ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 21 by nigel, Sat Feb 24 21:38:37 2007 UTC revision 569 by ph10, Sun Nov 7 16:14:50 2010 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Version 8.11 10-Oct-2010
5    ------------------------
7  Version 1.09 28 Apr-98  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".
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).
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.)
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").
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.]
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().
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.
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.
60    9.  Added --match-limit and --recursion-limit to pcregrep.
62    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
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.
67    12. Added an optional parentheses number to the -o and --only-matching options
68        of pcregrep.
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:
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).
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.)
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.
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.
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.
96    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
97        bad UTF-8 sequence and one that is incomplete.
100    Version 8.10 25-Jun-2010
101    ------------------------
103    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
104        THEN.
106    2.  (*ACCEPT) was not working when inside an atomic group.
108    3.  Inside a character class, \B is treated as a literal by default, but
109        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
110        causes the error). The code is unchanged, but I tidied the documentation.
112    4.  Inside a character class, PCRE always treated \R and \X as literals,
113        whereas Perl faults them if its -w option is set. I have changed PCRE so
114        that it faults them when PCRE_EXTRA is set.
116    5.  Added support for \N, which always matches any character other than
117        newline. (It is the same as "." when PCRE_DOTALL is not set.)
119    6.  When compiling pcregrep with newer versions of gcc which may have
120        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
121        declared with attribute warn_unused_result" were given. Just casting the
122        result to (void) does not stop the warnings; a more elaborate fudge is
123        needed. I've used a macro to implement this.
125    7.  Minor change to pcretest.c to avoid a compiler warning.
127    8.  Added four artifical Unicode properties to help with an option to make
128        \s etc use properties (see next item). The new properties are: Xan
129        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
131    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
132        use Unicode properties. (*UCP) at the start of a pattern can be used to set
133        this option. Modified pcretest to add /W to test this facility. Added
134        REG_UCP to make it available via the POSIX interface.
136    10. Added --line-buffered to pcregrep.
138    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
139        studied, and the match started with a letter with a code point greater than
140        127 whose first byte was different to the first byte of the other case of
141        the letter, the other case of this starting letter was not recognized
142        (#976).
144    12. If a pattern that was studied started with a repeated Unicode property
145        test, for example, \p{Nd}+, there was the theoretical possibility of
146        setting up an incorrect bitmap of starting bytes, but fortunately it could
147        not have actually happened in practice until change 8 above was made (it
148        added property types that matched character-matching opcodes).
150    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
151        possible starting bytes for non-anchored patterns.
153    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
154        \R, and also a number of cases that involve Unicode properties, both
155        explicit and implicit when PCRE_UCP is set.
157    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
158        input, it could crash or give wrong results if characters with values
159        greater than 0xc0 were present in the subject string. (Detail: it assumed
160        UTF-8 input when processing these items.)
162    16. Added a lot of (int) casts to avoid compiler warnings in systems where
163        size_t is 64-bit (#991).
165    17. Added a check for running out of memory when PCRE is compiled with
166        --disable-stack-for-recursion (#990).
168    18. If the last data line in a file for pcretest does not have a newline on
169        the end, a newline was missing in the output.
171    19. The default pcre_chartables.c file recognizes only ASCII characters (values
172        less than 128) in its various bitmaps. However, there is a facility for
173        generating tables according to the current locale when PCRE is compiled. It
174        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
175        characters, are recognized by isspace() and therefore were getting set in
176        these tables, and indeed these tables seem to approximate to ISO 8859. This
177        caused a problem in UTF-8 mode when pcre_study() was used to create a list
178        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
179        which of course cannot start UTF-8 characters. I have changed the code so
180        that only real ASCII characters (less than 128) and the correct starting
181        bytes for UTF-8 encodings are set for characters greater than 127 when in
182        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
183        altogether.)
185    20. Added the /T option to pcretest so as to be able to run tests with non-
186        standard character tables, thus making it possible to include the tests
187        used for 19 above in the standard set of tests.
189    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
190        reference to a subpattern the other side of a comment that contains an
191        opening parenthesis caused either an internal compiling error, or a
192        reference to the wrong subpattern.
195    Version 8.02 19-Mar-2010
196    ------------------------
198    1.  The Unicode data tables have been updated to Unicode 5.2.0.
200    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
201        configured.
203    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
204        original author of that file, following a query about its status.
206    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
207        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
209    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
210        quantifier applied to a forward-referencing subroutine call, could compile
211        incorrect code or give the error "internal error: previously-checked
212        referenced subpattern not found".
214    6.  Both MS Visual Studio and Symbian OS have problems with initializing
215        variables to point to external functions. For these systems, therefore,
216        pcre_malloc etc. are now initialized to local functions that call the
217        relevant global functions.
219    7.  There were two entries missing in the vectors called coptable and poptable
220        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
221        I've fixed the data, and added a kludgy way of testing at compile time that
222        the lengths are correct (equal to the number of opcodes).
224    8.  Following on from 7, I added a similar kludge to check the length of the
225        eint vector in pcreposix.c.
227    9.  Error texts for pcre_compile() are held as one long string to avoid too
228        much relocation at load time. To find a text, the string is searched,
229        counting zeros. There was no check for running off the end of the string,
230        which could happen if a new error number was added without updating the
231        string.
233    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
235    11. \K was not working if it appeared in an atomic group or in a group that
236        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
237        \K is "not well defined" if used in an assertion. PCRE now accepts it if
238        the assertion is positive, but not if it is negative.
240    12. Change 11 fortuitously reduced the size of the stack frame used in the
241        "match()" function of pcre_exec.c by one pointer. Forthcoming
242        implementation of support for (*MARK) will need an extra pointer on the
243        stack; I have reserved it now, so that the stack frame size does not
244        decrease.
246    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
247        item in branch that calls a recursion is a subroutine call - as in the
248        second branch in the above example - was incorrectly given the compile-
249        time error "recursive call could loop indefinitely" because pcre_compile()
250        was not correctly checking the subroutine for matching a non-empty string.
252    14. The checks for overrunning compiling workspace could trigger after an
253        overrun had occurred. This is a "should never occur" error, but it can be
254        triggered by pathological patterns such as hundreds of nested parentheses.
255        The checks now trigger 100 bytes before the end of the workspace.
257    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
260    Version 8.01 19-Jan-2010
261    ------------------------
263    1.  If a pattern contained a conditional subpattern with only one branch (in
264        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
265        computed the wrong minimum data length (which is of course zero for such
266        subpatterns). This could cause incorrect "no match" results.
268    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
269        the pattern is reset in the first branch, pcre_compile() failed with
270        "internal error: code overflow at offset...". This happened only when
271        the reset was to the original external option setting. (An optimization
272        abstracts leading options settings into an external setting, which was the
273        cause of this.)
275    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
276        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
277        assertion pattern did not match (meaning that the assertion was true), it
278        was incorrectly treated as false if the SKIP had been reached during the
279        matching. This also applied to assertions used as conditions.
281    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
282        assertion subpattern, including such a pattern used as a condition,
283        unpredictable results occurred, instead of the error return
286    5.  The C++ GlobalReplace function was not working like Perl for the special
287        situation when an empty string is matched. It now does the fancy magic
288        stuff that is necessary.
290    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
291        removed. (These were left over from very, very early versions of PCRE.)
293    7.  Some cosmetic changes to the code to make life easier when compiling it
294        as part of something else:
296        (a) Change DEBUG to PCRE_DEBUG.
298        (b) In pcre_compile(), rename the member of the "branch_chain" structure
299            called "current" as "current_branch", to prevent a collision with the
300            Linux macro when compiled as a kernel module.
302        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
303            prevent a collision with the Linux macro when compiled as a kernel
304            module.
306    8.  In pcre_compile() there are some checks for integer overflows that used to
307        cast potentially large values to (double). This has been changed to that
308        when building, a check for int64_t is made, and if it is found, it is used
309        instead, thus avoiding the use of floating point arithmetic. (There is no
310        other use of FP in PCRE.) If int64_t is not found, the fallback is to
311        double.
313    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
314        2005 (difference between two addresses compared to an unsigned value).
316    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
317        custom one, because of the following reported problem in Windows:
319          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
320              under Win32.
321          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
322              therefore missing the function definition.
323          - The compiler thus generates a "C" signature for the test function.
324          - The linker fails to find the "C" function.
325          - PCRE fails to configure if asked to do so against libbz2.
327    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
328        messages were output:
330          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
331          rerunning libtoolize, to keep the correct libtool macros in-tree.
332          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
334        I have done both of these things.
336    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
337        most of the time, it *can* run out if it is given a pattern that contains a
338        runaway infinite recursion. I updated the discussion in the pcrestack man
339        page.
341    13. Now that we have gone to the x.xx style of version numbers, the minor
342        version may start with zero. Using 08 or 09 is a bad idea because users
343        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
344        interpreted as invalid octal numbers. I've updated the previous comment in
345        configure.ac, and also added a check that gives an error if 08 or 09 are
346        used.
348    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
349        causing partial matching to fail when the end of the subject matched \W
350        in a UTF-8 pattern where \W was quantified with a minimum of 3.
352    15. There were some discrepancies between the declarations in pcre_internal.h
353        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
354        their definitions. The declarations used "const uschar *" and the
355        definitions used USPTR. Even though USPTR is normally defined as "const
356        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
357        reported that: "This difference in casting confuses some C++ compilers, for
358        example, SunCC recognizes above declarations as different functions and
359        generates broken code for hbpcre." I have changed the declarations to use
360        USPTR.
362    16. GNU libtool is named differently on some systems. The autogen.sh script now
363        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
364        (FreeBSD).
366    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
367        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
368        comment: "Figure out how to create a longlong from a string: strtoll and
369        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
370        instance, but it only takes 2 args instead of 3!"
372    18. A subtle bug concerned with back references has been fixed by a change of
373        specification, with a corresponding code fix. A pattern such as
374        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
375        refers, was giving matches when it shouldn't. For example, xa=xaaa would
376        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
377        same bug. Such groups have to be quantified to be useful, or contained
378        inside another quantified group. (If there's no repetition, the reference
379        can never match.) The problem arises because, having left the group and
380        moved on to the rest of the pattern, a later failure that backtracks into
381        the group uses the captured value from the final iteration of the group
382        rather than the correct earlier one. I have fixed this in PCRE by forcing
383        any group that contains a reference to itself to be an atomic group; that
384        is, there cannot be any backtracking into it once it has completed. This is
385        similar to recursive and subroutine calls.
388    Version 8.00 19-Oct-09
389    ----------------------
391    1.  The table for translating pcre_compile() error codes into POSIX error codes
392        was out-of-date, and there was no check on the pcre_compile() error code
393        being within the table. This could lead to an OK return being given in
394        error.
396    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
397        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
398        in a Windows environment.
400    3.  The pcregrep --count option prints the count for each file even when it is
401        zero, as does GNU grep. However, pcregrep was also printing all files when
402        --files-with-matches was added. Now, when both options are given, it prints
403        counts only for those files that have at least one match. (GNU grep just
404        prints the file name in this circumstance, but including the count seems
405        more useful - otherwise, why use --count?) Also ensured that the
406        combination -clh just lists non-zero counts, with no names.
408    4.  The long form of the pcregrep -F option was incorrectly implemented as
409        --fixed_strings instead of --fixed-strings. This is an incompatible change,
410        but it seems right to fix it, and I didn't think it was worth preserving
411        the old behaviour.
413    5.  The command line items --regex=pattern and --regexp=pattern were not
414        recognized by pcregrep, which required --regex pattern or --regexp pattern
415        (with a space rather than an '='). The man page documented the '=' forms,
416        which are compatible with GNU grep; these now work.
418    6.  No libpcreposix.pc file was created for pkg-config; there was just
419        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
421    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
422        when UCP support is not needed, by modifying the Python script that
423        generates it from Unicode data files. This should not matter if the module
424        is correctly used as a library, but I received one complaint about 50K of
425        unwanted data. My guess is that the person linked everything into his
426        program rather than using a library. Anyway, it does no harm.
428    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
429        was a minimum greater than 1 for a wide character in a possessive
430        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
431        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
432        character. Chaos in the form of incorrect output or a compiling loop could
433        result.
435    9.  The restrictions on what a pattern can contain when partial matching is
436        requested for pcre_exec() have been removed. All patterns can now be
437        partially matched by this function. In addition, if there are at least two
438        slots in the offset vector, the offset of the earliest inspected character
439        for the match and the offset of the end of the subject are set in them when
440        PCRE_ERROR_PARTIAL is returned.
442    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
443        synonymous with PCRE_PARTIAL, for backwards compatibility, and
444        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
445        and may be more useful for multi-segment matching.
447    11. Partial matching with pcre_exec() is now more intuitive. A partial match
448        used to be given if ever the end of the subject was reached; now it is
449        given only if matching could not proceed because another character was
450        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
451        string "Z", which now yields "no match" instead of "partial match". In the
452        case of pcre_dfa_exec(), "no match" is given if every matching path for the
453        final character ended with (*FAIL).
455    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
456        if the pattern had a "must contain" character that was already found in the
457        earlier partial match, unless partial matching was again requested. For
458        example, with the pattern /dog.(body)?/, the "must contain" character is
459        "g". If the first part-match was for the string "dog", restarting with
460        "sbody" failed. This bug has been fixed.
462    13. The string returned by pcre_dfa_exec() after a partial match has been
463        changed so that it starts at the first inspected character rather than the
464        first character of the match. This makes a difference only if the pattern
465        starts with a lookbehind assertion or \b or \B (\K is not supported by
466        pcre_dfa_exec()). It's an incompatible change, but it makes the two
467        matching functions compatible, and I think it's the right thing to do.
469    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
470        so that the demonstration program is easily available in environments where
471        PCRE has not been installed from source.
473    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
474        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
475        library.
477    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
478        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
479        is not the first non-POSIX option to be added. Clearly some people find
480        these options useful.
482    17. If a caller to the POSIX matching function regexec() passes a non-zero
483        value for nmatch with a NULL value for pmatch, the value of
484        nmatch is forced to zero.
486    18. RunGrepTest did not have a test for the availability of the -u option of
487        the diff command, as RunTest does. It now checks in the same way as
488        RunTest, and also checks for the -b option.
490    19. If an odd number of negated classes containing just a single character
491        interposed, within parentheses, between a forward reference to a named
492        subpattern and the definition of the subpattern, compilation crashed with
493        an internal error, complaining that it could not find the referenced
494        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
495        [The bug was that it was starting one character too far in when skipping
496        over the character class, thus treating the ] as data rather than
497        terminating the class. This meant it could skip too much.]
499    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
500        /g option in pcretest when the pattern contains \K, which makes it possible
501        to have an empty string match not at the start, even when the pattern is
502        anchored. Updated pcretest and pcredemo to use this option.
504    21. If the maximum number of capturing subpatterns in a recursion was greater
505        than the maximum at the outer level, the higher number was returned, but
506        with unset values at the outer level. The correct (outer level) value is
507        now given.
509    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
510        PCRE did not set those parentheses (unlike Perl). I have now found a way to
511        make it do so. The string so far is captured, making this feature
512        compatible with Perl.
514    23. The tests have been re-organized, adding tests 11 and 12, to make it
515        possible to check the Perl 5.10 features against Perl 5.10.
517    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
518        pattern matches a fixed length string. PCRE did not allow this; now it
519        does. Neither allows recursion.
521    25. I finally figured out how to implement a request to provide the minimum
522        length of subject string that was needed in order to match a given pattern.
523        (It was back references and recursion that I had previously got hung up
524        on.) This code has now been added to pcre_study(); it finds a lower bound
525        to the length of subject needed. It is not necessarily the greatest lower
526        bound, but using it to avoid searching strings that are too short does give
527        some useful speed-ups. The value is available to calling programs via
528        pcre_fullinfo().
530    26. While implementing 25, I discovered to my embarrassment that pcretest had
531        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
532        study optimizations had never been tested with that matching function.
533        Oops. What is worse, even when it was passed study data, there was a bug in
534        pcre_dfa_exec() that meant it never actually used it. Double oops. There
535        were also very few tests of studied patterns with pcre_dfa_exec().
537    27. If (?| is used to create subpatterns with duplicate numbers, they are now
538        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
539        on the other side of the coin, they are no longer allowed to have different
540        names, because these cannot be distinguished in PCRE, and this has caused
541        confusion. (This is a difference from Perl.)
543    28. When duplicate subpattern names are present (necessarily with different
544        numbers, as required by 27 above), and a test is made by name in a
545        conditional pattern, either for a subpattern having been matched, or for
546        recursion in such a pattern, all the associated numbered subpatterns are
547        tested, and the overall condition is true if the condition is true for any
548        one of them. This is the way Perl works, and is also more like the way
549        testing by number works.
552    Version 7.9 11-Apr-09
553    ---------------------
555    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
556        (pcretest), all targets were linked against these libraries. This included
557        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
558        libraries. This caused unwanted dependencies to be created. This problem
559        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
560        pcretest is linked with readline.
562    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
563        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
564        moved outside it again, because FALSE and TRUE are already defined in AIX,
565        but BOOL is not.
567    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
568        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
570    4.  The pcregrep documentation said spaces were inserted as well as colons (or
571        hyphens) following file names and line numbers when outputting matching
572        lines. This is not true; no spaces are inserted. I have also clarified the
573        wording for the --colour (or --color) option.
575    5.  In pcregrep, when --colour was used with -o, the list of matching strings
576        was not coloured; this is different to GNU grep, so I have changed it to be
577        the same.
579    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
580        each matching line was coloured. Now it goes on to look for further matches
581        of any of the test patterns, which is the same behaviour as GNU grep.
583    7.  A pattern that could match an empty string could cause pcregrep to loop; it
584        doesn't make sense to accept an empty string match in pcregrep, so I have
585        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
586        seems to be how GNU grep behaves.
588    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
589        start or after a newline", because the conditional assertion was not being
590        correctly handled. The rule now is that both the assertion and what follows
591        in the first alternative must satisfy the test.
593    9.  If auto-callout was enabled in a pattern with a conditional group whose
594        condition was an assertion, PCRE could crash during matching, both with
595        pcre_exec() and pcre_dfa_exec().
597    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
598        used for matching.
600    11. Unicode property support in character classes was not working for
601        characters (bytes) greater than 127 when not in UTF-8 mode.
603    12. Added the -M command line option to pcretest.
605    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
607    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
609    16. Added comments and documentation about mis-use of no_arg in the C++
610        wrapper.
612    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
613        from Martin Jerabek that uses macro names for all relevant character and
614        string constants.
616    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
617        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
618        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
619        these, but not everybody uses configure.
621    19. A conditional group that had only one branch was not being correctly
622        recognized as an item that could match an empty string. This meant that an
623        enclosing group might also not be so recognized, causing infinite looping
624        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
625        with the subject "ab", where knowledge that the repeated group can match
626        nothing is needed in order to break the loop.
628    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
629        exec(), but without supplying a callout function, matching went wrong.
631    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
632        leak if the size of the offset vector was greater than 30. When the vector
633        is smaller, the saved offsets during recursion go onto a local stack
634        vector, but for larger vectors malloc() is used. It was failing to free
635        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
636        error, in fact).
638    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
639        heapframe that is used only when UTF-8 support is enabled. This caused no
640        problem, but was untidy.
642    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
643        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
644        included within another project.
646    24. Steven Van Ingelgem's patches to add more options to the CMake support,
647        slightly modified by me:
649          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
650              not building pcregrep.
652          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
653              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
655    25. Forward references, both numeric and by name, in patterns that made use of
656        duplicate group numbers, could behave incorrectly or give incorrect errors,
657        because when scanning forward to find the reference group, PCRE was not
658        taking into account the duplicate group numbers. A pattern such as
659        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
661    26. Changed a few more instances of "const unsigned char *" to USPTR, making
662        the feature of a custom pointer more persuasive (as requested by a user).
664    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
665        pcretest.c, inside #ifndefs, because it seems they are sometimes already
666        pre-defined.
668    28. Added support for (*UTF8) at the start of a pattern.
670    29. Arrange for flags added by the "release type" setting in CMake to be shown
671        in the configuration summary.
674    Version 7.8 05-Sep-08
675    ---------------------
677    1.  Replaced UCP searching code with optimized version as implemented for Ad
678        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
679        stage table and inline lookup instead of a function, giving speed ups of 2
680        to 5 times on some simple patterns that I tested. Permission was given to
681        distribute the MultiStage2.py script that generates the tables (it's not in
682        the tarball, but is in the Subversion repository).
684    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
685        scripts.
687    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
688        a group with a zero qualifier. The result of the study could be incorrect,
689        or the function might crash, depending on the pattern.
691    4.  Caseless matching was not working for non-ASCII characters in back
692        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
693        It now works when Unicode Property Support is available.
695    5.  In pcretest, an escape such as \x{de} in the data was always generating
696        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
697        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
698        truncation.
700    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
702    7.  Added two (int) casts to pcregrep when printing the difference of two
703        pointers, in case they are 64-bit values.
705    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
706        test 2 if it fails.
708    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
709        and a #define of that name to empty if it is not externally set. This is to
710        allow users of MSVC to set it if necessary.
712    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
713        the convenience functions in the pcre_get.c source file.
715    11. An option change at the start of a pattern that had top-level alternatives
716        could cause overwriting and/or a crash. This command provoked a crash in
717        some environments:
719          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
721        This potential security problem was recorded as CVE-2008-2371.
723    12. For a pattern where the match had to start at the beginning or immediately
724        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
725        pcre_dfa_exec() could read past the end of the passed subject if there was
726        no match. To help with detecting such bugs (e.g. with valgrind), I modified
727        pcretest so that it places the subject at the end of its malloc-ed buffer.
729    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
730        exec() might read past the end of the data buffer in UTF-8 mode.
732    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
733        the data contained the byte 0x85 as part of a UTF-8 character within its
734        first line. This applied both to normal and DFA matching.
736    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
737        /^[^d]*?$/8 failed to match "abc".
739    16. Added a missing copyright notice to pcrecpp_internal.h.
741    17. Make it more clear in the documentation that values returned from
742        pcre_exec() in ovector are byte offsets, not character counts.
744    18. Tidied a few places to stop certain compilers from issuing warnings.
746    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
747        supplied by Stefan Weber. I made a further small update for 7.8 because
748        there is a change of source arrangements: the pcre_searchfuncs.c module is
749        replaced by pcre_ucd.c.
752    Version 7.7 07-May-08
753    ---------------------
755    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
756        a string to a long long, pretend we don't even have a long long." This is
757        done by checking for the strtoq, strtoll, and _strtoi64 functions.
759    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
760        pre-7.6 versions, which defined a global no_arg variable instead of putting
761        it in the RE class. (See also #8 below.)
763    3.  Remove a line of dead code, identified by coverity and reported by Nuno
764        Lopes.
766    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
768        (1) The include/exclude patterns were being applied to the whole pathnames
769            of files, instead of just to the final components.
771        (2) If there was more than one level of directory, the subdirectories were
772            skipped unless they satisfied the include/exclude conditions. This is
773            inconsistent with GNU grep (and could even be seen as contrary to the
774            pcregrep specification - which I improved to make it absolutely clear).
775            The action now is always to scan all levels of directory, and just
776            apply the include/exclude patterns to regular files.
778    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
779        --exclude_dir in the tests to avoid scanning .svn directories.
781    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
782        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
783        doesn't support NULs in patterns.
785    7.  Added some missing "const"s to declarations of static tables in
786        pcre_compile.c and pcre_dfa_exec.c.
788    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
789        caused by fix #2  above. (Subsequently also a second patch to fix the
790        first patch. And a third patch - this was a messy problem.)
792    9.  Applied Craig's patch to remove the use of push_back().
794    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
795        matching function regexec().
797    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
798        which, however, unlike Perl's \g{...}, are subroutine calls, not back
799        references. PCRE supports relative numbers with this syntax (I don't think
800        Oniguruma does).
802    12. Previously, a group with a zero repeat such as (...){0} was completely
803        omitted from the compiled regex. However, this means that if the group
804        was called as a subroutine from elsewhere in the pattern, things went wrong
805        (an internal error was given). Such groups are now left in the compiled
806        pattern, with a new opcode that causes them to be skipped at execution
807        time.
809    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
810        to the way PCRE behaves:
812        (a) A lone ] character is dis-allowed (Perl treats it as data).
814        (b) A back reference to an unmatched subpattern matches an empty string
815            (Perl fails the current match path).
817        (c) A data ] in a character class must be notated as \] because if the
818            first data character in a class is ], it defines an empty class. (In
819            Perl it is not possible to have an empty class.) The empty class []
820            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
821            The negative empty class [^] matches any one character, independently
822            of the DOTALL setting.
824    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
825        non-existent subpattern following a character class starting with ']' and
826        containing () gave an internal compiling error instead of "reference to
827        non-existent subpattern". Fortunately, when the pattern did exist, the
828        compiled code was correct. (When scanning forwards to check for the
829        existencd of the subpattern, it was treating the data ']' as terminating
830        the class, so got the count wrong. When actually compiling, the reference
831        was subsequently set up correctly.)
833    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
834        it was being rejected as not supported by pcre_dfa_exec(), even though
835        other assertions are supported. I have made pcre_dfa_exec() support
836        (*FAIL).
838    16. The implementation of 13c above involved the invention of a new opcode,
839        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
840        cannot be changed at match time, I realized I could make a small
841        improvement to matching performance by compiling OP_ALLANY instead of
842        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
843        on the OP_ANY path.
845    17. Compiling pcretest on Windows with readline support failed without the
846        following two fixes: (1) Make the unistd.h include conditional on
847        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
849    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
850        ncurses library to be included for pcretest when ReadLine support is
851        requested, but also to allow for it to be overridden. This patch came from
852        Daniel Bergström.
854    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
855        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
856        any errors with the current Unicode tables. Thanks to Peter Kankowski for
857        spotting this.
860    Version 7.6 28-Jan-08
861    ---------------------
863    1.  A character class containing a very large number of characters with
864        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
865        overflow.
867    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
868        HAVE_LONG_LONG is not defined.
870    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
871        bring them up to date and include new features. This patch includes:
873        - Fixed PH's badly added libz and libbz2 support.
874        - Fixed a problem with static linking.
875        - Added pcredemo. [But later removed - see 7 below.]
876        - Fixed dftables problem and added an option.
877        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
878            HAVE_LONG_LONG.
879        - Added readline support for pcretest.
880        - Added an listing of the option settings after cmake has run.
882    4.  A user submitted a patch to Makefile that makes it easy to create
883        "pcre.dll" under mingw when using Configure/Make. I added stuff to
884        Makefile.am that cause it to include this special target, without
885        affecting anything else. Note that the same mingw target plus all
886        the other distribution libraries and programs are now supported
887        when configuring with CMake (see 6 below) instead of with
888        Configure/Make.
890    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
891        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
892        exported in the Windows port". It has not yet been confirmed that the patch
893        solves the problem, but it does no harm.
895    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
896        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
897        with CMake, and also correct the comment about stack recursion.
899    7.  Remove the automatic building of pcredemo from the ./configure system and
900        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
901        of a program that users should build themselves after PCRE is installed, so
902        building it automatically is not really right. What is more, it gave
903        trouble in some build environments.
905    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
908    Version 7.5 10-Jan-08
909    ---------------------
911    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
912        values in parens when parsing an RE using the C++ wrapper."
914    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
915        Characters greater than 255 were excluded from the class instead of being
916        included.
918    3.  The same bug as (2) above applied to negated POSIX classes such as
919        [:^space:].
921    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
922        defined or documented. It seems to have been a typo for PCRE_STATIC, so
923        I have changed it.
925    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
926        first named subpattern) and a construct such as (?&a) would reference the
927        first named subpattern whose name started with "a" (in other words, the
928        length check was missing). Both these problems are fixed. "Subpattern name
929        expected" is now given for (?&) (a zero-length name), and this patch also
930        makes it give the same error for \k'' (previously it complained that that
931        was a reference to a non-existent subpattern).
933    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
934        this is right because (?- can be followed by option settings as well as by
935        digits. I have, however, made the messages clearer.
937    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
938        than the number used in the conditional) now cause a compile-time error.
939        This is actually not compatible with Perl, which accepts such patterns, but
940        treats the conditional as always being FALSE (as PCRE used to), but it
941        seems to me that giving a diagnostic is better.
943    8.  Change "alphameric" to the more common word "alphanumeric" in comments
944        and messages.
946    9.  Fix two occurrences of "backslash" in comments that should have been
947        "backspace".
949    10. Remove two redundant lines of code that can never be obeyed (their function
950        was moved elsewhere).
952    11. The program that makes PCRE's Unicode character property table had a bug
953        which caused it to generate incorrect table entries for sequences of
954        characters that have the same character type, but are in different scripts.
955        It amalgamated them into a single range, with the script of the first of
956        them. In other words, some characters were in the wrong script. There were
957        thirteen such cases, affecting characters in the following ranges:
959          U+002b0 - U+002c1
960          U+0060c - U+0060d
961          U+0061e - U+00612
962          U+0064b - U+0065e
963          U+0074d - U+0076d
964          U+01800 - U+01805
965          U+01d00 - U+01d77
966          U+01d9b - U+01dbf
967          U+0200b - U+0200f
968          U+030fc - U+030fe
969          U+03260 - U+0327f
970          U+0fb46 - U+0fbb1
971          U+10450 - U+1049d
973    12. The -o option (show only the matching part of a line) for pcregrep was not
974        compatible with GNU grep in that, if there was more than one match in a
975        line, it showed only the first of them. It now behaves in the same way as
976        GNU grep.
978    13. If the -o and -v options were combined for pcregrep, it printed a blank
979        line for every non-matching line. GNU grep prints nothing, and pcregrep now
980        does the same. The return code can be used to tell if there were any
981        non-matching lines.
983    14. Added --file-offsets and --line-offsets to pcregrep.
985    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
986        infinitely looping recursion. The bug was that positive lookaheads were not
987        being skipped when checking for a possible empty match (negative lookaheads
988        and both kinds of lookbehind were skipped).
990    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
991        inclusion of <windows.h> to before rather than after the definition of
992        INVALID_FILE_ATTRIBUTES (patch from David Byron).
994    17. Specifying a possessive quantifier with a specific limit for a Unicode
995        character property caused pcre_compile() to compile bad code, which led at
996        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
997        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
998        caused the error; without that there was no problem.
1000    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1002    19. Added --enable-pcretest-libreadline.
1004    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1005        RE::GlobalReplace(). As a result, the number of replacements returned was
1006        double what it should be. I removed one of the increments, but Craig sent a
1007        later patch that removed the other one (the right fix) and added unit tests
1008        that check the return values (which was not done before).
1010    21. Several CMake things:
1012        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1013            the names libpcre and libpcreposix, not just pcre and pcreposix.
1015        (2) The above change means that pcretest and pcregrep are now correctly
1016            linked with the newly-built libraries, not previously installed ones.
1020    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1021        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1022        UTF-8 newline character). The key issue is that the pattern starts .*;
1023        this means that the match must be either at the beginning, or after a
1024        newline. The bug was in the code for advancing after a failed match and
1025        checking that the new position followed a newline. It was not taking
1026        account of UTF-8 characters correctly.
1028    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1029        character classes. PCRE was not treating the sequence [:...:] as a
1030        character class unless the ... were all letters. Perl, however, seems to
1031        allow any characters between [: and :], though of course it rejects as
1032        unknown any "names" that contain non-letters, because all the known class
1033        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1034        for example, whereas PCRE did not - it did not recognize a POSIX character
1035        class. This seemed a bit dangerous, so the code has been changed to be
1036        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1037        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1038        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1039        Perl does, and where it didn't before.
1041    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1042        Windows environments %n is disabled by default.
1045    Version 7.4 21-Sep-07
1046    ---------------------
1048    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1049        means that a class such as [\s] counted as "explicit reference to CR or
1050        LF". That isn't really right - the whole point of the change was to try to
1051        help when there was an actual mention of one of the two characters. So now
1052        the change happens only if \r or \n (or a literal CR or LF) character is
1053        encountered.
1055    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1056        of both had grown to the point where there were only 3 bits left.
1057        Fortunately, there was spare space in the data structure, and so I have
1058        moved the internal flags into a new 16-bit field to free up more option
1059        bits.
1061    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1062        but did not set the internal JCHANGED flag - either of these is enough to
1063        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1064        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1065        start sets both bits.
1067    4.  Added options (at build time, compile time, exec time) to change \R from
1068        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1070    5.  doc/pcresyntax.html was missing from the distribution.
1072    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1073        compatibility, even though it is no longer used.
1075    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1076        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1077        windows.h file is present (where different names are used). [This was
1078        reversed later after testing - see 16 below.]
1080    8.  Changed all #include <config.h> to #include "config.h". There were also
1081        some further <pcre.h> cases that I changed to "pcre.h".
1083    9.  When pcregrep was used with the --colour option, it missed the line ending
1084        sequence off the lines that it output.
1086    10. It was pointed out to me that arrays of string pointers cause lots of
1087        relocations when a shared library is dynamically loaded. A technique of
1088        using a single long string with a table of offsets can drastically reduce
1089        these. I have refactored PCRE in four places to do this. The result is
1090        dramatic:
1092          Originally:                          290
1093          After changing UCP table:            187
1094          After changing error message table:   43
1095          After changing table of "verbs"       36
1096          After changing table of Posix names   22
1098        Thanks to the folks working on Gregex for glib for this insight.
1100    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1101        unicode-properties was also set.
1103    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1105    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1106        checked only for CRLF.
1108    14. Added casts to pcretest.c to avoid compiler warnings.
1110    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1112    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1113        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1114        entirely. This removes changes made in 7 above.
1116    17. The CMake files have been updated, and there is now more information about
1117        building with CMake in the NON-UNIX-USE document.
1120    Version 7.3 28-Aug-07
1121    ---------------------
1123     1. In the rejigging of the build system that eventually resulted in 7.1, the
1124        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1125        brackets there is not right, since it causes compilers to look for an
1126        installed pcre.h, not the version that is in the source that is being
1127        compiled (which of course may be different). I have changed it back to:
1129          #include "pcre.h"
1131        I have a vague recollection that the change was concerned with compiling in
1132        different directories, but in the new build system, that is taken care of
1133        by the VPATH setting the Makefile.
1135     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1136        when the subject happened to end in the byte 0x85 (e.g. if the last
1137        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1138        characters but of course it shouldn't be taken as a newline when it is part
1139        of another character. The bug was that, for an unlimited repeat of . in
1140        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1141        characters when looking for a newline.
1143     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1145     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1146        in debug output.
1148     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1149        long printing in the pcrecpp unittest when running under MinGW.
1151     6. ESC_K was left out of the EBCDIC table.
1153     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1154        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1155        limit also applies to "virtual nesting" when a pattern is recursive, and in
1156        this case 1000 isn't so big. I have been able to remove this limit at the
1157        expense of backing off one optimization in certain circumstances. Normally,
1158        when pcre_exec() would call its internal match() function recursively and
1159        immediately return the result unconditionally, it uses a "tail recursion"
1160        feature to save stack. However, when a subpattern that can match an empty
1161        string has an unlimited repetition quantifier, it no longer makes this
1162        optimization. That gives it a stack frame in which to save the data for
1163        checking that an empty string has been matched. Previously this was taken
1164        from the 1000-entry workspace that had been reserved. So now there is no
1165        explicit limit, but more stack is used.
1167     8. Applied Daniel's patches to solve problems with the import/export magic
1168        syntax that is required for Windows, and which was going wrong for the
1169        pcreposix and pcrecpp parts of the library. These were overlooked when this
1170        problem was solved for the main library.
1172     9. There were some crude static tests to avoid integer overflow when computing
1173        the size of patterns that contain repeated groups with explicit upper
1174        limits. As the maximum quantifier is 65535, the maximum group length was
1175        set at 30,000 so that the product of these two numbers did not overflow a
1176        32-bit integer. However, it turns out that people want to use groups that
1177        are longer than 30,000 bytes (though not repeat them that many times).
1178        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1179        made it possible to implement the integer overflow checks in a much more
1180        dynamic way, which I have now done. The artificial limitation on group
1181        length has been removed - we now have only the limit on the total length of
1182        the compiled pattern, which depends on the LINK_SIZE setting.
1184    10. Fixed a bug in the documentation for get/copy named substring when
1185        duplicate names are permitted. If none of the named substrings are set, the
1186        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1187        empty string.
1189    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1190        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1191        because the ] is interpreted as the first data character and the
1192        terminating ] is not found. PCRE has been made compatible with Perl in this
1193        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1194        cause memory overwriting.
1196    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1197        string has been matched (to stop an infinite loop). It was not recognizing
1198        a conditional subpattern that could match an empty string if that
1199        subpattern was within another subpattern. For example, it looped when
1200        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1201        condition was not nested. This bug has been fixed.
1203    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1204        past the start of the subject in the presence of bytes with the top bit
1205        set, for example "\x8aBCD".
1207    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1208        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1210    14. Optimized (?!) to (*FAIL).
1212    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1213        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1214        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1215        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1216        does: it's just the validity check that is more restrictive.
1218    16. Inserted checks for integer overflows during escape sequence (backslash)
1219        processing, and also fixed erroneous offset values for syntax errors during
1220        backslash processing.
1222    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1223        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1225    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1226        caused an overrun.
1228    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1229        something other than just ASCII characters) inside a group that had an
1230        unlimited repeat caused a loop at compile time (while checking to see
1231        whether the group could match an empty string).
1233    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1234        [\P{Any}] did so. (Error in the code for printing property names.)
1236    21. An orphan \E inside a character class could cause a crash.
1238    22. A repeated capturing bracket such as (A)? could cause a wild memory
1239        reference during compilation.
1241    23. There are several functions in pcre_compile() that scan along a compiled
1242        expression for various reasons (e.g. to see if it's fixed length for look
1243        behind). There were bugs in these functions when a repeated \p or \P was
1244        present in the pattern. These operators have additional parameters compared
1245        with \d, etc, and these were not being taken into account when moving along
1246        the compiled data. Specifically:
1248        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1249            length.
1251        (b) An item such as \pL+ within a repeated group could cause crashes or
1252            loops.
1254        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1255            "reference to non-existent subpattern" error.
1257        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1259    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1260        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1262    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1264    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1265        character were causing crashes (broken optimization).
1267    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1268        \p or \P) caused a compile-time loop.
1270    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1271        break. For example, the unstudied pattern [\r\n]A does not match the string
1272        "\r\nA" because change 7.0/46 below moves the current point on by two
1273        characters after failing to match at the start. However, the pattern \nA
1274        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1275        the same is true. There doesn't seem any very clean way out of this, but
1276        what I have chosen to do makes the common cases work: PCRE now takes note
1277        of whether there can be an explicit match for \r or \n anywhere in the
1278        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1279        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1280        pattern has explicit CR or LF references.
1282    29. Added (*CR) etc for changing newline setting at start of pattern.
1285    Version 7.2 19-Jun-07
1286    ---------------------
1288     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1289        which is apparently normally available under Windows.
1291     2. Re-jig the pcregrep tests with different newline settings in an attempt
1292        to make them independent of the local environment's newline setting.
1294     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1296     4. Some of the "internals" tests were previously cut out when the link size
1297        was not 2, because the output contained actual offsets. The recent new
1298        "Z" feature of pcretest means that these can be cut out, making the tests
1299        usable with all link sizes.
1301     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1302        stack recursion. This gives a massive performance boost under BSD, but just
1303        a small improvement under Linux. However, it saves one field in the frame
1304        in all cases.
1306     6. Added more features from the forthcoming Perl 5.10:
1308        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1309            recursion call. It refers to the nth most recently opened parentheses.
1311        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1312            to be opened parentheses.
1314        (c) Conditions that refer to capturing parentheses can be specified
1315            relatively, for example, (?(-2)... or (?(+3)...
1317        (d) \K resets the start of the current match so that everything before
1318            is not part of it.
1320        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1322        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1323            reference syntax.
1325        (g) (?| introduces a group in which the numbering of parentheses in each
1326            alternative starts with the same number.
1328        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1330     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1333     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1334        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1335        for detecting groups that can match an empty string.
1337     9. A pattern with a very large number of alternatives (more than several
1338        hundred) was running out of internal workspace during the pre-compile
1339        phase, where pcre_compile() figures out how much memory will be needed. A
1340        bit of new cunning has reduced the workspace needed for groups with
1341        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1342        workspace instead of running out of the 4096 that are available.
1344    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1346    11. Applied patch from Google to remove an optimization that didn't quite work.
1347        The report of the bug said:
1349          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1350          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1351          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1353    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1354        it matched the wrong number of bytes.
1357    Version 7.1 24-Apr-07
1358    ---------------------
1360     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1361        that is more "standard", making use of automake and other Autotools. There
1362        is some re-arrangement of the files and adjustment of comments consequent
1363        on this.
1365     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1366        for recursive directory scanning broke on some systems because the files
1367        are not scanned in any specific order and on different systems the order
1368        was different. A call to "sort" has been inserted into RunGrepTest for the
1369        approprate test as a short-term fix. In the longer term there may be an
1370        alternative.
1372     3. I had an email from Eric Raymond about problems translating some of PCRE's
1373        man pages to HTML (despite the fact that I distribute HTML pages, some
1374        people do their own conversions for various reasons). The problems
1375        concerned the use of low-level troff macros .br and .in. I have therefore
1376        removed all such uses from the man pages (some were redundant, some could
1377        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1378        HTML has been updated to handle .nf/.fi and to complain if it encounters
1379        .br or .in.
1381     4. Updated comments in configure.ac that get placed in config.h.in and also
1382        arranged for config.h to be included in the distribution, with the name
1383        config.h.generic, for the benefit of those who have to compile without
1384        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1386     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1387        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1388        makevp.bat for the current PCRE, using the additional files
1389        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1391     6. A Windows user reported a minor discrepancy with test 2, which turned out
1392        to be caused by a trailing space on an input line that had got lost in his
1393        copy. The trailing space was an accident, so I've just removed it.
1395     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1396        that is needed.
1398     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1399        as "const" (a) because they are and (b) because it helps the PHP
1400        maintainers who have recently made a script to detect big data structures
1401        in the php code that should be moved to the .rodata section. I remembered
1402        to update Builducptable as well, so it won't revert if ucptable.h is ever
1403        re-created.
1405     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1406        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1407        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1408        support is not required. This saves 1.5-2K of code, which is important in
1409        some applications.
1411        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1412        so as not to refer to the tables, even though these functions will never be
1413        called when UTF-8 support is disabled. Otherwise there are problems with a
1414        shared library.
1416    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1418        (a) It was defining its arguments as char * instead of void *.
1420        (b) It was assuming that all moves were upwards in memory; this was true
1421            a long time ago when I wrote it, but is no longer the case.
1423        The emulated memove() is provided for those environments that have neither
1424        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1425        is clearly not the case, as these two bugs were recently reported.
1427    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1428        and Detrail to create the HTML documentation, the .txt form of the man
1429        pages, and it removes trailing spaces from listed files. It also creates
1430        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1431        case, it wraps all the #defines with #ifndefs. This script should be run
1432        before "make dist".
1434    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1435        with Unicode property support.
1437        (a) For a maximizing quantifier, if the two different cases of the
1438            character were of different lengths in their UTF-8 codings (there are
1439            some cases like this - I found 11), and the matching function had to
1440            back up over a mixture of the two cases, it incorrectly assumed they
1441            were both the same length.
1443        (b) When PCRE was configured to use the heap rather than the stack for
1444            recursion during matching, it was not correctly preserving the data for
1445            the other case of a UTF-8 character when checking ahead for a match
1446            while processing a minimizing repeat. If the check also involved
1447            matching a wide character, but failed, corruption could cause an
1448            erroneous result when trying to check for a repeat of the original
1449            character.
1451    13. Some tidying changes to the testing mechanism:
1453        (a) The RunTest script now detects the internal link size and whether there
1454            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1455            values substituted by "configure". (The RunGrepTest script already did
1456            this for UTF-8.) The configure.ac script no longer substitutes the
1457            relevant variables.
1459        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1460            with length and offset values. This means that the output is different
1461            for different internal link sizes. Test 2 is skipped for link sizes
1462            other than 2 because of this, bypassing the problem. Unfortunately,
1463            there was also a test in test 3 (the locale tests) that used /B and
1464            failed for link sizes other than 2. Rather than cut the whole test out,
1465            I have added a new /Z option to pcretest that replaces the length and
1466            offset values with spaces. This is now used to make test 3 independent
1467            of link size. (Test 2 will be tidied up later.)
1469    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1470        segmentation fault instead of returning the appropriate error message.
1472    15. In multiline mode when the newline sequence was set to "any", the pattern
1473        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1474        This doesn't seem right; it now treats the CRLF combination as the line
1475        ending, and so does not match in that case. It's only a pattern such as ^$
1476        that would hit this one: something like ^ABC$ would have failed after \r
1477        and then tried again after \r\n.
1479    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1480        in an attempt to make files that differ only in their line terminators
1481        compare equal. This works on Linux.
1483    17. Under certain error circumstances pcregrep might try to free random memory
1484        as it exited. This is now fixed, thanks to valgrind.
1486    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1487        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1488        was because its rules for how to advance for /g after matching an empty
1489        string at the end of a line did not allow for this case. They now check for
1490        it specially.
1492    20. pcretest is supposed to handle patterns and data of any length, by
1493        extending its buffers when necessary. It was getting this wrong when the
1494        buffer for a data line had to be extended.
1496    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1497        CRLF as a newline sequence.
1499    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1500        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1501        I have nevertheless tidied it up.
1503    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1505    24. Added a man page for pcre-config.
1508    Version 7.0 19-Dec-06
1509    ---------------------
1511     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1512        moving to gcc 4.1.1.
1514     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1515        sys/time.h, which is documented as needed for this function. It doesn't
1516        seem to matter on Linux, but it showed up on some releases of OS X.
1518     3. It seems that there are systems where bytes whose values are greater than
1519        127 match isprint() in the "C" locale. The "C" locale should be the
1520        default when a C program starts up. In most systems, only ASCII printing
1521        characters match isprint(). This difference caused the output from pcretest
1522        to vary, making some of the tests fail. I have changed pcretest so that:
1524        (a) When it is outputting text in the compiled version of a pattern, bytes
1525            other than 32-126 are always shown as hex escapes.
1527        (b) When it is outputting text that is a matched part of a subject string,
1528            it does the same, unless a different locale has been set for the match
1529            (using the /L modifier). In this case, it uses isprint() to decide.
1531     4. Fixed a major bug that caused incorrect computation of the amount of memory
1532        required for a compiled pattern when options that changed within the
1533        pattern affected the logic of the preliminary scan that determines the
1534        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1535        that the computed length was too small. The symptoms of this bug were
1536        either the PCRE error "internal error: code overflow" from pcre_compile(),
1537        or a glibc crash with a message such as "pcretest: free(): invalid next
1538        size (fast)". Examples of patterns that provoked this bug (shown in
1539        pcretest format) are:
1541          /(?-x: )/x
1542          /(?x)(?-x: \s*#\s*)/
1543          /((?i)[\x{c0}])/8
1544          /(?i:[\x{c0}])/8
1546        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1547        is now done differently.
1549     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1550        wrapper classes; (b) implement a new function in the C++ scanner that is
1551        more efficient than the old way of doing things because it avoids levels of
1552        recursion in the regex matching; (c) add a paragraph to the documentation
1553        for the FullMatch() function.
1555     6. The escape sequence \n was being treated as whatever was defined as
1556        "newline". Not only was this contrary to the documentation, which states
1557        that \n is character 10 (hex 0A), but it also went horribly wrong when
1558        "newline" was defined as CRLF. This has been fixed.
1560     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1561        was being set to -1 for the "end of line" case (supposedly a value that no
1562        character can have). Though this value is never used (the check for end of
1563        line is "zero bytes in current character"), it caused compiler complaints.
1564        I've changed it to 0xffffffff.
1566     8. In pcre_version.c, the version string was being built by a sequence of
1567        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1568        string (as it is for production releases) called a macro with an empty
1569        argument. The C standard says the result of this is undefined. The gcc
1570        compiler treats it as an empty string (which was what was wanted) but it is
1571        reported that Visual C gives an error. The source has been hacked around to
1572        avoid this problem.
1574     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1575        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1576        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1577        of them did).
1579    10. Originally, pcretest opened its input and output without "b"; then I was
1580        told that "b" was needed in some environments, so it was added for release
1581        5.0 to both the input and output. (It makes no difference on Unix-like
1582        systems.) Later I was told that it is wrong for the input on Windows. I've
1583        now abstracted the modes into two macros, to make it easier to fiddle with
1584        them, and removed "b" from the input mode under Windows.
1586    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1588    12. Added -help and --help to pcretest as an official way of being reminded
1589        of the options.
1591    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1592        and pcrecpp.cc because they annoy compilers at high warning levels.
1594    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1596    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1597        scripts are not C programs :-) and which was not noticed because it works
1598        on Linux.
1600    16. pcretest is supposed to handle any length of pattern and data line (as one
1601        line or as a continued sequence of lines) by extending its input buffer if
1602        necessary. This feature was broken for very long pattern lines, leading to
1603        a string of junk being passed to pcre_compile() if the pattern was longer
1604        than about 50K.
1606    17. I have done a major re-factoring of the way pcre_compile() computes the
1607        amount of memory needed for a compiled pattern. Previously, there was code
1608        that made a preliminary scan of the pattern in order to do this. That was
1609        OK when PCRE was new, but as the facilities have expanded, it has become
1610        harder and harder to keep it in step with the real compile phase, and there
1611        have been a number of bugs (see for example, 4 above). I have now found a
1612        cunning way of running the real compile function in a "fake" mode that
1613        enables it to compute how much memory it would need, while actually only
1614        ever using a few hundred bytes of working memory and without too many
1615        tests of the mode. This should make future maintenance and development
1616        easier. A side effect of this work is that the limit of 200 on the nesting
1617        depth of parentheses has been removed (though this was never a serious
1618        limitation, I suspect). However, there is a downside: pcre_compile() now
1619        runs more slowly than before (30% or more, depending on the pattern). I
1620        hope this isn't a big issue. There is no effect on runtime performance.
1622    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1623        newline (only possible for the last line of a file) and it was a
1624        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1626    19. Added additional timing features to pcretest. (1) The -tm option now times
1627        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1628        separate command line item, by a number that specifies the number of
1629        repeats to use when timing. The default is 50000; this gives better
1630        precision, but takes uncomfortably long for very large patterns.
1632    20. Extended pcre_study() to be more clever in cases where a branch of a
1633        subpattern has no definite first character. For example, (a*|b*)[cd] would
1634        previously give no result from pcre_study(). Now it recognizes that the
1635        first character must be a, b, c, or d.
1637    21. There was an incorrect error "recursive call could loop indefinitely" if
1638        a subpattern (or the entire pattern) that was being tested for matching an
1639        empty string contained only one non-empty item after a nested subpattern.
1640        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1641        incorrectly, because the \d was being skipped in the check.
1643    22. The pcretest program now has a new pattern option /B and a command line
1644        option -b, which is equivalent to adding /B to every pattern. This causes
1645        it to show the compiled bytecode, without the additional information that
1646        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1647        is the same as /B/I).
1649    23. A new optimization is now able automatically to treat some sequences such
1650        as a*b as a*+b. More specifically, if something simple (such as a character
1651        or a simple class like \d) has an unlimited quantifier, and is followed by
1652        something that cannot possibly match the quantified thing, the quantifier
1653        is automatically "possessified".
1655    24. A recursive reference to a subpattern whose number was greater than 39
1656        went wrong under certain circumstances in UTF-8 mode. This bug could also
1657        have affected the operation of pcre_study().
1659    25. Realized that a little bit of performance could be had by replacing
1660        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1662    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1664    27. Possessive quantifiers such as a++ were previously implemented by turning
1665        them into atomic groups such as ($>a+). Now they have their own opcodes,
1666        which improves performance. This includes the automatically created ones
1667        from 23 above.
1669    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1670        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1671        the first matched character to be a colon. This applied both to named and
1672        numbered groups.
1674    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1676    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1677        building PCRE on Windows easier, so I have included it in the distribution.
1679    31. There is now a check in pcretest against a ridiculously large number being
1680        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1681        loop, the loop is abandoned.
1683    32. Forward references to subpatterns in conditions such as (?(2)...) where
1684        subpattern 2 is defined later cause pcre_compile() to search forwards in
1685        the pattern for the relevant set of parentheses. This search went wrong
1686        when there were unescaped parentheses in a character class, parentheses
1687        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1689    33. "Subroutine" calls and backreferences were previously restricted to
1690        referencing subpatterns earlier in the regex. This restriction has now
1691        been removed.
1693    34. Added a number of extra features that are going to be in Perl 5.10. On the
1694        whole, these are just syntactic alternatives for features that PCRE had
1695        previously implemented using the Python syntax or my own invention. The
1696        other formats are all retained for compatibility.
1698        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1699            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1700            also .NET compatible.
1702        (b) A recursion or subroutine call to a named group can now be defined as
1703            (?&name) as well as (?P>name).
1705        (c) A backreference to a named group can now be defined as \k<name> or
1706            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1707            5.10, are also .NET compatible.
1709        (d) A conditional reference to a named group can now use the syntax
1710            (?(<name>) or (?('name') as well as (?(name).
1712        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1713            groups (named and numbered) that are never evaluated inline, but can be
1714            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1715            is always false. There may be only one alternative in such a group.
1717        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1718            as the simple (?(R). The condition is true only if the most recent
1719            recursion is that of the given number or name. It does not search out
1720            through the entire recursion stack.
1722        (g) The escape \gN or \g{N} has been added, where N is a positive or
1723            negative number, specifying an absolute or relative reference.
1725    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1726        some "unreachable code" warnings.
1728    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1729        things, this adds five new scripts.
1731    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1732        There were also incompatibilities regarding the handling of \Q..\E inside
1733        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1734        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1736    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1737        matches an empty string, and forcibly breaks the loop. There were bugs in
1738        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1739        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1740        separate and independent bugs (that affected different cases) have been
1741        fixed.
1743    39. Refactored the code to abolish the use of different opcodes for small
1744        capturing bracket numbers. This is a tidy that I avoided doing when I
1745        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1746        The new approach is not only tidier, it makes it possible to reduce the
1747        memory needed to fix the previous bug (38).
1749    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1750        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1751        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1752        mode.
1754    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1755        report.
1757    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1758        copy construction and assignment in the C++ wrapper.
1760    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1761        couple of bugs that could have given wrong results in the "--newline=crlf"
1762        case.
1764    44. Added a number of casts and did some reorganization of signed/unsigned int
1765        variables following suggestions from Dair Grant. Also renamed the variable
1766        "this" as "item" because it is a C++ keyword.
1768    45. Arranged for dftables to add
1770          #include "pcre_internal.h"
1772        to pcre_chartables.c because without it, gcc 4.x may remove the array
1773        definition from the final binary if PCRE is built into a static library and
1774        dead code stripping is activated.
1776    46. For an unanchored pattern, if a match attempt fails at the start of a
1777        newline sequence, and the newline setting is CRLF or ANY, and the next two
1778        characters are CRLF, advance by two characters instead of one.
1781    Version 6.7 04-Jul-06
1782    ---------------------
1784     1. In order to handle tests when input lines are enormously long, pcretest has
1785        been re-factored so that it automatically extends its buffers when
1786        necessary. The code is crude, but this _is_ just a test program. The
1787        default size has been increased from 32K to 50K.
1789     2. The code in pcre_study() was using the value of the re argument before
1790        testing it for NULL. (Of course, in any sensible call of the function, it
1791        won't be NULL.)
1793     3. The memmove() emulation function in pcre_internal.h, which is used on
1794        systems that lack both memmove() and bcopy() - that is, hardly ever -
1795        was missing a "static" storage class specifier.
1797     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1798        containing an extended class (one that cannot be represented by a bitmap
1799        because it contains high-valued characters or Unicode property items, e.g.
1800        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1801        pattern, but PCRE should not loop if you do not (it no longer does).
1802        [Detail: two cases were found: (a) a repeated subpattern containing an
1803        extended class; (b) a recursive reference to a subpattern that followed a
1804        previous extended class. It wasn't skipping over the extended class
1805        correctly when UTF-8 mode was not set.]
1807     5. A negated single-character class was not being recognized as fixed-length
1808        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1809        compile error "lookbehind assertion is not fixed length".
1811     6. The RunPerlTest auxiliary script was showing an unexpected difference
1812        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1813        write a Perl script that can interpret lines of an input file either as
1814        byte characters or as UTF-8, which is what "perltest" was being required to
1815        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1816        can't do is switch easily at run time between having the "use utf8;" pragma
1817        or not. In the end, I fudged it by using the RunPerlTest script to insert
1818        "use utf8;" explicitly for the UTF-8 tests.
1820     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1821        the end of the subject string, contrary to the documentation and to what
1822        Perl does. This was true of both matching functions. Now it matches only at
1823        the start of the subject and immediately after *internal* newlines.
1825     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1826        a pointer to an int instead of a pointer to an unsigned long int. This
1827        caused problems on 64-bit systems.
1829     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1830        instance of the 'standard' template library not being so standard".
1832    10. There was no check on the number of named subpatterns nor the maximum
1833        length of a subpattern name. The product of these values is used to compute
1834        the size of the memory block for a compiled pattern. By supplying a very
1835        long subpattern name and a large number of named subpatterns, the size
1836        computation could be caused to overflow. This is now prevented by limiting
1837        the length of names to 32 characters, and the number of named subpatterns
1838        to 10,000.
1840    11. Subpatterns that are repeated with specific counts have to be replicated in
1841        the compiled pattern. The size of memory for this was computed from the
1842        length of the subpattern and the repeat count. The latter is limited to
1843        65535, but there was no limit on the former, meaning that integer overflow
1844        could in principle occur. The compiled length of a repeated subpattern is
1845        now limited to 30,000 bytes in order to prevent this.
1847    12. Added the optional facility to have named substrings with the same name.
1849    13. Added the ability to use a named substring as a condition, using the
1850        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1851        are numbers (not recommended). Forward references are permitted.
1853    14. Added forward references in named backreferences (if you see what I mean).
1855    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1856        pattern could run off the end of the subject. For example, the pattern
1857        "(?s)(.{1,5})"8 did this with the subject "ab".
1859    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1860        PCRE_CASELESS was set when matching characters that were quantified with ?
1861        or *.
1863    17. A character class other than a single negated character that had a minimum
1864        but no maximum quantifier - for example [ab]{6,} - was not handled
1865        correctly by pce_dfa_exec(). It would match only one character.
1867    18. A valid (though odd) pattern that looked like a POSIX character
1868        class but used an invalid character after [ (for example [[,abc,]]) caused
1869        pcre_compile() to give the error "Failed: internal error: code overflow" or
1870        in some cases to crash with a glibc free() error. This could even happen if
1871        the pattern terminated after [[ but there just happened to be a sequence of
1872        letters, a binary zero, and a closing ] in the memory that followed.
1874    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1875        over the years. Originally (before any Unicode support), just the bottom 8
1876        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1877        output from "man perlunicode" includes this:
1879          The regular expression compiler produces polymorphic opcodes.  That
1880          is, the pattern adapts to the data and automatically switches to
1881          the Unicode character scheme when presented with Unicode data--or
1882          instead uses a traditional byte scheme when presented with byte
1883          data.
1885        Sadly, a wide octal escape does not cause a switch, and in a string with
1886        no other multibyte characters, these octal escapes are treated as before.
1887        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1888        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1889        Unicode string.
1891        I have not perpetrated such confusion in PCRE. Up till now, it took just
1892        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1893        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1894        translate to the appropriate multibyte character.
1896    29. Applied some refactoring to reduce the number of warnings from Microsoft
1897        and Borland compilers. This has included removing the fudge introduced
1898        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1899        a warning about an unused variable.
1901    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1902        characters since release 4.0, because Perl (from release 5.004) does not.
1903        [Or at least, is documented not to: some releases seem to be in conflict
1904        with the documentation.] However, when a pattern was studied with
1905        pcre_study() and all its branches started with \s, PCRE still included VT
1906        as a possible starting character. Of course, this did no harm; it just
1907        caused an unnecessary match attempt.
1909    22. Removed a now-redundant internal flag bit that recorded the fact that case
1910        dependency changed within the pattern. This was once needed for "required
1911        byte" processing, but is no longer used. This recovers a now-scarce options
1912        bit. Also moved the least significant internal flag bit to the most-
1913        significant bit of the word, which was not previously used (hangover from
1914        the days when it was an int rather than a uint) to free up another bit for
1915        the future.
1917    23. Added support for CRLF line endings as well as CR and LF. As well as the
1918        default being selectable at build time, it can now be changed at runtime
1919        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1920        specify that it is scanning data with non-default line endings.
1922    24. Changed the definition of CXXLINK to make it agree with the definition of
1923        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1925    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1926        recursions. This makes a big different to stack usage for some patterns.
1928    26. If a subpattern containing a named recursion or subroutine reference such
1929        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1930        the space required for the compiled pattern went wrong and gave too small a
1931        value. Depending on the environment, this could lead to "Failed: internal
1932        error: code overflow at offset 49" or "glibc detected double free or
1933        corruption" errors.
1935    27. Applied patches from Google (a) to support the new newline modes and (b) to
1936        advance over multibyte UTF-8 characters in GlobalReplace.
1938    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1939        difference for some implementation of PCRE in some Windows version.
1941    29. Added some extra testing facilities to pcretest:
1943        \q<number>   in a data line sets the "match limit" value
1944        \Q<number>   in a data line sets the "match recursion limt" value
1945        -S <number>  sets the stack size, where <number> is in megabytes
1947        The -S option isn't available for Windows.
1950    Version 6.6 06-Feb-06
1951    ---------------------
1953     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1954        in pcreposix.h. I have copied the definition from pcre.h.
1956     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1957        because pcre.h is no longer a built file.
1959     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1960        not normally included in the compiled code.
1963    Version 6.5 01-Feb-06
1964    ---------------------
1966     1. When using the partial match feature with pcre_dfa_exec(), it was not
1967        anchoring the second and subsequent partial matches at the new starting
1968        point. This could lead to incorrect results. For example, with the pattern
1969        /1234/, partially matching against "123" and then "a4" gave a match.
1971     2. Changes to pcregrep:
1973        (a) All non-match returns from pcre_exec() were being treated as failures
1974            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1975            error message is output. Some extra information is given for the
1976            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1977            probably the only errors that are likely to be caused by users (by
1978            specifying a regex that has nested indefinite repeats, for instance).
1979            If there are more than 20 of these errors, pcregrep is abandoned.
1981        (b) A binary zero was treated as data while matching, but terminated the
1982            output line if it was written out. This has been fixed: binary zeroes
1983            are now no different to any other data bytes.
1985        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1986            used to set a locale for matching. The --locale=xxxx long option has
1987            been added (no short equivalent) to specify a locale explicitly on the
1988            pcregrep command, overriding the environment variables.
1990        (d) When -B was used with -n, some line numbers in the output were one less
1991            than they should have been.
1993        (e) Added the -o (--only-matching) option.
1995        (f) If -A or -C was used with -c (count only), some lines of context were
1996            accidentally printed for the final match.
1998        (g) Added the -H (--with-filename) option.
2000        (h) The combination of options -rh failed to suppress file names for files
2001            that were found from directory arguments.
2003        (i) Added the -D (--devices) and -d (--directories) options.
2005        (j) Added the -F (--fixed-strings) option.
2007        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2009        (l) Added the --colo(u)r option.
2011        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2012            is not present by default.
2014     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2015        items such as (?R) or (?1), when the recursion could match a number of
2016        alternatives. If it matched one of the alternatives, but subsequently,
2017        outside the recursion, there was a failure, the code tried to back up into
2018        the recursion. However, because of the way PCRE is implemented, this is not
2019        possible, and the result was an incorrect result from the match.
2021        In order to prevent this happening, the specification of recursion has
2022        been changed so that all such subpatterns are automatically treated as
2023        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2025     4. I had overlooked the fact that, in some locales, there are characters for
2026        which isalpha() is true but neither isupper() nor islower() are true. In
2027        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2028        and ordfeminine) are like this. This affected the treatment of \w and \W
2029        when they appeared in character classes, but not when they appeared outside
2030        a character class. The bit map for "word" characters is now created
2031        separately from the results of isalnum() instead of just taking it from the
2032        upper, lower, and digit maps. (Plus the underscore character, of course.)
2034     5. The above bug also affected the handling of POSIX character classes such as
2035        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2036        permanent tables. Instead, the bit maps for such a class were previously
2037        created as the appropriate unions of the upper, lower, and digit bitmaps.
2038        Now they are created by subtraction from the [[:word:]] class, which has
2039        its own bitmap.
2041     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2042        It is created by subtracting the vertical space characters (\x09, \x0a,
2043        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2044        subtraction was done in the overall bitmap for a character class, meaning
2045        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2046        be recognized. This bug has been fixed.
2048     7. Patches from the folks at Google:
2050          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2051          real life, but is still worth protecting against".
2053          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2054          regular expressions".
2056          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2057          have it.
2059          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2060          "configure" and the latter not, in order to fix a problem somebody had
2061          with compiling the Arg class on HP-UX.
2063          (e) Improve the error-handling of the C++ wrapper a little bit.
2065          (f) New tests for checking recursion limiting.
2067     8. The pcre_memmove() function, which is used only if the environment does not
2068        have a standard memmove() function (and is therefore rarely compiled),
2069        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2070        returning a result (though PCRE never actually uses the result).
2072     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2073        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2074        returned instead of calling malloc() with an overflowing number that would
2075        most likely cause subsequent chaos.
2077    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2079    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2080        with this option is matched, the nmatch and pmatch options of regexec() are
2081        ignored.
2083    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2084        provided in case anyone wants to the the POSIX interface with UTF-8
2085        strings.
2087    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2088        C++ linking (needed for some HP-UX environments).
2090    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2091        (unused parameter) and in the pcre_printint() function (omitted "default"
2092        switch label when the default is to do nothing).
2094    15. Added some code to make it possible, when PCRE is compiled as a C++
2095        library, to replace subject pointers for pcre_exec() with a smart pointer
2096        class, thus making it possible to process discontinuous strings.
2098    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2099        much the same function. They were added by different people who were trying
2100        to make PCRE easy to compile on non-Unix systems. It has been suggested
2101        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2102        for compiling on Windows systems. I have therefore replaced it with
2103        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2104        defaults to "extern" for C or "extern C" for C++, which works fine on
2105        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2106        SCOPE with something explicit in config.h. In addition:
2108        (a) pcreposix.h still had just "extern" instead of either of these macros;
2109            I have replaced it with PCRE_DATA_SCOPE.
2111        (b) Functions such as _pcre_xclass(), which are internal to the library,
2112            but external in the C sense, all had PCRE_EXPORT in their definitions.
2113            This is apparently wrong for the Windows case, so I have removed it.
2114            (It makes no difference on Unix-like systems.)
2116    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2117        of recursive calls to match(). This is different to MATCH_LIMIT because
2118        that limits the total number of calls to match(), not all of which increase
2119        the depth of recursion. Limiting the recursion depth limits the amount of
2120        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2121        when PCRE is compiled, and changed at run time. A patch from Google adds
2122        this functionality to the C++ interface.
2124    18. Changes to the handling of Unicode character properties:
2126        (a) Updated the table to Unicode 4.1.0.
2128        (b) Recognize characters that are not in the table as "Cn" (undefined).
2130        (c) I revised the way the table is implemented to a much improved format
2131            which includes recognition of ranges. It now supports the ranges that
2132            are defined in UnicodeData.txt, and it also amalgamates other
2133            characters into ranges. This has reduced the number of entries in the
2134            table from around 16,000 to around 3,000, thus reducing its size
2135            considerably. I realized I did not need to use a tree structure after
2136            all - a binary chop search is just as efficient. Having reduced the
2137            number of entries, I extended their size from 6 bytes to 8 bytes to
2138            allow for more data.
2140        (d) Added support for Unicode script names via properties such as \p{Han}.
2142    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2143        matching that character.
2145    20. When matching a repeated Unicode property with a minimum greater than zero,
2146        (for example \pL{2,}), PCRE could look past the end of the subject if it
2147        reached it while seeking the minimum number of characters. This could
2148        happen only if some of the characters were more than one byte long, because
2149        there is a check for at least the minimum number of bytes.
2151    21. Refactored the implementation of \p and \P so as to be more general, to
2152        allow for more different types of property in future. This has changed the
2153        compiled form incompatibly. Anybody with saved compiled patterns that use
2154        \p or \P will have to recompile them.
2156    22. Added "Any" and "L&" to the supported property types.
2158    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2159        but give a compile time error if the value is greater than 0xff.
2161    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2162        accidentally not being installed or uninstalled.
2164    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2165        made were to insert the current release number. This seemed silly, because
2166        it made things harder for people building PCRE on systems that don't run
2167        "configure". I have turned pcre.h into a distributed file, no longer built
2168        by "configure", with the version identification directly included. There is
2169        no longer a pcre.h.in file.
2171        However, this change necessitated a change to the pcre-config script as
2172        well. It is built from pcre-config.in, and one of the substitutions was the
2173        release number. I have updated configure.ac so that ./configure now finds
2174        the release number by grepping pcre.h.
2176    26. Added the ability to run the tests under valgrind.
2179    Version 6.4 05-Sep-05
2180    ---------------------
2182     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2183        "--" to be printed when multiple files were scanned, even when none of the
2184        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2185        consider it to be a bug, and have restored the previous behaviour.
2187     2. A couple of code tidies to get rid of compiler warnings.
2189     3. The pcretest program used to cheat by referring to symbols in the library
2190        whose names begin with _pcre_. These are internal symbols that are not
2191        really supposed to be visible externally, and in some environments it is
2192        possible to suppress them. The cheating is now confined to including
2193        certain files from the library's source, which is a bit cleaner.
2195     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2196        file's purpose clearer.
2198     5. Reorganized pcre_ucp_findchar().
2201    Version 6.3 15-Aug-05
2202    ---------------------
2204     1. The file libpcre.pc.in did not have general read permission in the tarball.
2206     2. There were some problems when building without C++ support:
2208        (a) If C++ support was not built, "make install" and "make test" still
2209            tried to test it.
2211        (b) There were problems when the value of CXX was explicitly set. Some
2212            changes have been made to try to fix these, and ...
2214        (c) --disable-cpp can now be used to explicitly disable C++ support.
2216        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2217            backslash in a target when C++ was disabled. This confuses some
2218            versions of "make", apparently. Using an intermediate variable solves
2219            this. (Same for CPP_LOBJ.)
2221     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2222        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2223        necessary on certain architectures.
2225     4. Added a setting of -export-symbols-regex to the link command to remove
2226        those symbols that are exported in the C sense, but actually are local
2227        within the library, and not documented. Their names all begin with
2228        "_pcre_". This is not a perfect job, because (a) we have to except some
2229        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2230        available (and never for static libraries). I have made a note to try to
2231        find a way round (a) in the future.
2234    Version 6.2 01-Aug-05
2235    ---------------------
2237     1. There was no test for integer overflow of quantifier values. A construction
2238        such as {1111111111111111} would give undefined results. What is worse, if
2239        a minimum quantifier for a parenthesized subpattern overflowed and became
2240        negative, the calculation of the memory size went wrong. This could have
2241        led to memory overwriting.
2243     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2245     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2246        operating environments where this matters.
2248     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2249        PCRE options from within the C++ wrapper.
2251     5. Named capturing subpatterns were not being correctly counted when a pattern
2252        was compiled. This caused two problems: (a) If there were more than 100
2253        such subpatterns, the calculation of the memory needed for the whole
2254        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2255        back references of the form \12, where the number was greater than 9, were
2256        not recognized as back references, even though there were sufficient
2257        previous subpatterns.
2259     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2260        versions of gcc, e.g. 2.95.4.
2263    Version 6.1 21-Jun-05
2264    ---------------------
2266     1. There was one reference to the variable "posix" in pcretest.c that was not
2267        surrounded by "#if !defined NOPOSIX".
2269     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2270        the cross-check on the old pcre_info() function, for the benefit of the
2271        cut-down version of PCRE that is currently imported into Exim.
2273     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2274        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2275        patterns that start (?i) but I don't think that matters. The use of (?i) is
2276        just an example; this all applies to the other options as well.
2278     4. Since libtool seems to echo the compile commands it is issuing, the output
2279        from "make" can be reduced a bit by putting "@" in front of each libtool
2280        compile command.
2282     5. Patch from the folks at Google for configure.in to be a bit more thorough
2283        in checking for a suitable C++ installation before trying to compile the
2284        C++ stuff. This should fix a reported problem when a compiler was present,
2285        but no suitable headers.
2287     6. The man pages all had just "PCRE" as their title. I have changed them to
2288        be the relevant file name. I have also arranged that these names are
2289        retained in the file doc/pcre.txt, which is a concatenation in text format
2290        of all the man pages except the little individual ones for each function.
2292     7. The NON-UNIX-USE file had not been updated for the different set of source
2293        files that come with release 6. I also added a few comments about the C++
2294        wrapper.
2297    Version 6.0 07-Jun-05
2298    ---------------------
2300     1. Some minor internal re-organization to help with my DFA experiments.
2302     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2303        didn't matter for the library itself when fully configured, but did matter
2304        when compiling without UCP support, or within Exim, where the ucp files are
2305        not imported.
2307     3. Refactoring of the library code to split up the various functions into
2308        different source modules. The addition of the new DFA matching code (see
2309        below) to a single monolithic source would have made it really too
2310        unwieldy, quite apart from causing all the code to be include in a
2311        statically linked application, when only some functions are used. This is
2312        relevant even without the DFA addition now that patterns can be compiled in
2313        one application and matched in another.
2315        The downside of splitting up is that there have to be some external
2316        functions and data tables that are used internally in different modules of
2317        the library but which are not part of the API. These have all had their
2318        names changed to start with "_pcre_" so that they are unlikely to clash
2319        with other external names.
2321     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2322        a different (DFA) algorithm. Although it is slower than the original
2323        function, it does have some advantages for certain types of matching
2324        problem.
2326     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2327        including restarting after a partial match.
2329     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2330        defined when compiling for Windows was sent to me. I have put it into the
2331        code, though I have no means of testing or verifying it.
2333     7. Added the pcre_refcount() auxiliary function.
2335     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2336        match before or at the first newline in the subject string. In pcretest,
2337        the /f option on a pattern can be used to set this.
2339     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2340        would behave wrongly. This has been present in PCRE since release 4.0.
2342    10. A number of changes to the pcregrep command:
2344        (a) Refactored how -x works; insert ^(...)$ instead of setting
2345            PCRE_ANCHORED and checking the length, in preparation for adding
2346            something similar for -w.
2348        (b) Added the -w (match as a word) option.
2350        (c) Refactored the way lines are read and buffered so as to have more
2351            than one at a time available.
2353        (d) Implemented a pcregrep test script.
2355        (e) Added the -M (multiline match) option. This allows patterns to match
2356            over several lines of the subject. The buffering ensures that at least
2357            8K, or the rest of the document (whichever is the shorter) is available
2358            for matching (and similarly the previous 8K for lookbehind assertions).
2360        (f) Changed the --help output so that it now says
2362              -w, --word-regex(p)
2364            instead of two lines, one with "regex" and the other with "regexp"
2365            because that confused at least one person since the short forms are the
2366            same. (This required a bit of code, as the output is generated
2367            automatically from a table. It wasn't just a text change.)
2369        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2370            option but starts with a hyphen. Could be a pattern or a path name
2371            starting with a hyphen, for instance.
2373        (h) "-" can be given as a file name to represent stdin.
2375        (i) When file names are being printed, "(standard input)" is used for
2376            the standard input, for compatibility with GNU grep. Previously
2377            "<stdin>" was used.
2379        (j) The option --label=xxx can be used to supply a name to be used for
2380            stdin when file names are being printed. There is no short form.
2382        (k) Re-factored the options decoding logic because we are going to add
2383            two more options that take data. Such options can now be given in four
2384            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2386        (l) Added the -A, -B, and -C options for requesting that lines of context
2387            around matches be printed.
2389        (m) Added the -L option to print the names of files that do not contain
2390            any matching lines, that is, the complement of -l.
2392        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2393            continue to scan other files.
2395        (o) The -s option was incorrectly implemented. For compatibility with other
2396            greps, it now suppresses the error message for a non-existent or non-
2397            accessible file (but not the return code). There is a new option called
2398            -q that suppresses the output of matching lines, which was what -s was
2399            previously doing.
2401        (p) Added --include and --exclude options to specify files for inclusion
2402            and exclusion when recursing.
2404    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2405        Hopefully, it now does.
2407    12. Missing cast in pcre_study().
2409    13. Added an "uninstall" target to the makefile.
2411    14. Replaced "extern" in the function prototypes in Makefile.in with
2412        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2413        world, but is set differently for Windows.
2415    15. Added a second compiling function called pcre_compile2(). The only
2416        difference is that it has an extra argument, which is a pointer to an
2417        integer error code. When there is a compile-time failure, this is set
2418        non-zero, in addition to the error test pointer being set to point to an
2419        error message. The new argument may be NULL if no error number is required
2420        (but then you may as well call pcre_compile(), which is now just a
2421        wrapper). This facility is provided because some applications need a
2422        numeric error indication, but it has also enabled me to tidy up the way
2423        compile-time errors are handled in the POSIX wrapper.
2425    16. Added VPATH=.libs to the makefile; this should help when building with one
2426        prefix path and installing with another. (Or so I'm told by someone who
2427        knows more about this stuff than I do.)
2429    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2430        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2431        match everything, including newlines. This is not POSIX-compatible, but
2432        somebody wanted the feature. From pcretest it can be activated by using
2433        both the P and the s flags.
2435    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2437    19. libpcre.pc was being incorrectly installed as executable.
2439    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2440        it now also looks for '\r' so that it will work unmodified on Windows.
2442    21. Added Google's contributed C++ wrapper to the distribution.
2444    22. Added some untidy missing memory free() calls in pcretest, to keep
2445        Electric Fence happy when testing.
2449    Version 5.0 13-Sep-04
2450    ---------------------
2452     1. Internal change: literal characters are no longer packed up into items
2453        containing multiple characters in a single byte-string. Each character
2454        is now matched using a separate opcode. However, there may be more than one
2455        byte in the character in UTF-8 mode.
2457     2. The pcre_callout_block structure has two new fields: pattern_position and
2458        next_item_length. These contain the offset in the pattern to the next match
2459        item, and its length, respectively.
2461     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2462        insertion of callouts before each pattern item. Added the /C option to
2463        pcretest to make use of this.
2465     4. On the advice of a Windows user, the lines
2467          #if defined(_WIN32) || defined(WIN32)
2468          _setmode( _fileno( stdout ), 0x8000 );
2469          #endif  /* defined(_WIN32) || defined(WIN32) */
2471        have been added to the source of pcretest. This apparently does useful
2472        magic in relation to line terminators.
2474     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2475        for the benefit of those environments where the "b" makes a difference.
2477     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2478        to know about it. I have put a hack into configure.in that adds in code
2479        to set GCC=yes if CC=icc. This seems to end up at a point in the
2480        generated configure script that is early enough to affect the setting of
2481        compiler options, which is what is needed, but I have no means of testing
2482        whether it really works. (The user who reported this had patched the
2483        generated configure script, which of course I cannot do.)
2485        LATER: After change 22 below (new libtool files), the configure script
2486        seems to know about icc (and also ecc). Therefore, I have commented out
2487        this hack in configure.in.
2489     7. Added support for pkg-config (2 patches were sent in).
2491     8. Negated POSIX character classes that used a combination of internal tables
2492        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2493        [[:^ascii]]. Typically, they would match almost any characters. The other
2494        POSIX classes were not broken in this way.
2496     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2497        to find the match, as PCRE was deluded into thinking that the match had to
2498        start at the start point or following a newline. The same bug applied to
2499        patterns with negative forward assertions or any backward assertions
2500        preceding ".*" at the start, unless the pattern required a fixed first
2501        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2503    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2504        starting at the last subject character, bytes beyond the end of the subject
2505        string were read.
2507    11. Renamed the variable "class" as "classbits" to make life easier for C++
2508        users. (Previously there was a macro definition, but it apparently wasn't
2509        enough.)
2511    12. Added the new field "tables" to the extra data so that tables can be passed
2512        in at exec time, or the internal tables can be re-selected. This allows
2513        a compiled regex to be saved and re-used at a later time by a different
2514        program that might have everything at different addresses.
2516    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2517        -R library as well as a -L library.
2519    14. The debugging options of pcretest (-d on the command line or D on a
2520        pattern) showed incorrect output for anything following an extended class
2521        that contained multibyte characters and which was followed by a quantifier.
2523    15. Added optional support for general category Unicode character properties
2524        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2525        support. It adds about 90K to the size of the library. The meanings of the
2526        inbuilt class escapes such as \d and \s have NOT been changed.
2528    16. Updated pcredemo.c to include calls to free() to release the memory for the
2529        compiled pattern.
2531    17. The generated file chartables.c was being created in the source directory
2532        instead of in the building directory. This caused the build to fail if the
2533        source directory was different from the building directory, and was
2534        read-only.
2536    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2537        file. No doubt somebody will tell me if they don't make sense... Also added
2538        Dan Mooney's comments about building on OpenVMS.
2540    19. Added support for partial matching via the PCRE_PARTIAL option for
2541        pcre_exec() and the \P data escape in pcretest.
2543    20. Extended pcretest with 3 new pattern features:
2545        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2546              write the compiled pattern to the file whose name is "rest-of-line".
2547              This is a straight binary dump of the data, with the saved pointer to
2548              the character tables forced to be NULL. The study data, if any, is
2549              written too. After writing, pcretest reads a new pattern.
2551        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2552              compiled pattern from the given file. There must not be any
2553              occurrences of "<" in the file name (pretty unlikely); if there are,
2554              pcretest will instead treat the initial "<" as a pattern delimiter.
2555              After reading in the pattern, pcretest goes on to read data lines as
2556              usual.
2558        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2559              and 16-bit fields in a compiled pattern, to simulate a pattern that
2560              was compiled on a host of opposite endianness.
2562    21. The pcre-exec() function can now cope with patterns that were compiled on
2563        hosts of opposite endianness, with this restriction:
2565          As for any compiled expression that is saved and used later, the tables
2566          pointer field cannot be preserved; the extra_data field in the arguments
2567          to pcre_exec() should be used to pass in a tables address if a value
2568          other than the default internal tables were used at compile time.
2570    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2571        now diagnosed as an error. Previously, most of the time, a negative number
2572        would have been treated as zero, but if in addition "ovector" was passed as
2573        NULL, a crash could occur.
2575    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2576        new versions from the libtool 1.5 distribution (the last one is a copy of
2577        a file called libtool.m4). This seems to have fixed the need to patch
2578        "configure" to support Darwin 1.3 (which I used to do). However, I still
2579        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2580        workstation).
2582    24. Changed the PCRE licence to be the more standard "BSD" licence.
2585    Version 4.5 01-Dec-03
2586    ---------------------
2588     1. There has been some re-arrangement of the code for the match() function so
2589        that it can be compiled in a version that does not call itself recursively.
2590        Instead, it keeps those local variables that need separate instances for
2591        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2592        needs to "recurse". Keeping track of where control must go is done by means
2593        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2594        hide most of the details from the main code, and operates only if
2595        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2596        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2597        operating.
2599        To make it easier for callers to provide specially tailored get/free
2600        functions for this usage, two new functions, pcre_stack_malloc, and
2601        pcre_stack_free, are used. They are always called in strict stacking order,
2602        and the size of block requested is always the same.
2604        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2605        PCRE has been compiled to use the stack or the heap for recursion. The
2606        -C option of pcretest uses this to show which version is compiled.
2608        A new data escape \S, is added to pcretest; it causes the amounts of store
2609        obtained and freed by both kinds of malloc/free at match time to be added
2610        to the output.
2612     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2613        what's available on my current Linux desktop machine.
2615     3. When matching a UTF-8 string, the test for a valid string at the start has
2616        been extended. If start_offset is not zero, PCRE now checks that it points
2617        to a byte that is the start of a UTF-8 character. If not, it returns
2618        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2619        this is necessary because there may be backward assertions in the pattern.
2620        When matching the same subject several times, it may save resources to use
2621        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2623     4. The code for checking the validity of UTF-8 strings has been tightened so
2624        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2625        containing "overlong sequences".
2627     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2628        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2629        should have been "&", but it just so happened that all the cases this let
2630        through by mistake were picked up later in the function.
2632     6. I had used a variable called "isblank" - this is a C99 function, causing
2633        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2635     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2636        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2637        so the version is shown; (c) stop "make test" echoing "./RunTest".
2639     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2641     9. The prototype for memmove() for systems that don't have it was using
2642        size_t, but the inclusion of the header that defines size_t was later. I've
2643        moved the #includes for the C headers earlier to avoid this.
2645    10. Added some adjustments to the code to make it easier to compiler on certain
2646        special systems:
2648          (a) Some "const" qualifiers were missing.
2649          (b) Added the macro EXPORT before all exported functions; by default this
2650              is defined to be empty.
2651          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2652              that it reads its output file name as an argument instead of writing
2653              to the standard output and assuming this can be redirected.
2655    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2656        class containing characters with values greater than 255, PCRE compilation
2657        went into a loop.
2659    12. A recursive reference to a subpattern that was within another subpattern
2660        that had a minimum quantifier of zero caused PCRE to crash. For example,
2661        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2662        recursion. If the recursively-called subpattern itself had a zero repeat,
2663        that was OK.
2665    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2666        buffer into which it was copied (for escape processing) was still set at
2667        1024, so long lines caused crashes.
2669    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2670        "internal error: code overflow...". This applied to any character class
2671        that was followed by a possessive quantifier.
2673    15. Modified the Makefile to add libpcre.la as a prerequisite for
2674        libpcreposix.la because I was told this is needed for a parallel build to
2675        work.
2677    16. If a pattern that contained .* following optional items at the start was
2678        studied, the wrong optimizing data was generated, leading to matching
2679        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2680        matching string must start with a or b or c. The correct conclusion for
2681        this pattern is that a match can start with any character.
2684    Version 4.4 13-Aug-03
2685    ---------------------
2687     1. In UTF-8 mode, a character class containing characters with values between
2688        127 and 255 was not handled correctly if the compiled pattern was studied.
2689        In fixing this, I have also improved the studying algorithm for such
2690        classes (slightly).
2692     2. Three internal functions had redundant arguments passed to them. Removal
2693        might give a very teeny performance improvement.
2695     3. Documentation bug: the value of the capture_top field in a callout is *one
2696        more than* the number of the hightest numbered captured substring.
2698     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2699        in incorrectly linking with a previously installed version. They now link
2700        explicitly with libpcre.la.
2702     5. configure.in no longer needs to recognize Cygwin specially.
2704     6. A problem in pcre.in for Windows platforms is fixed.
2706     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2707        pcretest, it used to include the size of the study block as part of its
2708        output. Unfortunately, the structure contains a field that has a different
2709        size on different hardware architectures. This meant that the tests that
2710        showed this size failed. As the block is currently always of a fixed size,
2711        this information isn't actually particularly useful in pcretest output, so
2712        I have just removed it.
2714     8. Three pre-processor statements accidentally did not start in column 1.
2715        Sadly, there are *still* compilers around that complain, even though
2716        standard C has not required this for well over a decade. Sigh.
2718     9. In pcretest, the code for checking callouts passed small integers in the
2719        callout_data field, which is a void * field. However, some picky compilers
2720        complained about the casts involved for this on 64-bit systems. Now
2721        pcretest passes the address of the small integer instead, which should get
2722        rid of the warnings.
2724    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2725        both compile and run time, and gives an error if an invalid UTF-8 sequence
2726        is found. There is a option for disabling this check in cases where the
2727        string is known to be correct and/or the maximum performance is wanted.
2729    11. In response to a bug report, I changed one line in Makefile.in from
2731            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2732        to
2733            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2735        to look similar to other lines, but I have no way of telling whether this
2736        is the right thing to do, as I do not use Windows. No doubt I'll get told
2737        if it's wrong...
2740    Version 4.3 21-May-03
2741    ---------------------
2743    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2744       Makefile.
2746    2. Some refactoring to improve the quality of the code:
2748       (i)   The utf8_table... variables are now declared "const".
2750       (ii)  The code for \cx, which used the "case flipping" table to upper case
2751             lower case letters, now just substracts 32. This is ASCII-specific,
2752             but the whole concept of \cx is ASCII-specific, so it seems
2753             reasonable.
2755       (iii) PCRE was using its character types table to recognize decimal and
2756             hexadecimal digits in the pattern. This is silly, because it handles
2757             only 0-9, a-f, and A-F, but the character types table is locale-
2758             specific, which means strange things might happen. A private
2759             table is now used for this - though it costs 256 bytes, a table is
2760             much faster than multiple explicit tests. Of course, the standard
2761             character types table is still used for matching digits in subject
2762             strings against \d.
2764       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2765             ending in _t are). So I've renamed it as ESC_tee.
2767    3. The first argument for regexec() in the POSIX wrapper should have been
2768       defined as "const".
2770    4. Changed pcretest to use malloc() for its buffers so that they can be
2771       Electric Fenced for debugging.
2773    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2774       to read one or more bytes before the start of the subject string. Often this
2775       had no effect on PCRE's behaviour, but in some circumstances it could
2776       provoke a segmentation fault.
2778    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2779       to try to read one or more bytes before the start of the subject string.
2781    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2782       UTF-8 support could misbehave in various ways if the subject string
2783       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2784       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2785       back over UTF-8 characters.)
2788    Version 4.2 14-Apr-03
2789    ---------------------
2791    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2793    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2794         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2795         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2796         [WIN_PREFIX]: new variable, "cyg" for Cygwin
2797         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2798           and BUILD_EXEEXT
2799         Note: automatic setting of the BUILD variables is not yet working
2800         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2801           compile-time but not at link-time
2802         [LINK]: use for linking executables only
2803         make different versions for Windows and non-Windows
2804         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2805           libraries
2806         [LINK_FOR_BUILD]: new variable
2807         [OBJEXT]: use throughout
2808         [EXEEXT]: use throughout
2809         <winshared>: new target
2810         <wininstall>: new target
2811         <dftables.o>: use native compiler
2812         <dftables>: use native linker
2813         <install>: handle Windows platform correctly
2814         <clean>: ditto
2815         <check>: ditto
2816         copy DLL to top builddir before testing
2818       As part of these changes, -no-undefined was removed again. This was reported
2819       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2820       in any case.
2822    3. Some tidies to get rid of compiler warnings:
2824       . In the match_data structure, match_limit was an unsigned long int, whereas
2825         match_call_count was an int. I've made them both unsigned long ints.
2827       . In pcretest the fact that a const uschar * doesn't automatically cast to
2828         a void * provoked a warning.
2830       . Turning on some more compiler warnings threw up some "shadow" variables
2831         and a few more missing casts.
2833    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2834       option, a class that contained a single character with a value between 128
2835       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2837    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2838       option, a class that contained several characters, but with at least one
2839       whose value was between 128 and 255 caused PCRE to crash.
2842    Version 4.1 12-Mar-03
2843    ---------------------
2845    1. Compiling with gcc -pedantic found a couple of places where casts were
2846    needed, and a string in dftables.c that was longer than standard compilers are
2847    required to support.
2849    2. Compiling with Sun's compiler found a few more places where the code could
2850    be tidied up in order to avoid warnings.
2852    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2853    first of these names is deprecated in the latest Autoconf in favour of the name
2854    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2855    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2856    analogy I have changed it to CFLAGS_FOR_BUILD.
2858    4. Added -no-undefined to the linking command in the Makefile, because this is
2859    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2860    linking step for the pcreposix library.
2862    5. PCRE was failing to diagnose the case of two named groups with the same
2863    name.
2865    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2866    literal character that is needed in the subject for a match, and scans along to
2867    ensure that it is present before embarking on the full matching process. This
2868    saves time in cases of nested unlimited repeats that are never going to match.
2869    Problem: the scan can take a lot of time if the subject is very long (e.g.
2870    megabytes), thus penalizing straightforward matches. It is now done only if the
2871    amount of subject to be scanned is less than 1000 bytes.
2873    7. A lesser problem with the same optimization is that it was recording the
2874    first character of an anchored pattern as "needed", thus provoking a search
2875    right along the subject, even when the first match of the pattern was going to
2876    fail. The "needed" character is now not set for anchored patterns, unless it
2877    follows something in the pattern that is of non-fixed length. Thus, it still
2878    fulfils its original purpose of finding quick non-matches in cases of nested
2879    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2882    Version 4.0 17-Feb-03
2883    ---------------------
2885    1. If a comment in an extended regex that started immediately after a meta-item
2886    extended to the end of string, PCRE compiled incorrect data. This could lead to
2887    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2889    2. Moved to autoconf 2.53 and libtool 1.4.2.
2891    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2892    the special perltest8 script is no longer needed - all the tests can be run
2893    from a single perltest script.
2895    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2896    by \s. It has now been removed in PCRE. This means it isn't recognized as
2897    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2898    class [:space:] *does* include VT, thereby creating a mess.
2900    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2901    space and tab.
2903    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2904    its new features into the main test script, reducing the number of scripts.
2906    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2907    were backward compatible, and made the (?i) apply to the whole pattern, as if
2908    /i were given. Now it behaves more logically, and applies the option setting
2909    only to what follows. PCRE has been changed to follow suit. However, if it
2910    finds options settings right at the start of the pattern, it extracts them into
2911    the global options, as before. Thus, they show up in the info data.
2913    8. Added support for the \Q...\E escape sequence. Characters in between are
2914    treated as literals. This is slightly different from Perl in that $ and @ are
2915    also handled as literals inside the quotes. In Perl, they will cause variable
2916    interpolation. Note the following examples:
2918        Pattern            PCRE matches      Perl matches
2920        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
2921        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2922        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2924    For compatibility with Perl, \Q...\E sequences are recognized inside character
2925    classes as well as outside them.
2927    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2928    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2929    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2930    signed/unsigned warnings.
2932    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2933    option for pcretest, so I've replaced it by a simple function that does just
2934    that job.
2936    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2937    "pcregrep -".
2939    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2940    Java package. This provides some syntactic sugar for simple cases of what my
2941    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2942    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2943    item, you can use this simplified notation. Note that only makes sense with
2944    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2945    greediness, whatever the setting of the PCRE_UNGREEDY option.
2947    13. A change of greediness default within a pattern was not taking effect at
2948    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2949    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2950    was abstracted outside.
2952    14. PCRE now supports the \G assertion. It is true when the current matching
2953    position is at the start point of the match. This differs from \A when the
2954    starting offset is non-zero. Used with the /g option of pcretest (or similar
2955    code), it works in the same way as it does for Perl's /g option. If all
2956    alternatives of a regex begin with \G, the expression is anchored to the start
2957    match position, and the "anchored" flag is set in the compiled expression.
2959    15. Some bugs concerning the handling of certain option changes within patterns
2960    have been fixed. These applied to options other than (?ims). For example,
2961    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2962    been the other way round. Some of this was related to change 7 above.
2964    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2965    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2966    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2967    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2969    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2970    mode. Unlike ".", it always matches newline, whatever the setting of
2971    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2972    assertions. Perl allows it, but it doesn't (in general) work because it can't
2973    calculate the length of the lookbehind. At least, that's the case for Perl
2974    5.8.0 - I've been told they are going to document that it doesn't work in
2975    future.
2977    18. Added an error diagnosis for escapes that PCRE does not support: these are
2978    \L, \l, \N, \P, \p, \U, \u, and \X.
2980    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2981    reading past the end of the pattern in cases such as /[abcd/.
2983    20. PCRE was getting more memory than necessary for patterns with classes that
2984    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2986    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2987    compiling PCRE for use with Virtual Pascal.
2989    22. Small fix to the Makefile to make it work properly if the build is done
2990    outside the source tree.
2992    23. Added a new extension: a condition to go with recursion. If a conditional
2993    subpattern starts with (?(R) the "true" branch is used if recursion has
2994    happened, whereas the "false" branch is used only at the top level.
2996    24. When there was a very long string of literal characters (over 255 bytes
2997    without UTF support, over 250 bytes with UTF support), the computation of how
2998    much memory was required could be incorrect, leading to segfaults or other
2999    strange effects.
3001    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
3002    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
3003    there was a subsequent back reference to those brackets. This meant that, for
3004    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
3005    possible to check for precisely this case. All we can do is abandon the
3006    optimization if .* occurs inside capturing brackets when there are any back
3007    references whatsoever. (See below for a better fix that came later.)
3009    26. The handling of the optimization for finding the first character of a
3010    non-anchored pattern, and for finding a character that is required later in the
3011    match were failing in some cases. This didn't break the matching; it just
3012    failed to optimize when it could. The way this is done has been re-implemented.
3014    27. Fixed typo in error message for invalid (?R item (it said "(?p").
3016    28. Added a new feature that provides some of the functionality that Perl
3017    provides with (?{...}). The facility is termed a "callout". The way it is done
3018    in PCRE is for the caller to provide an optional function, by setting
3019    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
3020    global variable. By default it is unset, which disables all calling out. To get
3021    the function called, the regex must include (?C) at appropriate points. This
3022    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
3023    This provides a means of identifying different callout points. When PCRE
3024    reaches such a point in the regex, if pcre_callout has been set, the external
3025    function is called. It is provided with data in a structure called
3026    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
3027    matching continues; if it returns a non-zero value, the match at the current
3028    point fails. However, backtracking will occur if possible. [This was changed
3029    later and other features added - see item 49 below.]
3031    29. pcretest is upgraded to test the callout functionality. It provides a
3032    callout function that displays information. By default, it shows the start of
3033    the match and the current position in the text. There are some new data escapes
3034    to vary what happens:
3036        \C+         in addition, show current contents of captured substrings
3037        \C-         do not supply a callout function
3038        \C!n        return 1 when callout number n is reached
3039        \C!n!m      return 1 when callout number n is reached for the mth time
3041    30. If pcregrep was called with the -l option and just a single file name, it
3042    output "<stdin>" if a match was found, instead of the file name.
3044    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
3045    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
3046    pcre_exec(). This saves a malloc/free per call. The default value of
3047    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
3048    when configuring.
3050    32. The default maximum size of a compiled pattern is 64K. There have been a
3051    few cases of people hitting this limit. The code now uses macros to handle the
3052    storing of links as offsets within the compiled pattern. It defaults to 2-byte
3053    links, but this can be changed to 3 or 4 bytes by --with-link-size when
3054    configuring. Tests 2 and 5 work only with 2-byte links because they output
3055    debugging information about compiled patterns.
3057    33. Internal code re-arrangements:
3059    (a) Moved the debugging function for printing out a compiled regex into
3060        its own source file (printint.c) and used #include to pull it into
3061        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
3062        separate copies.
3064    (b) Defined the list of op-code names for debugging as a macro in
3065        internal.h so that it is next to the definition of the opcodes.
3067    (c) Defined a table of op-code lengths for simpler skipping along compiled
3068        code. This is again a macro in internal.h so that it is next to the
3069        definition of the opcodes.
3071    34. Added support for recursive calls to individual subpatterns, along the
3072    lines of Robin Houston's patch (but implemented somewhat differently).
3074    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
3075    allow it to read and process whole directories in Win32. This code was
3076    contributed by Lionel Fourquaux; it has not been tested by me.
3078    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
3079    used to name a group. Names consist of alphanumerics and underscores, and must
3080    be unique. Back references use the syntax (?P=name) and recursive calls use
3081    (?P>name) which is a PCRE extension to the Python extension. Groups still have
3082    numbers. The function pcre_fullinfo() can be used after compilation to extract
3083    a name/number map. There are three relevant calls:
3085      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
3086      PCRE_INFO_NAMECOUNT            yields the number of entries
3087      PCRE_INFO_NAMETABLE            yields a pointer to the map.
3089    The map is a vector of fixed-size entries. The size of each entry depends on
3090    the length of the longest name used. The first two bytes of each entry are the
3091    group number, most significant byte first. There follows the corresponding
3092    name, zero terminated. The names are in alphabetical order.
3094    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
3095    case instead of 255. Making it the same both with and without UTF-8 support
3096    means that the same test output works with both.
3098    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
3099    calling malloc() with a zero argument.
3101    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
3102    optimization. I've improved things by keeping a bitmap of backreferences with
3103    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
3104    fact referenced, the optimization can be applied. It is unlikely that a
3105    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
3106    the match to follow \n) will appear inside brackets with a number greater than
3107    31, but if it does, any back reference > 31 suppresses the optimization.
3109    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
3110    of disabling numbered capturing parentheses. Any opening parenthesis that is
3111    not followed by ? behaves as if it were followed by ?: but named parentheses
3112    can still be used for capturing (and they will acquire numbers in the usual
3113    way).
3115    41. Redesigned the return codes from the match() function into yes/no/error so
3116    that errors can be passed back from deep inside the nested calls. A malloc
3117    failure while inside a recursive subpattern call now causes the
3118    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
3120    42. It is now possible to set a limit on the number of times the match()
3121    function is called in a call to pcre_exec(). This facility makes it possible to
3122    limit the amount of recursion and backtracking, though not in a directly
3123    obvious way, because the match() function is used in a number of different
3124    circumstances. The count starts from zero for each position in the subject
3125    string (for non-anchored patterns). The default limit is, for compatibility, a
3126    large number, namely 10 000 000. You can change this in two ways:
3128    (a) When configuring PCRE before making, you can use --with-match-limit=n
3129        to set a default value for the compiled library.
3131    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
3132        a different value is set. See 45 below.
3134    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3136    43. Added a new function pcre_config(int, void *) to enable run-time extraction
3137    of things that can be changed at compile time. The first argument specifies
3138    what is wanted and the second points to where the information is to be placed.
3139    The current list of available information is:
3141      PCRE_CONFIG_UTF8
3143    The output is an integer that is set to one if UTF-8 support is available;
3144    otherwise it is set to zero.
3148    The output is an integer that it set to the value of the code that is used for
3149    newline. It is either LF (10) or CR (13).
3153    The output is an integer that contains the number of bytes used for internal
3154    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3158    The output is an integer that contains the threshold above which the POSIX
3159    interface uses malloc() for output vectors. See item 31 above.
3163    The output is an unsigned integer that contains the default limit of the number
3164    of match() calls in a pcre_exec() execution. See 42 above.
3166    44. pcretest has been upgraded by the addition of the -C option. This causes it
3167    to extract all the available output from the new pcre_config() function, and to
3168    output it. The program then exits immediately.
3170    45. A need has arisen to pass over additional data with calls to pcre_exec() in
3171    order to support additional features. One way would have been to define
3172    pcre_exec2() (for example) with extra arguments, but this would not have been
3173    extensible, and would also have required all calls to the original function to
3174    be mapped to the new one. Instead, I have chosen to extend the mechanism that
3175    is used for passing in "extra" data from pcre_study().
3177    The pcre_extra structure is now exposed and defined in pcre.h. It currently
3178    contains the following fields:
3180      flags         a bitmap indicating which of the following fields are set
3181      study_data    opaque data from pcre_study()
3182      match_limit   a way of specifying a limit on match() calls for a specific
3183                      call to pcre_exec()
3184      callout_data  data for callouts (see 49 below)
3186    The flag bits are also defined in pcre.h, and are
3192    The pcre_study() function now returns one of these new pcre_extra blocks, with
3193    the actual study data pointed to by the study_data field, and the
3194    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3195    before. That is, this change is entirely upwards-compatible and requires no
3196    change to existing code.
3198    If you want to pass in additional data to pcre_exec(), you can either place it
3199    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3200    block.
3202    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3203    data string contains the escape sequence \M, pcretest calls pcre_exec() several
3204    times with different match limits, until it finds the minimum value needed for
3205    pcre_exec() to complete. The value is then output. This can be instructive; for
3206    most simple matches the number is quite small, but for pathological cases it
3207    gets very large very quickly.
3209    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3210    returns the size of the data block pointed to by the study_data field in a
3211    pcre_extra block, that is, the value that was passed as the argument to
3212    pcre_malloc() when PCRE was getting memory in which to place the information
3213    created by pcre_study(). The fourth argument should point to a size_t variable.
3214    pcretest has been extended so that this information is shown after a successful
3215    pcre_study() call when information about the compiled regex is being displayed.
3217    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3218    because what follows is always an absolute path. (Later: it turns out that this
3219    is more than cosmetic for MinGW, because it doesn't like empty path
3220    components.)
3222    49. Some changes have been made to the callout feature (see 28 above):
3224    (i)  A callout function now has three choices for what it returns:
3226           0  =>  success, carry on matching
3227         > 0  =>  failure at this point, but backtrack if possible
3228         < 0  =>  serious error, return this value from pcre_exec()
3230         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3231         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3232         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3233         use by callout functions. It will never be used by PCRE itself.
3235    (ii) The pcre_extra structure (see 45 above) has a void * field called
3236         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3237         pcre_callout_block structure has a field of the same name. The contents of
3238         the field passed in the pcre_extra structure are passed to the callout
3239         function in the corresponding field in the callout block. This makes it
3240         easier to use the same callout-containing regex from multiple threads. For
3241         testing, the pcretest program has a new data escape
3243           \C*n        pass the number n (may be negative) as callout_data
3245         If the callout function in pcretest receives a non-zero value as
3246         callout_data, it returns that value.
3248    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3249    there were some redundant $(CFLAGS) in commands that are now specified as
3250    $(LINK), which already includes $(CFLAGS).
3252    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3253    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3254    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3255    one-byte characters throughout. Note that case-insensitive matching applies
3256    only to characters whose values are less than 256. PCRE doesn't support the
3257    notion of cases for higher-valued characters.
3259    (i)   A character class whose characters are all within 0-255 is handled as
3260          a bit map, and the map is inverted for negative classes. Previously, a
3261          character > 255 always failed to match such a class; however it should
3262          match if the class was a negative one (e.g. [^ab]). This has been fixed.
3264    (ii)  A negated character class with a single character < 255 is coded as
3265          "not this character" (OP_NOT). This wasn't working properly when the test
3266          character was multibyte, either singly or repeated.
3268    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3269          mode, for example: \x{100}{2,3}.
3271    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3272          singly or repeated) now correctly test multibyte characters. However,
3273          PCRE doesn't recognize any characters with values greater than 255 as
3274          digits, spaces, or word characters. Such characters always match \D, \S,
3275          and \W, and never match \d, \s, or \w.
3277    (v)   Classes may now contain characters and character ranges with values
3278          greater than 255. For example: [ab\x{100}-\x{400}].
3280    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
3281          PCRE in UTF-8 mode.
3283    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3284    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3285    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3286    value.)
3288    53. The single man page has become too large. I have therefore split it up into
3289    a number of separate man pages. These also give rise to individual HTML pages;
3290    these are now put in a separate directory, and there is an index.html page that
3291    lists them all. Some hyperlinking between the pages has been installed.
3293    54. Added convenience functions for handling named capturing parentheses.
3295    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3296    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3297    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3298    are faulted.
3300    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3301    calling configure. These values are used when compiling the dftables.c program
3302    which is run to generate the source of the default character tables. They
3303    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3304    you will need to set these values.
3306    57. Updated the building process for Windows DLL, as provided by Fred Cox.
3309    Version 3.9 02-Jan-02
3310    ---------------------
3312    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3314    2. If --disable-static was given, the building process failed when trying to
3315    build pcretest and pcregrep. (For some reason it was using libtool to compile
3316    them, which is not right, as they aren't part of the library.)
3319    Version 3.8 18-Dec-01
3320    ---------------------
3322    1. The experimental UTF-8 code was completely screwed up. It was packing the
3323    bytes in the wrong order. How dumb can you get?
3326    Version 3.7 29-Oct-01
3327    ---------------------
3329    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3330    This caused pcretest, when used on the test data, to segfault. Unfortunately,
3331    this didn't happen under Solaris 8, where I normally test things.
3333    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3334    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3335    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3336    this fix an hour or so after the initial 3.7 release.)
3339    Version 3.6 23-Oct-01
3340    ---------------------
3342    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3343    offsets passed as NULL with zero offset count.
3345    2. The config.guess and config.sub files had not been updated when I moved to
3346    the latest autoconf.
3349    Version 3.5 15-Aug-01
3350    ---------------------
3352    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
3353    had been forgotten.
3355    2. By using declared but undefined structures, we can avoid using "void"
3356    definitions in pcre.h while keeping the internal definitions of the structures
3357    private.
3359    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
3360    user point of view, this means that both static and shared libraries are built
3361    by default, but this can be individually controlled. More of the work of
3362    handling this static/shared cases is now inside libtool instead of PCRE's make
3363    file.
3365    4. The pcretest utility is now installed along with pcregrep because it is
3366    useful for users (to test regexs) and by doing this, it automatically gets
3367    relinked by libtool. The documentation has been turned into a man page, so
3368    there are now .1, .txt, and .html versions in /doc.
3370    5. Upgrades to pcregrep:
3371       (i)   Added long-form option names like gnu grep.
3372       (ii)  Added --help to list all options with an explanatory phrase.
3373       (iii) Added -r, --recursive to recurse into sub-directories.
3374       (iv)  Added -f, --file to read patterns from a file.
3376    6. pcre_exec() was referring to its "code" argument before testing that
3377    argument for NULL (and giving an error if it was NULL).
3379    7. Upgraded Makefile.in to allow for compiling in a different directory from
3380    the source directory.
3382    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
3383    options bits, the pointer it was passed was to an int instead of to an unsigned
3384    long int. This mattered only on 64-bit systems.
3386    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3387    generated) instead of pcre.in, which it its source. Also made the same change
3388    in several of the .c files.
3390    10. A new release of gcc defines printf() as a macro, which broke pcretest
3391    because it had an ifdef in the middle of a string argument for printf(). Fixed
3392    by using separate calls to printf().
3394    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
3395    script, to force use of CR or LF instead of \n in the source. On non-Unix
3396    systems, the value can be set in config.h.
3398    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
3399    absolute limit. Changed the text of the error message to make this clear, and
3400    likewise updated the man page.
3402    13. The limit of 99 on the number of capturing subpatterns has been removed.
3403    The new limit is 65535, which I hope will not be a "real" limit.
3406    Version 3.4 22-Aug-00
3407    ---------------------
3409    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
3411    2. Diagnose condition (?(0) as an error instead of crashing on matching.
3414    Version 3.3 01-Aug-00
3415    ---------------------
3417    1. If an octal character was given, but the value was greater than \377, it
3418    was not getting masked to the least significant bits, as documented. This could
3419    lead to crashes in some systems.
3421    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
3422    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
3424    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
3425    These just pass their arguments on to (pcre_free)(), but they are provided
3426    because some uses of PCRE bind it to non-C systems that can call its functions,
3427    but cannot call free() or pcre_free() directly.
3429    4. Add "make test" as a synonym for "make check". Corrected some comments in
3430    the Makefile.
3432    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
3433    Makefile.
3435    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
3436    command called pgrep for grepping around the active processes.
3438    7. Added the beginnings of support for UTF-8 character strings.
3440    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
3441    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
3442    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
3443    out for the ar command.)
3446    Version 3.2 12-May-00
3447    ---------------------
3449    This is purely a bug fixing release.
3451    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
3452    of ZA. This was just one example of several cases that could provoke this bug,
3453    which was introduced by change 9 of version 2.00. The code for breaking
3454    infinite loops after an iteration that matches an empty string was't working
3455    correctly.
3457    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3458    when matched against abbab (for example). After matching an empty string, it
3459    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3460    caused it to match further down the string than it should.
3462    3. The code contained an inclusion of sys/types.h. It isn't clear why this
3463    was there because it doesn't seem to be needed, and it causes trouble on some
3464    systems, as it is not a Standard C header. It has been removed.
3466    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3467    were reported on the Macintosh. The changes were from
3469      while ((c = *(++ptr)) != 0 && c != '\n');
3470    to
3471      while ((c = *(++ptr)) != 0 && c != '\n') ;
3473    Totally extraordinary, but if that's what it takes...
3475    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3476    available. Added HAVE_BCOPY and an autoconf test for it; if neither
3477    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3478    assumes the way PCRE uses memmove() (always moving upwards).
3480    6. PCRE is being used in one environment where strchr() is not available. There
3481    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3482    faster code anyway.
3485    Version 3.1 09-Feb-00
3486    ---------------------
3488    The only change in this release is the fixing of some bugs in Makefile.in for
3489    the "install" target:
3491    (1) It was failing to install pcreposix.h.
3493    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3496    Version 3.0 01-Feb-00
3497    ---------------------
3499    1. Add support for the /+ modifier to perltest (to output $` like it does in
3500    pcretest).
3502    2. Add support for the /g modifier to perltest.
3504    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3505    matches null strings.
3507    4. Fix perltest so that it doesn't do unwanted things when fed an empty
3508    pattern. Perl treats empty patterns specially - it reuses the most recent
3509    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3510    effect.
3512    5. The POSIX interface was broken in that it was just handing over the POSIX
3513    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3514    required a bigger vector, with some working space on the end. This means that
3515    the POSIX wrapper now has to get and free some memory, and copy the results.
3517    6. Added some simple autoconf support, placing the test data and the
3518    documentation in separate directories, re-organizing some of the
3519    information files, and making it build pcre-config (a GNU standard). Also added
3520    libtool support for building PCRE as a shared library, which is now the
3521    default.
3523    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3524    09 are not valid octal constants. Single digits will be used for minor values
3525    less than 10.
3527    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3528    existing programs that set these in the POSIX interface can use PCRE without
3529    modification.
3531    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3532    return all that pcre_info() returns, plus additional data. The pcre_info()
3533    function is retained for compatibility, but is considered to be obsolete.
3535    10. Added experimental recursion feature (?R) to handle one common case that
3536    Perl 5.6 will be able to do with (?p{...}).
3538    11. Added support for POSIX character classes like [:alpha:], which Perl is
3539    adopting.
3542    Version 2.08 31-Aug-99
3543    ----------------------
3545    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3546    trying to match at the startoffset position, but instead was moving forward to
3547    the next newline as if a previous match had failed.
3549    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3550    and could get into a loop if a null string was matched other than at the start
3551    of the subject.
3553    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3554    be distinguished at compile time, and for completeness also added PCRE_DATE.
3556    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3557    in GnuWin32 environments.
3560    Version 2.07 29-Jul-99
3561    ----------------------
3563    1. The documentation is now supplied in plain text form and HTML as well as in
3564    the form of man page sources.
3566    2. C++ compilers don't like assigning (void *) values to other pointer types.
3567    In particular this affects malloc(). Although there is no problem in Standard
3568    C, I've put in casts to keep C++ compilers happy.
3570    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call