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

Diff of /code/trunk/ChangeLog

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

revision 25 by nigel, Sat Feb 24 21:38:45 2007 UTC revision 576 by ph10, Sun Nov 21 18:45:10 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    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
87        as a starting offset was within the subject string. There is now a new
88        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
89        negative or greater than the length of the string. In order to test this,
90        pcretest is extended to allow the setting of negative starting offsets.
91    
92    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
93        starting offset points to the beginning of a UTF-8 character was
94        unnecessarily clumsy. I tidied it up.
95    
96    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
97        bad UTF-8 sequence and one that is incomplete.
98    
99    17. Nobody had reported that the --include_dir option, which was added in
100        release 7.7 should have been called --include-dir (hyphen, not underscore)
101        for compatibility with GNU grep. I have changed it to --include-dir, but
102        left --include_dir as an undocumented synonym, and the same for
103        --exclude-dir, though that is not available in GNU grep, at least as of
104        release 2.5.4.
105    
106    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
107        characters from a string of bytes) have been redefined so as not to use
108        loops, in order to improve performance in some environments. At the same
109        time, I abstracted some of the common code into auxiliary macros to save
110        repetition (this should not affect the compiled code).
111    
112    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
113        compile-time error is now given if \c is not followed by an ASCII
114        character, that is, a byte less than 128. (In EBCDIC mode, the code is
115        different, and any byte value is allowed.)
116    
117    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
118        START_OPTIMIZE option, which is now allowed at compile time - but just
119        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
120        to pcregrep and other applications that have no direct access to PCRE
121        options. The new /Y option in pcretest sets this option when calling
122        pcre_compile().
123    
124    
125    Version 8.10 25-Jun-2010
126    ------------------------
127    
128    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
129        THEN.
130    
131    2.  (*ACCEPT) was not working when inside an atomic group.
132    
133    3.  Inside a character class, \B is treated as a literal by default, but
134        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
135        causes the error). The code is unchanged, but I tidied the documentation.
136    
137    4.  Inside a character class, PCRE always treated \R and \X as literals,
138        whereas Perl faults them if its -w option is set. I have changed PCRE so
139        that it faults them when PCRE_EXTRA is set.
140    
141    5.  Added support for \N, which always matches any character other than
142        newline. (It is the same as "." when PCRE_DOTALL is not set.)
143    
144    6.  When compiling pcregrep with newer versions of gcc which may have
145        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
146        declared with attribute warn_unused_result" were given. Just casting the
147        result to (void) does not stop the warnings; a more elaborate fudge is
148        needed. I've used a macro to implement this.
149    
150    7.  Minor change to pcretest.c to avoid a compiler warning.
151    
152    8.  Added four artifical Unicode properties to help with an option to make
153        \s etc use properties (see next item). The new properties are: Xan
154        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
155    
156    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
157        use Unicode properties. (*UCP) at the start of a pattern can be used to set
158        this option. Modified pcretest to add /W to test this facility. Added
159        REG_UCP to make it available via the POSIX interface.
160    
161    10. Added --line-buffered to pcregrep.
162    
163    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
164        studied, and the match started with a letter with a code point greater than
165        127 whose first byte was different to the first byte of the other case of
166        the letter, the other case of this starting letter was not recognized
167        (#976).
168    
169    12. If a pattern that was studied started with a repeated Unicode property
170        test, for example, \p{Nd}+, there was the theoretical possibility of
171        setting up an incorrect bitmap of starting bytes, but fortunately it could
172        not have actually happened in practice until change 8 above was made (it
173        added property types that matched character-matching opcodes).
174    
175    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
176        possible starting bytes for non-anchored patterns.
177    
178    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
179        \R, and also a number of cases that involve Unicode properties, both
180        explicit and implicit when PCRE_UCP is set.
181    
182    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
183        input, it could crash or give wrong results if characters with values
184        greater than 0xc0 were present in the subject string. (Detail: it assumed
185        UTF-8 input when processing these items.)
186    
187    16. Added a lot of (int) casts to avoid compiler warnings in systems where
188        size_t is 64-bit (#991).
189    
190    17. Added a check for running out of memory when PCRE is compiled with
191        --disable-stack-for-recursion (#990).
192    
193    18. If the last data line in a file for pcretest does not have a newline on
194        the end, a newline was missing in the output.
195    
196    19. The default pcre_chartables.c file recognizes only ASCII characters (values
197        less than 128) in its various bitmaps. However, there is a facility for
198        generating tables according to the current locale when PCRE is compiled. It
199        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
200        characters, are recognized by isspace() and therefore were getting set in
201        these tables, and indeed these tables seem to approximate to ISO 8859. This
202        caused a problem in UTF-8 mode when pcre_study() was used to create a list
203        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
204        which of course cannot start UTF-8 characters. I have changed the code so
205        that only real ASCII characters (less than 128) and the correct starting
206        bytes for UTF-8 encodings are set for characters greater than 127 when in
207        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
208        altogether.)
209    
210    20. Added the /T option to pcretest so as to be able to run tests with non-
211        standard character tables, thus making it possible to include the tests
212        used for 19 above in the standard set of tests.
213    
214    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
215        reference to a subpattern the other side of a comment that contains an
216        opening parenthesis caused either an internal compiling error, or a
217        reference to the wrong subpattern.
218    
219    
220    Version 8.02 19-Mar-2010
221    ------------------------
222    
223    1.  The Unicode data tables have been updated to Unicode 5.2.0.
224    
225    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
226        configured.
227    
228    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
229        original author of that file, following a query about its status.
230    
231    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
232        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
233    
234    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
235        quantifier applied to a forward-referencing subroutine call, could compile
236        incorrect code or give the error "internal error: previously-checked
237        referenced subpattern not found".
238    
239    6.  Both MS Visual Studio and Symbian OS have problems with initializing
240        variables to point to external functions. For these systems, therefore,
241        pcre_malloc etc. are now initialized to local functions that call the
242        relevant global functions.
243    
244    7.  There were two entries missing in the vectors called coptable and poptable
245        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
246        I've fixed the data, and added a kludgy way of testing at compile time that
247        the lengths are correct (equal to the number of opcodes).
248    
249    8.  Following on from 7, I added a similar kludge to check the length of the
250        eint vector in pcreposix.c.
251    
252    9.  Error texts for pcre_compile() are held as one long string to avoid too
253        much relocation at load time. To find a text, the string is searched,
254        counting zeros. There was no check for running off the end of the string,
255        which could happen if a new error number was added without updating the
256        string.
257    
258    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
259    
260    11. \K was not working if it appeared in an atomic group or in a group that
261        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
262        \K is "not well defined" if used in an assertion. PCRE now accepts it if
263        the assertion is positive, but not if it is negative.
264    
265    12. Change 11 fortuitously reduced the size of the stack frame used in the
266        "match()" function of pcre_exec.c by one pointer. Forthcoming
267        implementation of support for (*MARK) will need an extra pointer on the
268        stack; I have reserved it now, so that the stack frame size does not
269        decrease.
270    
271    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
272        item in branch that calls a recursion is a subroutine call - as in the
273        second branch in the above example - was incorrectly given the compile-
274        time error "recursive call could loop indefinitely" because pcre_compile()
275        was not correctly checking the subroutine for matching a non-empty string.
276    
277    14. The checks for overrunning compiling workspace could trigger after an
278        overrun had occurred. This is a "should never occur" error, but it can be
279        triggered by pathological patterns such as hundreds of nested parentheses.
280        The checks now trigger 100 bytes before the end of the workspace.
281    
282    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
283    
284    
285    Version 8.01 19-Jan-2010
286    ------------------------
287    
288    1.  If a pattern contained a conditional subpattern with only one branch (in
289        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
290        computed the wrong minimum data length (which is of course zero for such
291        subpatterns). This could cause incorrect "no match" results.
292    
293    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
294        the pattern is reset in the first branch, pcre_compile() failed with
295        "internal error: code overflow at offset...". This happened only when
296        the reset was to the original external option setting. (An optimization
297        abstracts leading options settings into an external setting, which was the
298        cause of this.)
299    
300    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
301        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
302        assertion pattern did not match (meaning that the assertion was true), it
303        was incorrectly treated as false if the SKIP had been reached during the
304        matching. This also applied to assertions used as conditions.
305    
306    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
307        assertion subpattern, including such a pattern used as a condition,
308        unpredictable results occurred, instead of the error return
309        PCRE_ERROR_DFA_UITEM.
310    
311    5.  The C++ GlobalReplace function was not working like Perl for the special
312        situation when an empty string is matched. It now does the fancy magic
313        stuff that is necessary.
314    
315    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
316        removed. (These were left over from very, very early versions of PCRE.)
317    
318    7.  Some cosmetic changes to the code to make life easier when compiling it
319        as part of something else:
320    
321        (a) Change DEBUG to PCRE_DEBUG.
322    
323        (b) In pcre_compile(), rename the member of the "branch_chain" structure
324            called "current" as "current_branch", to prevent a collision with the
325            Linux macro when compiled as a kernel module.
326    
327        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
328            prevent a collision with the Linux macro when compiled as a kernel
329            module.
330    
331    8.  In pcre_compile() there are some checks for integer overflows that used to
332        cast potentially large values to (double). This has been changed to that
333        when building, a check for int64_t is made, and if it is found, it is used
334        instead, thus avoiding the use of floating point arithmetic. (There is no
335        other use of FP in PCRE.) If int64_t is not found, the fallback is to
336        double.
337    
338    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
339        2005 (difference between two addresses compared to an unsigned value).
340    
341    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
342        custom one, because of the following reported problem in Windows:
343    
344          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
345              under Win32.
346          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
347              therefore missing the function definition.
348          - The compiler thus generates a "C" signature for the test function.
349          - The linker fails to find the "C" function.
350          - PCRE fails to configure if asked to do so against libbz2.
351    
352    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
353        messages were output:
354    
355          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
356          rerunning libtoolize, to keep the correct libtool macros in-tree.
357          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
358    
359        I have done both of these things.
360    
361    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
362        most of the time, it *can* run out if it is given a pattern that contains a
363        runaway infinite recursion. I updated the discussion in the pcrestack man
364        page.
365    
366    13. Now that we have gone to the x.xx style of version numbers, the minor
367        version may start with zero. Using 08 or 09 is a bad idea because users
368        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
369        interpreted as invalid octal numbers. I've updated the previous comment in
370        configure.ac, and also added a check that gives an error if 08 or 09 are
371        used.
372    
373    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
374        causing partial matching to fail when the end of the subject matched \W
375        in a UTF-8 pattern where \W was quantified with a minimum of 3.
376    
377    15. There were some discrepancies between the declarations in pcre_internal.h
378        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
379        their definitions. The declarations used "const uschar *" and the
380        definitions used USPTR. Even though USPTR is normally defined as "const
381        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
382        reported that: "This difference in casting confuses some C++ compilers, for
383        example, SunCC recognizes above declarations as different functions and
384        generates broken code for hbpcre." I have changed the declarations to use
385        USPTR.
386    
387    16. GNU libtool is named differently on some systems. The autogen.sh script now
388        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
389        (FreeBSD).
390    
391    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
392        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
393        comment: "Figure out how to create a longlong from a string: strtoll and
394        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
395        instance, but it only takes 2 args instead of 3!"
396    
397    18. A subtle bug concerned with back references has been fixed by a change of
398        specification, with a corresponding code fix. A pattern such as
399        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
400        refers, was giving matches when it shouldn't. For example, xa=xaaa would
401        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
402        same bug. Such groups have to be quantified to be useful, or contained
403        inside another quantified group. (If there's no repetition, the reference
404        can never match.) The problem arises because, having left the group and
405        moved on to the rest of the pattern, a later failure that backtracks into
406        the group uses the captured value from the final iteration of the group
407        rather than the correct earlier one. I have fixed this in PCRE by forcing
408        any group that contains a reference to itself to be an atomic group; that
409        is, there cannot be any backtracking into it once it has completed. This is
410        similar to recursive and subroutine calls.
411    
412    
413    Version 8.00 19-Oct-09
414    ----------------------
415    
416    1.  The table for translating pcre_compile() error codes into POSIX error codes
417        was out-of-date, and there was no check on the pcre_compile() error code
418        being within the table. This could lead to an OK return being given in
419        error.
420    
421    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
422        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
423        in a Windows environment.
424    
425    3.  The pcregrep --count option prints the count for each file even when it is
426        zero, as does GNU grep. However, pcregrep was also printing all files when
427        --files-with-matches was added. Now, when both options are given, it prints
428        counts only for those files that have at least one match. (GNU grep just
429        prints the file name in this circumstance, but including the count seems
430        more useful - otherwise, why use --count?) Also ensured that the
431        combination -clh just lists non-zero counts, with no names.
432    
433    4.  The long form of the pcregrep -F option was incorrectly implemented as
434        --fixed_strings instead of --fixed-strings. This is an incompatible change,
435        but it seems right to fix it, and I didn't think it was worth preserving
436        the old behaviour.
437    
438    5.  The command line items --regex=pattern and --regexp=pattern were not
439        recognized by pcregrep, which required --regex pattern or --regexp pattern
440        (with a space rather than an '='). The man page documented the '=' forms,
441        which are compatible with GNU grep; these now work.
442    
443    6.  No libpcreposix.pc file was created for pkg-config; there was just
444        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
445    
446    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
447        when UCP support is not needed, by modifying the Python script that
448        generates it from Unicode data files. This should not matter if the module
449        is correctly used as a library, but I received one complaint about 50K of
450        unwanted data. My guess is that the person linked everything into his
451        program rather than using a library. Anyway, it does no harm.
452    
453    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
454        was a minimum greater than 1 for a wide character in a possessive
455        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
456        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
457        character. Chaos in the form of incorrect output or a compiling loop could
458        result.
459    
460    9.  The restrictions on what a pattern can contain when partial matching is
461        requested for pcre_exec() have been removed. All patterns can now be
462        partially matched by this function. In addition, if there are at least two
463        slots in the offset vector, the offset of the earliest inspected character
464        for the match and the offset of the end of the subject are set in them when
465        PCRE_ERROR_PARTIAL is returned.
466    
467    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
468        synonymous with PCRE_PARTIAL, for backwards compatibility, and
469        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
470        and may be more useful for multi-segment matching.
471    
472    11. Partial matching with pcre_exec() is now more intuitive. A partial match
473        used to be given if ever the end of the subject was reached; now it is
474        given only if matching could not proceed because another character was
475        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
476        string "Z", which now yields "no match" instead of "partial match". In the
477        case of pcre_dfa_exec(), "no match" is given if every matching path for the
478        final character ended with (*FAIL).
479    
480    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
481        if the pattern had a "must contain" character that was already found in the
482        earlier partial match, unless partial matching was again requested. For
483        example, with the pattern /dog.(body)?/, the "must contain" character is
484        "g". If the first part-match was for the string "dog", restarting with
485        "sbody" failed. This bug has been fixed.
486    
487    13. The string returned by pcre_dfa_exec() after a partial match has been
488        changed so that it starts at the first inspected character rather than the
489        first character of the match. This makes a difference only if the pattern
490        starts with a lookbehind assertion or \b or \B (\K is not supported by
491        pcre_dfa_exec()). It's an incompatible change, but it makes the two
492        matching functions compatible, and I think it's the right thing to do.
493    
494    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
495        so that the demonstration program is easily available in environments where
496        PCRE has not been installed from source.
497    
498    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
499        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
500        library.
501    
502    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
503        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
504        is not the first non-POSIX option to be added. Clearly some people find
505        these options useful.
506    
507    17. If a caller to the POSIX matching function regexec() passes a non-zero
508        value for nmatch with a NULL value for pmatch, the value of
509        nmatch is forced to zero.
510    
511    18. RunGrepTest did not have a test for the availability of the -u option of
512        the diff command, as RunTest does. It now checks in the same way as
513        RunTest, and also checks for the -b option.
514    
515    19. If an odd number of negated classes containing just a single character
516        interposed, within parentheses, between a forward reference to a named
517        subpattern and the definition of the subpattern, compilation crashed with
518        an internal error, complaining that it could not find the referenced
519        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
520        [The bug was that it was starting one character too far in when skipping
521        over the character class, thus treating the ] as data rather than
522        terminating the class. This meant it could skip too much.]
523    
524    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
525        /g option in pcretest when the pattern contains \K, which makes it possible
526        to have an empty string match not at the start, even when the pattern is
527        anchored. Updated pcretest and pcredemo to use this option.
528    
529    21. If the maximum number of capturing subpatterns in a recursion was greater
530        than the maximum at the outer level, the higher number was returned, but
531        with unset values at the outer level. The correct (outer level) value is
532        now given.
533    
534    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
535        PCRE did not set those parentheses (unlike Perl). I have now found a way to
536        make it do so. The string so far is captured, making this feature
537        compatible with Perl.
538    
539    23. The tests have been re-organized, adding tests 11 and 12, to make it
540        possible to check the Perl 5.10 features against Perl 5.10.
541    
542    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
543        pattern matches a fixed length string. PCRE did not allow this; now it
544        does. Neither allows recursion.
545    
546    25. I finally figured out how to implement a request to provide the minimum
547        length of subject string that was needed in order to match a given pattern.
548        (It was back references and recursion that I had previously got hung up
549        on.) This code has now been added to pcre_study(); it finds a lower bound
550        to the length of subject needed. It is not necessarily the greatest lower
551        bound, but using it to avoid searching strings that are too short does give
552        some useful speed-ups. The value is available to calling programs via
553        pcre_fullinfo().
554    
555    26. While implementing 25, I discovered to my embarrassment that pcretest had
556        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
557        study optimizations had never been tested with that matching function.
558        Oops. What is worse, even when it was passed study data, there was a bug in
559        pcre_dfa_exec() that meant it never actually used it. Double oops. There
560        were also very few tests of studied patterns with pcre_dfa_exec().
561    
562    27. If (?| is used to create subpatterns with duplicate numbers, they are now
563        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
564        on the other side of the coin, they are no longer allowed to have different
565        names, because these cannot be distinguished in PCRE, and this has caused
566        confusion. (This is a difference from Perl.)
567    
568    28. When duplicate subpattern names are present (necessarily with different
569        numbers, as required by 27 above), and a test is made by name in a
570        conditional pattern, either for a subpattern having been matched, or for
571        recursion in such a pattern, all the associated numbered subpatterns are
572        tested, and the overall condition is true if the condition is true for any
573        one of them. This is the way Perl works, and is also more like the way
574        testing by number works.
575    
576    
577    Version 7.9 11-Apr-09
578    ---------------------
579    
580    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
581        (pcretest), all targets were linked against these libraries. This included
582        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
583        libraries. This caused unwanted dependencies to be created. This problem
584        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
585        pcretest is linked with readline.
586    
587    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
588        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
589        moved outside it again, because FALSE and TRUE are already defined in AIX,
590        but BOOL is not.
591    
592    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
593        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
594    
595    4.  The pcregrep documentation said spaces were inserted as well as colons (or
596        hyphens) following file names and line numbers when outputting matching
597        lines. This is not true; no spaces are inserted. I have also clarified the
598        wording for the --colour (or --color) option.
599    
600    5.  In pcregrep, when --colour was used with -o, the list of matching strings
601        was not coloured; this is different to GNU grep, so I have changed it to be
602        the same.
603    
604    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
605        each matching line was coloured. Now it goes on to look for further matches
606        of any of the test patterns, which is the same behaviour as GNU grep.
607    
608    7.  A pattern that could match an empty string could cause pcregrep to loop; it
609        doesn't make sense to accept an empty string match in pcregrep, so I have
610        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
611        seems to be how GNU grep behaves.
612    
613    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
614        start or after a newline", because the conditional assertion was not being
615        correctly handled. The rule now is that both the assertion and what follows
616        in the first alternative must satisfy the test.
617    
618    9.  If auto-callout was enabled in a pattern with a conditional group whose
619        condition was an assertion, PCRE could crash during matching, both with
620        pcre_exec() and pcre_dfa_exec().
621    
622    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
623        used for matching.
624    
625    11. Unicode property support in character classes was not working for
626        characters (bytes) greater than 127 when not in UTF-8 mode.
627    
628    12. Added the -M command line option to pcretest.
629    
630    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
631    
632    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
633    
634    16. Added comments and documentation about mis-use of no_arg in the C++
635        wrapper.
636    
637    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
638        from Martin Jerabek that uses macro names for all relevant character and
639        string constants.
640    
641    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
642        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
643        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
644        these, but not everybody uses configure.
645    
646    19. A conditional group that had only one branch was not being correctly
647        recognized as an item that could match an empty string. This meant that an
648        enclosing group might also not be so recognized, causing infinite looping
649        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
650        with the subject "ab", where knowledge that the repeated group can match
651        nothing is needed in order to break the loop.
652    
653    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
654        exec(), but without supplying a callout function, matching went wrong.
655    
656    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
657        leak if the size of the offset vector was greater than 30. When the vector
658        is smaller, the saved offsets during recursion go onto a local stack
659        vector, but for larger vectors malloc() is used. It was failing to free
660        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
661        error, in fact).
662    
663    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
664        heapframe that is used only when UTF-8 support is enabled. This caused no
665        problem, but was untidy.
666    
667    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
668        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
669        included within another project.
670    
671    24. Steven Van Ingelgem's patches to add more options to the CMake support,
672        slightly modified by me:
673    
674          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
675              not building pcregrep.
676    
677          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
678              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
679    
680    25. Forward references, both numeric and by name, in patterns that made use of
681        duplicate group numbers, could behave incorrectly or give incorrect errors,
682        because when scanning forward to find the reference group, PCRE was not
683        taking into account the duplicate group numbers. A pattern such as
684        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
685    
686    26. Changed a few more instances of "const unsigned char *" to USPTR, making
687        the feature of a custom pointer more persuasive (as requested by a user).
688    
689    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
690        pcretest.c, inside #ifndefs, because it seems they are sometimes already
691        pre-defined.
692    
693    28. Added support for (*UTF8) at the start of a pattern.
694    
695    29. Arrange for flags added by the "release type" setting in CMake to be shown
696        in the configuration summary.
697    
698    
699    Version 7.8 05-Sep-08
700    ---------------------
701    
702    1.  Replaced UCP searching code with optimized version as implemented for Ad
703        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
704        stage table and inline lookup instead of a function, giving speed ups of 2
705        to 5 times on some simple patterns that I tested. Permission was given to
706        distribute the MultiStage2.py script that generates the tables (it's not in
707        the tarball, but is in the Subversion repository).
708    
709    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
710        scripts.
711    
712    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
713        a group with a zero qualifier. The result of the study could be incorrect,
714        or the function might crash, depending on the pattern.
715    
716    4.  Caseless matching was not working for non-ASCII characters in back
717        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
718        It now works when Unicode Property Support is available.
719    
720    5.  In pcretest, an escape such as \x{de} in the data was always generating
721        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
722        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
723        truncation.
724    
725    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
726    
727    7.  Added two (int) casts to pcregrep when printing the difference of two
728        pointers, in case they are 64-bit values.
729    
730    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
731        test 2 if it fails.
732    
733    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
734        and a #define of that name to empty if it is not externally set. This is to
735        allow users of MSVC to set it if necessary.
736    
737    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
738        the convenience functions in the pcre_get.c source file.
739    
740    11. An option change at the start of a pattern that had top-level alternatives
741        could cause overwriting and/or a crash. This command provoked a crash in
742        some environments:
743    
744          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
745    
746        This potential security problem was recorded as CVE-2008-2371.
747    
748    12. For a pattern where the match had to start at the beginning or immediately
749        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
750        pcre_dfa_exec() could read past the end of the passed subject if there was
751        no match. To help with detecting such bugs (e.g. with valgrind), I modified
752        pcretest so that it places the subject at the end of its malloc-ed buffer.
753    
754    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
755        exec() might read past the end of the data buffer in UTF-8 mode.
756    
757    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
758        the data contained the byte 0x85 as part of a UTF-8 character within its
759        first line. This applied both to normal and DFA matching.
760    
761    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
762        /^[^d]*?$/8 failed to match "abc".
763    
764    16. Added a missing copyright notice to pcrecpp_internal.h.
765    
766    17. Make it more clear in the documentation that values returned from
767        pcre_exec() in ovector are byte offsets, not character counts.
768    
769    18. Tidied a few places to stop certain compilers from issuing warnings.
770    
771    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
772        supplied by Stefan Weber. I made a further small update for 7.8 because
773        there is a change of source arrangements: the pcre_searchfuncs.c module is
774        replaced by pcre_ucd.c.
775    
776    
777    Version 7.7 07-May-08
778    ---------------------
779    
780    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
781        a string to a long long, pretend we don't even have a long long." This is
782        done by checking for the strtoq, strtoll, and _strtoi64 functions.
783    
784    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
785        pre-7.6 versions, which defined a global no_arg variable instead of putting
786        it in the RE class. (See also #8 below.)
787    
788    3.  Remove a line of dead code, identified by coverity and reported by Nuno
789        Lopes.
790    
791    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
792    
793        (1) The include/exclude patterns were being applied to the whole pathnames
794            of files, instead of just to the final components.
795    
796        (2) If there was more than one level of directory, the subdirectories were
797            skipped unless they satisfied the include/exclude conditions. This is
798            inconsistent with GNU grep (and could even be seen as contrary to the
799            pcregrep specification - which I improved to make it absolutely clear).
800            The action now is always to scan all levels of directory, and just
801            apply the include/exclude patterns to regular files.
802    
803    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
804        --exclude_dir in the tests to avoid scanning .svn directories.
805    
806    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
807        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
808        doesn't support NULs in patterns.
809    
810    7.  Added some missing "const"s to declarations of static tables in
811        pcre_compile.c and pcre_dfa_exec.c.
812    
813    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
814        caused by fix #2  above. (Subsequently also a second patch to fix the
815        first patch. And a third patch - this was a messy problem.)
816    
817    9.  Applied Craig's patch to remove the use of push_back().
818    
819    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
820        matching function regexec().
821    
822    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
823        which, however, unlike Perl's \g{...}, are subroutine calls, not back
824        references. PCRE supports relative numbers with this syntax (I don't think
825        Oniguruma does).
826    
827    12. Previously, a group with a zero repeat such as (...){0} was completely
828        omitted from the compiled regex. However, this means that if the group
829        was called as a subroutine from elsewhere in the pattern, things went wrong
830        (an internal error was given). Such groups are now left in the compiled
831        pattern, with a new opcode that causes them to be skipped at execution
832        time.
833    
834    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
835        to the way PCRE behaves:
836    
837        (a) A lone ] character is dis-allowed (Perl treats it as data).
838    
839        (b) A back reference to an unmatched subpattern matches an empty string
840            (Perl fails the current match path).
841    
842        (c) A data ] in a character class must be notated as \] because if the
843            first data character in a class is ], it defines an empty class. (In
844            Perl it is not possible to have an empty class.) The empty class []
845            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
846            The negative empty class [^] matches any one character, independently
847            of the DOTALL setting.
848    
849    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
850        non-existent subpattern following a character class starting with ']' and
851        containing () gave an internal compiling error instead of "reference to
852        non-existent subpattern". Fortunately, when the pattern did exist, the
853        compiled code was correct. (When scanning forwards to check for the
854        existencd of the subpattern, it was treating the data ']' as terminating
855        the class, so got the count wrong. When actually compiling, the reference
856        was subsequently set up correctly.)
857    
858    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
859        it was being rejected as not supported by pcre_dfa_exec(), even though
860        other assertions are supported. I have made pcre_dfa_exec() support
861        (*FAIL).
862    
863    16. The implementation of 13c above involved the invention of a new opcode,
864        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
865        cannot be changed at match time, I realized I could make a small
866        improvement to matching performance by compiling OP_ALLANY instead of
867        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
868        on the OP_ANY path.
869    
870    17. Compiling pcretest on Windows with readline support failed without the
871        following two fixes: (1) Make the unistd.h include conditional on
872        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
873    
874    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
875        ncurses library to be included for pcretest when ReadLine support is
876        requested, but also to allow for it to be overridden. This patch came from
877        Daniel Bergström.
878    
879    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
880        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
881        any errors with the current Unicode tables. Thanks to Peter Kankowski for
882        spotting this.
883    
884    
885    Version 7.6 28-Jan-08
886    ---------------------
887    
888    1.  A character class containing a very large number of characters with
889        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
890        overflow.
891    
892    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
893        HAVE_LONG_LONG is not defined.
894    
895    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
896        bring them up to date and include new features. This patch includes:
897    
898        - Fixed PH's badly added libz and libbz2 support.
899        - Fixed a problem with static linking.
900        - Added pcredemo. [But later removed - see 7 below.]
901        - Fixed dftables problem and added an option.
902        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
903            HAVE_LONG_LONG.
904        - Added readline support for pcretest.
905        - Added an listing of the option settings after cmake has run.
906    
907    4.  A user submitted a patch to Makefile that makes it easy to create
908        "pcre.dll" under mingw when using Configure/Make. I added stuff to
909        Makefile.am that cause it to include this special target, without
910        affecting anything else. Note that the same mingw target plus all
911        the other distribution libraries and programs are now supported
912        when configuring with CMake (see 6 below) instead of with
913        Configure/Make.
914    
915    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
916        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
917        exported in the Windows port". It has not yet been confirmed that the patch
918        solves the problem, but it does no harm.
919    
920    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
921        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
922        with CMake, and also correct the comment about stack recursion.
923    
924    7.  Remove the automatic building of pcredemo from the ./configure system and
925        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
926        of a program that users should build themselves after PCRE is installed, so
927        building it automatically is not really right. What is more, it gave
928        trouble in some build environments.
929    
930    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
931    
932    
933    Version 7.5 10-Jan-08
934    ---------------------
935    
936    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
937        values in parens when parsing an RE using the C++ wrapper."
938    
939    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
940        Characters greater than 255 were excluded from the class instead of being
941        included.
942    
943    3.  The same bug as (2) above applied to negated POSIX classes such as
944        [:^space:].
945    
946    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
947        defined or documented. It seems to have been a typo for PCRE_STATIC, so
948        I have changed it.
949    
950    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
951        first named subpattern) and a construct such as (?&a) would reference the
952        first named subpattern whose name started with "a" (in other words, the
953        length check was missing). Both these problems are fixed. "Subpattern name
954        expected" is now given for (?&) (a zero-length name), and this patch also
955        makes it give the same error for \k'' (previously it complained that that
956        was a reference to a non-existent subpattern).
957    
958    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
959        this is right because (?- can be followed by option settings as well as by
960        digits. I have, however, made the messages clearer.
961    
962    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
963        than the number used in the conditional) now cause a compile-time error.
964        This is actually not compatible with Perl, which accepts such patterns, but
965        treats the conditional as always being FALSE (as PCRE used to), but it
966        seems to me that giving a diagnostic is better.
967    
968    8.  Change "alphameric" to the more common word "alphanumeric" in comments
969        and messages.
970    
971    9.  Fix two occurrences of "backslash" in comments that should have been
972        "backspace".
973    
974    10. Remove two redundant lines of code that can never be obeyed (their function
975        was moved elsewhere).
976    
977    11. The program that makes PCRE's Unicode character property table had a bug
978        which caused it to generate incorrect table entries for sequences of
979        characters that have the same character type, but are in different scripts.
980        It amalgamated them into a single range, with the script of the first of
981        them. In other words, some characters were in the wrong script. There were
982        thirteen such cases, affecting characters in the following ranges:
983    
984          U+002b0 - U+002c1
985          U+0060c - U+0060d
986          U+0061e - U+00612
987          U+0064b - U+0065e
988          U+0074d - U+0076d
989          U+01800 - U+01805
990          U+01d00 - U+01d77
991          U+01d9b - U+01dbf
992          U+0200b - U+0200f
993          U+030fc - U+030fe
994          U+03260 - U+0327f
995          U+0fb46 - U+0fbb1
996          U+10450 - U+1049d
997    
998    12. The -o option (show only the matching part of a line) for pcregrep was not
999        compatible with GNU grep in that, if there was more than one match in a
1000        line, it showed only the first of them. It now behaves in the same way as
1001        GNU grep.
1002    
1003    13. If the -o and -v options were combined for pcregrep, it printed a blank
1004        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1005        does the same. The return code can be used to tell if there were any
1006        non-matching lines.
1007    
1008    14. Added --file-offsets and --line-offsets to pcregrep.
1009    
1010    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1011        infinitely looping recursion. The bug was that positive lookaheads were not
1012        being skipped when checking for a possible empty match (negative lookaheads
1013        and both kinds of lookbehind were skipped).
1014    
1015    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1016        inclusion of <windows.h> to before rather than after the definition of
1017        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1018    
1019    17. Specifying a possessive quantifier with a specific limit for a Unicode
1020        character property caused pcre_compile() to compile bad code, which led at
1021        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1022        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1023        caused the error; without that there was no problem.
1024    
1025    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1026    
1027    19. Added --enable-pcretest-libreadline.
1028    
1029    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1030        RE::GlobalReplace(). As a result, the number of replacements returned was
1031        double what it should be. I removed one of the increments, but Craig sent a
1032        later patch that removed the other one (the right fix) and added unit tests
1033        that check the return values (which was not done before).
1034    
1035    21. Several CMake things:
1036    
1037        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1038            the names libpcre and libpcreposix, not just pcre and pcreposix.
1039    
1040        (2) The above change means that pcretest and pcregrep are now correctly
1041            linked with the newly-built libraries, not previously installed ones.
1042    
1043        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
1044    
1045    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1046        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1047        UTF-8 newline character). The key issue is that the pattern starts .*;
1048        this means that the match must be either at the beginning, or after a
1049        newline. The bug was in the code for advancing after a failed match and
1050        checking that the new position followed a newline. It was not taking
1051        account of UTF-8 characters correctly.
1052    
1053    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1054        character classes. PCRE was not treating the sequence [:...:] as a
1055        character class unless the ... were all letters. Perl, however, seems to
1056        allow any characters between [: and :], though of course it rejects as
1057        unknown any "names" that contain non-letters, because all the known class
1058        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1059        for example, whereas PCRE did not - it did not recognize a POSIX character
1060        class. This seemed a bit dangerous, so the code has been changed to be
1061        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1062        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1063        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1064        Perl does, and where it didn't before.
1065    
1066    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1067        Windows environments %n is disabled by default.
1068    
1069    
1070    Version 7.4 21-Sep-07
1071    ---------------------
1072    
1073    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1074        means that a class such as [\s] counted as "explicit reference to CR or
1075        LF". That isn't really right - the whole point of the change was to try to
1076        help when there was an actual mention of one of the two characters. So now
1077        the change happens only if \r or \n (or a literal CR or LF) character is
1078        encountered.
1079    
1080    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1081        of both had grown to the point where there were only 3 bits left.
1082        Fortunately, there was spare space in the data structure, and so I have
1083        moved the internal flags into a new 16-bit field to free up more option
1084        bits.
1085    
1086    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1087        but did not set the internal JCHANGED flag - either of these is enough to
1088        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1089        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1090        start sets both bits.
1091    
1092    4.  Added options (at build time, compile time, exec time) to change \R from
1093        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1094    
1095    5.  doc/pcresyntax.html was missing from the distribution.
1096    
1097    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1098        compatibility, even though it is no longer used.
1099    
1100    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1101        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1102        windows.h file is present (where different names are used). [This was
1103        reversed later after testing - see 16 below.]
1104    
1105    8.  Changed all #include <config.h> to #include "config.h". There were also
1106        some further <pcre.h> cases that I changed to "pcre.h".
1107    
1108    9.  When pcregrep was used with the --colour option, it missed the line ending
1109        sequence off the lines that it output.
1110    
1111    10. It was pointed out to me that arrays of string pointers cause lots of
1112        relocations when a shared library is dynamically loaded. A technique of
1113        using a single long string with a table of offsets can drastically reduce
1114        these. I have refactored PCRE in four places to do this. The result is
1115        dramatic:
1116    
1117          Originally:                          290
1118          After changing UCP table:            187
1119          After changing error message table:   43
1120          After changing table of "verbs"       36
1121          After changing table of Posix names   22
1122    
1123        Thanks to the folks working on Gregex for glib for this insight.
1124    
1125    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1126        unicode-properties was also set.
1127    
1128    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1129    
1130    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1131        checked only for CRLF.
1132    
1133    14. Added casts to pcretest.c to avoid compiler warnings.
1134    
1135    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1136    
1137    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1138        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1139        entirely. This removes changes made in 7 above.
1140    
1141    17. The CMake files have been updated, and there is now more information about
1142        building with CMake in the NON-UNIX-USE document.
1143    
1144    
1145    Version 7.3 28-Aug-07
1146    ---------------------
1147    
1148     1. In the rejigging of the build system that eventually resulted in 7.1, the
1149        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1150        brackets there is not right, since it causes compilers to look for an
1151        installed pcre.h, not the version that is in the source that is being
1152        compiled (which of course may be different). I have changed it back to:
1153    
1154          #include "pcre.h"
1155    
1156        I have a vague recollection that the change was concerned with compiling in
1157        different directories, but in the new build system, that is taken care of
1158        by the VPATH setting the Makefile.
1159    
1160     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1161        when the subject happened to end in the byte 0x85 (e.g. if the last
1162        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1163        characters but of course it shouldn't be taken as a newline when it is part
1164        of another character. The bug was that, for an unlimited repeat of . in
1165        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1166        characters when looking for a newline.
1167    
1168     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1169    
1170     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1171        in debug output.
1172    
1173     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1174        long printing in the pcrecpp unittest when running under MinGW.
1175    
1176     6. ESC_K was left out of the EBCDIC table.
1177    
1178     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1179        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1180        limit also applies to "virtual nesting" when a pattern is recursive, and in
1181        this case 1000 isn't so big. I have been able to remove this limit at the
1182        expense of backing off one optimization in certain circumstances. Normally,
1183        when pcre_exec() would call its internal match() function recursively and
1184        immediately return the result unconditionally, it uses a "tail recursion"
1185        feature to save stack. However, when a subpattern that can match an empty
1186        string has an unlimited repetition quantifier, it no longer makes this
1187        optimization. That gives it a stack frame in which to save the data for
1188        checking that an empty string has been matched. Previously this was taken
1189        from the 1000-entry workspace that had been reserved. So now there is no
1190        explicit limit, but more stack is used.
1191    
1192     8. Applied Daniel's patches to solve problems with the import/export magic
1193        syntax that is required for Windows, and which was going wrong for the
1194        pcreposix and pcrecpp parts of the library. These were overlooked when this
1195        problem was solved for the main library.
1196    
1197     9. There were some crude static tests to avoid integer overflow when computing
1198        the size of patterns that contain repeated groups with explicit upper
1199        limits. As the maximum quantifier is 65535, the maximum group length was
1200        set at 30,000 so that the product of these two numbers did not overflow a
1201        32-bit integer. However, it turns out that people want to use groups that
1202        are longer than 30,000 bytes (though not repeat them that many times).
1203        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1204        made it possible to implement the integer overflow checks in a much more
1205        dynamic way, which I have now done. The artificial limitation on group
1206        length has been removed - we now have only the limit on the total length of
1207        the compiled pattern, which depends on the LINK_SIZE setting.
1208    
1209    10. Fixed a bug in the documentation for get/copy named substring when
1210        duplicate names are permitted. If none of the named substrings are set, the
1211        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1212        empty string.
1213    
1214    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1215        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1216        because the ] is interpreted as the first data character and the
1217        terminating ] is not found. PCRE has been made compatible with Perl in this
1218        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1219        cause memory overwriting.
1220    
1221    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1222        string has been matched (to stop an infinite loop). It was not recognizing
1223        a conditional subpattern that could match an empty string if that
1224        subpattern was within another subpattern. For example, it looped when
1225        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1226        condition was not nested. This bug has been fixed.
1227    
1228    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1229        past the start of the subject in the presence of bytes with the top bit
1230        set, for example "\x8aBCD".
1231    
1232    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1233        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1234    
1235    14. Optimized (?!) to (*FAIL).
1236    
1237    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1238        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1239        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1240        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1241        does: it's just the validity check that is more restrictive.
1242    
1243    16. Inserted checks for integer overflows during escape sequence (backslash)
1244        processing, and also fixed erroneous offset values for syntax errors during
1245        backslash processing.
1246    
1247    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1248        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1249    
1250    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1251        caused an overrun.
1252    
1253    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1254        something other than just ASCII characters) inside a group that had an
1255        unlimited repeat caused a loop at compile time (while checking to see
1256        whether the group could match an empty string).
1257    
1258    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1259        [\P{Any}] did so. (Error in the code for printing property names.)
1260    
1261    21. An orphan \E inside a character class could cause a crash.
1262    
1263    22. A repeated capturing bracket such as (A)? could cause a wild memory
1264        reference during compilation.
1265    
1266    23. There are several functions in pcre_compile() that scan along a compiled
1267        expression for various reasons (e.g. to see if it's fixed length for look
1268        behind). There were bugs in these functions when a repeated \p or \P was
1269        present in the pattern. These operators have additional parameters compared
1270        with \d, etc, and these were not being taken into account when moving along
1271        the compiled data. Specifically:
1272    
1273        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1274            length.
1275    
1276        (b) An item such as \pL+ within a repeated group could cause crashes or
1277            loops.
1278    
1279        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1280            "reference to non-existent subpattern" error.
1281    
1282        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1283    
1284    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1285        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1286    
1287    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1288    
1289    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1290        character were causing crashes (broken optimization).
1291    
1292    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1293        \p or \P) caused a compile-time loop.
1294    
1295    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1296        break. For example, the unstudied pattern [\r\n]A does not match the string
1297        "\r\nA" because change 7.0/46 below moves the current point on by two
1298        characters after failing to match at the start. However, the pattern \nA
1299        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1300        the same is true. There doesn't seem any very clean way out of this, but
1301        what I have chosen to do makes the common cases work: PCRE now takes note
1302        of whether there can be an explicit match for \r or \n anywhere in the
1303        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1304        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1305        pattern has explicit CR or LF references.
1306    
1307    29. Added (*CR) etc for changing newline setting at start of pattern.
1308    
1309    
1310    Version 7.2 19-Jun-07
1311    ---------------------
1312    
1313     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1314        which is apparently normally available under Windows.
1315    
1316     2. Re-jig the pcregrep tests with different newline settings in an attempt
1317        to make them independent of the local environment's newline setting.
1318    
1319     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1320    
1321     4. Some of the "internals" tests were previously cut out when the link size
1322        was not 2, because the output contained actual offsets. The recent new
1323        "Z" feature of pcretest means that these can be cut out, making the tests
1324        usable with all link sizes.
1325    
1326     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1327        stack recursion. This gives a massive performance boost under BSD, but just
1328        a small improvement under Linux. However, it saves one field in the frame
1329        in all cases.
1330    
1331     6. Added more features from the forthcoming Perl 5.10:
1332    
1333        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1334            recursion call. It refers to the nth most recently opened parentheses.
1335    
1336        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1337            to be opened parentheses.
1338    
1339        (c) Conditions that refer to capturing parentheses can be specified
1340            relatively, for example, (?(-2)... or (?(+3)...
1341    
1342        (d) \K resets the start of the current match so that everything before
1343            is not part of it.
1344    
1345        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1346    
1347        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1348            reference syntax.
1349    
1350        (g) (?| introduces a group in which the numbering of parentheses in each
1351            alternative starts with the same number.
1352    
1353        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1354    
1355     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1356        PCRE_INFO_JCHANGED.
1357    
1358     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1359        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1360        for detecting groups that can match an empty string.
1361    
1362     9. A pattern with a very large number of alternatives (more than several
1363        hundred) was running out of internal workspace during the pre-compile
1364        phase, where pcre_compile() figures out how much memory will be needed. A
1365        bit of new cunning has reduced the workspace needed for groups with
1366        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1367        workspace instead of running out of the 4096 that are available.
1368    
1369    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1370    
1371    11. Applied patch from Google to remove an optimization that didn't quite work.
1372        The report of the bug said:
1373    
1374          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1375          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1376          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1377    
1378    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1379        it matched the wrong number of bytes.
1380    
1381    
1382    Version 7.1 24-Apr-07
1383    ---------------------
1384    
1385     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1386        that is more "standard", making use of automake and other Autotools. There
1387        is some re-arrangement of the files and adjustment of comments consequent
1388        on this.
1389    
1390     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1391        for recursive directory scanning broke on some systems because the files
1392        are not scanned in any specific order and on different systems the order
1393        was different. A call to "sort" has been inserted into RunGrepTest for the
1394        approprate test as a short-term fix. In the longer term there may be an
1395        alternative.
1396    
1397     3. I had an email from Eric Raymond about problems translating some of PCRE's
1398        man pages to HTML (despite the fact that I distribute HTML pages, some
1399        people do their own conversions for various reasons). The problems
1400        concerned the use of low-level troff macros .br and .in. I have therefore
1401        removed all such uses from the man pages (some were redundant, some could
1402        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1403        HTML has been updated to handle .nf/.fi and to complain if it encounters
1404        .br or .in.
1405    
1406     4. Updated comments in configure.ac that get placed in config.h.in and also
1407        arranged for config.h to be included in the distribution, with the name
1408        config.h.generic, for the benefit of those who have to compile without
1409        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1410    
1411     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1412        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1413        makevp.bat for the current PCRE, using the additional files
1414        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1415    
1416     6. A Windows user reported a minor discrepancy with test 2, which turned out
1417        to be caused by a trailing space on an input line that had got lost in his
1418        copy. The trailing space was an accident, so I've just removed it.
1419    
1420     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1421        that is needed.
1422    
1423     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1424        as "const" (a) because they are and (b) because it helps the PHP
1425        maintainers who have recently made a script to detect big data structures
1426        in the php code that should be moved to the .rodata section. I remembered
1427        to update Builducptable as well, so it won't revert if ucptable.h is ever
1428        re-created.
1429    
1430     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1431        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1432        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1433        support is not required. This saves 1.5-2K of code, which is important in
1434        some applications.
1435    
1436        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1437        so as not to refer to the tables, even though these functions will never be
1438        called when UTF-8 support is disabled. Otherwise there are problems with a
1439        shared library.
1440    
1441    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1442    
1443        (a) It was defining its arguments as char * instead of void *.
1444    
1445        (b) It was assuming that all moves were upwards in memory; this was true
1446            a long time ago when I wrote it, but is no longer the case.
1447    
1448        The emulated memove() is provided for those environments that have neither
1449        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1450        is clearly not the case, as these two bugs were recently reported.
1451    
1452    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1453        and Detrail to create the HTML documentation, the .txt form of the man
1454        pages, and it removes trailing spaces from listed files. It also creates
1455        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1456        case, it wraps all the #defines with #ifndefs. This script should be run
1457        before "make dist".
1458    
1459    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1460        with Unicode property support.
1461    
1462        (a) For a maximizing quantifier, if the two different cases of the
1463            character were of different lengths in their UTF-8 codings (there are
1464            some cases like this - I found 11), and the matching function had to
1465            back up over a mixture of the two cases, it incorrectly assumed they
1466            were both the same length.
1467    
1468        (b) When PCRE was configured to use the heap rather than the stack for
1469            recursion during matching, it was not correctly preserving the data for
1470            the other case of a UTF-8 character when checking ahead for a match
1471            while processing a minimizing repeat. If the check also involved
1472            matching a wide character, but failed, corruption could cause an
1473            erroneous result when trying to check for a repeat of the original
1474            character.
1475    
1476    13. Some tidying changes to the testing mechanism:
1477    
1478        (a) The RunTest script now detects the internal link size and whether there
1479            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1480            values substituted by "configure". (The RunGrepTest script already did
1481            this for UTF-8.) The configure.ac script no longer substitutes the
1482            relevant variables.
1483    
1484        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1485            with length and offset values. This means that the output is different
1486            for different internal link sizes. Test 2 is skipped for link sizes
1487            other than 2 because of this, bypassing the problem. Unfortunately,
1488            there was also a test in test 3 (the locale tests) that used /B and
1489            failed for link sizes other than 2. Rather than cut the whole test out,
1490            I have added a new /Z option to pcretest that replaces the length and
1491            offset values with spaces. This is now used to make test 3 independent
1492            of link size. (Test 2 will be tidied up later.)
1493    
1494    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1495        segmentation fault instead of returning the appropriate error message.
1496    
1497    15. In multiline mode when the newline sequence was set to "any", the pattern
1498        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1499        This doesn't seem right; it now treats the CRLF combination as the line
1500        ending, and so does not match in that case. It's only a pattern such as ^$
1501        that would hit this one: something like ^ABC$ would have failed after \r
1502        and then tried again after \r\n.
1503    
1504    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1505        in an attempt to make files that differ only in their line terminators
1506        compare equal. This works on Linux.
1507    
1508    17. Under certain error circumstances pcregrep might try to free random memory
1509        as it exited. This is now fixed, thanks to valgrind.
1510    
1511    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1512        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1513        was because its rules for how to advance for /g after matching an empty
1514        string at the end of a line did not allow for this case. They now check for
1515        it specially.
1516    
1517    20. pcretest is supposed to handle patterns and data of any length, by
1518        extending its buffers when necessary. It was getting this wrong when the
1519        buffer for a data line had to be extended.
1520    
1521    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1522        CRLF as a newline sequence.
1523    
1524    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1525        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1526        I have nevertheless tidied it up.
1527    
1528    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1529    
1530    24. Added a man page for pcre-config.
1531    
1532    
1533    Version 7.0 19-Dec-06
1534    ---------------------
1535    
1536     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1537        moving to gcc 4.1.1.
1538    
1539     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1540        sys/time.h, which is documented as needed for this function. It doesn't
1541        seem to matter on Linux, but it showed up on some releases of OS X.
1542    
1543     3. It seems that there are systems where bytes whose values are greater than
1544        127 match isprint() in the "C" locale. The "C" locale should be the
1545        default when a C program starts up. In most systems, only ASCII printing
1546        characters match isprint(). This difference caused the output from pcretest
1547        to vary, making some of the tests fail. I have changed pcretest so that:
1548    
1549        (a) When it is outputting text in the compiled version of a pattern, bytes
1550            other than 32-126 are always shown as hex escapes.
1551    
1552        (b) When it is outputting text that is a matched part of a subject string,
1553            it does the same, unless a different locale has been set for the match
1554            (using the /L modifier). In this case, it uses isprint() to decide.
1555    
1556     4. Fixed a major bug that caused incorrect computation of the amount of memory
1557        required for a compiled pattern when options that changed within the
1558        pattern affected the logic of the preliminary scan that determines the
1559        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1560        that the computed length was too small. The symptoms of this bug were
1561        either the PCRE error "internal error: code overflow" from pcre_compile(),
1562        or a glibc crash with a message such as "pcretest: free(): invalid next
1563        size (fast)". Examples of patterns that provoked this bug (shown in
1564        pcretest format) are:
1565    
1566          /(?-x: )/x
1567          /(?x)(?-x: \s*#\s*)/
1568          /((?i)[\x{c0}])/8
1569          /(?i:[\x{c0}])/8
1570    
1571        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1572        is now done differently.
1573    
1574     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1575        wrapper classes; (b) implement a new function in the C++ scanner that is
1576        more efficient than the old way of doing things because it avoids levels of
1577        recursion in the regex matching; (c) add a paragraph to the documentation
1578        for the FullMatch() function.
1579    
1580     6. The escape sequence \n was being treated as whatever was defined as
1581        "newline". Not only was this contrary to the documentation, which states
1582        that \n is character 10 (hex 0A), but it also went horribly wrong when
1583        "newline" was defined as CRLF. This has been fixed.
1584    
1585     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1586        was being set to -1 for the "end of line" case (supposedly a value that no
1587        character can have). Though this value is never used (the check for end of
1588        line is "zero bytes in current character"), it caused compiler complaints.
1589        I've changed it to 0xffffffff.
1590    
1591     8. In pcre_version.c, the version string was being built by a sequence of
1592        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1593        string (as it is for production releases) called a macro with an empty
1594        argument. The C standard says the result of this is undefined. The gcc
1595        compiler treats it as an empty string (which was what was wanted) but it is
1596        reported that Visual C gives an error. The source has been hacked around to
1597        avoid this problem.
1598    
1599     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1600        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1601        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1602        of them did).
1603    
1604    10. Originally, pcretest opened its input and output without "b"; then I was
1605        told that "b" was needed in some environments, so it was added for release
1606        5.0 to both the input and output. (It makes no difference on Unix-like
1607        systems.) Later I was told that it is wrong for the input on Windows. I've
1608        now abstracted the modes into two macros, to make it easier to fiddle with
1609        them, and removed "b" from the input mode under Windows.
1610    
1611    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1612    
1613    12. Added -help and --help to pcretest as an official way of being reminded
1614        of the options.
1615    
1616    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1617        and pcrecpp.cc because they annoy compilers at high warning levels.
1618    
1619    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1620    
1621    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1622        scripts are not C programs :-) and which was not noticed because it works
1623        on Linux.
1624    
1625    16. pcretest is supposed to handle any length of pattern and data line (as one
1626        line or as a continued sequence of lines) by extending its input buffer if
1627        necessary. This feature was broken for very long pattern lines, leading to
1628        a string of junk being passed to pcre_compile() if the pattern was longer
1629        than about 50K.
1630    
1631    17. I have done a major re-factoring of the way pcre_compile() computes the
1632        amount of memory needed for a compiled pattern. Previously, there was code
1633        that made a preliminary scan of the pattern in order to do this. That was
1634        OK when PCRE was new, but as the facilities have expanded, it has become
1635        harder and harder to keep it in step with the real compile phase, and there
1636        have been a number of bugs (see for example, 4 above). I have now found a
1637        cunning way of running the real compile function in a "fake" mode that
1638        enables it to compute how much memory it would need, while actually only
1639        ever using a few hundred bytes of working memory and without too many
1640        tests of the mode. This should make future maintenance and development
1641        easier. A side effect of this work is that the limit of 200 on the nesting
1642        depth of parentheses has been removed (though this was never a serious
1643        limitation, I suspect). However, there is a downside: pcre_compile() now
1644        runs more slowly than before (30% or more, depending on the pattern). I
1645        hope this isn't a big issue. There is no effect on runtime performance.
1646    
1647    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1648        newline (only possible for the last line of a file) and it was a
1649        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1650    
1651    19. Added additional timing features to pcretest. (1) The -tm option now times
1652        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1653        separate command line item, by a number that specifies the number of
1654        repeats to use when timing. The default is 50000; this gives better
1655        precision, but takes uncomfortably long for very large patterns.
1656    
1657    20. Extended pcre_study() to be more clever in cases where a branch of a
1658        subpattern has no definite first character. For example, (a*|b*)[cd] would
1659        previously give no result from pcre_study(). Now it recognizes that the
1660        first character must be a, b, c, or d.
1661    
1662    21. There was an incorrect error "recursive call could loop indefinitely" if
1663        a subpattern (or the entire pattern) that was being tested for matching an
1664        empty string contained only one non-empty item after a nested subpattern.
1665        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1666        incorrectly, because the \d was being skipped in the check.
1667    
1668    22. The pcretest program now has a new pattern option /B and a command line
1669        option -b, which is equivalent to adding /B to every pattern. This causes
1670        it to show the compiled bytecode, without the additional information that
1671        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1672        is the same as /B/I).
1673    
1674    23. A new optimization is now able automatically to treat some sequences such
1675        as a*b as a*+b. More specifically, if something simple (such as a character
1676        or a simple class like \d) has an unlimited quantifier, and is followed by
1677        something that cannot possibly match the quantified thing, the quantifier
1678        is automatically "possessified".
1679    
1680    24. A recursive reference to a subpattern whose number was greater than 39
1681        went wrong under certain circumstances in UTF-8 mode. This bug could also
1682        have affected the operation of pcre_study().
1683    
1684    25. Realized that a little bit of performance could be had by replacing
1685        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1686    
1687    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1688    
1689    27. Possessive quantifiers such as a++ were previously implemented by turning
1690        them into atomic groups such as ($>a+). Now they have their own opcodes,
1691        which improves performance. This includes the automatically created ones
1692        from 23 above.
1693    
1694    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1695        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1696        the first matched character to be a colon. This applied both to named and
1697        numbered groups.
1698    
1699    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1700    
1701    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1702        building PCRE on Windows easier, so I have included it in the distribution.
1703    
1704    31. There is now a check in pcretest against a ridiculously large number being
1705        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1706        loop, the loop is abandoned.
1707    
1708    32. Forward references to subpatterns in conditions such as (?(2)...) where
1709        subpattern 2 is defined later cause pcre_compile() to search forwards in
1710        the pattern for the relevant set of parentheses. This search went wrong
1711        when there were unescaped parentheses in a character class, parentheses
1712        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1713    
1714    33. "Subroutine" calls and backreferences were previously restricted to
1715        referencing subpatterns earlier in the regex. This restriction has now
1716        been removed.
1717    
1718    34. Added a number of extra features that are going to be in Perl 5.10. On the
1719        whole, these are just syntactic alternatives for features that PCRE had
1720        previously implemented using the Python syntax or my own invention. The
1721        other formats are all retained for compatibility.
1722    
1723        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1724            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1725            also .NET compatible.
1726    
1727        (b) A recursion or subroutine call to a named group can now be defined as
1728            (?&name) as well as (?P>name).
1729    
1730        (c) A backreference to a named group can now be defined as \k<name> or
1731            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1732            5.10, are also .NET compatible.
1733    
1734        (d) A conditional reference to a named group can now use the syntax
1735            (?(<name>) or (?('name') as well as (?(name).
1736    
1737        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1738            groups (named and numbered) that are never evaluated inline, but can be
1739            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1740            is always false. There may be only one alternative in such a group.
1741    
1742        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1743            as the simple (?(R). The condition is true only if the most recent
1744            recursion is that of the given number or name. It does not search out
1745            through the entire recursion stack.
1746    
1747        (g) The escape \gN or \g{N} has been added, where N is a positive or
1748            negative number, specifying an absolute or relative reference.
1749    
1750    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1751        some "unreachable code" warnings.
1752    
1753    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1754        things, this adds five new scripts.
1755    
1756    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1757        There were also incompatibilities regarding the handling of \Q..\E inside
1758        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1759        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1760    
1761    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1762        matches an empty string, and forcibly breaks the loop. There were bugs in
1763        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1764        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1765        separate and independent bugs (that affected different cases) have been
1766        fixed.
1767    
1768    39. Refactored the code to abolish the use of different opcodes for small
1769        capturing bracket numbers. This is a tidy that I avoided doing when I
1770        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1771        The new approach is not only tidier, it makes it possible to reduce the
1772        memory needed to fix the previous bug (38).
1773    
1774    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1775        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1776        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1777        mode.
1778    
1779    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1780        report.
1781    
1782    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1783        copy construction and assignment in the C++ wrapper.
1784    
1785    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1786        couple of bugs that could have given wrong results in the "--newline=crlf"
1787        case.
1788    
1789    44. Added a number of casts and did some reorganization of signed/unsigned int
1790        variables following suggestions from Dair Grant. Also renamed the variable
1791        "this" as "item" because it is a C++ keyword.
1792    
1793    45. Arranged for dftables to add
1794    
1795          #include "pcre_internal.h"
1796    
1797        to pcre_chartables.c because without it, gcc 4.x may remove the array
1798        definition from the final binary if PCRE is built into a static library and
1799        dead code stripping is activated.
1800    
1801    46. For an unanchored pattern, if a match attempt fails at the start of a
1802        newline sequence, and the newline setting is CRLF or ANY, and the next two
1803        characters are CRLF, advance by two characters instead of one.
1804    
1805    
1806    Version 6.7 04-Jul-06
1807    ---------------------
1808    
1809     1. In order to handle tests when input lines are enormously long, pcretest has
1810        been re-factored so that it automatically extends its buffers when
1811        necessary. The code is crude, but this _is_ just a test program. The
1812        default size has been increased from 32K to 50K.
1813    
1814     2. The code in pcre_study() was using the value of the re argument before
1815        testing it for NULL. (Of course, in any sensible call of the function, it
1816        won't be NULL.)
1817    
1818     3. The memmove() emulation function in pcre_internal.h, which is used on
1819        systems that lack both memmove() and bcopy() - that is, hardly ever -
1820        was missing a "static" storage class specifier.
1821    
1822     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1823        containing an extended class (one that cannot be represented by a bitmap
1824        because it contains high-valued characters or Unicode property items, e.g.
1825        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1826        pattern, but PCRE should not loop if you do not (it no longer does).
1827        [Detail: two cases were found: (a) a repeated subpattern containing an
1828        extended class; (b) a recursive reference to a subpattern that followed a
1829        previous extended class. It wasn't skipping over the extended class
1830        correctly when UTF-8 mode was not set.]
1831    
1832     5. A negated single-character class was not being recognized as fixed-length
1833        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1834        compile error "lookbehind assertion is not fixed length".
1835    
1836     6. The RunPerlTest auxiliary script was showing an unexpected difference
1837        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1838        write a Perl script that can interpret lines of an input file either as
1839        byte characters or as UTF-8, which is what "perltest" was being required to
1840        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1841        can't do is switch easily at run time between having the "use utf8;" pragma
1842        or not. In the end, I fudged it by using the RunPerlTest script to insert
1843        "use utf8;" explicitly for the UTF-8 tests.
1844    
1845     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1846        the end of the subject string, contrary to the documentation and to what
1847        Perl does. This was true of both matching functions. Now it matches only at
1848        the start of the subject and immediately after *internal* newlines.
1849    
1850     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1851        a pointer to an int instead of a pointer to an unsigned long int. This
1852        caused problems on 64-bit systems.
1853    
1854     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1855        instance of the 'standard' template library not being so standard".
1856    
1857    10. There was no check on the number of named subpatterns nor the maximum
1858        length of a subpattern name. The product of these values is used to compute
1859        the size of the memory block for a compiled pattern. By supplying a very
1860        long subpattern name and a large number of named subpatterns, the size
1861        computation could be caused to overflow. This is now prevented by limiting
1862        the length of names to 32 characters, and the number of named subpatterns
1863        to 10,000.
1864    
1865    11. Subpatterns that are repeated with specific counts have to be replicated in
1866        the compiled pattern. The size of memory for this was computed from the
1867        length of the subpattern and the repeat count. The latter is limited to
1868        65535, but there was no limit on the former, meaning that integer overflow
1869        could in principle occur. The compiled length of a repeated subpattern is
1870        now limited to 30,000 bytes in order to prevent this.
1871    
1872    12. Added the optional facility to have named substrings with the same name.
1873    
1874    13. Added the ability to use a named substring as a condition, using the
1875        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1876        are numbers (not recommended). Forward references are permitted.
1877    
1878    14. Added forward references in named backreferences (if you see what I mean).
1879    
1880    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1881        pattern could run off the end of the subject. For example, the pattern
1882        "(?s)(.{1,5})"8 did this with the subject "ab".
1883    
1884    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1885        PCRE_CASELESS was set when matching characters that were quantified with ?
1886        or *.
1887    
1888    17. A character class other than a single negated character that had a minimum
1889        but no maximum quantifier - for example [ab]{6,} - was not handled
1890        correctly by pce_dfa_exec(). It would match only one character.
1891    
1892    18. A valid (though odd) pattern that looked like a POSIX character
1893        class but used an invalid character after [ (for example [[,abc,]]) caused
1894        pcre_compile() to give the error "Failed: internal error: code overflow" or
1895        in some cases to crash with a glibc free() error. This could even happen if
1896        the pattern terminated after [[ but there just happened to be a sequence of
1897        letters, a binary zero, and a closing ] in the memory that followed.
1898    
1899    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1900        over the years. Originally (before any Unicode support), just the bottom 8
1901        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1902        output from "man perlunicode" includes this:
1903    
1904          The regular expression compiler produces polymorphic opcodes.  That
1905          is, the pattern adapts to the data and automatically switches to
1906          the Unicode character scheme when presented with Unicode data--or
1907          instead uses a traditional byte scheme when presented with byte
1908          data.
1909    
1910        Sadly, a wide octal escape does not cause a switch, and in a string with
1911        no other multibyte characters, these octal escapes are treated as before.
1912        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1913        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1914        Unicode string.
1915    
1916        I have not perpetrated such confusion in PCRE. Up till now, it took just
1917        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1918        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1919        translate to the appropriate multibyte character.
1920    
1921    29. Applied some refactoring to reduce the number of warnings from Microsoft
1922        and Borland compilers. This has included removing the fudge introduced
1923        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1924        a warning about an unused variable.
1925    
1926    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1927        characters since release 4.0, because Perl (from release 5.004) does not.
1928        [Or at least, is documented not to: some releases seem to be in conflict
1929        with the documentation.] However, when a pattern was studied with
1930        pcre_study() and all its branches started with \s, PCRE still included VT
1931        as a possible starting character. Of course, this did no harm; it just
1932        caused an unnecessary match attempt.
1933    
1934    22. Removed a now-redundant internal flag bit that recorded the fact that case
1935        dependency changed within the pattern. This was once needed for "required
1936        byte" processing, but is no longer used. This recovers a now-scarce options
1937        bit. Also moved the least significant internal flag bit to the most-
1938        significant bit of the word, which was not previously used (hangover from
1939        the days when it was an int rather than a uint) to free up another bit for
1940        the future.
1941    
1942    23. Added support for CRLF line endings as well as CR and LF. As well as the
1943        default being selectable at build time, it can now be changed at runtime
1944        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1945        specify that it is scanning data with non-default line endings.
1946    
1947    24. Changed the definition of CXXLINK to make it agree with the definition of
1948        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1949    
1950    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1951        recursions. This makes a big different to stack usage for some patterns.
1952    
1953    26. If a subpattern containing a named recursion or subroutine reference such
1954        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1955        the space required for the compiled pattern went wrong and gave too small a
1956        value. Depending on the environment, this could lead to "Failed: internal
1957        error: code overflow at offset 49" or "glibc detected double free or
1958        corruption" errors.
1959    
1960    27. Applied patches from Google (a) to support the new newline modes and (b) to
1961        advance over multibyte UTF-8 characters in GlobalReplace.
1962    
1963    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1964        difference for some implementation of PCRE in some Windows version.
1965    
1966    29. Added some extra testing facilities to pcretest:
1967    
1968        \q<number>   in a data line sets the "match limit" value
1969        \Q<number>   in a data line sets the "match recursion limt" value
1970        -S <number>  sets the stack size, where <number> is in megabytes
1971    
1972        The -S option isn't available for Windows.
1973    
1974    
1975    Version 6.6 06-Feb-06
1976    ---------------------
1977    
1978     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1979        in pcreposix.h. I have copied the definition from pcre.h.
1980    
1981     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1982        because pcre.h is no longer a built file.
1983    
1984     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1985        not normally included in the compiled code.
1986    
1987    
1988    Version 6.5 01-Feb-06
1989    ---------------------
1990    
1991     1. When using the partial match feature with pcre_dfa_exec(), it was not
1992        anchoring the second and subsequent partial matches at the new starting
1993        point. This could lead to incorrect results. For example, with the pattern
1994        /1234/, partially matching against "123" and then "a4" gave a match.
1995    
1996     2. Changes to pcregrep:
1997    
1998        (a) All non-match returns from pcre_exec() were being treated as failures
1999            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2000            error message is output. Some extra information is given for the
2001            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2002            probably the only errors that are likely to be caused by users (by
2003            specifying a regex that has nested indefinite repeats, for instance).
2004            If there are more than 20 of these errors, pcregrep is abandoned.
2005    
2006        (b) A binary zero was treated as data while matching, but terminated the
2007            output line if it was written out. This has been fixed: binary zeroes
2008            are now no different to any other data bytes.
2009    
2010        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2011            used to set a locale for matching. The --locale=xxxx long option has
2012            been added (no short equivalent) to specify a locale explicitly on the
2013            pcregrep command, overriding the environment variables.
2014    
2015        (d) When -B was used with -n, some line numbers in the output were one less
2016            than they should have been.
2017    
2018        (e) Added the -o (--only-matching) option.
2019    
2020        (f) If -A or -C was used with -c (count only), some lines of context were
2021            accidentally printed for the final match.
2022    
2023        (g) Added the -H (--with-filename) option.
2024    
2025        (h) The combination of options -rh failed to suppress file names for files
2026            that were found from directory arguments.
2027    
2028        (i) Added the -D (--devices) and -d (--directories) options.
2029    
2030        (j) Added the -F (--fixed-strings) option.
2031    
2032        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2033    
2034        (l) Added the --colo(u)r option.
2035    
2036        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2037            is not present by default.
2038    
2039     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2040        items such as (?R) or (?1), when the recursion could match a number of
2041        alternatives. If it matched one of the alternatives, but subsequently,
2042        outside the recursion, there was a failure, the code tried to back up into
2043        the recursion. However, because of the way PCRE is implemented, this is not
2044        possible, and the result was an incorrect result from the match.
2045    
2046        In order to prevent this happening, the specification of recursion has
2047        been changed so that all such subpatterns are automatically treated as
2048        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2049    
2050     4. I had overlooked the fact that, in some locales, there are characters for
2051        which isalpha() is true but neither isupper() nor islower() are true. In
2052        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2053        and ordfeminine) are like this. This affected the treatment of \w and \W
2054        when they appeared in character classes, but not when they appeared outside
2055        a character class. The bit map for "word" characters is now created
2056        separately from the results of isalnum() instead of just taking it from the
2057        upper, lower, and digit maps. (Plus the underscore character, of course.)
2058    
2059     5. The above bug also affected the handling of POSIX character classes such as
2060        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2061        permanent tables. Instead, the bit maps for such a class were previously
2062        created as the appropriate unions of the upper, lower, and digit bitmaps.
2063        Now they are created by subtraction from the [[:word:]] class, which has
2064        its own bitmap.
2065    
2066     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2067        It is created by subtracting the vertical space characters (\x09, \x0a,
2068        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2069        subtraction was done in the overall bitmap for a character class, meaning
2070        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2071        be recognized. This bug has been fixed.
2072    
2073     7. Patches from the folks at Google:
2074    
2075          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2076          real life, but is still worth protecting against".
2077    
2078          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2079          regular expressions".
2080    
2081          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2082          have it.
2083    
2084          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2085          "configure" and the latter not, in order to fix a problem somebody had
2086          with compiling the Arg class on HP-UX.
2087    
2088          (e) Improve the error-handling of the C++ wrapper a little bit.
2089    
2090          (f) New tests for checking recursion limiting.
2091    
2092     8. The pcre_memmove() function, which is used only if the environment does not
2093        have a standard memmove() function (and is therefore rarely compiled),
2094        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2095        returning a result (though PCRE never actually uses the result).
2096    
2097     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2098        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2099        returned instead of calling malloc() with an overflowing number that would
2100        most likely cause subsequent chaos.
2101    
2102    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2103    
2104    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2105        with this option is matched, the nmatch and pmatch options of regexec() are
2106        ignored.
2107    
2108    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2109        provided in case anyone wants to the the POSIX interface with UTF-8
2110        strings.
2111    
2112    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2113        C++ linking (needed for some HP-UX environments).
2114    
2115    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2116        (unused parameter) and in the pcre_printint() function (omitted "default"
2117        switch label when the default is to do nothing).
2118    
2119    15. Added some code to make it possible, when PCRE is compiled as a C++
2120        library, to replace subject pointers for pcre_exec() with a smart pointer
2121        class, thus making it possible to process discontinuous strings.
2122    
2123    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2124        much the same function. They were added by different people who were trying
2125        to make PCRE easy to compile on non-Unix systems. It has been suggested
2126        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2127        for compiling on Windows systems. I have therefore replaced it with
2128        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2129        defaults to "extern" for C or "extern C" for C++, which works fine on
2130        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2131        SCOPE with something explicit in config.h. In addition:
2132    
2133        (a) pcreposix.h still had just "extern" instead of either of these macros;
2134            I have replaced it with PCRE_DATA_SCOPE.
2135    
2136        (b) Functions such as _pcre_xclass(), which are internal to the library,
2137            but external in the C sense, all had PCRE_EXPORT in their definitions.
2138            This is apparently wrong for the Windows case, so I have removed it.
2139            (It makes no difference on Unix-like systems.)
2140    
2141    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2142        of recursive calls to match(). This is different to MATCH_LIMIT because
2143        that limits the total number of calls to match(), not all of which increase
2144        the depth of recursion. Limiting the recursion depth limits the amount of
2145        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2146        when PCRE is compiled, and changed at run time. A patch from Google adds
2147        this functionality to the C++ interface.
2148    
2149    18. Changes to the handling of Unicode character properties:
2150    
2151        (a) Updated the table to Unicode 4.1.0.
2152    
2153        (b) Recognize characters that are not in the table as "Cn" (undefined).
2154    
2155        (c) I revised the way the table is implemented to a much improved format
2156            which includes recognition of ranges. It now supports the ranges that
2157            are defined in UnicodeData.txt, and it also amalgamates other
2158            characters into ranges. This has reduced the number of entries in the
2159            table from around 16,000 to around 3,000, thus reducing its size
2160            considerably. I realized I did not need to use a tree structure after
2161            all - a binary chop search is just as efficient. Having reduced the
2162            number of entries, I extended their size from 6 bytes to 8 bytes to
2163            allow for more data.
2164    
2165        (d) Added support for Unicode script names via properties such as \p{Han}.
2166    
2167    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2168        matching that character.
2169    
2170    20. When matching a repeated Unicode property with a minimum greater than zero,
2171        (for example \pL{2,}), PCRE could look past the end of the subject if it
2172        reached it while seeking the minimum number of characters. This could
2173        happen only if some of the characters were more than one byte long, because
2174        there is a check for at least the minimum number of bytes.
2175    
2176    21. Refactored the implementation of \p and \P so as to be more general, to
2177        allow for more different types of property in future. This has changed the
2178        compiled form incompatibly. Anybody with saved compiled patterns that use
2179        \p or \P will have to recompile them.
2180    
2181    22. Added "Any" and "L&" to the supported property types.
2182    
2183    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2184        but give a compile time error if the value is greater than 0xff.
2185    
2186    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2187        accidentally not being installed or uninstalled.
2188    
2189    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2190        made were to insert the current release number. This seemed silly, because
2191        it made things harder for people building PCRE on systems that don't run
2192        "configure". I have turned pcre.h into a distributed file, no longer built
2193        by "configure", with the version identification directly included. There is
2194        no longer a pcre.h.in file.
2195    
2196        However, this change necessitated a change to the pcre-config script as
2197        well. It is built from pcre-config.in, and one of the substitutions was the
2198        release number. I have updated configure.ac so that ./configure now finds
2199        the release number by grepping pcre.h.
2200    
2201    26. Added the ability to run the tests under valgrind.
2202    
2203    
2204    Version 6.4 05-Sep-05
2205    ---------------------
2206    
2207     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2208        "--" to be printed when multiple files were scanned, even when none of the
2209        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2210        consider it to be a bug, and have restored the previous behaviour.
2211    
2212     2. A couple of code tidies to get rid of compiler warnings.
2213    
2214     3. The pcretest program used to cheat by referring to symbols in the library
2215        whose names begin with _pcre_. These are internal symbols that are not
2216        really supposed to be visible externally, and in some environments it is
2217        possible to suppress them. The cheating is now confined to including
2218        certain files from the library's source, which is a bit cleaner.
2219    
2220     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2221        file's purpose clearer.
2222    
2223     5. Reorganized pcre_ucp_findchar().
2224    
2225    
2226    Version 6.3 15-Aug-05
2227    ---------------------
2228    
2229     1. The file libpcre.pc.in did not have general read permission in the tarball.
2230    
2231     2. There were some problems when building without C++ support:
2232    
2233        (a) If C++ support was not built, "make install" and "make test" still
2234            tried to test it.
2235    
2236        (b) There were problems when the value of CXX was explicitly set. Some
2237            changes have been made to try to fix these, and ...
2238    
2239        (c) --disable-cpp can now be used to explicitly disable C++ support.
2240    
2241        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2242            backslash in a target when C++ was disabled. This confuses some
2243            versions of "make", apparently. Using an intermediate variable solves
2244            this. (Same for CPP_LOBJ.)
2245    
2246     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2247        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2248        necessary on certain architectures.
2249    
2250     4. Added a setting of -export-symbols-regex to the link command to remove
2251        those symbols that are exported in the C sense, but actually are local
2252        within the library, and not documented. Their names all begin with
2253        "_pcre_". This is not a perfect job, because (a) we have to except some
2254        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2255        available (and never for static libraries). I have made a note to try to
2256        find a way round (a) in the future.
2257    
2258    
2259    Version 6.2 01-Aug-05
2260    ---------------------
2261    
2262     1. There was no test for integer overflow of quantifier values. A construction
2263        such as {1111111111111111} would give undefined results. What is worse, if
2264        a minimum quantifier for a parenthesized subpattern overflowed and became
2265        negative, the calculation of the memory size went wrong. This could have
2266        led to memory overwriting.
2267    
2268     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2269    
2270     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2271        operating environments where this matters.
2272    
2273     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2274        PCRE options from within the C++ wrapper.
2275    
2276     5. Named capturing subpatterns were not being correctly counted when a pattern
2277        was compiled. This caused two problems: (a) If there were more than 100
2278        such subpatterns, the calculation of the memory needed for the whole
2279        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2280        back references of the form \12, where the number was greater than 9, were
2281        not recognized as back references, even though there were sufficient
2282        previous subpatterns.
2283    
2284     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2285        versions of gcc, e.g. 2.95.4.
2286    
2287    
2288    Version 6.1 21-Jun-05
2289    ---------------------
2290    
2291     1. There was one reference to the variable "posix" in pcretest.c that was not
2292        surrounded by "#if !defined NOPOSIX".
2293    
2294     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2295        the cross-check on the old pcre_info() function, for the benefit of the
2296        cut-down version of PCRE that is currently imported into Exim.
2297    
2298     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2299        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2300        patterns that start (?i) but I don't think that matters. The use of (?i) is
2301        just an example; this all applies to the other options as well.
2302    
2303     4. Since libtool seems to echo the compile commands it is issuing, the output
2304        from "make" can be reduced a bit by putting "@" in front of each libtool
2305        compile command.
2306    
2307     5. Patch from the folks at Google for configure.in to be a bit more thorough
2308        in checking for a suitable C++ installation before trying to compile the
2309        C++ stuff. This should fix a reported problem when a compiler was present,
2310        but no suitable headers.
2311    
2312     6. The man pages all had just "PCRE" as their title. I have changed them to
2313        be the relevant file name. I have also arranged that these names are
2314        retained in the file doc/pcre.txt, which is a concatenation in text format
2315        of all the man pages except the little individual ones for each function.
2316    
2317     7. The NON-UNIX-USE file had not been updated for the different set of source
2318        files that come with release 6. I also added a few comments about the C++
2319        wrapper.
2320    
2321    
2322    Version 6.0 07-Jun-05
2323    ---------------------
2324    
2325     1. Some minor internal re-organization to help with my DFA experiments.
2326    
2327     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2328        didn't matter for the library itself when fully configured, but did matter
2329        when compiling without UCP support, or within Exim, where the ucp files are
2330        not imported.
2331    
2332     3. Refactoring of the library code to split up the various functions into
2333        different source modules. The addition of the new DFA matching code (see
2334        below) to a single monolithic source would have made it really too
2335        unwieldy, quite apart from causing all the code to be include in a
2336        statically linked application, when only some functions are used. This is
2337        relevant even without the DFA addition now that patterns can be compiled in
2338        one application and matched in another.
2339    
2340        The downside of splitting up is that there have to be some external
2341        functions and data tables that are used internally in different modules of
2342        the library but which are not part of the API. These have all had their
2343        names changed to start with "_pcre_" so that they are unlikely to clash
2344        with other external names.
2345    
2346     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2347        a different (DFA) algorithm. Although it is slower than the original
2348        function, it does have some advantages for certain types of matching
2349        problem.
2350    
2351     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2352        including restarting after a partial match.
2353    
2354     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2355        defined when compiling for Windows was sent to me. I have put it into the
2356        code, though I have no means of testing or verifying it.
2357    
2358     7. Added the pcre_refcount() auxiliary function.
2359    
2360     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2361        match before or at the first newline in the subject string. In pcretest,
2362        the /f option on a pattern can be used to set this.
2363    
2364     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2365        would behave wrongly. This has been present in PCRE since release 4.0.
2366    
2367    10. A number of changes to the pcregrep command:
2368    
2369        (a) Refactored how -x works; insert ^(...)$ instead of setting
2370            PCRE_ANCHORED and checking the length, in preparation for adding
2371            something similar for -w.
2372    
2373        (b) Added the -w (match as a word) option.
2374    
2375        (c) Refactored the way lines are read and buffered so as to have more
2376            than one at a time available.
2377    
2378        (d) Implemented a pcregrep test script.
2379    
2380        (e) Added the -M (multiline match) option. This allows patterns to match
2381            over several lines of the subject. The buffering ensures that at least
2382            8K, or the rest of the document (whichever is the shorter) is available
2383            for matching (and similarly the previous 8K for lookbehind assertions).
2384    
2385        (f) Changed the --help output so that it now says
2386    
2387              -w, --word-regex(p)
2388    
2389            instead of two lines, one with "regex" and the other with "regexp"
2390            because that confused at least one person since the short forms are the
2391            same. (This required a bit of code, as the output is generated
2392            automatically from a table. It wasn't just a text change.)
2393    
2394        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2395            option but starts with a hyphen. Could be a pattern or a path name
2396            starting with a hyphen, for instance.
2397    
2398        (h) "-" can be given as a file name to represent stdin.
2399    
2400        (i) When file names are being printed, "(standard input)" is used for
2401            the standard input, for compatibility with GNU grep. Previously
2402            "<stdin>" was used.
2403    
2404        (j) The option --label=xxx can be used to supply a name to be used for
2405            stdin when file names are being printed. There is no short form.
2406    
2407        (k) Re-factored the options decoding logic because we are going to add
2408            two more options that take data. Such options can now be given in four
2409            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2410    
2411        (l) Added the -A, -B, and -C options for requesting that lines of context
2412            around matches be printed.
2413    
2414        (m) Added the -L option to print the names of files that do not contain
2415            any matching lines, that is, the complement of -l.
2416    
2417        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2418            continue to scan other files.
2419    
2420        (o) The -s option was incorrectly implemented. For compatibility with other
2421            greps, it now suppresses the error message for a non-existent or non-
2422            accessible file (but not the return code). There is a new option called
2423            -q that suppresses the output of matching lines, which was what -s was
2424            previously doing.
2425    
2426        (p) Added --include and --exclude options to specify files for inclusion
2427            and exclusion when recursing.
2428    
2429    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2430        Hopefully, it now does.
2431    
2432    12. Missing cast in pcre_study().
2433    
2434    13. Added an "uninstall" target to the makefile.
2435    
2436    14. Replaced "extern" in the function prototypes in Makefile.in with
2437        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2438        world, but is set differently for Windows.
2439    
2440    15. Added a second compiling function called pcre_compile2(). The only
2441        difference is that it has an extra argument, which is a pointer to an
2442        integer error code. When there is a compile-time failure, this is set
2443        non-zero, in addition to the error test pointer being set to point to an
2444        error message. The new argument may be NULL if no error number is required
2445        (but then you may as well call pcre_compile(), which is now just a
2446        wrapper). This facility is provided because some applications need a
2447        numeric error indication, but it has also enabled me to tidy up the way
2448        compile-time errors are handled in the POSIX wrapper.
2449    
2450    16. Added VPATH=.libs to the makefile; this should help when building with one
2451        prefix path and installing with another. (Or so I'm told by someone who
2452        knows more about this stuff than I do.)
2453    
2454    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2455        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2456        match everything, including newlines. This is not POSIX-compatible, but
2457        somebody wanted the feature. From pcretest it can be activated by using
2458        both the P and the s flags.
2459    
2460    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2461    
2462    19. libpcre.pc was being incorrectly installed as executable.
2463    
2464    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2465        it now also looks for '\r' so that it will work unmodified on Windows.
2466    
2467    21. Added Google's contributed C++ wrapper to the distribution.
2468    
2469    22. Added some untidy missing memory free() calls in pcretest, to keep
2470        Electric Fence happy when testing.
2471    
2472    
2473    
2474    Version 5.0 13-Sep-04
2475    ---------------------
2476    
2477     1. Internal change: literal characters are no longer packed up into items
2478        containing multiple characters in a single byte-string. Each character
2479        is now matched using a separate opcode. However, there may be more than one
2480        byte in the character in UTF-8 mode.
2481    
2482     2. The pcre_callout_block structure has two new fields: pattern_position and
2483        next_item_length. These contain the offset in the pattern to the next match
2484        item, and its length, respectively.
2485    
2486     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2487        insertion of callouts before each pattern item. Added the /C option to
2488        pcretest to make use of this.
2489    
2490     4. On the advice of a Windows user, the lines
2491    
2492          #if defined(_WIN32) || defined(WIN32)
2493          _setmode( _fileno( stdout ), 0x8000 );
2494          #endif  /* defined(_WIN32) || defined(WIN32) */
2495    
2496        have been added to the source of pcretest. This apparently does useful
2497        magic in relation to line terminators.
2498    
2499     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2500        for the benefit of those environments where the "b" makes a difference.
2501    
2502     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2503        to know about it. I have put a hack into configure.in that adds in code
2504        to set GCC=yes if CC=icc. This seems to end up at a point in the
2505        generated configure script that is early enough to affect the setting of
2506        compiler options, which is what is needed, but I have no means of testing
2507        whether it really works. (The user who reported this had patched the
2508        generated configure script, which of course I cannot do.)
2509    
2510        LATER: After change 22 below (new libtool files), the configure script
2511        seems to know about icc (and also ecc). Therefore, I have commented out
2512        this hack in configure.in.
2513    
2514     7. Added support for pkg-config (2 patches were sent in).
2515    
2516     8. Negated POSIX character classes that used a combination of internal tables
2517        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2518        [[:^ascii]]. Typically, they would match almost any characters. The other
2519        POSIX classes were not broken in this way.
2520    
2521     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2522        to find the match, as PCRE was deluded into thinking that the match had to
2523        start at the start point or following a newline. The same bug applied to
2524        patterns with negative forward assertions or any backward assertions
2525        preceding ".*" at the start, unless the pattern required a fixed first
2526        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2527    
2528    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2529        starting at the last subject character, bytes beyond the end of the subject
2530        string were read.
2531    
2532    11. Renamed the variable "class" as "classbits" to make life easier for C++
2533        users. (Previously there was a macro definition, but it apparently wasn't
2534        enough.)
2535    
2536    12. Added the new field "tables" to the extra data so that tables can be passed
2537        in at exec time, or the internal tables can be re-selected. This allows
2538        a compiled regex to be saved and re-used at a later time by a different
2539        program that might have everything at different addresses.
2540    
2541    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2542        -R library as well as a -L library.
2543    
2544    14. The debugging options of pcretest (-d on the command line or D on a
2545        pattern) showed incorrect output for anything following an extended class
2546        that contained multibyte characters and which was followed by a quantifier.
2547    
2548    15. Added optional support for general category Unicode character properties
2549        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2550        support. It adds about 90K to the size of the library. The meanings of the
2551        inbuilt class escapes such as \d and \s have NOT been changed.
2552    
2553    16. Updated pcredemo.c to include calls to free() to release the memory for the
2554        compiled pattern.
2555    
2556    17. The generated file chartables.c was being created in the source directory
2557        instead of in the building directory. This caused the build to fail if the
2558        source directory was different from the building directory, and was
2559        read-only.
2560    
2561    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2562        file. No doubt somebody will tell me if they don't make sense... Also added
2563        Dan Mooney's comments about building on OpenVMS.
2564    
2565    19. Added support for partial matching via the PCRE_PARTIAL option for
2566        pcre_exec() and the \P data escape in pcretest.
2567    
2568    20. Extended pcretest with 3 new pattern features:
2569    
2570        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2571              write the compiled pattern to the file whose name is "rest-of-line".
2572              This is a straight binary dump of the data, with the saved pointer to
2573              the character tables forced to be NULL. The study data, if any, is
2574              written too. After writing, pcretest reads a new pattern.
2575    
2576        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2577              compiled pattern from the given file. There must not be any
2578              occurrences of "<" in the file name (pretty unlikely); if there are,
2579              pcretest will instead treat the initial "<" as a pattern delimiter.
2580              After reading in the pattern, pcretest goes on to read data lines as
2581              usual.
2582    
2583        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2584              and 16-bit fields in a compiled pattern, to simulate a pattern that
2585              was compiled on a host of opposite endianness.
2586    
2587    21. The pcre-exec() function can now cope with patterns that were compiled on
2588        hosts of opposite endianness, with this restriction:
2589    
2590          As for any compiled expression that is saved and used later, the tables
2591          pointer field cannot be preserved; the extra_data field in the arguments
2592          to pcre_exec() should be used to pass in a tables address if a value
2593          other than the default internal tables were used at compile time.
2594    
2595    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2596        now diagnosed as an error. Previously, most of the time, a negative number
2597        would have been treated as zero, but if in addition "ovector" was passed as
2598        NULL, a crash could occur.
2599    
2600    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2601        new versions from the libtool 1.5 distribution (the last one is a copy of
2602        a file called libtool.m4). This seems to have fixed the need to patch
2603        "configure" to support Darwin 1.3 (which I used to do). However, I still
2604        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2605        workstation).
2606    
2607    24. Changed the PCRE licence to be the more standard "BSD" licence.
2608    
2609    
2610    Version 4.5 01-Dec-03
2611    ---------------------
2612    
2613     1. There has been some re-arrangement of the code for the match() function so
2614        that it can be compiled in a version that does not call itself recursively.
2615        Instead, it keeps those local variables that need separate instances for
2616        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2617        needs to "recurse". Keeping track of where control must go is done by means
2618        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2619        hide most of the details from the main code, and operates only if
2620        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2621        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2622        operating.
2623    
2624        To make it easier for callers to provide specially tailored get/free
2625        functions for this usage, two new functions, pcre_stack_malloc, and
2626        pcre_stack_free, are used. They are always called in strict stacking order,
2627        and the size of block requested is always the same.
2628    
2629        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2630        PCRE has been compiled to use the stack or the heap for recursion. The
2631        -C option of pcretest uses this to show which version is compiled.
2632    
2633        A new data escape \S, is added to pcretest; it causes the amounts of store
2634        obtained and freed by both kinds of malloc/free at match time to be added
2635        to the output.
2636    
2637     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2638        what's available on my current Linux desktop machine.
2639    
2640     3. When matching a UTF-8 string, the test for a valid string at the start has
2641        been extended. If start_offset is not zero, PCRE now checks that it points
2642        to a byte that is the start of a UTF-8 character. If not, it returns
2643        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2644        this is necessary because there may be backward assertions in the pattern.
2645        When matching the same subject several times, it may save resources to use
2646        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2647    
2648     4. The code for checking the validity of UTF-8 strings has been tightened so
2649        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2650        containing "overlong sequences".
2651    
2652     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2653        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2654        should have been "&", but it just so happened that all the cases this let
2655        through by mistake were picked up later in the function.
2656    
2657     6. I had used a variable called "isblank" - this is a C99 function, causing
2658        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2659    
2660     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2661        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2662        so the version is shown; (c) stop "make test" echoing "./RunTest".
2663    
2664     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2665    
2666     9. The prototype for memmove() for systems that don't have it was using
2667        size_t, but the inclusion of the header that defines size_t was later. I've
2668        moved the #includes for the C headers earlier to avoid this.
2669    
2670    10. Added some adjustments to the code to make it easier to compiler on certain
2671        special systems:
2672    
2673          (a) Some "const" qualifiers were missing.
2674          (b) Added the macro EXPORT before all exported functions; by default this
2675              is defined to be empty.
2676          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2677              that it reads its output file name as an argument instead of writing
2678              to the standard output and assuming this can be redirected.
2679    
2680    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2681        class containing characters with values greater than 255, PCRE compilation
2682        went into a loop.
2683    
2684    12. A recursive reference to a subpattern that was within another subpattern
2685        that had a minimum quantifier of zero caused PCRE to crash. For example,
2686        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2687        recursion. If the recursively-called subpattern itself had a zero repeat,
2688        that was OK.
2689    
2690    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2691        buffer into which it was copied (for escape processing) was still set at
2692        1024, so long lines caused crashes.
2693    
2694    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2695        "internal error: code overflow...". This applied to any character class
2696        that was followed by a possessive quantifier.
2697    
2698    15. Modified the Makefile to add libpcre.la as a prerequisite for
2699        libpcreposix.la because I was told this is needed for a parallel build to
2700        work.
2701    
2702    16. If a pattern that contained .* following optional items at the start was
2703        studied, the wrong optimizing data was generated, leading to matching
2704        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2705        matching string must start with a or b or c. The correct conclusion for
2706        this pattern is that a match can start with any character.
2707    
2708    
2709    Version 4.4 13-Aug-03
2710    ---------------------
2711    
2712     1. In UTF-8 mode, a character class containing characters with values between
2713        127 and 255 was not handled correctly if the compiled pattern was studied.
2714        In fixing this, I have also improved the studying algorithm for such
2715        classes (slightly).
2716    
2717     2. Three internal functions had redundant arguments passed to them. Removal
2718        might give a very teeny performance improvement.
2719    
2720     3. Documentation bug: the value of the capture_top field in a callout is *one
2721        more than* the number of the hightest numbered captured substring.
2722    
2723     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2724        in incorrectly linking with a previously installed version. They now link
2725        explicitly with libpcre.la.
2726    
2727     5. configure.in no longer needs to recognize Cygwin specially.
2728    
2729     6. A problem in pcre.in for Windows platforms is fixed.
2730    
2731     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2732        pcretest, it used to include the size of the study block as part of its
2733        output. Unfortunately, the structure contains a field that has a different
2734        size on different hardware architectures. This meant that the tests that
2735        showed this size failed. As the block is currently always of a fixed size,
2736        this information isn't actually particularly useful in pcretest output, so
2737        I have just removed it.
2738    
2739     8. Three pre-processor statements accidentally did not start in column 1.
2740        Sadly, there are *still* compilers around that complain, even though
2741        standard C has not required this for well over a decade. Sigh.
2742    
2743     9. In pcretest, the code for checking callouts passed small integers in the
2744        callout_data field, which is a void * field. However, some picky compilers
2745        complained about the casts involved for this on 64-bit systems. Now
2746        pcretest passes the address of the small integer instead, which should get
2747        rid of the warnings.
2748    
2749    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2750        both compile and run time, and gives an error if an invalid UTF-8 sequence
2751        is found. There is a option for disabling this check in cases where the
2752        string is known to be correct and/or the maximum performance is wanted.
2753    
2754    11. In response to a bug report, I changed one line in Makefile.in from
2755    
2756            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2757        to
2758            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2759    
2760        to look similar to other lines, but I have no way of telling whether this
2761        is the right thing to do, as I do not use Windows. No doubt I'll get told
2762        if it's wrong...
2763    
2764    
2765    Version 4.3 21-May-03
2766    ---------------------
2767    
2768    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2769       Makefile.
2770    
2771    2. Some refactoring to improve the quality of the code:
2772    
2773       (i)   The utf8_table... variables are now declared "const".
2774    
2775       (ii)  The code for \cx, which used the "case flipping" table to upper case
2776             lower case letters, now just substracts 32. This is ASCII-specific,
2777             but the whole concept of \cx is ASCII-specific, so it seems
2778             reasonable.
2779    
2780       (iii) PCRE was using its character types table to recognize decimal and
2781             hexadecimal digits in the pattern. This is silly, because it handles
2782             only 0-9, a-f, and A-F, but the character types table is locale-
2783             specific, which means strange things might happen. A private
2784             table is now used for this - though it costs 256 bytes, a table is
2785             much faster than multiple explicit tests. Of course, the standard
2786             character types table is still used for matching digits in subject
2787             strings against \d.
2788    
2789       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2790             ending in _t are). So I've renamed it as ESC_tee.
2791    
2792    3. The first argument for regexec() in the POSIX wrapper should have been
2793       defined as "const".
2794    
2795    4. Changed pcretest to use malloc() for its buffers so that they can be
2796       Electric Fenced for debugging.
2797    
2798    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2799       to read one or more bytes before the start of the subject string. Often this
2800       had no effect on PCRE's behaviour, but in some circumstances it could
2801       provoke a segmentation fault.
2802    
2803    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2804       to try to read one or more bytes before the start of the subject string.
2805    
2806    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2807       UTF-8 support could misbehave in various ways if the subject string
2808       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2809       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2810       back over UTF-8 characters.)
2811    
2812    
2813    Version 4.2 14-Apr-03
2814    ---------------------
2815    
2816    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2817    
2818    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2819         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2820         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2821         [WIN_PREFIX]: new variable, "cyg" for Cygwin
2822         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2823           and BUILD_EXEEXT
2824         Note: automatic setting of the BUILD variables is not yet working
2825         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2826           compile-time but not at link-time
2827         [LINK]: use for linking executables only
2828         make different versions for Windows and non-Windows
2829         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2830           libraries
2831         [LINK_FOR_BUILD]: new variable
2832         [OBJEXT]: use throughout
2833         [EXEEXT]: use throughout
2834         <winshared>: new target
2835         <wininstall>: new target
2836         <dftables.o>: use native compiler
2837         <dftables>: use native linker
2838         <install>: handle Windows platform correctly
2839         <clean>: ditto
2840         <check>: ditto
2841         copy DLL to top builddir before testing
2842    
2843       As part of these changes, -no-undefined was removed again. This was reported
2844       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2845       in any case.
2846    
2847    3. Some tidies to get rid of compiler warnings:
2848    
2849       . In the match_data structure, match_limit was an unsigned long int, whereas
2850         match_call_count was an int. I've made them both unsigned long ints.
2851    
2852       . In pcretest the fact that a const uschar * doesn't automatically cast to
2853         a void * provoked a warning.
2854    
2855       . Turning on some more compiler warnings threw up some "shadow" variables
2856         and a few more missing casts.
2857    
2858    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2859       option, a class that contained a single character with a value between 128
2860       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2861    
2862    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2863       option, a class that contained several characters, but with at least one
2864       whose value was between 128 and 255 caused PCRE to crash.
2865    
2866    
2867    Version 4.1 12-Mar-03
2868    ---------------------
2869    
2870    1. Compiling with gcc -pedantic found a couple of places where casts were
2871    needed, and a string in dftables.c that was longer than standard compilers are
2872    required to support.
2873    
2874    2. Compiling with Sun's compiler found a few more places where the code could
2875    be tidied up in order to avoid warnings.
2876    
2877    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2878    first of these names is deprecated in the latest Autoconf in favour of the name
2879    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2880    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2881    analogy I have changed it to CFLAGS_FOR_BUILD.
2882    
2883    4. Added -no-undefined to the linking command in the Makefile, because this is
2884    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2885    linking step for the pcreposix library.
2886    
2887    5. PCRE was failing to diagnose the case of two named groups with the same
2888    name.
2889    
2890    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2891    literal character that is needed in the subject for a match, and scans along to
2892    ensure that it is present before embarking on the full matching process. This
2893    saves time in cases of nested unlimited repeats that are never going to match.
2894    Problem: the scan can take a lot of time if the subject is very long (e.g.
2895    megabytes), thus penalizing straightforward matches. It is now done only if the
2896    amount of subject to be scanned is less than 1000 bytes.
2897    
2898    7. A lesser problem with the same optimization is that it was recording the
2899    first character of an anchored pattern as "needed", thus provoking a search
2900    right along the subject, even when the first match of the pattern was going to
2901    fail. The "needed" character is now not set for anchored patterns, unless it
2902    follows something in the pattern that is of non-fixed length. Thus, it still
2903    fulfils its original purpose of finding quick non-matches in cases of nested
2904    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2905    
2906    
2907    Version 4.0 17-Feb-03
2908    ---------------------
2909    
2910    1. If a comment in an extended regex that started immediately after a meta-item
2911    extended to the end of string, PCRE compiled incorrect data. This could lead to
2912    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2913    
2914    2. Moved to autoconf 2.53 and libtool 1.4.2.
2915    
2916    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2917    the special perltest8 script is no longer needed - all the tests can be run
2918    from a single perltest script.
2919    
2920    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2921    by \s. It has now been removed in PCRE. This means it isn't recognized as
2922    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2923    class [:space:] *does* include VT, thereby creating a mess.
2924    
2925    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2926    space and tab.
2927    
2928    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2929    its new features into the main test script, reducing the number of scripts.
2930    
2931    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2932    were backward compatible, and made the (?i) apply to the whole pattern, as if
2933    /i were given. Now it behaves more logically, and applies the option setting
2934    only to what follows. PCRE has been changed to follow suit. However, if it
2935    finds options settings right at the start of the pattern, it extracts them into
2936    the global options, as before. Thus, they show up in the info data.
2937    
2938    8. Added support for the \Q...\E escape sequence. Characters in between are
2939    treated as literals. This is slightly different from Perl in that $ and @ are
2940    also handled as literals inside the quotes. In Perl, they will cause variable
2941    interpolation. Note the following examples:
2942    
2943        Pattern            PCRE matches      Perl matches
2944    
2945        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
2946        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2947        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2948    
2949    For compatibility with Perl, \Q...\E sequences are recognized inside character
2950    classes as well as outside them.
2951    
2952    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2953    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2954    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2955    signed/unsigned warnings.
2956    
2957    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2958    option for pcretest, so I've replaced it by a simple function that does just
2959    that job.
2960    
2961    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2962    "pcregrep -".
2963    
2964    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2965    Java package. This provides some syntactic sugar for simple cases of what my
2966    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2967    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2968    item, you can use this simplified notation. Note that only makes sense with
2969    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2970    greediness, whatever the setting of the PCRE_UNGREEDY option.
2971    
2972    13. A change of greediness default within a pattern was not taking effect at
2973    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2974    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2975    was abstracted outside.
2976    
2977    14. PCRE now supports the \G assertion. It is true when the current matching
2978    position is at the start point of the match. This differs from \A when the
2979    starting offset is non-zero. Used with the /g option of pcretest (or similar
2980    code), it works in the same way as it does for Perl's /g option. If all
2981    alternatives of a regex begin with \G, the expression is anchored to the start
2982    match position, and the "anchored" flag is set in the compiled expression.
2983    
2984    15. Some bugs concerning the handling of certain option changes within patterns
2985    have been fixed. These applied to options other than (?ims). For example,
2986    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2987    been the other way round. Some of this was related to change 7 above.
2988    
2989    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2990    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2991    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2992    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2993    
2994    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2995    mode. Unlike ".", it always matches newline, whatever the setting of
2996    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2997    assertions. Perl allows it, but it doesn't (in general) work because it can't
2998    calculate the length of the lookbehind. At least, that's the case for Perl
2999    5.8.0 - I've been told they are going to document that it doesn't work in
3000    future.
3001    
3002    18. Added an error diagnosis for escapes that PCRE does not support: these are
3003    \L, \l, \N, \P, \p, \U, \u, and \X.
3004    
3005    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
3006    reading past the end of the pattern in cases such as /[abcd/.
3007    
3008    20. PCRE was getting more memory than necessary for patterns with classes that
3009    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
3010    
3011    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
3012    compiling PCRE for use with Virtual Pascal.
3013    
3014    22. Small fix to the Makefile to make it work properly if the build is done
3015    outside the source tree.
3016    
3017    23. Added a new extension: a condition to go with recursion. If a conditional
3018    subpattern starts with (?(R) the "true" branch is used if recursion has
3019    happened, whereas the "false" branch is used only at the top level.
3020    
3021    24. When there was a very long string of literal characters (over 255 bytes
3022    without UTF support, over 250 bytes with UTF support), the computation of how
3023    much memory was required could be incorrect, leading to segfaults or other
3024    strange effects.
3025    
3026    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
3027    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
3028    there was a subsequent back reference to those brackets. This meant that, for
3029    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
3030    possible to check for precisely this case. All we can do is abandon the
3031    optimization if .* occurs inside capturing brackets when there are any back
3032    references whatsoever. (See below for a better fix that came later.)
3033    
3034    26. The handling of the optimization for finding the first character of a
3035    non-anchored pattern, and for finding a character that is required later in the
3036    match were failing in some cases. This didn't break the matching; it just
3037    failed to optimize when it could. The way this is done has been re-implemented.
3038    
3039    27. Fixed typo in error message for invalid (?R item (it said "(?p").
3040    
3041    28. Added a new feature that provides some of the functionality that Perl
3042    provides with (?{...}). The facility is termed a "callout". The way it is done
3043    in PCRE is for the caller to provide an optional function, by setting
3044    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
3045    global variable. By default it is unset, which disables all calling out. To get
3046    the function called, the regex must include (?C) at appropriate points. This
3047    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
3048    This provides a means of identifying different callout points. When PCRE
3049    reaches such a point in the regex, if pcre_callout has been set, the external
3050    function is called. It is provided with data in a structure called
3051    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
3052    matching continues; if it returns a non-zero value, the match at the current
3053    point fails. However, backtracking will occur if possible. [This was changed
3054    later and other features added - see item 49 below.]
3055    
3056    29. pcretest is upgraded to test the callout functionality. It provides a
3057    callout function that displays information. By default, it shows the start of
3058    the match and the current position in the text. There are some new data escapes
3059    to vary what happens:
3060    
3061        \C+         in addition, show current contents of captured substrings
3062        \C-         do not supply a callout function
3063        \C!n        return 1 when callout number n is reached
3064        \C!n!m      return 1 when callout number n is reached for the mth time
3065    
3066    30. If pcregrep was called with the -l option and just a single file name, it
3067    output "<stdin>" if a match was found, instead of the file name.
3068    
3069    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
3070    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
3071    pcre_exec(). This saves a malloc/free per call. The default value of
3072    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
3073    when configuring.
3074    
3075    32. The default maximum size of a compiled pattern is 64K. There have been a
3076    few cases of people hitting this limit. The code now uses macros to handle the
3077    storing of links as offsets within the compiled pattern. It defaults to 2-byte
3078    links, but this can be changed to 3 or 4 bytes by --with-link-size when
3079    configuring. Tests 2 and 5 work only with 2-byte links because they output
3080    debugging information about compiled patterns.
3081    
3082    33. Internal code re-arrangements:
3083    
3084    (a) Moved the debugging function for printing out a compiled regex into
3085        its own source file (printint.c) and used #include to pull it into
3086        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
3087        separate copies.
3088    
3089    (b) Defined the list of op-code names for debugging as a macro in
3090        internal.h so that it is next to the definition of the opcodes.
3091    
3092    (c) Defined a table of op-code lengths for simpler skipping along compiled
3093        code. This is again a macro in internal.h so that it is next to the
3094        definition of the opcodes.
3095    
3096    34. Added support for recursive calls to individual subpatterns, along the
3097    lines of Robin Houston's patch (but implemented somewhat differently).
3098    
3099    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
3100    allow it to read and process whole directories in Win32. This code was
3101    contributed by Lionel Fourquaux; it has not been tested by me.
3102    
3103    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
3104    used to name a group. Names consist of alphanumerics and underscores, and must
3105    be unique. Back references use the syntax (?P=name) and recursive calls use
3106    (?P>name) which is a PCRE extension to the Python extension. Groups still have
3107    numbers. The function pcre_fullinfo() can be used after compilation to extract
3108    a name/number map. There are three relevant calls:
3109    
3110      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
3111      PCRE_INFO_NAMECOUNT            yields the number of entries
3112      PCRE_INFO_NAMETABLE            yields a pointer to the map.
3113    
3114    The map is a vector of fixed-size entries. The size of each entry depends on
3115    the length of the longest name used. The first two bytes of each entry are the
3116    group number, most significant byte first. There follows the corresponding
3117    name, zero terminated. The names are in alphabetical order.
3118    
3119    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
3120    case instead of 255. Making it the same both with and without UTF-8 support
3121    means that the same test output works with both.
3122    
3123    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
3124    calling malloc() with a zero argument.
3125    
3126    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
3127    optimization. I've improved things by keeping a bitmap of backreferences with
3128    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
3129    fact referenced, the optimization can be applied. It is unlikely that a
3130    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
3131    the match to follow \n) will appear inside brackets with a number greater than
3132    31, but if it does, any back reference > 31 suppresses the optimization.
3133    
3134    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
3135    of disabling numbered capturing parentheses. Any opening parenthesis that is
3136    not followed by ? behaves as if it were followed by ?: but named parentheses
3137    can still be used for capturing (and they will acquire numbers in the usual
3138    way).
3139    
3140    41. Redesigned the return codes from the match() function into yes/no/error so
3141    that errors can be passed back from deep inside the nested calls. A malloc
3142    failure while inside a recursive subpattern call now causes the
3143    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
3144    
3145    42. It is now possible to set a limit on the number of times the match()
3146    function is called in a call to pcre_exec(). This facility makes it possible to
3147    limit the amount of recursion and backtracking, though not in a directly
3148    obvious way, because the match() function is used in a number of different
3149    circumstances. The count starts from zero for each position in the subject
3150    string (for non-anchored patterns). The default limit is, for compatibility, a
3151    large number, namely 10 000 000. You can change this in two ways:
3152    
3153    (a) When configuring PCRE before making, you can use --with-match-limit=n
3154        to set a default value for the compiled library.
3155    
3156    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
3157        a different value is set. See 45 below.
3158    
3159    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3160    
3161    43. Added a new function pcre_config(int, void *) to enable run-time extraction
3162    of things that can be changed at compile time. The first argument specifies
3163    what is wanted and the second points to where the information is to be placed.
3164    The current list of available information is:
3165    
3166      PCRE_CONFIG_UTF8
3167    
3168    The output is an integer that is set to one if UTF-8 support is available;
3169    otherwise it is set to zero.
3170    
3171      PCRE_CONFIG_NEWLINE
3172    
3173    The output is an integer that it set to the value of the code that is used for
3174    newline. It is either LF (10) or CR (13).
3175    
3176      PCRE_CONFIG_LINK_SIZE
3177    
3178    The output is an integer that contains the number of bytes used for internal
3179    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3180    
3181      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
3182    
3183    The output is an integer that contains the threshold above which the POSIX
3184    interface uses malloc() for output vectors. See item 31 above.
3185    
3186      PCRE_CONFIG_MATCH_LIMIT
3187    
3188    The output is an unsigned integer that contains the default limit of the number
3189    of match() calls in a pcre_exec() execution. See 42 above.
3190    
3191    44. pcretest has been upgraded by the addition of the -C option. This causes it
3192    to extract all the available output from the new pcre_config() function, and to
3193    output it. The program then exits immediately.
3194    
3195    45. A need has arisen to pass over additional data with calls to pcre_exec() in
3196    order to support additional features. One way would have been to define
3197    pcre_exec2() (for example) with extra arguments, but this would not have been
3198    extensible, and would also have required all calls to the original function to
3199    be mapped to the new one. Instead, I have chosen to extend the mechanism that
3200    is used for passing in "extra" data from pcre_study().
3201    
3202    The pcre_extra structure is now exposed and defined in pcre.h. It currently
3203    contains the following fields:
3204    
3205      flags         a bitmap indicating which of the following fields are set
3206      study_data    opaque data from pcre_study()
3207      match_limit   a way of specifying a limit on match() calls for a specific
3208                      call to pcre_exec()
3209      callout_data  data for callouts (see 49 below)
3210    
3211    The flag bits are also defined in pcre.h, and are
3212    
3213      PCRE_EXTRA_STUDY_DATA
3214      PCRE_EXTRA_MATCH_LIMIT
3215      PCRE_EXTRA_CALLOUT_DATA
3216    
3217    The pcre_study() function now returns one of these new pcre_extra blocks, with
3218    the actual study data pointed to by the study_data field, and the
3219    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3220    before. That is, this change is entirely upwards-compatible and requires no
3221    change to existing code.
3222    
3223    If you want to pass in additional data to pcre_exec(), you can either place it
3224    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3225    block.
3226    
3227    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3228    data string contains the escape sequence \M, pcretest calls pcre_exec() several
3229    times with different match limits, until it finds the minimum value needed for
3230    pcre_exec() to complete. The value is then output. This can be instructive; for
3231    most simple matches the number is quite small, but for pathological cases it
3232    gets very large very quickly.
3233    
3234    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3235    returns the size of the data block pointed to by the study_data field in a
3236    pcre_extra block, that is, the value that was passed as the argument to
3237    pcre_malloc() when PCRE was getting memory in which to place the information
3238    created by pcre_study(). The fourth argument should point to a size_t variable.
3239    pcretest has been extended so that this information is shown after a successful
3240    pcre_study() call when information about the compiled regex is being displayed.
3241    
3242    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3243    because what follows is always an absolute path. (Later: it turns out that this
3244    is more than cosmetic for MinGW, because it doesn't like empty path
3245    components.)
3246    
3247    49. Some changes have been made to the callout feature (see 28 above):
3248    
3249    (i)  A callout function now has three choices for what it returns:
3250    
3251           0  =>  success, carry on matching
3252         > 0  =>  failure at this point, but backtrack if possible
3253         < 0  =>  serious error, return this value from pcre_exec()
3254    
3255         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3256         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3257         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3258         use by callout functions. It will never be used by PCRE itself.
3259    
3260    (ii) The pcre_extra structure (see 45 above) has a void * field called
3261         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3262         pcre_callout_block structure has a field of the same name. The contents of
3263         the field passed in the pcre_extra structure are passed to the callout
3264         function in the corresponding field in the callout block. This makes it
3265         easier to use the same callout-containing regex from multiple threads. For
3266         testing, the pcretest program has a new data escape
3267    
3268           \C*n        pass the number n (may be negative) as callout_data
3269    
3270         If the callout function in pcretest receives a non-zero value as
3271         callout_data, it returns that value.
3272    
3273    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3274    there were some redundant $(CFLAGS) in commands that are now specified as
3275    $(LINK), which already includes $(CFLAGS).
3276    
3277    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3278    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3279    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3280    one-byte characters throughout. Note that case-insensitive matching applies
3281    only to characters whose values are less than 256. PCRE doesn't support the
3282    notion of cases for higher-valued characters.
3283    
3284    (i)   A character class whose characters are all within 0-255 is handled as
3285          a bit map, and the map is inverted for negative classes. Previously, a
3286          character > 255 always failed to match such a class; however it should
3287          match if the class was a negative one (e.g. [^ab]). This has been fixed.
3288    
3289    (ii)  A negated character class with a single character < 255 is coded as
3290          "not this character" (OP_NOT). This wasn't working properly when the test
3291          character was multibyte, either singly or repeated.
3292    
3293    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3294          mode, for example: \x{100}{2,3}.
3295    
3296    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3297          singly or repeated) now correctly test multibyte characters. However,
3298          PCRE doesn't recognize any characters with values greater than 255 as
3299          digits, spaces, or word characters. Such characters always match \D, \S,
3300          and \W, and never match \d, \s, or \w.
3301    
3302    (v)   Classes may now contain characters and character ranges with values
3303          greater than 255. For example: [ab\x{100}-\x{400}].
3304    
3305    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
3306          PCRE in UTF-8 mode.
3307    
3308    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3309    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3310    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3311    value.)
3312    
3313    53. The single man page has become too large. I have therefore split it up into
3314    a number of separate man pages. These also give rise to individual HTML pages;
3315    these are now put in a separate directory, and there is an index.html page that
3316    lists them all. Some hyperlinking between the pages has been installed.
3317    
3318    54. Added convenience functions for handling named capturing parentheses.
3319    
3320    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3321    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3322    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3323    are faulted.
3324    
3325    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3326    calling configure. These values are used when compiling the dftables.c program
3327    which is run to generate the source of the default character tables. They