifier, if the two different cases of the
1472            character were of different lengths in their UTF-8 codings (there are
1473            some cases like this - I found 11), and the matching function had to
1474            back up over a mixture of the two cases, it incorrectly assumed they
1475            were both the same length.
1476    
1477        (b) When PCRE was configured to use the heap rather than the stack for
1478            recursion during matching, it was not correctly preserving the data for
1479            the other case of a UTF-8 character when checking ahead for a match
1480            while processing a minimizing repeat. If the check also involved
1481            matching a wide character, but failed, corruption could cause an
1482            erroneous result when trying to check for a repeat of the original
1483            character.
1484    
1485    13. Some tidying changes to the testing mechanism:
1486    
1487        (a) The RunTest script now detects the internal link size and whether there
1488            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1489            values substituted by "configure". (The RunGrepTest script already did
1490            this for UTF-8.) The configure.ac script no longer substitutes the
1491            relevant variables.
1492    
1493        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1494            with length and offset values. This means that the output is different
1495            for different internal link sizes. Test 2 is skipped for link sizes
1496            other than 2 because of this, bypassing the problem. Unfortunately,
1497            there was also a test in test 3 (the locale tests) that used /B and
1498            failed for link sizes other than 2. Rather than cut the whole test out,
1499            I have added a new /Z option to pcretest that replaces the length and
1500            offset values with spaces. This is now used to make test 3 independent
1501            of link size. (Test 2 will be tidied up later.)
1502    
1503    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1504        segmentation fault instead of returning the appropriate error message.
1505    
1506    15. In multiline mode when the newline sequence was set to "any", the pattern
1507        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1508        This doesn't seem right; it now treats the CRLF combination as the line
1509        ending, and so does not match in that case. It's only a pattern such as ^$
1510        that would hit this one: something like ^ABC$ would have failed after \r
1511        and then tried again after \r\n.
1512    
1513    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1514        in an attempt to make files that differ only in their line terminators
1515        compare equal. This works on Linux.
1516    
1517    17. Under certain error circumstances pcregrep might try to free random memory
1518        as it exited. This is now fixed, thanks to valgrind.
1519    
1520    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1521        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1522        was because its rules for how to advance for /g after matching an empty
1523        string at the end of a line did not allow for this case. They now check for
1524        it specially.
1525    
1526    20. pcretest is supposed to handle patterns and data of any length, by
1527        extending its buffers when necessary. It was getting this wrong when the
1528        buffer for a data line had to be extended.
1529    
1530    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1531        CRLF as a newline sequence.
1532    
1533    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1534        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1535        I have nevertheless tidied it up.
1536    
1537    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1538    
1539    24. Added a man page for pcre-config.
1540    
1541    
1542    Version 7.0 19-Dec-06
1543    ---------------------
1544    
1545     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1546        moving to gcc 4.1.1.
1547    
1548     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1549        sys/time.h, which is documented as needed for this function. It doesn't
1550        seem to matter on Linux, but it showed up on some releases of OS X.
1551    
1552     3. It seems that there are systems where bytes whose values are greater than
1553        127 match isprint() in the "C" locale. The "C" locale should be the
1554        default when a C program starts up. In most systems, only ASCII printing
1555        characters match isprint(). This difference caused the output from pcretest
1556        to vary, making some of the tests fail. I have changed pcretest so that:
1557    
1558        (a) When it is outputting text in the compiled version of a pattern, bytes
1559            other than 32-126 are always shown as hex escapes.
1560    
1561        (b) When it is outputting text that is a matched part of a subject string,
1562            it does the same, unless a different locale has been set for the match
1563            (using the /L modifier). In this case, it uses isprint() to decide.
1564    
1565     4. Fixed a major bug that caused incorrect computation of the amount of memory
1566        required for a compiled pattern when options that changed within the
1567        pattern affected the logic of the preliminary scan that determines the
1568        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1569        that the computed length was too small. The symptoms of this bug were
1570        either the PCRE error "internal error: code overflow" from pcre_compile(),
1571        or a glibc crash with a message such as "pcretest: free(): invalid next
1572        size (fast)". Examples of patterns that provoked this bug (shown in
1573        pcretest format) are:
1574    
1575          /(?-x: )/x
1576          /(?x)(?-x: \s*#\s*)/
1577          /((?i)[\x{c0}])/8
1578          /(?i:[\x{c0}])/8
1579    
1580        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1581        is now done differently.
1582    
1583     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1584        wrapper classes; (b) implement a new function in the C++ scanner that is
1585        more efficient than the old way of doing things because it avoids levels of
1586        recursion in the regex matching; (c) add a paragraph to the documentation
1587        for the FullMatch() function.
1588    
1589     6. The escape sequence \n was being treated as whatever was defined as
1590        "newline". Not only was this contrary to the documentation, which states
1591        that \n is character 10 (hex 0A), but it also went horribly wrong when
1592        "newline" was defined as CRLF. This has been fixed.
1593    
1594     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1595        was being set to -1 for the "end of line" case (supposedly a value that no
1596        character can have). Though this value is never used (the check for end of
1597        line is "zero bytes in current character"), it caused compiler complaints.
1598        I've changed it to 0xffffffff.
1599    
1600     8. In pcre_version.c, the version string was being built by a sequence of
1601        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1602        string (as it is for production releases) called a macro with an empty
1603        argument. The C standard says the result of this is undefined. The gcc
1604        compiler treats it as an empty string (which was what was wanted) but it is
1605        reported that Visual C gives an error. The source has been hacked around to
1606        avoid this problem.
1607    
1608     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1609        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1610        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1611        of them did).
1612    
1613    10. Originally, pcretest opened its input and output without "b"; then I was
1614        told that "b" was needed in some environments, so it was added for release
1615        5.0 to both the input and output. (It makes no difference on Unix-like
1616        systems.) Later I was told that it is wrong for the input on Windows. I've
1617        now abstracted the modes into two macros, to make it easier to fiddle with
1618        them, and removed "b" from the input mode under Windows.
1619    
1620    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1621    
1622    12. Added -help and --help to pcretest as an official way of being reminded
1623        of the options.
1624    
1625    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1626        and pcrecpp.cc because they annoy compilers at high warning levels.
1627    
1628    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1629    
1630    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1631        scripts are not C programs :-) and which was not noticed because it works
1632        on Linux.
1633    
1634    16. pcretest is supposed to handle any length of pattern and data line (as one
1635        line or as a continued sequence of lines) by extending its input buffer if
1636        necessary. This feature was broken for very long pattern lines, leading to
1637        a string of junk being passed to pcre_compile() if the pattern was longer
1638        than about 50K.
1639    
1640    17. I have done a major re-factoring of the way pcre_compile() computes the
1641        amount of memory needed for a compiled pattern. Previously, there was code
1642        that made a preliminary scan of the pattern in order to do this. That was
1643        OK when PCRE was new, but as the facilities have expanded, it has become
1644        harder and harder to keep it in step with the real compile phase, and there
1645        have been a number of bugs (see for example, 4 above). I have now found a
1646        cunning way of running the real compile function in a "fake" mode that
1647        enables it to compute how much memory it would need, while actually only
1648        ever using a few hundred bytes of working memory and without too many
1649        tests of the mode. This should make future maintenance and development
1650        easier. A side effect of this work is that the limit of 200 on the nesting
1651        depth of parentheses has been removed (though this was never a serious
1652        limitation, I suspect). However, there is a downside: pcre_compile() now
1653        runs more slowly than before (30% or more, depending on the pattern). I
1654        hope this isn't a big issue. There is no effect on runtime performance.
1655    
1656    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1657        newline (only possible for the last line of a file) and it was a
1658        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1659    
1660    19. Added additional timing features to pcretest. (1) The -tm option now times
1661        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1662        separate command line item, by a number that specifies the number of
1663        repeats to use when timing. The default is 50000; this gives better
1664        precision, but takes uncomfortably long for very large patterns.
1665    
1666    20. Extended pcre_study() to be more clever in cases where a branch of a
1667        subpattern has no definite first character. For example, (a*|b*)[cd] would
1668        previously give no result from pcre_study(). Now it recognizes that the
1669        first character must be a, b, c, or d.
1670    
1671    21. There was an incorrect error "recursive call could loop indefinitely" if
1672        a subpattern (or the entire pattern) that was being tested for matching an
1673        empty string contained only one non-empty item after a nested subpattern.
1674        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1675        incorrectly, because the \d was being skipped in the check.
1676    
1677    22. The pcretest program now has a new pattern option /B and a command line
1678        option -b, which is equivalent to adding /B to every pattern. This causes
1679        it to show the compiled bytecode, without the additional information that
1680        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1681        is the same as /B/I).
1682    
1683    23. A new optimization is now able automatically to treat some sequences such
1684        as a*b as a*+b. More specifically, if something simple (such as a character
1685        or a simple class like \d) has an unlimited quantifier, and is followed by
1686        something that cannot possibly match the quantified thing, the quantifier
1687        is automatically "possessified".
1688    
1689    24. A recursive reference to a subpattern whose number was greater than 39
1690        went wrong under certain circumstances in UTF-8 mode. This bug could also
1691        have affected the operation of pcre_study().
1692    
1693    25. Realized that a little bit of performance could be had by replacing
1694        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1695    
1696    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1697    
1698    27. Possessive quantifiers such as a++ were previously implemented by turning
1699        them into atomic groups such as ($>a+). Now they have their own opcodes,
1700        which improves performance. This includes the automatically created ones
1701        from 23 above.
1702    
1703    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1704        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1705        the first matched character to be a colon. This applied both to named and
1706        numbered groups.
1707    
1708    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1709    
1710    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1711        building PCRE on Windows easier, so I have included it in the distribution.
1712    
1713    31. There is now a check in pcretest against a ridiculously large number being
1714        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1715        loop, the loop is abandoned.
1716    
1717    32. Forward references to subpatterns in conditions such as (?(2)...) where
1718        subpattern 2 is defined later cause pcre_compile() to search forwards in
1719        the pattern for the relevant set of parentheses. This search went wrong
1720        when there were unescaped parentheses in a character class, parentheses
1721        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1722    
1723    33. "Subroutine" calls and backreferences were previously restricted to
1724        referencing subpatterns earlier in the regex. This restriction has now
1725        been removed.
1726    
1727    34. Added a number of extra features that are going to be in Perl 5.10. On the
1728        whole, these are just syntactic alternatives for features that PCRE had
1729        previously implemented using the Python syntax or my own invention. The
1730        other formats are all retained for compatibility.
1731    
1732        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1733            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1734            also .NET compatible.
1735    
1736        (b) A recursion or subroutine call to a named group can now be defined as
1737            (?&name) as well as (?P>name).
1738    
1739        (c) A backreference to a named group can now be defined as \k<name> or
1740            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1741            5.10, are also .NET compatible.
1742    
1743        (d) A conditional reference to a named group can now use the syntax
1744            (?(<name>) or (?('name') as well as (?(name).
1745    
1746        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1747            groups (named and numbered) that are never evaluated inline, but can be
1748            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1749            is always false. There may be only one alternative in such a group.
1750    
1751        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1752            as the simple (?(R). The condition is true only if the most recent
1753            recursion is that of the given number or name. It does not search out
1754            through the entire recursion stack.
1755    
1756        (g) The escape \gN or \g{N} has been added, where N is a positive or
1757            negative number, specifying an absolute or relative reference.
1758    
1759    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1760        some "unreachable code" warnings.
1761    
1762    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1763        things, this adds five new scripts.
1764    
1765    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1766        There were also incompatibilities regarding the handling of \Q..\E inside
1767        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1768        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1769    
1770    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1771        matches an empty string, and forcibly breaks the loop. There were bugs in
1772        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1773        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1774        separate and independent bugs (that affected different cases) have been
1775        fixed.
1776    
1777    39. Refactored the code to abolish the use of different opcodes for small
1778        capturing bracket numbers. This is a tidy that I avoided doing when I
1779        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1780        The new approach is not only tidier, it makes it possible to reduce the
1781        memory needed to fix the previous bug (38).
1782    
1783    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1784        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1785        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1786        mode.
1787    
1788    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1789        report.
1790    
1791    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1792        copy construction and assignment in the C++ wrapper.
1793    
1794    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1795        couple of bugs that could have given wrong results in the "--newline=crlf"
1796        case.
1797    
1798    44. Added a number of casts and did some reorganization of signed/unsigned int
1799        variables following suggestions from Dair Grant. Also renamed the variable
1800        "this" as "item" because it is a C++ keyword.
1801    
1802    45. Arranged for dftables to add
1803    
1804          #include "pcre_internal.h"
1805    
1806        to pcre_chartables.c because without it, gcc 4.x may remove the array
1807        definition from the final binary if PCRE is built into a static library and
1808        dead code stripping is activated.
1809    
1810    46. For an unanchored pattern, if a match attempt fails at the start of a
1811        newline sequence, and the newline setting is CRLF or ANY, and the next two
1812        characters are CRLF, advance by two characters instead of one.
1813    
1814    
1815    Version 6.7 04-Jul-06
1816    ---------------------
1817    
1818     1. In order to handle tests when input lines are enormously long, pcretest has
1819        been re-factored so that it automatically extends its buffers when
1820        necessary. The code is crude, but this _is_ just a test program. The
1821        default size has been increased from 32K to 50K.
1822    
1823     2. The code in pcre_study() was using the value of the re argument before
1824        testing it for NULL. (Of course, in any sensible call of the function, it
1825        won't be NULL.)
1826    
1827     3. The memmove() emulation function in pcre_internal.h, which is used on
1828        systems that lack both memmove() and bcopy() - that is, hardly ever -
1829        was missing a "static" storage class specifier.
1830    
1831     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1832        containing an extended class (one that cannot be represented by a bitmap
1833        because it contains high-valued characters or Unicode property items, e.g.
1834        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1835        pattern, but PCRE should not loop if you do not (it no longer does).
1836        [Detail: two cases were found: (a) a repeated subpattern containing an
1837        extended class; (b) a recursive reference to a subpattern that followed a
1838        previous extended class. It wasn't skipping over the extended class
1839        correctly when UTF-8 mode was not set.]
1840    
1841     5. A negated single-character class was not being recognized as fixed-length
1842        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1843        compile error "lookbehind assertion is not fixed length".
1844    
1845     6. The RunPerlTest auxiliary script was showing an unexpected difference
1846        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1847        write a Perl script that can interpret lines of an input file either as
1848        byte characters or as UTF-8, which is what "perltest" was being required to
1849        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1850        can't do is switch easily at run time between having the "use utf8;" pragma
1851        or not. In the end, I fudged it by using the RunPerlTest script to insert
1852        "use utf8;" explicitly for the UTF-8 tests.
1853    
1854     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1855        the end of the subject string, contrary to the documentation and to what
1856        Perl does. This was true of both matching functions. Now it matches only at
1857        the start of the subject and immediately after *internal* newlines.
1858    
1859     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1860        a pointer to an int instead of a pointer to an unsigned long int. This
1861        caused problems on 64-bit systems.
1862    
1863     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1864        instance of the 'standard' template library not being so standard".
1865    
1866    10. There was no check on the number of named subpatterns nor the maximum
1867        length of a subpattern name. The product of these values is used to compute
1868        the size of the memory block for a compiled pattern. By supplying a very
1869        long subpattern name and a large number of named subpatterns, the size
1870        computation could be caused to overflow. This is now prevented by limiting
1871        the length of names to 32 characters, and the number of named subpatterns
1872        to 10,000.
1873    
1874    11. Subpatterns that are repeated with specific counts have to be replicated in
1875        the compiled pattern. The size of memory for this was computed from the
1876        length of the subpattern and the repeat count. The latter is limited to
1877        65535, but there was no limit on the former, meaning that integer overflow
1878        could in principle occur. The compiled length of a repeated subpattern is
1879        now limited to 30,000 bytes in order to prevent this.
1880    
1881    12. Added the optional facility to have named substrings with the same name.
1882    
1883    13. Added the ability to use a named substring as a condition, using the
1884        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1885        are numbers (not recommended). Forward references are permitted.
1886    
1887    14. Added forward references in named backreferences (if you see what I mean).
1888    
1889    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1890        pattern could run off the end of the subject. For example, the pattern
1891        "(?s)(.{1,5})"8 did this with the subject "ab".
1892    
1893    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1894        PCRE_CASELESS was set when matching characters that were quantified with ?
1895        or *.
1896    
1897    17. A character class other than a single negated character that had a minimum
1898        but no maximum quantifier - for example [ab]{6,} - was not handled
1899        correctly by pce_dfa_exec(). It would match only one character.
1900    
1901    18. A valid (though odd) pattern that looked like a POSIX character
1902        class but used an invalid character after [ (for example [[,abc,]]) caused
1903        pcre_compile() to give the error "Failed: internal error: code overflow" or
1904        in some cases to crash with a glibc free() error. This could even happen if
1905        the pattern terminated after [[ but there just happened to be a sequence of
1906        letters, a binary zero, and a closing ] in the memory that followed.
1907    
1908    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1909        over the years. Originally (before any Unicode support), just the bottom 8
1910        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1911        output from "man perlunicode" includes this:
1912    
1913          The regular expression compiler produces polymorphic opcodes.  That
1914          is, the pattern adapts to the data and automatically switches to
1915          the Unicode character scheme when presented with Unicode data--or
1916          instead uses a traditional byte scheme when presented with byte
1917          data.
1918    
1919        Sadly, a wide octal escape does not cause a switch, and in a string with
1920        no other multibyte characters, these octal escapes are treated as before.
1921        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1922        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1923        Unicode string.
1924    
1925        I have not perpetrated such confusion in PCRE. Up till now, it took just
1926        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1927        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1928        translate to the appropriate multibyte character.
1929    
1930    29. Applied some refactoring to reduce the number of warnings from Microsoft
1931        and Borland compilers. This has included removing the fudge introduced
1932        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1933        a warning about an unused variable.
1934    
1935    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1936        characters since release 4.0, because Perl (from release 5.004) does not.
1937        [Or at least, is documented not to: some releases seem to be in conflict
1938        with the documentation.] However, when a pattern was studied with
1939        pcre_study() and all its branches started with \s, PCRE still included VT
1940        as a possible starting character. Of course, this did no harm; it just
1941        caused an unnecessary match attempt.
1942    
1943    22. Removed a now-redundant internal flag bit that recorded the fact that case
1944        dependency changed within the pattern. This was once needed for "required
1945        byte" processing, but is no longer used. This recovers a now-scarce options
1946        bit. Also moved the least significant internal flag bit to the most-
1947        significant bit of the word, which was not previously used (hangover from
1948        the days when it was an int rather than a uint) to free up another bit for
1949        the future.
1950    
1951    23. Added support for CRLF line endings as well as CR and LF. As well as the
1952        default being selectable at build time, it can now be changed at runtime
1953        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1954        specify that it is scanning data with non-default line endings.
1955    
1956    24. Changed the definition of CXXLINK to make it agree with the definition of
1957        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1958    
1959    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1960        recursions. This makes a big different to stack usage for some patterns.
1961    
1962    26. If a subpattern containing a named recursion or subroutine reference such
1963        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1964        the space required for the compiled pattern went wrong and gave too small a
1965        value. Depending on the environment, this could lead to "Failed: internal
1966        error: code overflow at offset 49" or "glibc detected double free or
1967        corruption" errors.
1968    
1969    27. Applied patches from Google (a) to support the new newline modes and (b) to
1970        advance over multibyte UTF-8 characters in GlobalReplace.
1971    
1972    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1973        difference for some implementation of PCRE in some Windows version.
1974    
1975    29. Added some extra testing facilities to pcretest:
1976    
1977        \q<number>   in a data line sets the "match limit" value
1978        \Q<number>   in a data line sets the "match recursion limt" value
1979        -S <number>  sets the stack size, where <number> is in megabytes
1980    
1981        The -S option isn't available for Windows.
1982    
1983    
1984    Version 6.6 06-Feb-06
1985    ---------------------
1986    
1987     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1988        in pcreposix.h. I have copied the definition from pcre.h.
1989    
1990     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1991        because pcre.h is no longer a built file.
1992    
1993     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1994        not normally included in the compiled code.
1995    
1996    
1997    Version 6.5 01-Feb-06
1998    ---------------------
1999    
2000     1. When using the partial match feature with pcre_dfa_exec(), it was not
2001        anchoring the second and subsequent partial matches at the new starting
2002        point. This could lead to incorrect results. For example, with the pattern
2003        /1234/, partially matching against "123" and then "a4" gave a match.
2004    
2005     2. Changes to pcregrep:
2006    
2007        (a) All non-match returns from pcre_exec() were being treated as failures
2008            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2009            error message is output. Some extra information is given for the
2010            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2011            probably the only errors that are likely to be caused by users (by
2012            specifying a regex that has nested indefinite repeats, for instance).
2013            If there are more than 20 of these errors, pcregrep is abandoned.
2014    
2015        (b) A binary zero was treated as data while matching, but terminated the
2016            output line if it was written out. This has been fixed: binary zeroes
2017            are now no different to any other data bytes.
2018    
2019        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2020            used to set a locale for matching. The --locale=xxxx long option has
2021            been added (no short equivalent) to specify a locale explicitly on the
2022            pcregrep command, overriding the environment variables.
2023    
2024        (d) When -B was used with -n, some line numbers in the output were one less
2025            than they should have been.
2026    
2027        (e) Added the -o (--only-matching) option.
2028    
2029        (f) If -A or -C was used with -c (count only), some lines of context were
2030            accidentally printed for the final match.
2031    
2032        (g) Added the -H (--with-filename) option.
2033    
2034        (h) The combination of options -rh failed to suppress file names for files
2035            that were found from directory arguments.
2036    
2037        (i) Added the -D (--devices) and -d (--directories) options.
2038    
2039        (j) Added the -F (--fixed-strings) option.
2040    
2041        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2042    
2043        (l) Added the --colo(u)r option.
2044    
2045        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2046            is not present by default.
2047    
2048     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2049        items such as (?R) or (?1), when the recursion could match a number of
2050        alternatives. If it matched one of the alternatives, but subsequently,
2051        outside the recursion, there was a failure, the code tried to back up into
2052        the recursion. However, because of the way PCRE is implemented, this is not
2053        possible, and the result was an incorrect result from the match.
2054    
2055        In order to prevent this happening, the specification of recursion has
2056        been changed so that all such subpatterns are automatically treated as
2057        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2058    
2059     4. I had overlooked the fact that, in some locales, there are characters for
2060        which isalpha() is true but neither isupper() nor islower() are true. In
2061        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2062        and ordfeminine) are like this. This affected the treatment of \w and \W
2063        when they appeared in character classes, but not when they appeared outside
2064        a character class. The bit map for "word" characters is now created
2065        separately from the results of isalnum() instead of just taking it from the
2066        upper, lower, and digit maps. (Plus the underscore character, of course.)
2067    
2068     5. The above bug also affected the handling of POSIX character classes such as
2069        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2070        permanent tables. Instead, the bit maps for such a class were previously
2071        created as the appropriate unions of the upper, lower, and digit bitmaps.
2072        Now they are created by subtraction from the [[:word:]] class, which has
2073        its own bitmap.
2074    
2075     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2076        It is created by subtracting the vertical space characters (\x09, \x0a,
2077        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2078        subtraction was done in the overall bitmap for a character class, meaning
2079        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2080        be recognized. This bug has been fixed.
2081    
2082     7. Patches from the folks at Google:
2083    
2084          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2085          real life, but is still worth protecting against".
2086    
2087          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2088          regular expressions".
2089    
2090          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2091          have it.
2092    
2093          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2094          "configure" and the latter not, in order to fix a problem somebody had
2095          with compiling the Arg class on HP-UX.
2096    
2097          (e) Improve the error-handling of the C++ wrapper a little bit.
2098    
2099          (f) New tests for checking recursion limiting.
2100    
2101     8. The pcre_memmove() function, which is used only if the environment does not
2102        have a standard memmove() function (and is therefore rarely compiled),
2103        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2104        returning a result (though PCRE never actually uses the result).
2105    
2106     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2107        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2108        returned instead of calling malloc() with an overflowing number that would
2109        most likely cause subsequent chaos.
2110    
2111    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2112    
2113    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2114        with this option is matched, the nmatch and pmatch options of regexec() are
2115        ignored.
2116    
2117    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2118        provided in case anyone wants to the the POSIX interface with UTF-8
2119        strings.
2120    
2121    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2122        C++ linking (needed for some HP-UX environments).
2123    
2124    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2125        (unused parameter) and in the pcre_printint() function (omitted "default"
2126        switch label when the default is to do nothing).
2127    
2128    15. Added some code to make it possible, when PCRE is compiled as a C++
2129        library, to replace subject pointers for pcre_exec() with a smart pointer
2130        class, thus making it possible to process discontinuous strings.
2131    
2132    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2133        much the same function. They were added by different people who were trying
2134        to make PCRE easy to compile on non-Unix systems. It has been suggested
2135        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2136        for compiling on Windows systems. I have therefore replaced it with
2137        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2138        defaults to "extern" for C or "extern C" for C++, which works fine on
2139        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2140        SCOPE with something explicit in config.h. In addition:
2141    
2142        (a) pcreposix.h still had just "extern" instead of either of these macros;
2143            I have replaced it with PCRE_DATA_SCOPE.
2144    
2145        (b) Functions such as _pcre_xclass(), which are internal to the library,
2146            but external in the C sense, all had PCRE_EXPORT in their definitions.
2147            This is apparently wrong for the Windows case, so I have removed it.
2148            (It makes no difference on Unix-like systems.)
2149    
2150    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2151        of recursive calls to match(). This is different to MATCH_LIMIT because
2152        that limits the total number of calls to match(), not all of which increase
2153        the depth of recursion. Limiting the recursion depth limits the amount of
2154        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2155        when PCRE is compiled, and changed at run time. A patch from Google adds
2156        this functionality to the C++ interface.
2157    
2158    18. Changes to the handling of Unicode character properties:
2159    
2160        (a) Updated the table to Unicode 4.1.0.
2161    
2162        (b) Recognize characters that are not in the table as "Cn" (undefined).
2163    
2164        (c) I revised the way the table is implemented to a much improved format
2165            which includes recognition of ranges. It now supports the ranges that
2166            are defined in UnicodeData.txt, and it also amalgamates other
2167            characters into ranges. This has reduced the number of entries in the
2168            table from around 16,000 to around 3,000, thus reducing its size
2169            considerably. I realized I did not need to use a tree structure after
2170            all - a binary chop search is just as efficient. Having reduced the
2171            number of entries, I extended their size from 6 bytes to 8 bytes to
2172            allow for more data.
2173    
2174        (d) Added support for Unicode script names via properties such as \p{Han}.
2175    
2176    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2177        matching that character.
2178    
2179    20. When matching a repeated Unicode property with a minimum greater than zero,
2180        (for example \pL{2,}), PCRE could look past the end of the subject if it
2181        reached it while seeking the minimum number of characters. This could
2182        happen only if some of the characters were more than one byte long, because
2183        there is a check for at least the minimum number of bytes.
2184    
2185    21. Refactored the implementation of \p and \P so as to be more general, to
2186        allow for more different types of property in future. This has changed the
2187        compiled form incompatibly. Anybody with saved compiled patterns that use
2188        \p or \P will have to recompile them.
2189    
2190    22. Added "Any" and "L&" to the supported property types.
2191    
2192    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2193        but give a compile time error if the value is greater than 0xff.
2194    
2195    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2196        accidentally not being installed or uninstalled.
2197    
2198    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2199        made were to insert the current release number. This seemed silly, because
2200        it made things harder for people building PCRE on systems that don't run
2201        "configure". I have turned pcre.h into a distributed file, no longer built
2202        by "configure", with the version identification directly included. There is
2203        no longer a pcre.h.in file.
2204    
2205        However, this change necessitated a change to the pcre-config script as
2206        well. It is built from pcre-config.in, and one of the substitutions was the
2207        release number. I have updated configure.ac so that ./configure now finds
2208        the release number by grepping pcre.h.
2209    
2210    26. Added the ability to run the tests under valgrind.
2211    
2212    
2213    Version 6.4 05-Sep-05
2214    ---------------------
2215    
2216     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2217        "--" to be printed when multiple files were scanned, even when none of the
2218        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2219        consider it to be a bug, and have restored the previous behaviour.
2220    
2221     2. A couple of code tidies to get rid of compiler warnings.
2222    
2223     3. The pcretest program used to cheat by referring to symbols in the library
2224        whose names begin with _pcre_. These are internal symbols that are not
2225        really supposed to be visible externally, and in some environments it is
2226        possible to suppress them. The cheating is now confined to including
2227        certain files from the library's source, which is a bit cleaner.
2228    
2229     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2230        file's purpose clearer.
2231    
2232     5. Reorganized pcre_ucp_findchar().
2233    
2234    
2235    Version 6.3 15-Aug-05
2236    ---------------------
2237    
2238     1. The file libpcre.pc.in did not have general read permission in the tarball.
2239    
2240     2. There were some problems when building without C++ support:
2241    
2242        (a) If C++ support was not built, "make install" and "make test" still
2243            tried to test it.
2244    
2245        (b) There were problems when the value of CXX was explicitly set. Some
2246            changes have been made to try to fix these, and ...
2247    
2248        (c) --disable-cpp can now be used to explicitly disable C++ support.
2249    
2250        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2251            backslash in a target when C++ was disabled. This confuses some
2252            versions of "make", apparently. Using an intermediate variable solves
2253            this. (Same for CPP_LOBJ.)
2254    
2255     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2256        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2257        necessary on certain architectures.
2258    
2259     4. Added a setting of -export-symbols-regex to the link command to remove
2260        those symbols that are exported in the C sense, but actually are local
2261        within the library, and not documented. Their names all begin with
2262        "_pcre_". This is not a perfect job, because (a) we have to except some
2263        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2264        available (and never for static libraries). I have made a note to try to
2265        find a way round (a) in the future.
2266    
2267    
2268    Version 6.2 01-Aug-05
2269    ---------------------
2270    
2271     1. There was no test for integer overflow of quantifier values. A construction
2272        such as {1111111111111111} would give undefined results. What is worse, if
2273        a minimum quantifier for a parenthesized subpattern overflowed and became
2274        negative, the calculation of the memory size went wrong. This could have
2275        led to memory overwriting.
2276    
2277     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2278    
2279     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2280        operating environments where this matters.
2281    
2282     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2283        PCRE options from within the C++ wrapper.
2284    
2285     5. Named capturing subpatterns were not being correctly counted when a pattern
2286        was compiled. This caused two problems: (a) If there were more than 100
2287        such subpatterns, the calculation of the memory needed for the whole
2288        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2289        back references of the form \12, where the number was greater than 9, were
2290        not recognized as back references, even though there were sufficient
2291        previous subpatterns.
2292    
2293     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2294        versions of gcc, e.g. 2.95.4.
2295    
2296    
2297    Version 6.1 21-Jun-05
2298    ---------------------
2299    
2300     1. There was one reference to the variable "posix" in pcretest.c that was not
2301        surrounded by "#if !defined NOPOSIX".
2302    
2303     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2304        the cross-check on the old pcre_info() function, for the benefit of the
2305        cut-down version of PCRE that is currently imported into Exim.
2306    
2307     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2308        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2309        patterns that start (?i) but I don't think that matters. The use of (?i) is
2310        just an example; this all applies to the other options as well.
2311    
2312     4. Since libtool seems to echo the compile commands it is issuing, the output
2313        from "make" can be reduced a bit by putting "@" in front of each libtool
2314        compile command.
2315    
2316     5. Patch from the folks at Google for configure.in to be a bit more thorough
2317        in checking for a suitable C++ installation before trying to compile the
2318        C++ stuff. This should fix a reported problem when a compiler was present,
2319        but no suitable headers.
2320    
2321     6. The man pages all had just "PCRE" as their title. I have changed them to
2322        be the relevant file name. I have also arranged that these names are
2323        retained in the file doc/pcre.txt, which is a concatenation in text format
2324        of all the man pages except the little individual ones for each function.
2325    
2326     7. The NON-UNIX-USE file had not been updated for the different set of source
2327        files that come with release 6. I also added a few comments about the C++
2328        wrapper.
2329    
2330    
2331    Version 6.0 07-Jun-05
2332    ---------------------
2333    
2334     1. Some minor internal re-organization to help with my DFA experiments.
2335    
2336     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2337        didn't matter for the library itself when fully configured, but did matter
2338        when compiling without UCP support, or within Exim, where the ucp files are
2339        not imported.
2340    
2341     3. Refactoring of the library code to split up the various functions into
2342        different source modules. The addition of the new DFA matching code (see
2343        below) to a single monolithic source would have made it really too
2344        unwieldy, quite apart from causing all the code to be include in a
2345        statically linked application, when only some functions are used. This is
2346        relevant even without the DFA addition now that patterns can be compiled in
2347        one application and matched in another.
2348    
2349        The downside of splitting up is that there have to be some external
2350        functions and data tables that are used internally in different modules of
2351        the library but which are not part of the API. These have all had their
2352        names changed to start with "_pcre_" so that they are unlikely to clash
2353        with other external names.
2354    
2355     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2356        a different (DFA) algorithm. Although it is slower than the original
2357        function, it does have some advantages for certain types of matching
2358        problem.
2359    
2360     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2361        including restarting after a partial match.
2362    
2363     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2364        defined when compiling for Windows was sent to me. I have put it into the
2365        code, though I have no means of testing or verifying it.
2366    
2367     7. Added the pcre_refcount() auxiliary function.
2368    
2369     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2370        match before or at the first newline in the subject string. In pcretest,
2371        the /f option on a pattern can be used to set this.
2372    
2373     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2374        would behave wrongly. This has been present in PCRE since release 4.0.
2375    
2376    10. A number of changes to the pcregrep command:
2377    
2378        (a) Refactored how -x works; insert ^(...)$ instead of setting
2379            PCRE_ANCHORED and checking the length, in preparation for adding
2380            something similar for -w.
2381    
2382        (b) Added the -w (match as a word) option.
2383    
2384        (c) Refactored the way lines are read and buffered so as to have more
2385            than one at a time available.
2386    
2387        (d) Implemented a pcregrep test script.
2388    
2389        (e) Added the -M (multiline match) option. This allows patterns to match
2390            over several lines of the subject. The buffering ensures that at least
2391            8K, or the rest of the document (whichever is the shorter) is available
2392            for matching (and similarly the previous 8K for lookbehind assertions).
2393    
2394        (f) Changed the --help output so that it now says
2395    
2396              -w, --word-regex(p)
2397    
2398            instead of two lines, one with "regex" and the other with "regexp"
2399            because that confused at least one person since the short forms are the
2400            same. (This required a bit of code, as the output is generated
2401            automatically from a table. It wasn't just a text change.)
2402    
2403        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2404            option but starts with a hyphen. Could be a pattern or a path name
2405            starting with a hyphen, for instance.
2406    
2407        (h) "-" can be given as a file name to represent stdin.
2408    
2409        (i) When file names are being printed, "(standard input)" is used for
2410            the standard input, for compatibility with GNU grep. Previously
2411            "<stdin>" was used.
2412    
2413        (j) The option --label=xxx can be used to supply a name to be used for
2414            stdin when file names are being printed. There is no short form.
2415    
2416        (k) Re-factored the options decoding logic because we are going to add
2417            two more options that take data. Such options can now be given in four
2418            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2419    
2420        (l) Added the -A, -B, and -C options for requesting that lines of context
2421            around matches be printed.
2422    
2423        (m) Added the -L option to print the names of files that do not contain
2424            any matching lines, that is, the complement of -l.
2425    
2426        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2427            continue to scan other files.
2428    
2429        (o) The -s option was incorrectly implemented. For compatibility with other
2430            greps, it now suppresses the error message for a non-existent or non-
2431            accessible file (but not the return code). There is a new option called
2432            -q that suppresses the output of matching lines, which was what -s was
2433            previously doing.
2434    
2435        (p) Added --include and --exclude options to specify files for inclusion
2436            and exclusion when recursing.
2437    
2438    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2439        Hopefully, it now does.
2440    
2441    12. Missing cast in pcre_study().
2442    
2443    13. Added an "uninstall" target to the makefile.
2444    
2445    14. Replaced "extern" in the function prototypes in Makefile.in with
2446        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2447        world, but is set differently for Windows.
2448    
2449    15. Added a second compiling function called pcre_compile2(). The only
2450        difference is that it has an extra argument, which is a pointer to an
2451        integer error code. When there is a compile-time failure, this is set
2452        non-zero, in addition to the error test pointer being set to point to an
2453        error message. The new argument may be NULL if no error number is required
2454        (but then you may as well call pcre_compile(), which is now just a
2455        wrapper). This facility is provided because some applications need a
2456        numeric error indication, but it has also enabled me to tidy up the way
2457        compile-time errors are handled in the POSIX wrapper.
2458    
2459    16. Added VPATH=.libs to the makefile; this should help when building with one
2460        prefix path and installing with another. (Or so I'm told by someone who
2461        knows more about this stuff than I do.)
2462    
2463    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2464        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2465        match everything, including newlines. This is not POSIX-compatible, but
2466        somebody wanted the feature. From pcretest it can be activated by using
2467        both the P and the s flags.
2468    
2469    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2470    
2471    19. libpcre.pc was being incorrectly installed as executable.
2472    
2473    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2474        it now also looks for '\r' so that it will work unmodified on Windows.
2475    
2476    21. Added Google's contributed C++ wrapper to the distribution.
2477    
2478    22. Added some untidy missing memory free() calls in pcretest, to keep
2479        Electric Fence happy when testing.
2480    
2481    
2482    
2483    Version 5.0 13-Sep-04
2484    ---------------------
2485    
2486     1. Internal change: literal characters are no longer packed up into items
2487        containing multiple characters in a single byte-string. Each character
2488        is now matched using a separate opcode. However, there may be more than one
2489        byte in the character in UTF-8 mode.
2490    
2491     2. The pcre_callout_block structure has two new fields: pattern_position and
2492        next_item_length. These contain the offset in the pattern to the next match
2493        item, and its length, respectively.
2494    
2495     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2496        insertion of callouts before each pattern item. Added the /C option to
2497        pcretest to make use of this.
2498    
2499     4. On the advice of a Windows user, the lines
2500    
2501          #if defined(_WIN32) || defined(WIN32)
2502          _setmode( _fileno( stdout ), 0x8000 );
2503          #endif  /* defined(_WIN32) || defined(WIN32) */
2504    
2505        have been added to the source of pcretest. This apparently does useful
2506        magic in relation to line terminators.
2507    
2508     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2509        for the benefit of those environments where the "b" makes a difference.
2510    
2511     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2512        to know about it. I have put a hack into configure.in that adds in code
2513        to set GCC=yes if CC=icc. This seems to end up at a point in the
2514        generated configure script that is early enough to affect the setting of
2515        compiler options, which is what is needed, but I have no means of testing
2516        whether it really works. (The user who reported this had patched the
2517        generated configure script, which of course I cannot do.)
2518    
2519        LATER: After change 22 below (new libtool files), the configure script
2520        seems to know about icc (and also ecc). Therefore, I have commented out
2521        this hack in configure.in.
2522    
2523     7. Added support for pkg-config (2 patches were sent in).
2524    
2525     8. Negated POSIX character classes that used a combination of internal tables
2526        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2527        [[:^ascii]]. Typically, they would match almost any characters. The other
2528        POSIX classes were not broken in this way.
2529    
2530     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2531        to find the match, as PCRE was deluded into thinking that the match had to
2532        start at the start point or following a newline. The same bug applied to
2533        patterns with negative forward assertions or any backward assertions
2534        preceding ".*" at the start, unless the pattern required a fixed first
2535        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2536    
2537    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2538        starting at the last subject character, bytes beyond the end of the subject
2539        string were read.
2540    
2541    11. Renamed the variable "class" as "classbits" to make life easier for C++
2542        users. (Previously there was a macro definition, but it apparently wasn't
2543        enough.)
2544    
2545    12. Added the new field "tables" to the extra data so that tables can be passed
2546        in at exec time, or the internal tables can be re-selected. This allows
2547        a compiled regex to be saved and re-used at a later time by a different
2548        program that might have everything at different addresses.
2549    
2550    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2551        -R library as well as a -L library.
2552    
2553    14. The debugging options of pcretest (-d on the command line or D on a
2554        pattern) showed incorrect output for anything following an extended class
2555        that contained multibyte characters and which was followed by a quantifier.
2556    
2557    15. Added optional support for general category Unicode character properties
2558        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2559        support. It adds about 90K to the size of the library. The meanings of the
2560        inbuilt class escapes such as \d and \s have NOT been changed.
2561    
2562    16. Updated pcredemo.c to include calls to free() to release the memory for the
2563        compiled pattern.
2564    
2565    17. The generated file chartables.c was being created in the source directory
2566        instead of in the building directory. This caused the build to fail if the
2567        source directory was different from the building directory, and was
2568        read-only.
2569    
2570    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2571        file. No doubt somebody will tell me if they don't make sense... Also added
2572        Dan Mooney's comments about building on OpenVMS.
2573    
2574    19. Added support for partial matching via the PCRE_PARTIAL option for
2575        pcre_exec() and the \P data escape in pcretest.
2576    
2577    20. Extended pcretest with 3 new pattern features:
2578    
2579        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2580              write the compiled pattern to the file whose name is "rest-of-line".
2581              This is a straight binary dump of the data, with the saved pointer to
2582              the character tables forced to be NULL. The study data, if any, is
2583              written too. After writing, pcretest reads a new pattern.
2584    
2585        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2586              compiled pattern from the given file. There must not be any
2587              occurrences of "<" in the file name (pretty unlikely); if there are,
2588              pcretest will instead treat the initial "<" as a pattern delimiter.
2589              After reading in the pattern, pcretest goes on to read data lines as
2590              usual.
2591    
2592        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2593              and 16-bit fields in a compiled pattern, to simulate a pattern that
2594              was compiled on a host of opposite endianness.
2595    
2596    21. The pcre-exec() function can now cope with patterns that were compiled on
2597        hosts of opposite endianness, with this restriction:
2598    
2599          As for any compiled expression that is saved and used later, the tables
2600          pointer field cannot be preserved; the extra_data field in the arguments
2601          to pcre_exec() should be used to pass in a tables address if a value
2602          other than the default internal tables were used at compile time.
2603    
2604    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2605        now diagnosed as an error. Previously, most of the time, a negative number
2606        would have been treated as zero, but if in addition "ovector" was passed as
2607        NULL, a crash could occur.
2608    
2609    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2610        new versions from the libtool 1.5 distribution (the last one is a copy of
2611        a file called libtool.m4). This seems to have fixed the need to patch
2612        "configure" to support Darwin 1.3 (which I used to do). However, I still
2613        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2614        workstation).
2615    
2616    24. Changed the PCRE licence to be the more standard "BSD" licence.
2617    
2618    
2619    Version 4.5 01-Dec-03
2620    ---------------------
2621    
2622     1. There has been some re-arrangement of the code for the match() function so
2623        that it can be compiled in a version that does not call itself recursively.
2624        Instead, it keeps those local variables that need separate instances for
2625        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2626        needs to "recurse". Keeping track of where control must go is done by means
2627        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2628        hide most of the details from the main code, and operates only if
2629        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2630        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2631        operating.
2632    
2633        To make it easier for callers to provide specially tailored get/free
2634        functions for this usage, two new functions, pcre_stack_malloc, and
2635        pcre_stack_free, are used. They are always called in strict stacking order,
2636        and the size of block requested is always the same.
2637    
2638        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2639        PCRE has been compiled to use the stack or the heap for recursion. The
2640        -C option of pcretest uses this to show which version is compiled.
2641    
2642        A new data escape \S, is added to pcretest; it causes the amounts of store
2643        obtained and freed by both kinds of malloc/free at match time to be added
2644        to the output.
2645    
2646     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2647        what's available on my current Linux desktop machine.
2648    
2649     3. When matching a UTF-8 string, the test for a valid string at the start has
2650        been extended. If start_offset is not zero, PCRE now checks that it points
2651        to a byte that is the start of a UTF-8 character. If not, it returns
2652        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2653        this is necessary because there may be backward assertions in the pattern.
2654        When matching the same subject several times, it may save resources to use
2655        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2656    
2657     4. The code for checking the validity of UTF-8 strings has been tightened so
2658        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2659        containing "overlong sequences".
2660    
2661     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2662        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2663        should have been "&", but it just so happened that all the cases this let
2664        through by mistake were picked up later in the function.
2665    
2666     6. I had used a variable called "isblank" - this is a C99 function, causing
2667        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2668    
2669     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2670        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2671        so the version is shown; (c) stop "make test" echoing "./RunTest".
2672    
2673     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2674    
2675     9. The prototype for memmove() for systems that don't have it was using
2676        size_t, but the inclusion of the header that defines size_t was later. I've
2677        moved the #includes for the C headers earlier to avoid this.
2678    
2679    10. Added some adjustments to the code to make it easier to compiler on certain
2680        special systems:
2681    
2682          (a) Some "const" qualifiers were missing.
2683          (b) Added the macro EXPORT before all exported functions; by default this
2684              is defined to be empty.
2685          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2686              that it reads its output file name as an argument instead of writing
2687              to the standard output and assuming this can be redirected.
2688    
2689    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2690        class containing characters with values greater than 255, PCRE compilation
2691        went into a loop.
2692    
2693    12. A recursive reference to a subpattern that was within another subpattern
2694        that had a minimum quantifier of zero caused PCRE to crash. For example,
2695        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2696        recursion. If the recursively-called subpattern itself had a zero repeat,
2697        that was OK.
2698    
2699    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2700        buffer into which it was copied (for escape processing) was still set at
2701        1024, so long lines caused crashes.
2702    
2703    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2704        "internal error: code overflow...". This applied to any character class
2705        that was followed by a possessive quantifier.
2706    
2707    15. Modified the Makefile to add libpcre.la as a prerequisite for
2708        libpcreposix.la because I was told this is needed for a parallel build to
2709        work.
2710    
2711    16. If a pattern that contained .* following optional items at the start was
2712        studied, the wrong optimizing data was generated, leading to matching
2713        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2714        matching string must start with a or b or c. The correct conclusion for
2715        this pattern is that a match can start with any character.
2716    
2717    
2718  Version 4.4 13-Aug-03  Version 4.4 13-Aug-03
2719  ---------------------  ---------------------
2720    

Legend:
Removed from v.71  
changed lines
  Added in v.581

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12