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

Diff of /code/trunk/ChangeLog

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

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

Legend:
Removed from v.69  
changed lines
  Added in v.566

  ViewVC Help
Powered by ViewVC 1.1.5