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

Diff of /code/trunk/ChangeLog

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

revision 69 by nigel, Sat Feb 24 21:40:18 2007 UTC revision 606 by ph10, Mon Jun 6 17:46:22 2011 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 8.13 30-Apr-2011
5    ------------------------
6    
7    1.  The Unicode data tables have been updated to Unicode 6.0.0.
8    
9    2.  Two minor typos in pcre_internal.h have been fixed.
10    
11    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
12        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
13        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
14    
15    4.  There were a number of related bugs in the code for matching backrefences
16        caselessly in UTF-8 mode when codes for the characters concerned were
17        different numbers of bytes. For example, U+023A and U+2C65 are an upper
18        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
19        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
20        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
21        2-byte code at the end of the subject (it thought there wasn't enough data
22        left).
23    
24    5.  Comprehensive information about what went wrong is now returned by
25        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
26        as the output vector has at least 2 elements. The offset of the start of
27        the failing character and a reason code are placed in the vector.
28    
29    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
30        now returned is for the first byte of the failing character, instead of the
31        last byte inspected. This is an incompatible change, but I hope it is small
32        enough not to be a problem. It makes the returned offset consistent with
33        pcre_exec() and pcre_dfa_exec().
34    
35    7.  pcretest now gives a text phrase as well as the error number when
36        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
37        failure, the offset and reason code are output.
38    
39    8.  When \R was used with a maximizing quantifier it failed to skip backwards
40        over a \r\n pair if the subsequent match failed. Instead, it just skipped
41        back over a single character (\n). This seems wrong (because it treated the
42        two characters as a single entity when going forwards), conflicts with the
43        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
44        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
45        has been changed.
46    
47    9.  Some internal refactoring has changed the processing so that the handling
48        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
49        time (the PCRE_DOTALL option was changed this way some time ago: version
50        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
51        which was always a bit of a fudge. It also means that there is one less
52        argument for the match() function, which reduces its stack requirements
53        slightly. This change also fixes an incompatibility with Perl: the pattern
54        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
55    
56    10. More internal refactoring has drastically reduced the number of recursive
57        calls to match() for possessively repeated groups such as (abc)++ when
58        using pcre_exec().
59    
60    11. While implementing 10, a number of bugs in the handling of groups were
61        discovered and fixed:
62    
63        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
64        (a|)*(?1) gave a compile-time internal error.
65        ((a|)+)+  did not notice that the outer group could match an empty string.
66        (^a|^)+   was not marked as anchored.
67        (.*a|.*)+ was not marked as matching at start or after a newline.
68    
69    12. Yet more internal refactoring has removed another argument from the match()
70        function. Special calls to this function are now indicated by setting a
71        value in a variable in the "match data" data block.
72    
73    13. Be more explicit in pcre_study() instead of relying on "default" for
74        opcodes that mean there is no starting character; this means that when new
75        ones are added and accidentally left out of pcre_study(), testing should
76        pick them up.
77    
78    14. The -s option of pcretest has been documented for ages as being an old
79        synonym of -m (show memory usage). I have changed it to mean "force study
80        for every regex", that is, assume /S for every regex. This is similar to -i
81        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
82        using it. It makes it easier to run collection of tests with study enabled,
83        and thereby test pcre_study() more easily.
84    
85    
86    Version 8.12 15-Jan-2011
87    ------------------------
88    
89    1.  Fixed some typos in the markup of the man pages, and wrote a script that
90        checks for such things as part of the documentation building process.
91    
92    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
93        --match-limit and --recursion-limit options (added for 8.11). In
94        particular, this made one of the standard tests fail. (The integer value
95        went into the wrong half of a long int.)
96    
97    3.  If the --colour option was given to pcregrep with -v (invert match), it
98        did strange things, either producing crazy output, or crashing. It should,
99        of course, ignore a request for colour when reporting lines that do not
100        match.
101    
102    4.  Another pcregrep bug caused similar problems if --colour was specified with
103        -M (multiline) and the pattern match finished with a line ending.
104    
105    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
106        matched in multiline mode, the following line was shown as part of the
107        match. This seems wrong, so I have changed it.
108    
109    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
110        the check for further matches in the same line (so they could be coloured)
111        to overrun the end of the current line. If another match was found, it was
112        incorrectly shown (and then shown again when found in the next line).
113    
114    7.  If pcregrep was compiled under Windows, there was a reference to the
115        function pcregrep_exit() before it was defined. I am assuming this was
116        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
117        reported by a user. I've moved the definition above the reference.
118    
119    
120    Version 8.11 10-Dec-2010
121    ------------------------
122    
123    1.  (*THEN) was not working properly if there were untried alternatives prior
124        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
125        backtracked to try for "b" instead of moving to the next alternative branch
126        at the same level (in this case, to look for "c"). The Perl documentation
127        is clear that when (*THEN) is backtracked onto, it goes to the "next
128        alternative in the innermost enclosing group".
129    
130    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
131        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
132        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
133        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
134        (*THEN).
135    
136    3.  If \s appeared in a character class, it removed the VT character from
137        the class, even if it had been included by some previous item, for example
138        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
139        of \s, but is part of the POSIX "space" class.)
140    
141    4.  A partial match never returns an empty string (because you can always
142        match an empty string at the end of the subject); however the checking for
143        an empty string was starting at the "start of match" point. This has been
144        changed to the "earliest inspected character" point, because the returned
145        data for a partial match starts at this character. This means that, for
146        example, /(?<=abc)def/ gives a partial match for the subject "abc"
147        (previously it gave "no match").
148    
149    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
150        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
151        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
152        has an implication that the given string is incomplete (because a partial
153        match is preferred over a full match). For this reason, these items now
154        give a partial match in this situation. [Aside: previously, the one case
155        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
156        match rather than a full match, which was wrong by the old rules, but is
157        now correct.]
158    
159    6.  There was a bug in the handling of #-introduced comments, recognized when
160        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
161        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
162        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
163        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
164        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
165        places in pcre_compile().
166    
167    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
168        comments when looking ahead for named forward references to subpatterns,
169        the only newline sequence it recognized was NL. It now handles newlines
170        according to the set newline convention.
171    
172    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
173        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
174        cater for a lack of strerror(). These oversights have been fixed.
175    
176    9.  Added --match-limit and --recursion-limit to pcregrep.
177    
178    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
179    
180    11. When the -o option was used, pcregrep was setting a return code of 1, even
181        when matches were found, and --line-buffered was not being honoured.
182    
183    12. Added an optional parentheses number to the -o and --only-matching options
184        of pcregrep.
185    
186    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
187        can match an empty string. The code to do it in pcretest and pcredemo
188        needed fixing:
189    
190        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
191            only one byte after an empty string match just before CRLF (this case
192            just got forgotten; "any" and "anycrlf" were OK).
193    
194        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
195            mode when an empty string match preceded an ASCII character followed by
196            a non-ASCII character. (The code for advancing by one character rather
197            than one byte was nonsense.)
198    
199        (c) The pcredemo.c sample program did not have any code at all to handle
200            the cases when CRLF is a valid newline sequence.
201    
202    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
203        as a starting offset was within the subject string. There is now a new
204        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
205        negative or greater than the length of the string. In order to test this,
206        pcretest is extended to allow the setting of negative starting offsets.
207    
208    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
209        starting offset points to the beginning of a UTF-8 character was
210        unnecessarily clumsy. I tidied it up.
211    
212    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
213        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
214    
215    17. Nobody had reported that the --include_dir option, which was added in
216        release 7.7 should have been called --include-dir (hyphen, not underscore)
217        for compatibility with GNU grep. I have changed it to --include-dir, but
218        left --include_dir as an undocumented synonym, and the same for
219        --exclude-dir, though that is not available in GNU grep, at least as of
220        release 2.5.4.
221    
222    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
223        characters from a string of bytes) have been redefined so as not to use
224        loops, in order to improve performance in some environments. At the same
225        time, I abstracted some of the common code into auxiliary macros to save
226        repetition (this should not affect the compiled code).
227    
228    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
229        compile-time error is now given if \c is not followed by an ASCII
230        character, that is, a byte less than 128. (In EBCDIC mode, the code is
231        different, and any byte value is allowed.)
232    
233    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
234        START_OPTIMIZE option, which is now allowed at compile time - but just
235        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
236        to pcregrep and other applications that have no direct access to PCRE
237        options. The new /Y option in pcretest sets this option when calling
238        pcre_compile().
239    
240    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
241        back references. Groups containing recursive back references were forced to
242        be atomic by that change, but in the case of named groups, the amount of
243        memory required was incorrectly computed, leading to "Failed: internal
244        error: code overflow". This has been fixed.
245    
246    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
247        pcretest.c, to avoid build problems in some Borland environments.
248    
249    
250    Version 8.10 25-Jun-2010
251    ------------------------
252    
253    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
254        THEN.
255    
256    2.  (*ACCEPT) was not working when inside an atomic group.
257    
258    3.  Inside a character class, \B is treated as a literal by default, but
259        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
260        causes the error). The code is unchanged, but I tidied the documentation.
261    
262    4.  Inside a character class, PCRE always treated \R and \X as literals,
263        whereas Perl faults them if its -w option is set. I have changed PCRE so
264        that it faults them when PCRE_EXTRA is set.
265    
266    5.  Added support for \N, which always matches any character other than
267        newline. (It is the same as "." when PCRE_DOTALL is not set.)
268    
269    6.  When compiling pcregrep with newer versions of gcc which may have
270        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
271        declared with attribute warn_unused_result" were given. Just casting the
272        result to (void) does not stop the warnings; a more elaborate fudge is
273        needed. I've used a macro to implement this.
274    
275    7.  Minor change to pcretest.c to avoid a compiler warning.
276    
277    8.  Added four artifical Unicode properties to help with an option to make
278        \s etc use properties (see next item). The new properties are: Xan
279        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
280    
281    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
282        use Unicode properties. (*UCP) at the start of a pattern can be used to set
283        this option. Modified pcretest to add /W to test this facility. Added
284        REG_UCP to make it available via the POSIX interface.
285    
286    10. Added --line-buffered to pcregrep.
287    
288    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
289        studied, and the match started with a letter with a code point greater than
290        127 whose first byte was different to the first byte of the other case of
291        the letter, the other case of this starting letter was not recognized
292        (#976).
293    
294    12. If a pattern that was studied started with a repeated Unicode property
295        test, for example, \p{Nd}+, there was the theoretical possibility of
296        setting up an incorrect bitmap of starting bytes, but fortunately it could
297        not have actually happened in practice until change 8 above was made (it
298        added property types that matched character-matching opcodes).
299    
300    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
301        possible starting bytes for non-anchored patterns.
302    
303    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
304        \R, and also a number of cases that involve Unicode properties, both
305        explicit and implicit when PCRE_UCP is set.
306    
307    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
308        input, it could crash or give wrong results if characters with values
309        greater than 0xc0 were present in the subject string. (Detail: it assumed
310        UTF-8 input when processing these items.)
311    
312    16. Added a lot of (int) casts to avoid compiler warnings in systems where
313        size_t is 64-bit (#991).
314    
315    17. Added a check for running out of memory when PCRE is compiled with
316        --disable-stack-for-recursion (#990).
317    
318    18. If the last data line in a file for pcretest does not have a newline on
319        the end, a newline was missing in the output.
320    
321    19. The default pcre_chartables.c file recognizes only ASCII characters (values
322        less than 128) in its various bitmaps. However, there is a facility for
323        generating tables according to the current locale when PCRE is compiled. It
324        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
325        characters, are recognized by isspace() and therefore were getting set in
326        these tables, and indeed these tables seem to approximate to ISO 8859. This
327        caused a problem in UTF-8 mode when pcre_study() was used to create a list
328        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
329        which of course cannot start UTF-8 characters. I have changed the code so
330        that only real ASCII characters (less than 128) and the correct starting
331        bytes for UTF-8 encodings are set for characters greater than 127 when in
332        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
333        altogether.)
334    
335    20. Added the /T option to pcretest so as to be able to run tests with non-
336        standard character tables, thus making it possible to include the tests
337        used for 19 above in the standard set of tests.
338    
339    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
340        reference to a subpattern the other side of a comment that contains an
341        opening parenthesis caused either an internal compiling error, or a
342        reference to the wrong subpattern.
343    
344    
345    Version 8.02 19-Mar-2010
346    ------------------------
347    
348    1.  The Unicode data tables have been updated to Unicode 5.2.0.
349    
350    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
351        configured.
352    
353    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
354        original author of that file, following a query about its status.
355    
356    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
357        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
358    
359    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
360        quantifier applied to a forward-referencing subroutine call, could compile
361        incorrect code or give the error "internal error: previously-checked
362        referenced subpattern not found".
363    
364    6.  Both MS Visual Studio and Symbian OS have problems with initializing
365        variables to point to external functions. For these systems, therefore,
366        pcre_malloc etc. are now initialized to local functions that call the
367        relevant global functions.
368    
369    7.  There were two entries missing in the vectors called coptable and poptable
370        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
371        I've fixed the data, and added a kludgy way of testing at compile time that
372        the lengths are correct (equal to the number of opcodes).
373    
374    8.  Following on from 7, I added a similar kludge to check the length of the
375        eint vector in pcreposix.c.
376    
377    9.  Error texts for pcre_compile() are held as one long string to avoid too
378        much relocation at load time. To find a text, the string is searched,
379        counting zeros. There was no check for running off the end of the string,
380        which could happen if a new error number was added without updating the
381        string.
382    
383    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
384    
385    11. \K was not working if it appeared in an atomic group or in a group that
386        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
387        \K is "not well defined" if used in an assertion. PCRE now accepts it if
388        the assertion is positive, but not if it is negative.
389    
390    12. Change 11 fortuitously reduced the size of the stack frame used in the
391        "match()" function of pcre_exec.c by one pointer. Forthcoming
392        implementation of support for (*MARK) will need an extra pointer on the
393        stack; I have reserved it now, so that the stack frame size does not
394        decrease.
395    
396    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
397        item in branch that calls a recursion is a subroutine call - as in the
398        second branch in the above example - was incorrectly given the compile-
399        time error "recursive call could loop indefinitely" because pcre_compile()
400        was not correctly checking the subroutine for matching a non-empty string.
401    
402    14. The checks for overrunning compiling workspace could trigger after an
403        overrun had occurred. This is a "should never occur" error, but it can be
404        triggered by pathological patterns such as hundreds of nested parentheses.
405        The checks now trigger 100 bytes before the end of the workspace.
406    
407    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
408    
409    
410    Version 8.01 19-Jan-2010
411    ------------------------
412    
413    1.  If a pattern contained a conditional subpattern with only one branch (in
414        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
415        computed the wrong minimum data length (which is of course zero for such
416        subpatterns). This could cause incorrect "no match" results.
417    
418    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
419        the pattern is reset in the first branch, pcre_compile() failed with
420        "internal error: code overflow at offset...". This happened only when
421        the reset was to the original external option setting. (An optimization
422        abstracts leading options settings into an external setting, which was the
423        cause of this.)
424    
425    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
426        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
427        assertion pattern did not match (meaning that the assertion was true), it
428        was incorrectly treated as false if the SKIP had been reached during the
429        matching. This also applied to assertions used as conditions.
430    
431    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
432        assertion subpattern, including such a pattern used as a condition,
433        unpredictable results occurred, instead of the error return
434        PCRE_ERROR_DFA_UITEM.
435    
436    5.  The C++ GlobalReplace function was not working like Perl for the special
437        situation when an empty string is matched. It now does the fancy magic
438        stuff that is necessary.
439    
440    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
441        removed. (These were left over from very, very early versions of PCRE.)
442    
443    7.  Some cosmetic changes to the code to make life easier when compiling it
444        as part of something else:
445    
446        (a) Change DEBUG to PCRE_DEBUG.
447    
448        (b) In pcre_compile(), rename the member of the "branch_chain" structure
449            called "current" as "current_branch", to prevent a collision with the
450            Linux macro when compiled as a kernel module.
451    
452        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
453            prevent a collision with the Linux macro when compiled as a kernel
454            module.
455    
456    8.  In pcre_compile() there are some checks for integer overflows that used to
457        cast potentially large values to (double). This has been changed to that
458        when building, a check for int64_t is made, and if it is found, it is used
459        instead, thus avoiding the use of floating point arithmetic. (There is no
460        other use of FP in PCRE.) If int64_t is not found, the fallback is to
461        double.
462    
463    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
464        2005 (difference between two addresses compared to an unsigned value).
465    
466    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
467        custom one, because of the following reported problem in Windows:
468    
469          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
470              under Win32.
471          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
472              therefore missing the function definition.
473          - The compiler thus generates a "C" signature for the test function.
474          - The linker fails to find the "C" function.
475          - PCRE fails to configure if asked to do so against libbz2.
476    
477    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
478        messages were output:
479    
480          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
481          rerunning libtoolize, to keep the correct libtool macros in-tree.
482          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
483    
484        I have done both of these things.
485    
486    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
487        most of the time, it *can* run out if it is given a pattern that contains a
488        runaway infinite recursion. I updated the discussion in the pcrestack man
489        page.
490    
491    13. Now that we have gone to the x.xx style of version numbers, the minor
492        version may start with zero. Using 08 or 09 is a bad idea because users
493        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
494        interpreted as invalid octal numbers. I've updated the previous comment in
495        configure.ac, and also added a check that gives an error if 08 or 09 are
496        used.
497    
498    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
499        causing partial matching to fail when the end of the subject matched \W
500        in a UTF-8 pattern where \W was quantified with a minimum of 3.
501    
502    15. There were some discrepancies between the declarations in pcre_internal.h
503        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
504        their definitions. The declarations used "const uschar *" and the
505        definitions used USPTR. Even though USPTR is normally defined as "const
506        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
507        reported that: "This difference in casting confuses some C++ compilers, for
508        example, SunCC recognizes above declarations as different functions and
509        generates broken code for hbpcre." I have changed the declarations to use
510        USPTR.
511    
512    16. GNU libtool is named differently on some systems. The autogen.sh script now
513        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
514        (FreeBSD).
515    
516    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
517        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
518        comment: "Figure out how to create a longlong from a string: strtoll and
519        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
520        instance, but it only takes 2 args instead of 3!"
521    
522    18. A subtle bug concerned with back references has been fixed by a change of
523        specification, with a corresponding code fix. A pattern such as
524        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
525        refers, was giving matches when it shouldn't. For example, xa=xaaa would
526        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
527        same bug. Such groups have to be quantified to be useful, or contained
528        inside another quantified group. (If there's no repetition, the reference
529        can never match.) The problem arises because, having left the group and
530        moved on to the rest of the pattern, a later failure that backtracks into
531        the group uses the captured value from the final iteration of the group
532        rather than the correct earlier one. I have fixed this in PCRE by forcing
533        any group that contains a reference to itself to be an atomic group; that
534        is, there cannot be any backtracking into it once it has completed. This is
535        similar to recursive and subroutine calls.
536    
537    
538    Version 8.00 19-Oct-09
539    ----------------------
540    
541    1.  The table for translating pcre_compile() error codes into POSIX error codes
542        was out-of-date, and there was no check on the pcre_compile() error code
543        being within the table. This could lead to an OK return being given in
544        error.
545    
546    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
547        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
548        in a Windows environment.
549    
550    3.  The pcregrep --count option prints the count for each file even when it is
551        zero, as does GNU grep. However, pcregrep was also printing all files when
552        --files-with-matches was added. Now, when both options are given, it prints
553        counts only for those files that have at least one match. (GNU grep just
554        prints the file name in this circumstance, but including the count seems
555        more useful - otherwise, why use --count?) Also ensured that the
556        combination -clh just lists non-zero counts, with no names.
557    
558    4.  The long form of the pcregrep -F option was incorrectly implemented as
559        --fixed_strings instead of --fixed-strings. This is an incompatible change,
560        but it seems right to fix it, and I didn't think it was worth preserving
561        the old behaviour.
562    
563    5.  The command line items --regex=pattern and --regexp=pattern were not
564        recognized by pcregrep, which required --regex pattern or --regexp pattern
565        (with a space rather than an '='). The man page documented the '=' forms,
566        which are compatible with GNU grep; these now work.
567    
568    6.  No libpcreposix.pc file was created for pkg-config; there was just
569        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
570    
571    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
572        when UCP support is not needed, by modifying the Python script that
573        generates it from Unicode data files. This should not matter if the module
574        is correctly used as a library, but I received one complaint about 50K of
575        unwanted data. My guess is that the person linked everything into his
576        program rather than using a library. Anyway, it does no harm.
577    
578    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
579        was a minimum greater than 1 for a wide character in a possessive
580        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
581        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
582        character. Chaos in the form of incorrect output or a compiling loop could
583        result.
584    
585    9.  The restrictions on what a pattern can contain when partial matching is
586        requested for pcre_exec() have been removed. All patterns can now be
587        partially matched by this function. In addition, if there are at least two
588        slots in the offset vector, the offset of the earliest inspected character
589        for the match and the offset of the end of the subject are set in them when
590        PCRE_ERROR_PARTIAL is returned.
591    
592    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
593        synonymous with PCRE_PARTIAL, for backwards compatibility, and
594        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
595        and may be more useful for multi-segment matching.
596    
597    11. Partial matching with pcre_exec() is now more intuitive. A partial match
598        used to be given if ever the end of the subject was reached; now it is
599        given only if matching could not proceed because another character was
600        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
601        string "Z", which now yields "no match" instead of "partial match". In the
602        case of pcre_dfa_exec(), "no match" is given if every matching path for the
603        final character ended with (*FAIL).
604    
605    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
606        if the pattern had a "must contain" character that was already found in the
607        earlier partial match, unless partial matching was again requested. For
608        example, with the pattern /dog.(body)?/, the "must contain" character is
609        "g". If the first part-match was for the string "dog", restarting with
610        "sbody" failed. This bug has been fixed.
611    
612    13. The string returned by pcre_dfa_exec() after a partial match has been
613        changed so that it starts at the first inspected character rather than the
614        first character of the match. This makes a difference only if the pattern
615        starts with a lookbehind assertion or \b or \B (\K is not supported by
616        pcre_dfa_exec()). It's an incompatible change, but it makes the two
617        matching functions compatible, and I think it's the right thing to do.
618    
619    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
620        so that the demonstration program is easily available in environments where
621        PCRE has not been installed from source.
622    
623    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
624        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
625        library.
626    
627    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
628        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
629        is not the first non-POSIX option to be added. Clearly some people find
630        these options useful.
631    
632    17. If a caller to the POSIX matching function regexec() passes a non-zero
633        value for nmatch with a NULL value for pmatch, the value of
634        nmatch is forced to zero.
635    
636    18. RunGrepTest did not have a test for the availability of the -u option of
637        the diff command, as RunTest does. It now checks in the same way as
638        RunTest, and also checks for the -b option.
639    
640    19. If an odd number of negated classes containing just a single character
641        interposed, within parentheses, between a forward reference to a named
642        subpattern and the definition of the subpattern, compilation crashed with
643        an internal error, complaining that it could not find the referenced
644        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
645        [The bug was that it was starting one character too far in when skipping
646        over the character class, thus treating the ] as data rather than
647        terminating the class. This meant it could skip too much.]
648    
649    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
650        /g option in pcretest when the pattern contains \K, which makes it possible
651        to have an empty string match not at the start, even when the pattern is
652        anchored. Updated pcretest and pcredemo to use this option.
653    
654    21. If the maximum number of capturing subpatterns in a recursion was greater
655        than the maximum at the outer level, the higher number was returned, but
656        with unset values at the outer level. The correct (outer level) value is
657        now given.
658    
659    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
660        PCRE did not set those parentheses (unlike Perl). I have now found a way to
661        make it do so. The string so far is captured, making this feature
662        compatible with Perl.
663    
664    23. The tests have been re-organized, adding tests 11 and 12, to make it
665        possible to check the Perl 5.10 features against Perl 5.10.
666    
667    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
668        pattern matches a fixed length string. PCRE did not allow this; now it
669        does. Neither allows recursion.
670    
671    25. I finally figured out how to implement a request to provide the minimum
672        length of subject string that was needed in order to match a given pattern.
673        (It was back references and recursion that I had previously got hung up
674        on.) This code has now been added to pcre_study(); it finds a lower bound
675        to the length of subject needed. It is not necessarily the greatest lower
676        bound, but using it to avoid searching strings that are too short does give
677        some useful speed-ups. The value is available to calling programs via
678        pcre_fullinfo().
679    
680    26. While implementing 25, I discovered to my embarrassment that pcretest had
681        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
682        study optimizations had never been tested with that matching function.
683        Oops. What is worse, even when it was passed study data, there was a bug in
684        pcre_dfa_exec() that meant it never actually used it. Double oops. There
685        were also very few tests of studied patterns with pcre_dfa_exec().
686    
687    27. If (?| is used to create subpatterns with duplicate numbers, they are now
688        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
689        on the other side of the coin, they are no longer allowed to have different
690        names, because these cannot be distinguished in PCRE, and this has caused
691        confusion. (This is a difference from Perl.)
692    
693    28. When duplicate subpattern names are present (necessarily with different
694        numbers, as required by 27 above), and a test is made by name in a
695        conditional pattern, either for a subpattern having been matched, or for
696        recursion in such a pattern, all the associated numbered subpatterns are
697        tested, and the overall condition is true if the condition is true for any
698        one of them. This is the way Perl works, and is also more like the way
699        testing by number works.
700    
701    
702    Version 7.9 11-Apr-09
703    ---------------------
704    
705    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
706        (pcretest), all targets were linked against these libraries. This included
707        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
708        libraries. This caused unwanted dependencies to be created. This problem
709        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
710        pcretest is linked with readline.
711    
712    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
713        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
714        moved outside it again, because FALSE and TRUE are already defined in AIX,
715        but BOOL is not.
716    
717    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
718        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
719    
720    4.  The pcregrep documentation said spaces were inserted as well as colons (or
721        hyphens) following file names and line numbers when outputting matching
722        lines. This is not true; no spaces are inserted. I have also clarified the
723        wording for the --colour (or --color) option.
724    
725    5.  In pcregrep, when --colour was used with -o, the list of matching strings
726        was not coloured; this is different to GNU grep, so I have changed it to be
727        the same.
728    
729    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
730        each matching line was coloured. Now it goes on to look for further matches
731        of any of the test patterns, which is the same behaviour as GNU grep.
732    
733    7.  A pattern that could match an empty string could cause pcregrep to loop; it
734        doesn't make sense to accept an empty string match in pcregrep, so I have
735        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
736        seems to be how GNU grep behaves.
737    
738    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
739        start or after a newline", because the conditional assertion was not being
740        correctly handled. The rule now is that both the assertion and what follows
741        in the first alternative must satisfy the test.
742    
743    9.  If auto-callout was enabled in a pattern with a conditional group whose
744        condition was an assertion, PCRE could crash during matching, both with
745        pcre_exec() and pcre_dfa_exec().
746    
747    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
748        used for matching.
749    
750    11. Unicode property support in character classes was not working for
751        characters (bytes) greater than 127 when not in UTF-8 mode.
752    
753    12. Added the -M command line option to pcretest.
754    
755    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
756    
757    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
758    
759    16. Added comments and documentation about mis-use of no_arg in the C++
760        wrapper.
761    
762    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
763        from Martin Jerabek that uses macro names for all relevant character and
764        string constants.
765    
766    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
767        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
768        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
769        these, but not everybody uses configure.
770    
771    19. A conditional group that had only one branch was not being correctly
772        recognized as an item that could match an empty string. This meant that an
773        enclosing group might also not be so recognized, causing infinite looping
774        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
775        with the subject "ab", where knowledge that the repeated group can match
776        nothing is needed in order to break the loop.
777    
778    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
779        exec(), but without supplying a callout function, matching went wrong.
780    
781    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
782        leak if the size of the offset vector was greater than 30. When the vector
783        is smaller, the saved offsets during recursion go onto a local stack
784        vector, but for larger vectors malloc() is used. It was failing to free
785        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
786        error, in fact).
787    
788    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
789        heapframe that is used only when UTF-8 support is enabled. This caused no
790        problem, but was untidy.
791    
792    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
793        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
794        included within another project.
795    
796    24. Steven Van Ingelgem's patches to add more options to the CMake support,
797        slightly modified by me:
798    
799          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
800              not building pcregrep.
801    
802          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
803              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
804    
805    25. Forward references, both numeric and by name, in patterns that made use of
806        duplicate group numbers, could behave incorrectly or give incorrect errors,
807        because when scanning forward to find the reference group, PCRE was not
808        taking into account the duplicate group numbers. A pattern such as
809        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
810    
811    26. Changed a few more instances of "const unsigned char *" to USPTR, making
812        the feature of a custom pointer more persuasive (as requested by a user).
813    
814    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
815        pcretest.c, inside #ifndefs, because it seems they are sometimes already
816        pre-defined.
817    
818    28. Added support for (*UTF8) at the start of a pattern.
819    
820    29. Arrange for flags added by the "release type" setting in CMake to be shown
821        in the configuration summary.
822    
823    
824    Version 7.8 05-Sep-08
825    ---------------------
826    
827    1.  Replaced UCP searching code with optimized version as implemented for Ad
828        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
829        stage table and inline lookup instead of a function, giving speed ups of 2
830        to 5 times on some simple patterns that I tested. Permission was given to
831        distribute the MultiStage2.py script that generates the tables (it's not in
832        the tarball, but is in the Subversion repository).
833    
834    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
835        scripts.
836    
837    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
838        a group with a zero qualifier. The result of the study could be incorrect,
839        or the function might crash, depending on the pattern.
840    
841    4.  Caseless matching was not working for non-ASCII characters in back
842        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
843        It now works when Unicode Property Support is available.
844    
845    5.  In pcretest, an escape such as \x{de} in the data was always generating
846        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
847        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
848        truncation.
849    
850    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
851    
852    7.  Added two (int) casts to pcregrep when printing the difference of two
853        pointers, in case they are 64-bit values.
854    
855    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
856        test 2 if it fails.
857    
858    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
859        and a #define of that name to empty if it is not externally set. This is to
860        allow users of MSVC to set it if necessary.
861    
862    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
863        the convenience functions in the pcre_get.c source file.
864    
865    11. An option change at the start of a pattern that had top-level alternatives
866        could cause overwriting and/or a crash. This command provoked a crash in
867        some environments:
868    
869          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
870    
871        This potential security problem was recorded as CVE-2008-2371.
872    
873    12. For a pattern where the match had to start at the beginning or immediately
874        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
875        pcre_dfa_exec() could read past the end of the passed subject if there was
876        no match. To help with detecting such bugs (e.g. with valgrind), I modified
877        pcretest so that it places the subject at the end of its malloc-ed buffer.
878    
879    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
880        exec() might read past the end of the data buffer in UTF-8 mode.
881    
882    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
883        the data contained the byte 0x85 as part of a UTF-8 character within its
884        first line. This applied both to normal and DFA matching.
885    
886    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
887        /^[^d]*?$/8 failed to match "abc".
888    
889    16. Added a missing copyright notice to pcrecpp_internal.h.
890    
891    17. Make it more clear in the documentation that values returned from
892        pcre_exec() in ovector are byte offsets, not character counts.
893    
894    18. Tidied a few places to stop certain compilers from issuing warnings.
895    
896    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
897        supplied by Stefan Weber. I made a further small update for 7.8 because
898        there is a change of source arrangements: the pcre_searchfuncs.c module is
899        replaced by pcre_ucd.c.
900    
901    
902    Version 7.7 07-May-08
903    ---------------------
904    
905    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
906        a string to a long long, pretend we don't even have a long long." This is
907        done by checking for the strtoq, strtoll, and _strtoi64 functions.
908    
909    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
910        pre-7.6 versions, which defined a global no_arg variable instead of putting
911        it in the RE class. (See also #8 below.)
912    
913    3.  Remove a line of dead code, identified by coverity and reported by Nuno
914        Lopes.
915    
916    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
917    
918        (1) The include/exclude patterns were being applied to the whole pathnames
919            of files, instead of just to the final components.
920    
921        (2) If there was more than one level of directory, the subdirectories were
922            skipped unless they satisfied the include/exclude conditions. This is
923            inconsistent with GNU grep (and could even be seen as contrary to the
924            pcregrep specification - which I improved to make it absolutely clear).
925            The action now is always to scan all levels of directory, and just
926            apply the include/exclude patterns to regular files.
927    
928    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
929        --exclude_dir in the tests to avoid scanning .svn directories.
930    
931    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
932        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
933        doesn't support NULs in patterns.
934    
935    7.  Added some missing "const"s to declarations of static tables in
936        pcre_compile.c and pcre_dfa_exec.c.
937    
938    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
939        caused by fix #2  above. (Subsequently also a second patch to fix the
940        first patch. And a third patch - this was a messy problem.)
941    
942    9.  Applied Craig's patch to remove the use of push_back().
943    
944    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
945        matching function regexec().
946    
947    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
948        which, however, unlike Perl's \g{...}, are subroutine calls, not back
949        references. PCRE supports relative numbers with this syntax (I don't think
950        Oniguruma does).
951    
952    12. Previously, a group with a zero repeat such as (...){0} was completely
953        omitted from the compiled regex. However, this means that if the group
954        was called as a subroutine from elsewhere in the pattern, things went wrong
955        (an internal error was given). Such groups are now left in the compiled
956        pattern, with a new opcode that causes them to be skipped at execution
957        time.
958    
959    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
960        to the way PCRE behaves:
961    
962        (a) A lone ] character is dis-allowed (Perl treats it as data).
963    
964        (b) A back reference to an unmatched subpattern matches an empty string
965            (Perl fails the current match path).
966    
967        (c) A data ] in a character class must be notated as \] because if the
968            first data character in a class is ], it defines an empty class. (In
969            Perl it is not possible to have an empty class.) The empty class []
970            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
971            The negative empty class [^] matches any one character, independently
972            of the DOTALL setting.
973    
974    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
975        non-existent subpattern following a character class starting with ']' and
976        containing () gave an internal compiling error instead of "reference to
977        non-existent subpattern". Fortunately, when the pattern did exist, the
978        compiled code was correct. (When scanning forwards to check for the
979        existencd of the subpattern, it was treating the data ']' as terminating
980        the class, so got the count wrong. When actually compiling, the reference
981        was subsequently set up correctly.)
982    
983    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
984        it was being rejected as not supported by pcre_dfa_exec(), even though
985        other assertions are supported. I have made pcre_dfa_exec() support
986        (*FAIL).
987    
988    16. The implementation of 13c above involved the invention of a new opcode,
989        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
990        cannot be changed at match time, I realized I could make a small
991        improvement to matching performance by compiling OP_ALLANY instead of
992        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
993        on the OP_ANY path.
994    
995    17. Compiling pcretest on Windows with readline support failed without the
996        following two fixes: (1) Make the unistd.h include conditional on
997        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
998    
999    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1000        ncurses library to be included for pcretest when ReadLine support is
1001        requested, but also to allow for it to be overridden. This patch came from
1002        Daniel Bergström.
1003    
1004    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1005        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1006        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1007        spotting this.
1008    
1009    
1010    Version 7.6 28-Jan-08
1011    ---------------------
1012    
1013    1.  A character class containing a very large number of characters with
1014        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1015        overflow.
1016    
1017    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1018        HAVE_LONG_LONG is not defined.
1019    
1020    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1021        bring them up to date and include new features. This patch includes:
1022    
1023        - Fixed PH's badly added libz and libbz2 support.
1024        - Fixed a problem with static linking.
1025        - Added pcredemo. [But later removed - see 7 below.]
1026        - Fixed dftables problem and added an option.
1027        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1028            HAVE_LONG_LONG.
1029        - Added readline support for pcretest.
1030        - Added an listing of the option settings after cmake has run.
1031    
1032    4.  A user submitted a patch to Makefile that makes it easy to create
1033        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1034        Makefile.am that cause it to include this special target, without
1035        affecting anything else. Note that the same mingw target plus all
1036        the other distribution libraries and programs are now supported
1037        when configuring with CMake (see 6 below) instead of with
1038        Configure/Make.
1039    
1040    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1041        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1042        exported in the Windows port". It has not yet been confirmed that the patch
1043        solves the problem, but it does no harm.
1044    
1045    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1046        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1047        with CMake, and also correct the comment about stack recursion.
1048    
1049    7.  Remove the automatic building of pcredemo from the ./configure system and
1050        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1051        of a program that users should build themselves after PCRE is installed, so
1052        building it automatically is not really right. What is more, it gave
1053        trouble in some build environments.
1054    
1055    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1056    
1057    
1058    Version 7.5 10-Jan-08
1059    ---------------------
1060    
1061    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1062        values in parens when parsing an RE using the C++ wrapper."
1063    
1064    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1065        Characters greater than 255 were excluded from the class instead of being
1066        included.
1067    
1068    3.  The same bug as (2) above applied to negated POSIX classes such as
1069        [:^space:].
1070    
1071    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1072        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1073        I have changed it.
1074    
1075    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1076        first named subpattern) and a construct such as (?&a) would reference the
1077        first named subpattern whose name started with "a" (in other words, the
1078        length check was missing). Both these problems are fixed. "Subpattern name
1079        expected" is now given for (?&) (a zero-length name), and this patch also
1080        makes it give the same error for \k'' (previously it complained that that
1081        was a reference to a non-existent subpattern).
1082    
1083    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1084        this is right because (?- can be followed by option settings as well as by
1085        digits. I have, however, made the messages clearer.
1086    
1087    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1088        than the number used in the conditional) now cause a compile-time error.
1089        This is actually not compatible with Perl, which accepts such patterns, but
1090        treats the conditional as always being FALSE (as PCRE used to), but it
1091        seems to me that giving a diagnostic is better.
1092    
1093    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1094        and messages.
1095    
1096    9.  Fix two occurrences of "backslash" in comments that should have been
1097        "backspace".
1098    
1099    10. Remove two redundant lines of code that can never be obeyed (their function
1100        was moved elsewhere).
1101    
1102    11. The program that makes PCRE's Unicode character property table had a bug
1103        which caused it to generate incorrect table entries for sequences of
1104        characters that have the same character type, but are in different scripts.
1105        It amalgamated them into a single range, with the script of the first of
1106        them. In other words, some characters were in the wrong script. There were
1107        thirteen such cases, affecting characters in the following ranges:
1108    
1109          U+002b0 - U+002c1
1110          U+0060c - U+0060d
1111          U+0061e - U+00612
1112          U+0064b - U+0065e
1113          U+0074d - U+0076d
1114          U+01800 - U+01805
1115          U+01d00 - U+01d77
1116          U+01d9b - U+01dbf
1117          U+0200b - U+0200f
1118          U+030fc - U+030fe
1119          U+03260 - U+0327f
1120          U+0fb46 - U+0fbb1
1121          U+10450 - U+1049d
1122    
1123    12. The -o option (show only the matching part of a line) for pcregrep was not
1124        compatible with GNU grep in that, if there was more than one match in a
1125        line, it showed only the first of them. It now behaves in the same way as
1126        GNU grep.
1127    
1128    13. If the -o and -v options were combined for pcregrep, it printed a blank
1129        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1130        does the same. The return code can be used to tell if there were any
1131        non-matching lines.
1132    
1133    14. Added --file-offsets and --line-offsets to pcregrep.
1134    
1135    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1136        infinitely looping recursion. The bug was that positive lookaheads were not
1137        being skipped when checking for a possible empty match (negative lookaheads
1138        and both kinds of lookbehind were skipped).
1139    
1140    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1141        inclusion of <windows.h> to before rather than after the definition of
1142        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1143    
1144    17. Specifying a possessive quantifier with a specific limit for a Unicode
1145        character property caused pcre_compile() to compile bad code, which led at
1146        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1147        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1148        caused the error; without that there was no problem.
1149    
1150    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1151    
1152    19. Added --enable-pcretest-libreadline.
1153    
1154    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1155        RE::GlobalReplace(). As a result, the number of replacements returned was
1156        double what it should be. I removed one of the increments, but Craig sent a
1157        later patch that removed the other one (the right fix) and added unit tests
1158        that check the return values (which was not done before).
1159    
1160    21. Several CMake things:
1161    
1162        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1163            the names libpcre and libpcreposix, not just pcre and pcreposix.
1164    
1165        (2) The above change means that pcretest and pcregrep are now correctly
1166            linked with the newly-built libraries, not previously installed ones.
1167    
1168        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
1169    
1170    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1171        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1172        UTF-8 newline character). The key issue is that the pattern starts .*;
1173        this means that the match must be either at the beginning, or after a
1174        newline. The bug was in the code for advancing after a failed match and
1175        checking that the new position followed a newline. It was not taking
1176        account of UTF-8 characters correctly.
1177    
1178    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1179        character classes. PCRE was not treating the sequence [:...:] as a
1180        character class unless the ... were all letters. Perl, however, seems to
1181        allow any characters between [: and :], though of course it rejects as
1182        unknown any "names" that contain non-letters, because all the known class
1183        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1184        for example, whereas PCRE did not - it did not recognize a POSIX character
1185        class. This seemed a bit dangerous, so the code has been changed to be
1186        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1187        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1188        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1189        Perl does, and where it didn't before.
1190    
1191    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1192        Windows environments %n is disabled by default.
1193    
1194    
1195    Version 7.4 21-Sep-07
1196    ---------------------
1197    
1198    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1199        means that a class such as [\s] counted as "explicit reference to CR or
1200        LF". That isn't really right - the whole point of the change was to try to
1201        help when there was an actual mention of one of the two characters. So now
1202        the change happens only if \r or \n (or a literal CR or LF) character is
1203        encountered.
1204    
1205    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1206        of both had grown to the point where there were only 3 bits left.
1207        Fortunately, there was spare space in the data structure, and so I have
1208        moved the internal flags into a new 16-bit field to free up more option
1209        bits.
1210    
1211    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1212        but did not set the internal JCHANGED flag - either of these is enough to
1213        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1214        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1215        start sets both bits.
1216    
1217    4.  Added options (at build time, compile time, exec time) to change \R from
1218        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1219    
1220    5.  doc/pcresyntax.html was missing from the distribution.
1221    
1222    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1223        compatibility, even though it is no longer used.
1224    
1225    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1226        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1227        windows.h file is present (where different names are used). [This was
1228        reversed later after testing - see 16 below.]
1229    
1230    8.  Changed all #include <config.h> to #include "config.h". There were also
1231        some further <pcre.h> cases that I changed to "pcre.h".
1232    
1233    9.  When pcregrep was used with the --colour option, it missed the line ending
1234        sequence off the lines that it output.
1235    
1236    10. It was pointed out to me that arrays of string pointers cause lots of
1237        relocations when a shared library is dynamically loaded. A technique of
1238        using a single long string with a table of offsets can drastically reduce
1239        these. I have refactored PCRE in four places to do this. The result is
1240        dramatic:
1241    
1242          Originally:                          290
1243          After changing UCP table:            187
1244          After changing error message table:   43
1245          After changing table of "verbs"       36
1246          After changing table of Posix names   22
1247    
1248        Thanks to the folks working on Gregex for glib for this insight.
1249    
1250    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1251        unicode-properties was also set.
1252    
1253    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1254    
1255    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1256        checked only for CRLF.
1257    
1258    14. Added casts to pcretest.c to avoid compiler warnings.
1259    
1260    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1261    
1262    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1263        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1264        entirely. This removes changes made in 7 above.
1265    
1266    17. The CMake files have been updated, and there is now more information about
1267        building with CMake in the NON-UNIX-USE document.
1268    
1269    
1270    Version 7.3 28-Aug-07
1271    ---------------------
1272    
1273     1. In the rejigging of the build system that eventually resulted in 7.1, the
1274        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1275        brackets there is not right, since it causes compilers to look for an
1276        installed pcre.h, not the version that is in the source that is being
1277        compiled (which of course may be different). I have changed it back to:
1278    
1279          #include "pcre.h"
1280    
1281        I have a vague recollection that the change was concerned with compiling in
1282        different directories, but in the new build system, that is taken care of
1283        by the VPATH setting the Makefile.
1284    
1285     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1286        when the subject happened to end in the byte 0x85 (e.g. if the last
1287        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1288        characters but of course it shouldn't be taken as a newline when it is part
1289        of another character. The bug was that, for an unlimited repeat of . in
1290        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1291        characters when looking for a newline.
1292    
1293     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1294    
1295     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1296        in debug output.
1297    
1298     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1299        long printing in the pcrecpp unittest when running under MinGW.
1300    
1301     6. ESC_K was left out of the EBCDIC table.
1302    
1303     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1304        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1305        limit also applies to "virtual nesting" when a pattern is recursive, and in
1306        this case 1000 isn't so big. I have been able to remove this limit at the
1307        expense of backing off one optimization in certain circumstances. Normally,
1308        when pcre_exec() would call its internal match() function recursively and
1309        immediately return the result unconditionally, it uses a "tail recursion"
1310        feature to save stack. However, when a subpattern that can match an empty
1311        string has an unlimited repetition quantifier, it no longer makes this
1312        optimization. That gives it a stack frame in which to save the data for
1313        checking that an empty string has been matched. Previously this was taken
1314        from the 1000-entry workspace that had been reserved. So now there is no
1315        explicit limit, but more stack is used.
1316    
1317     8. Applied Daniel's patches to solve problems with the import/export magic
1318        syntax that is required for Windows, and which was going wrong for the
1319        pcreposix and pcrecpp parts of the library. These were overlooked when this
1320        problem was solved for the main library.
1321    
1322     9. There were some crude static tests to avoid integer overflow when computing
1323        the size of patterns that contain repeated groups with explicit upper
1324        limits. As the maximum quantifier is 65535, the maximum group length was
1325        set at 30,000 so that the product of these two numbers did not overflow a
1326        32-bit integer. However, it turns out that people want to use groups that
1327        are longer than 30,000 bytes (though not repeat them that many times).
1328        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1329        made it possible to implement the integer overflow checks in a much more
1330        dynamic way, which I have now done. The artificial limitation on group
1331        length has been removed - we now have only the limit on the total length of
1332        the compiled pattern, which depends on the LINK_SIZE setting.
1333    
1334    10. Fixed a bug in the documentation for get/copy named substring when
1335        duplicate names are permitted. If none of the named substrings are set, the
1336        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1337        empty string.
1338    
1339    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1340        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1341        because the ] is interpreted as the first data character and the
1342        terminating ] is not found. PCRE has been made compatible with Perl in this
1343        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1344        cause memory overwriting.
1345    
1346    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1347        string has been matched (to stop an infinite loop). It was not recognizing
1348        a conditional subpattern that could match an empty string if that
1349        subpattern was within another subpattern. For example, it looped when
1350        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1351        condition was not nested. This bug has been fixed.
1352    
1353    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1354        past the start of the subject in the presence of bytes with the top bit
1355        set, for example "\x8aBCD".
1356    
1357    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1358        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1359    
1360    14. Optimized (?!) to (*FAIL).
1361    
1362    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1363        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1364        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1365        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1366        does: it's just the validity check that is more restrictive.
1367    
1368    16. Inserted checks for integer overflows during escape sequence (backslash)
1369        processing, and also fixed erroneous offset values for syntax errors during
1370        backslash processing.
1371    
1372    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1373        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1374    
1375    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1376        caused an overrun.
1377    
1378    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1379        something other than just ASCII characters) inside a group that had an
1380        unlimited repeat caused a loop at compile time (while checking to see
1381        whether the group could match an empty string).
1382    
1383    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1384        [\P{Any}] did so. (Error in the code for printing property names.)
1385    
1386    21. An orphan \E inside a character class could cause a crash.
1387    
1388    22. A repeated capturing bracket such as (A)? could cause a wild memory
1389        reference during compilation.
1390    
1391    23. There are several functions in pcre_compile() that scan along a compiled
1392        expression for various reasons (e.g. to see if it's fixed length for look
1393        behind). There were bugs in these functions when a repeated \p or \P was
1394        present in the pattern. These operators have additional parameters compared
1395        with \d, etc, and these were not being taken into account when moving along
1396        the compiled data. Specifically:
1397    
1398        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1399            length.
1400    
1401        (b) An item such as \pL+ within a repeated group could cause crashes or
1402            loops.
1403    
1404        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1405            "reference to non-existent subpattern" error.
1406    
1407        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1408    
1409    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1410        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1411    
1412    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1413    
1414    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1415        character were causing crashes (broken optimization).
1416    
1417    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1418        \p or \P) caused a compile-time loop.
1419    
1420    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1421        break. For example, the unstudied pattern [\r\n]A does not match the string
1422        "\r\nA" because change 7.0/46 below moves the current point on by two
1423        characters after failing to match at the start. However, the pattern \nA
1424        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1425        the same is true. There doesn't seem any very clean way out of this, but
1426        what I have chosen to do makes the common cases work: PCRE now takes note
1427        of whether there can be an explicit match for \r or \n anywhere in the
1428        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1429        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1430        pattern has explicit CR or LF references.
1431    
1432    29. Added (*CR) etc for changing newline setting at start of pattern.
1433    
1434    
1435    Version 7.2 19-Jun-07
1436    ---------------------
1437    
1438     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1439        which is apparently normally available under Windows.
1440    
1441     2. Re-jig the pcregrep tests with different newline settings in an attempt
1442        to make them independent of the local environment's newline setting.
1443    
1444     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1445    
1446     4. Some of the "internals" tests were previously cut out when the link size
1447        was not 2, because the output contained actual offsets. The recent new
1448        "Z" feature of pcretest means that these can be cut out, making the tests
1449        usable with all link sizes.
1450    
1451     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1452        stack recursion. This gives a massive performance boost under BSD, but just
1453        a small improvement under Linux. However, it saves one field in the frame
1454        in all cases.
1455    
1456     6. Added more features from the forthcoming Perl 5.10:
1457    
1458        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1459            recursion call. It refers to the nth most recently opened parentheses.
1460    
1461        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1462            to be opened parentheses.
1463    
1464        (c) Conditions that refer to capturing parentheses can be specified
1465            relatively, for example, (?(-2)... or (?(+3)...
1466    
1467        (d) \K resets the start of the current match so that everything before
1468            is not part of it.
1469    
1470        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1471    
1472        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1473            reference syntax.
1474    
1475        (g) (?| introduces a group in which the numbering of parentheses in each
1476            alternative starts with the same number.
1477    
1478        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1479    
1480     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1481        PCRE_INFO_JCHANGED.
1482    
1483     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1484        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1485        for detecting groups that can match an empty string.
1486    
1487     9. A pattern with a very large number of alternatives (more than several
1488        hundred) was running out of internal workspace during the pre-compile
1489        phase, where pcre_compile() figures out how much memory will be needed. A
1490        bit of new cunning has reduced the workspace needed for groups with
1491        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1492        workspace instead of running out of the 4096 that are available.
1493    
1494    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1495    
1496    11. Applied patch from Google to remove an optimization that didn't quite work.
1497        The report of the bug said:
1498    
1499          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1500          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1501          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1502    
1503    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1504        it matched the wrong number of bytes.
1505    
1506    
1507    Version 7.1 24-Apr-07
1508    ---------------------
1509    
1510     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1511        that is more "standard", making use of automake and other Autotools. There
1512        is some re-arrangement of the files and adjustment of comments consequent
1513        on this.
1514    
1515     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1516        for recursive directory scanning broke on some systems because the files
1517        are not scanned in any specific order and on different systems the order
1518        was different. A call to "sort" has been inserted into RunGrepTest for the
1519        approprate test as a short-term fix. In the longer term there may be an
1520        alternative.
1521    
1522     3. I had an email from Eric Raymond about problems translating some of PCRE's
1523        man pages to HTML (despite the fact that I distribute HTML pages, some
1524        people do their own conversions for various reasons). The problems
1525        concerned the use of low-level troff macros .br and .in. I have therefore
1526        removed all such uses from the man pages (some were redundant, some could
1527        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1528        HTML has been updated to handle .nf/.fi and to complain if it encounters
1529        .br or .in.
1530    
1531     4. Updated comments in configure.ac that get placed in config.h.in and also
1532        arranged for config.h to be included in the distribution, with the name
1533        config.h.generic, for the benefit of those who have to compile without
1534        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1535    
1536     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1537        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1538        makevp.bat for the current PCRE, using the additional files
1539        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1540    
1541     6. A Windows user reported a minor discrepancy with test 2, which turned out
1542        to be caused by a trailing space on an input line that had got lost in his
1543        copy. The trailing space was an accident, so I've just removed it.
1544    
1545     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1546        that is needed.
1547    
1548     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1549        as "const" (a) because they are and (b) because it helps the PHP
1550        maintainers who have recently made a script to detect big data structures
1551        in the php code that should be moved to the .rodata section. I remembered
1552        to update Builducptable as well, so it won't revert if ucptable.h is ever
1553        re-created.
1554    
1555     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1556        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1557        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1558        support is not required. This saves 1.5-2K of code, which is important in
1559        some applications.
1560    
1561        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1562        so as not to refer to the tables, even though these functions will never be
1563        called when UTF-8 support is disabled. Otherwise there are problems with a
1564        shared library.
1565    
1566    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1567    
1568        (a) It was defining its arguments as char * instead of void *.
1569    
1570        (b) It was assuming that all moves were upwards in memory; this was true
1571            a long time ago when I wrote it, but is no longer the case.
1572    
1573        The emulated memove() is provided for those environments that have neither
1574        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1575        is clearly not the case, as these two bugs were recently reported.
1576    
1577    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1578        and Detrail to create the HTML documentation, the .txt form of the man
1579        pages, and it removes trailing spaces from listed files. It also creates
1580        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1581        case, it wraps all the #defines with #ifndefs. This script should be run
1582        before "make dist".
1583    
1584    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1585        with Unicode property support.
1586    
1587        (a) For a maximizing quantifier, if the two different cases of the
1588            character were of different lengths in their UTF-8 codings (there are
1589            some cases like this - I found 11), and the matching function had to
1590            back up over a mixture of the two cases, it incorrectly assumed they
1591            were both the same length.
1592    
1593        (b) When PCRE was configured to use the heap rather than the stack for
1594            recursion during matching, it was not correctly preserving the data for
1595            the other case of a UTF-8 character when checking ahead for a match
1596            while processing a minimizing repeat. If the check also involved
1597            matching a wide character, but failed, corruption could cause an
1598            erroneous result when trying to check for a repeat of the original
1599            character.
1600    
1601    13. Some tidying changes to the testing mechanism:
1602    
1603        (a) The RunTest script now detects the internal link size and whether there
1604            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1605            values substituted by "configure". (The RunGrepTest script already did
1606            this for UTF-8.) The configure.ac script no longer substitutes the
1607            relevant variables.
1608    
1609        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1610            with length and offset values. This means that the output is different
1611            for different internal link sizes. Test 2 is skipped for link sizes
1612            other than 2 because of this, bypassing the problem. Unfortunately,
1613            there was also a test in test 3 (the locale tests) that used /B and
1614            failed for link sizes other than 2. Rather than cut the whole test out,
1615            I have added a new /Z option to pcretest that replaces the length and
1616            offset values with spaces. This is now used to make test 3 independent
1617            of link size. (Test 2 will be tidied up later.)
1618    
1619    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1620        segmentation fault instead of returning the appropriate error message.
1621    
1622    15. In multiline mode when the newline sequence was set to "any", the pattern
1623        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1624        This doesn't seem right; it now treats the CRLF combination as the line
1625        ending, and so does not match in that case. It's only a pattern such as ^$
1626        that would hit this one: something like ^ABC$ would have failed after \r
1627        and then tried again after \r\n.
1628    
1629    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1630        in an attempt to make files that differ only in their line terminators
1631        compare equal. This works on Linux.
1632    
1633    17. Under certain error circumstances pcregrep might try to free random memory
1634        as it exited. This is now fixed, thanks to valgrind.
1635    
1636    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1637        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1638        was because its rules for how to advance for /g after matching an empty
1639        string at the end of a line did not allow for this case. They now check for
1640        it specially.
1641    
1642    20. pcretest is supposed to handle patterns and data of any length, by
1643        extending its buffers when necessary. It was getting this wrong when the
1644        buffer for a data line had to be extended.
1645    
1646    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1647        CRLF as a newline sequence.
1648    
1649    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1650        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1651        I have nevertheless tidied it up.
1652    
1653    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1654    
1655    24. Added a man page for pcre-config.
1656    
1657    
1658    Version 7.0 19-Dec-06
1659    ---------------------
1660    
1661     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1662        moving to gcc 4.1.1.
1663    
1664     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1665        sys/time.h, which is documented as needed for this function. It doesn't
1666        seem to matter on Linux, but it showed up on some releases of OS X.
1667    
1668     3. It seems that there are systems where bytes whose values are greater than
1669        127 match isprint() in the "C" locale. The "C" locale should be the
1670        default when a C program starts up. In most systems, only ASCII printing
1671        characters match isprint(). This difference caused the output from pcretest
1672        to vary, making some of the tests fail. I have changed pcretest so that:
1673    
1674        (a) When it is outputting text in the compiled version of a pattern, bytes
1675            other than 32-126 are always shown as hex escapes.
1676    
1677        (b) When it is outputting text that is a matched part of a subject string,
1678            it does the same, unless a different locale has been set for the match
1679            (using the /L modifier). In this case, it uses isprint() to decide.
1680    
1681     4. Fixed a major bug that caused incorrect computation of the amount of memory
1682        required for a compiled pattern when options that changed within the
1683        pattern affected the logic of the preliminary scan that determines the
1684        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1685        that the computed length was too small. The symptoms of this bug were
1686        either the PCRE error "internal error: code overflow" from pcre_compile(),
1687        or a glibc crash with a message such as "pcretest: free(): invalid next
1688        size (fast)". Examples of patterns that provoked this bug (shown in
1689        pcretest format) are:
1690    
1691          /(?-x: )/x
1692          /(?x)(?-x: \s*#\s*)/
1693          /((?i)[\x{c0}])/8
1694          /(?i:[\x{c0}])/8
1695    
1696        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1697        is now done differently.
1698    
1699     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1700        wrapper classes; (b) implement a new function in the C++ scanner that is
1701        more efficient than the old way of doing things because it avoids levels of
1702        recursion in the regex matching; (c) add a paragraph to the documentation
1703        for the FullMatch() function.
1704    
1705     6. The escape sequence \n was being treated as whatever was defined as
1706        "newline". Not only was this contrary to the documentation, which states
1707        that \n is character 10 (hex 0A), but it also went horribly wrong when
1708        "newline" was defined as CRLF. This has been fixed.
1709    
1710     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1711        was being set to -1 for the "end of line" case (supposedly a value that no
1712        character can have). Though this value is never used (the check for end of
1713        line is "zero bytes in current character"), it caused compiler complaints.
1714        I've changed it to 0xffffffff.
1715    
1716     8. In pcre_version.c, the version string was being built by a sequence of
1717        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1718        string (as it is for production releases) called a macro with an empty
1719        argument. The C standard says the result of this is undefined. The gcc
1720        compiler treats it as an empty string (which was what was wanted) but it is
1721        reported that Visual C gives an error. The source has been hacked around to
1722        avoid this problem.
1723    
1724     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1725        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1726        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1727        of them did).
1728    
1729    10. Originally, pcretest opened its input and output without "b"; then I was
1730        told that "b" was needed in some environments, so it was added for release
1731        5.0 to both the input and output. (It makes no difference on Unix-like
1732        systems.) Later I was told that it is wrong for the input on Windows. I've
1733        now abstracted the modes into two macros, to make it easier to fiddle with
1734        them, and removed "b" from the input mode under Windows.
1735    
1736    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1737    
1738    12. Added -help and --help to pcretest as an official way of being reminded
1739        of the options.
1740    
1741    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1742        and pcrecpp.cc because they annoy compilers at high warning levels.
1743    
1744    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1745    
1746    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1747        scripts are not C programs :-) and which was not noticed because it works
1748        on Linux.
1749    
1750    16. pcretest is supposed to handle any length of pattern and data line (as one
1751        line or as a continued sequence of lines) by extending its input buffer if
1752        necessary. This feature was broken for very long pattern lines, leading to
1753        a string of junk being passed to pcre_compile() if the pattern was longer
1754        than about 50K.
1755    
1756    17. I have done a major re-factoring of the way pcre_compile() computes the
1757        amount of memory needed for a compiled pattern. Previously, there was code
1758        that made a preliminary scan of the pattern in order to do this. That was
1759        OK when PCRE was new, but as the facilities have expanded, it has become
1760        harder and harder to keep it in step with the real compile phase, and there
1761        have been a number of bugs (see for example, 4 above). I have now found a
1762        cunning way of running the real compile function in a "fake" mode that
1763        enables it to compute how much memory it would need, while actually only
1764        ever using a few hundred bytes of working memory and without too many
1765        tests of the mode. This should make future maintenance and development
1766        easier. A side effect of this work is that the limit of 200 on the nesting
1767        depth of parentheses has been removed (though this was never a serious
1768        limitation, I suspect). However, there is a downside: pcre_compile() now
1769        runs more slowly than before (30% or more, depending on the pattern). I
1770        hope this isn't a big issue. There is no effect on runtime performance.
1771    
1772    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1773        newline (only possible for the last line of a file) and it was a
1774        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1775    
1776    19. Added additional timing features to pcretest. (1) The -tm option now times
1777        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1778        separate command line item, by a number that specifies the number of
1779        repeats to use when timing. The default is 50000; this gives better
1780        precision, but takes uncomfortably long for very large patterns.
1781    
1782    20. Extended pcre_study() to be more clever in cases where a branch of a
1783        subpattern has no definite first character. For example, (a*|b*)[cd] would
1784        previously give no result from pcre_study(). Now it recognizes that the
1785        first character must be a, b, c, or d.
1786    
1787    21. There was an incorrect error "recursive call could loop indefinitely" if
1788        a subpattern (or the entire pattern) that was being tested for matching an
1789        empty string contained only one non-empty item after a nested subpattern.
1790        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1791        incorrectly, because the \d was being skipped in the check.
1792    
1793    22. The pcretest program now has a new pattern option /B and a command line
1794        option -b, which is equivalent to adding /B to every pattern. This causes
1795        it to show the compiled bytecode, without the additional information that
1796        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1797        is the same as /B/I).
1798    
1799    23. A new optimization is now able automatically to treat some sequences such
1800        as a*b as a*+b. More specifically, if something simple (such as a character
1801        or a simple class like \d) has an unlimited quantifier, and is followed by
1802        something that cannot possibly match the quantified thing, the quantifier
1803        is automatically "possessified".
1804    
1805    24. A recursive reference to a subpattern whose number was greater than 39
1806        went wrong under certain circumstances in UTF-8 mode. This bug could also
1807        have affected the operation of pcre_study().
1808    
1809    25. Realized that a little bit of performance could be had by replacing
1810        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1811    
1812    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1813    
1814    27. Possessive quantifiers such as a++ were previously implemented by turning
1815        them into atomic groups such as ($>a+). Now they have their own opcodes,
1816        which improves performance. This includes the automatically created ones
1817        from 23 above.
1818    
1819    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1820        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1821        the first matched character to be a colon. This applied both to named and
1822        numbered groups.
1823    
1824    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1825    
1826    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1827        building PCRE on Windows easier, so I have included it in the distribution.
1828    
1829    31. There is now a check in pcretest against a ridiculously large number being
1830        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1831        loop, the loop is abandoned.
1832    
1833    32. Forward references to subpatterns in conditions such as (?(2)...) where
1834        subpattern 2 is defined later cause pcre_compile() to search forwards in
1835        the pattern for the relevant set of parentheses. This search went wrong
1836        when there were unescaped parentheses in a character class, parentheses
1837        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1838    
1839    33. "Subroutine" calls and backreferences were previously restricted to
1840        referencing subpatterns earlier in the regex. This restriction has now
1841        been removed.
1842    
1843    34. Added a number of extra features that are going to be in Perl 5.10. On the
1844        whole, these are just syntactic alternatives for features that PCRE had
1845        previously implemented using the Python syntax or my own invention. The
1846        other formats are all retained for compatibility.
1847    
1848        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1849            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1850            also .NET compatible.
1851    
1852        (b) A recursion or subroutine call to a named group can now be defined as
1853            (?&name) as well as (?P>name).
1854    
1855        (c) A backreference to a named group can now be defined as \k<name> or
1856            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1857            5.10, are also .NET compatible.
1858    
1859        (d) A conditional reference to a named group can now use the syntax
1860            (?(<name>) or (?('name') as well as (?(name).
1861    
1862        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1863            groups (named and numbered) that are never evaluated inline, but can be
1864            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1865            is always false. There may be only one alternative in such a group.
1866    
1867        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1868            as the simple (?(R). The condition is true only if the most recent
1869            recursion is that of the given number or name. It does not search out
1870            through the entire recursion stack.
1871    
1872        (g) The escape \gN or \g{N} has been added, where N is a positive or
1873            negative number, specifying an absolute or relative reference.
1874    
1875    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1876        some "unreachable code" warnings.
1877    
1878    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1879        things, this adds five new scripts.
1880    
1881    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1882        There were also incompatibilities regarding the handling of \Q..\E inside
1883        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1884        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1885    
1886    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1887        matches an empty string, and forcibly breaks the loop. There were bugs in
1888        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1889        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1890        separate and independent bugs (that affected different cases) have been
1891        fixed.
1892    
1893    39. Refactored the code to abolish the use of different opcodes for small
1894        capturing bracket numbers. This is a tidy that I avoided doing when I
1895        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1896        The new approach is not only tidier, it makes it possible to reduce the
1897        memory needed to fix the previous bug (38).
1898    
1899    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1900        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1901        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1902        mode.
1903    
1904    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1905        report.
1906    
1907    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1908        copy construction and assignment in the C++ wrapper.
1909    
1910    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1911        couple of bugs that could have given wrong results in the "--newline=crlf"
1912        case.
1913    
1914    44. Added a number of casts and did some reorganization of signed/unsigned int
1915        variables following suggestions from Dair Grant. Also renamed the variable
1916        "this" as "item" because it is a C++ keyword.
1917    
1918    45. Arranged for dftables to add
1919    
1920          #include "pcre_internal.h"
1921    
1922        to pcre_chartables.c because without it, gcc 4.x may remove the array
1923        definition from the final binary if PCRE is built into a static library and
1924        dead code stripping is activated.
1925    
1926    46. For an unanchored pattern, if a match attempt fails at the start of a
1927        newline sequence, and the newline setting is CRLF or ANY, and the next two
1928        characters are CRLF, advance by two characters instead of one.
1929    
1930    
1931    Version 6.7 04-Jul-06
1932    ---------------------
1933    
1934     1. In order to handle tests when input lines are enormously long, pcretest has
1935        been re-factored so that it automatically extends its buffers when
1936        necessary. The code is crude, but this _is_ just a test program. The
1937        default size has been increased from 32K to 50K.
1938    
1939     2. The code in pcre_study() was using the value of the re argument before
1940        testing it for NULL. (Of course, in any sensible call of the function, it
1941        won't be NULL.)
1942    
1943     3. The memmove() emulation function in pcre_internal.h, which is used on
1944        systems that lack both memmove() and bcopy() - that is, hardly ever -
1945        was missing a "static" storage class specifier.
1946    
1947     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1948        containing an extended class (one that cannot be represented by a bitmap
1949        because it contains high-valued characters or Unicode property items, e.g.
1950        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1951        pattern, but PCRE should not loop if you do not (it no longer does).
1952        [Detail: two cases were found: (a) a repeated subpattern containing an
1953        extended class; (b) a recursive reference to a subpattern that followed a
1954        previous extended class. It wasn't skipping over the extended class
1955        correctly when UTF-8 mode was not set.]
1956    
1957     5. A negated single-character class was not being recognized as fixed-length
1958        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1959        compile error "lookbehind assertion is not fixed length".
1960    
1961     6. The RunPerlTest auxiliary script was showing an unexpected difference
1962        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1963        write a Perl script that can interpret lines of an input file either as
1964        byte characters or as UTF-8, which is what "perltest" was being required to
1965        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1966        can't do is switch easily at run time between having the "use utf8;" pragma
1967        or not. In the end, I fudged it by using the RunPerlTest script to insert
1968        "use utf8;" explicitly for the UTF-8 tests.
1969    
1970     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1971        the end of the subject string, contrary to the documentation and to what
1972        Perl does. This was true of both matching functions. Now it matches only at
1973        the start of the subject and immediately after *internal* newlines.
1974    
1975     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1976        a pointer to an int instead of a pointer to an unsigned long int. This
1977        caused problems on 64-bit systems.
1978    
1979     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1980        instance of the 'standard' template library not being so standard".
1981    
1982    10. There was no check on the number of named subpatterns nor the maximum
1983        length of a subpattern name. The product of these values is used to compute
1984        the size of the memory block for a compiled pattern. By supplying a very
1985        long subpattern name and a large number of named subpatterns, the size
1986        computation could be caused to overflow. This is now prevented by limiting
1987        the length of names to 32 characters, and the number of named subpatterns
1988        to 10,000.
1989    
1990    11. Subpatterns that are repeated with specific counts have to be replicated in
1991        the compiled pattern. The size of memory for this was computed from the
1992        length of the subpattern and the repeat count. The latter is limited to
1993        65535, but there was no limit on the former, meaning that integer overflow
1994        could in principle occur. The compiled length of a repeated subpattern is
1995        now limited to 30,000 bytes in order to prevent this.
1996    
1997    12. Added the optional facility to have named substrings with the same name.
1998    
1999    13. Added the ability to use a named substring as a condition, using the
2000        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2001        are numbers (not recommended). Forward references are permitted.
2002    
2003    14. Added forward references in named backreferences (if you see what I mean).
2004    
2005    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2006        pattern could run off the end of the subject. For example, the pattern
2007        "(?s)(.{1,5})"8 did this with the subject "ab".
2008    
2009    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2010        PCRE_CASELESS was set when matching characters that were quantified with ?
2011        or *.
2012    
2013    17. A character class other than a single negated character that had a minimum
2014        but no maximum quantifier - for example [ab]{6,} - was not handled
2015        correctly by pce_dfa_exec(). It would match only one character.
2016    
2017    18. A valid (though odd) pattern that looked like a POSIX character
2018        class but used an invalid character after [ (for example [[,abc,]]) caused
2019        pcre_compile() to give the error "Failed: internal error: code overflow" or
2020        in some cases to crash with a glibc free() error. This could even happen if
2021        the pattern terminated after [[ but there just happened to be a sequence of
2022        letters, a binary zero, and a closing ] in the memory that followed.
2023    
2024    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2025        over the years. Originally (before any Unicode support), just the bottom 8
2026        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2027        output from "man perlunicode" includes this:
2028    
2029          The regular expression compiler produces polymorphic opcodes.  That
2030          is, the pattern adapts to the data and automatically switches to
2031          the Unicode character scheme when presented with Unicode data--or
2032          instead uses a traditional byte scheme when presented with byte
2033          data.
2034    
2035        Sadly, a wide octal escape does not cause a switch, and in a string with
2036        no other multibyte characters, these octal escapes are treated as before.
2037        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
2038        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2039        Unicode string.
2040    
2041        I have not perpetrated such confusion in PCRE. Up till now, it took just
2042        the bottom 8 bits, as in old Perl. I have now made octal escapes with
2043        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2044        translate to the appropriate multibyte character.
2045    
2046    29. Applied some refactoring to reduce the number of warnings from Microsoft
2047        and Borland compilers. This has included removing the fudge introduced
2048        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2049        a warning about an unused variable.
2050    
2051    21. PCRE has not included VT (character 0x0b) in the set of whitespace
2052        characters since release 4.0, because Perl (from release 5.004) does not.
2053        [Or at least, is documented not to: some releases seem to be in conflict
2054        with the documentation.] However, when a pattern was studied with
2055        pcre_study() and all its branches started with \s, PCRE still included VT
2056        as a possible starting character. Of course, this did no harm; it just
2057        caused an unnecessary match attempt.
2058    
2059    22. Removed a now-redundant internal flag bit that recorded the fact that case
2060        dependency changed within the pattern. This was once needed for "required
2061        byte" processing, but is no longer used. This recovers a now-scarce options
2062        bit. Also moved the least significant internal flag bit to the most-
2063        significant bit of the word, which was not previously used (hangover from
2064        the days when it was an int rather than a uint) to free up another bit for
2065        the future.
2066    
2067    23. Added support for CRLF line endings as well as CR and LF. As well as the
2068        default being selectable at build time, it can now be changed at runtime
2069        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2070        specify that it is scanning data with non-default line endings.
2071    
2072    24. Changed the definition of CXXLINK to make it agree with the definition of
2073        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2074    
2075    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2076        recursions. This makes a big different to stack usage for some patterns.
2077    
2078    26. If a subpattern containing a named recursion or subroutine reference such
2079        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2080        the space required for the compiled pattern went wrong and gave too small a
2081        value. Depending on the environment, this could lead to "Failed: internal
2082        error: code overflow at offset 49" or "glibc detected double free or
2083        corruption" errors.
2084    
2085    27. Applied patches from Google (a) to support the new newline modes and (b) to
2086        advance over multibyte UTF-8 characters in GlobalReplace.
2087    
2088    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2089        difference for some implementation of PCRE in some Windows version.
2090    
2091    29. Added some extra testing facilities to pcretest:
2092    
2093        \q<number>   in a data line sets the "match limit" value
2094        \Q<number>   in a data line sets the "match recursion limt" value
2095        -S <number>  sets the stack size, where <number> is in megabytes
2096    
2097        The -S option isn't available for Windows.
2098    
2099    
2100    Version 6.6 06-Feb-06
2101    ---------------------
2102    
2103     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2104        in pcreposix.h. I have copied the definition from pcre.h.
2105    
2106     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2107        because pcre.h is no longer a built file.
2108    
2109     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2110        not normally included in the compiled code.
2111    
2112    
2113    Version 6.5 01-Feb-06
2114    ---------------------
2115    
2116     1. When using the partial match feature with pcre_dfa_exec(), it was not
2117        anchoring the second and subsequent partial matches at the new starting
2118        point. This could lead to incorrect results. For example, with the pattern
2119        /1234/, partially matching against "123" and then "a4" gave a match.
2120    
2121     2. Changes to pcregrep:
2122    
2123        (a) All non-match returns from pcre_exec() were being treated as failures
2124            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2125            error message is output. Some extra information is given for the
2126            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2127            probably the only errors that are likely to be caused by users (by
2128            specifying a regex that has nested indefinite repeats, for instance).
2129            If there are more than 20 of these errors, pcregrep is abandoned.
2130    
2131        (b) A binary zero was treated as data while matching, but terminated the
2132            output line if it was written out. This has been fixed: binary zeroes
2133            are now no different to any other data bytes.
2134    
2135        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2136            used to set a locale for matching. The --locale=xxxx long option has
2137            been added (no short equivalent) to specify a locale explicitly on the
2138            pcregrep command, overriding the environment variables.
2139    
2140        (d) When -B was used with -n, some line numbers in the output were one less
2141            than they should have been.
2142    
2143        (e) Added the -o (--only-matching) option.
2144    
2145        (f) If -A or -C was used with -c (count only), some lines of context were
2146            accidentally printed for the final match.
2147    
2148        (g) Added the -H (--with-filename) option.
2149    
2150        (h) The combination of options -rh failed to suppress file names for files
2151            that were found from directory arguments.
2152    
2153        (i) Added the -D (--devices) and -d (--directories) options.
2154    
2155        (j) Added the -F (--fixed-strings) option.
2156    
2157        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2158    
2159        (l) Added the --colo(u)r option.
2160    
2161        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2162            is not present by default.
2163    
2164     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2165        items such as (?R) or (?1), when the recursion could match a number of
2166        alternatives. If it matched one of the alternatives, but subsequently,
2167        outside the recursion, there was a failure, the code tried to back up into
2168        the recursion. However, because of the way PCRE is implemented, this is not
2169        possible, and the result was an incorrect result from the match.
2170    
2171        In order to prevent this happening, the specification of recursion has
2172        been changed so that all such subpatterns are automatically treated as
2173        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2174    
2175     4. I had overlooked the fact that, in some locales, there are characters for
2176        which isalpha() is true but neither isupper() nor islower() are true. In
2177        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2178        and ordfeminine) are like this. This affected the treatment of \w and \W
2179        when they appeared in character classes, but not when they appeared outside
2180        a character class. The bit map for "word" characters is now created
2181        separately from the results of isalnum() instead of just taking it from the
2182        upper, lower, and digit maps. (Plus the underscore character, of course.)
2183    
2184     5. The above bug also affected the handling of POSIX character classes such as
2185        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2186        permanent tables. Instead, the bit maps for such a class were previously
2187        created as the appropriate unions of the upper, lower, and digit bitmaps.
2188        Now they are created by subtraction from the [[:word:]] class, which has
2189        its own bitmap.
2190    
2191     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2192        It is created by subtracting the vertical space characters (\x09, \x0a,
2193        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2194        subtraction was done in the overall bitmap for a character class, meaning
2195        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2196        be recognized. This bug has been fixed.
2197    
2198     7. Patches from the folks at Google:
2199    
2200          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2201          real life, but is still worth protecting against".
2202    
2203          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2204          regular expressions".
2205    
2206          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2207          have it.
2208    
2209          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2210          "configure" and the latter not, in order to fix a problem somebody had
2211          with compiling the Arg class on HP-UX.
2212    
2213          (e) Improve the error-handling of the C++ wrapper a little bit.
2214    
2215          (f) New tests for checking recursion limiting.
2216    
2217     8. The pcre_memmove() function, which is used only if the environment does not
2218        have a standard memmove() function (and is therefore rarely compiled),
2219        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2220        returning a result (though PCRE never actually uses the result).
2221    
2222     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2223        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2224        returned instead of calling malloc() with an overflowing number that would
2225        most likely cause subsequent chaos.
2226    
2227    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2228    
2229    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2230        with this option is matched, the nmatch and pmatch options of regexec() are
2231        ignored.
2232    
2233    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2234        provided in case anyone wants to the the POSIX interface with UTF-8
2235        strings.
2236    
2237    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2238        C++ linking (needed for some HP-UX environments).
2239    
2240    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2241        (unused parameter) and in the pcre_printint() function (omitted "default"
2242        switch label when the default is to do nothing).
2243    
2244    15. Added some code to make it possible, when PCRE is compiled as a C++
2245        library, to replace subject pointers for pcre_exec() with a smart pointer
2246        class, thus making it possible to process discontinuous strings.
2247    
2248    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2249        much the same function. They were added by different people who were trying
2250        to make PCRE easy to compile on non-Unix systems. It has been suggested
2251        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2252        for compiling on Windows systems. I have therefore replaced it with
2253        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2254        defaults to "extern" for C or "extern C" for C++, which works fine on
2255        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2256        SCOPE with something explicit in config.h. In addition:
2257    
2258        (a) pcreposix.h still had just "extern" instead of either of these macros;
2259            I have replaced it with PCRE_DATA_SCOPE.
2260    
2261        (b) Functions such as _pcre_xclass(), which are internal to the library,
2262            but external in the C sense, all had PCRE_EXPORT in their definitions.
2263            This is apparently wrong for the Windows case, so I have removed it.
2264            (It makes no difference on Unix-like systems.)
2265    
2266    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2267        of recursive calls to match(). This is different to MATCH_LIMIT because
2268        that limits the total number of calls to match(), not all of which increase
2269        the depth of recursion. Limiting the recursion depth limits the amount of
2270        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2271        when PCRE is compiled, and changed at run time. A patch from Google adds
2272        this functionality to the C++ interface.
2273    
2274    18. Changes to the handling of Unicode character properties:
2275    
2276        (a) Updated the table to Unicode 4.1.0.
2277    
2278        (b) Recognize characters that are not in the table as "Cn" (undefined).
2279    
2280        (c) I revised the way the table is implemented to a much improved format
2281            which includes recognition of ranges. It now supports the ranges that
2282            are defined in UnicodeData.txt, and it also amalgamates other
2283            characters into ranges. This has reduced the number of entries in the
2284            table from around 16,000 to around 3,000, thus reducing its size
2285            considerably. I realized I did not need to use a tree structure after
2286            all - a binary chop search is just as efficient. Having reduced the
2287            number of entries, I extended their size from 6 bytes to 8 bytes to
2288            allow for more data.
2289    
2290        (d) Added support for Unicode script names via properties such as \p{Han}.
2291    
2292    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2293        matching that character.
2294    
2295    20. When matching a repeated Unicode property with a minimum greater than zero,
2296        (for example \pL{2,}), PCRE could look past the end of the subject if it
2297        reached it while seeking the minimum number of characters. This could
2298        happen only if some of the characters were more than one byte long, because
2299        there is a check for at least the minimum number of bytes.
2300    
2301    21. Refactored the implementation of \p and \P so as to be more general, to
2302        allow for more different types of property in future. This has changed the
2303        compiled form incompatibly. Anybody with saved compiled patterns that use
2304        \p or \P will have to recompile them.
2305    
2306    22. Added "Any" and "L&" to the supported property types.
2307    
2308    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2309        but give a compile time error if the value is greater than 0xff.
2310    
2311    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2312        accidentally not being installed or uninstalled.
2313    
2314    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2315        made were to insert the current release number. This seemed silly, because
2316        it made things harder for people building PCRE on systems that don't run
2317        "configure". I have turned pcre.h into a distributed file, no longer built
2318        by "configure", with the version identification directly included. There is
2319        no longer a pcre.h.in file.
2320    
2321        However, this change necessitated a change to the pcre-config script as
2322        well. It is built from pcre-config.in, and one of the substitutions was the
2323        release number. I have updated configure.ac so that ./configure now finds
2324        the release number by grepping pcre.h.
2325    
2326    26. Added the ability to run the tests under valgrind.
2327    
2328    
2329    Version 6.4 05-Sep-05
2330    ---------------------
2331    
2332     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2333        "--" to be printed when multiple files were scanned, even when none of the
2334        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2335        consider it to be a bug, and have restored the previous behaviour.
2336    
2337     2. A couple of code tidies to get rid of compiler warnings.
2338    
2339     3. The pcretest program used to cheat by referring to symbols in the library
2340        whose names begin with _pcre_. These are internal symbols that are not
2341        really supposed to be visible externally, and in some environments it is
2342        possible to suppress them. The cheating is now confined to including
2343        certain files from the library's source, which is a bit cleaner.
2344    
2345     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2346        file's purpose clearer.
2347    
2348     5. Reorganized pcre_ucp_findchar().
2349    
2350    
2351    Version 6.3 15-Aug-05
2352    ---------------------
2353    
2354     1. The file libpcre.pc.in did not have general read permission in the tarball.
2355    
2356     2. There were some problems when building without C++ support:
2357    
2358        (a) If C++ support was not built, "make install" and "make test" still
2359            tried to test it.
2360    
2361        (b) There were problems when the value of CXX was explicitly set. Some
2362            changes have been made to try to fix these, and ...
2363    
2364        (c) --disable-cpp can now be used to explicitly disable C++ support.
2365    
2366        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2367            backslash in a target when C++ was disabled. This confuses some
2368            versions of "make", apparently. Using an intermediate variable solves
2369            this. (Same for CPP_LOBJ.)
2370    
2371     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2372        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2373        necessary on certain architectures.
2374    
2375     4. Added a setting of -export-symbols-regex to the link command to remove
2376        those symbols that are exported in the C sense, but actually are local
2377        within the library, and not documented. Their names all begin with
2378        "_pcre_". This is not a perfect job, because (a) we have to except some
2379        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2380        available (and never for static libraries). I have made a note to try to
2381        find a way round (a) in the future.
2382    
2383    
2384    Version 6.2 01-Aug-05
2385    ---------------------
2386    
2387     1. There was no test for integer overflow of quantifier values. A construction
2388        such as {1111111111111111} would give undefined results. What is worse, if
2389        a minimum quantifier for a parenthesized subpattern overflowed and became
2390        negative, the calculation of the memory size went wrong. This could have
2391        led to memory overwriting.
2392    
2393     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2394    
2395     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2396        operating environments where this matters.
2397    
2398     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2399        PCRE options from within the C++ wrapper.
2400    
2401     5. Named capturing subpatterns were not being correctly counted when a pattern
2402        was compiled. This caused two problems: (a) If there were more than 100
2403        such subpatterns, the calculation of the memory needed for the whole
2404        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2405        back references of the form \12, where the number was greater than 9, were
2406        not recognized as back references, even though there were sufficient
2407        previous subpatterns.
2408    
2409     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2410        versions of gcc, e.g. 2.95.4.
2411    
2412    
2413    Version 6.1 21-Jun-05
2414    ---------------------
2415    
2416     1. There was one reference to the variable "posix" in pcretest.c that was not
2417        surrounded by "#if !defined NOPOSIX".
2418    
2419     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2420        the cross-check on the old pcre_info() function, for the benefit of the
2421        cut-down version of PCRE that is currently imported into Exim.
2422    
2423     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2424        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2425        patterns that start (?i) but I don't think that matters. The use of (?i) is
2426        just an example; this all applies to the other options as well.
2427    
2428     4. Since libtool seems to echo the compile commands it is issuing, the output
2429        from "make" can be reduced a bit by putting "@" in front of each libtool
2430        compile command.
2431    
2432     5. Patch from the folks at Google for configure.in to be a bit more thorough
2433        in checking for a suitable C++ installation before trying to compile the
2434        C++ stuff. This should fix a reported problem when a compiler was present,
2435        but no suitable headers.
2436    
2437     6. The man pages all had just "PCRE" as their title. I have changed them to
2438        be the relevant file name. I have also arranged that these names are
2439        retained in the file doc/pcre.txt, which is a concatenation in text format
2440        of all the man pages except the little individual ones for each function.
2441    
2442     7. The NON-UNIX-USE file had not been updated for the different set of source
2443        files that come with release 6. I also added a few comments about the C++
2444        wrapper.
2445    
2446    
2447    Version 6.0 07-Jun-05
2448    ---------------------
2449    
2450     1. Some minor internal re-organization to help with my DFA experiments.
2451    
2452     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2453        didn't matter for the library itself when fully configured, but did matter
2454        when compiling without UCP support, or within Exim, where the ucp files are
2455        not imported.
2456    
2457     3. Refactoring of the library code to split up the various functions into
2458        different source modules. The addition of the new DFA matching code (see
2459        below) to a single monolithic source would have made it really too
2460        unwieldy, quite apart from causing all the code to be include in a
2461        statically linked application, when only some functions are used. This is
2462        relevant even without the DFA addition now that patterns can be compiled in
2463        one application and matched in another.
2464    
2465        The downside of splitting up is that there have to be some external
2466        functions and data tables that are used internally in different modules of
2467        the library but which are not part of the API. These have all had their
2468        names changed to start with "_pcre_" so that they are unlikely to clash
2469        with other external names.
2470    
2471     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2472        a different (DFA) algorithm. Although it is slower than the original
2473        function, it does have some advantages for certain types of matching
2474        problem.
2475    
2476     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2477        including restarting after a partial match.
2478    
2479     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2480        defined when compiling for Windows was sent to me. I have put it into the
2481        code, though I have no means of testing or verifying it.
2482    
2483     7. Added the pcre_refcount() auxiliary function.
2484    
2485     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2486        match before or at the first newline in the subject string. In pcretest,
2487        the /f option on a pattern can be used to set this.
2488    
2489     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2490        would behave wrongly. This has been present in PCRE since release 4.0.
2491    
2492    10. A number of changes to the pcregrep command:
2493    
2494        (a) Refactored how -x works; insert ^(...)$ instead of setting
2495            PCRE_ANCHORED and checking the length, in preparation for adding
2496            something similar for -w.
2497    
2498        (b) Added the -w (match as a word) option.
2499    
2500        (c) Refactored the way lines are read and buffered so as to have more
2501            than one at a time available.
2502    
2503        (d) Implemented a pcregrep test script.
2504    
2505        (e) Added the -M (multiline match) option. This allows patterns to match
2506            over several lines of the subject. The buffering ensures that at least
2507            8K, or the rest of the document (whichever is the shorter) is available
2508            for matching (and similarly the previous 8K for lookbehind assertions).
2509    
2510        (f) Changed the --help output so that it now says
2511    
2512              -w, --word-regex(p)
2513    
2514            instead of two lines, one with "regex" and the other with "regexp"
2515            because that confused at least one person since the short forms are the
2516            same. (This required a bit of code, as the output is generated
2517            automatically from a table. It wasn't just a text change.)
2518    
2519        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2520            option but starts with a hyphen. Could be a pattern or a path name
2521            starting with a hyphen, for instance.
2522    
2523        (h) "-" can be given as a file name to represent stdin.
2524    
2525        (i) When file names are being printed, "(standard input)" is used for
2526            the standard input, for compatibility with GNU grep. Previously
2527            "<stdin>" was used.
2528    
2529        (j) The option --label=xxx can be used to supply a name to be used for
2530            stdin when file names are being printed. There is no short form.
2531    
2532        (k) Re-factored the options decoding logic because we are going to add
2533            two more options that take data. Such options can now be given in four
2534            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2535    
2536        (l) Added the -A, -B, and -C options for requesting that lines of context
2537            around matches be printed.
2538    
2539        (m) Added the -L option to print the names of files that do not contain
2540            any matching lines, that is, the complement of -l.
2541    
2542        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2543            continue to scan other files.
2544    
2545        (o) The -s option was incorrectly implemented. For compatibility with other
2546            greps, it now suppresses the error message for a non-existent or non-
2547            accessible file (but not the return code). There is a new option called
2548            -q that suppresses the output of matching lines, which was what -s was
2549            previously doing.
2550    
2551        (p) Added --include and --exclude options to specify files for inclusion
2552            and exclusion when recursing.
2553    
2554    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2555        Hopefully, it now does.
2556    
2557    12. Missing cast in pcre_study().
2558    
2559    13. Added an "uninstall" target to the makefile.
2560    
2561    14. Replaced "extern" in the function prototypes in Makefile.in with
2562        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2563        world, but is set differently for Windows.
2564    
2565    15. Added a second compiling function called pcre_compile2(). The only
2566        difference is that it has an extra argument, which is a pointer to an
2567        integer error code. When there is a compile-time failure, this is set
2568        non-zero, in addition to the error test pointer being set to point to an
2569        error message. The new argument may be NULL if no error number is required
2570        (but then you may as well call pcre_compile(), which is now just a
2571        wrapper). This facility is provided because some applications need a
2572        numeric error indication, but it has also enabled me to tidy up the way
2573        compile-time errors are handled in the POSIX wrapper.
2574    
2575    16. Added VPATH=.libs to the makefile; this should help when building with one
2576        prefix path and installing with another. (Or so I'm told by someone who
2577        knows more about this stuff than I do.)
2578    
2579    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2580        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2581        match everything, including newlines. This is not POSIX-compatible, but
2582        somebody wanted the feature. From pcretest it can be activated by using
2583        both the P and the s flags.
2584    
2585    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2586    
2587    19. libpcre.pc was being incorrectly installed as executable.
2588    
2589    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2590        it now also looks for '\r' so that it will work unmodified on Windows.
2591    
2592    21. Added Google's contributed C++ wrapper to the distribution.
2593    
2594    22. Added some untidy missing memory free() calls in pcretest, to keep
2595        Electric Fence happy when testing.
2596    
2597    
2598    
2599    Version 5.0 13-Sep-04
2600    ---------------------
2601    
2602     1. Internal change: literal characters are no longer packed up into items
2603        containing multiple characters in a single byte-string. Each character
2604        is now matched using a separate opcode. However, there may be more than one
2605        byte in the character in UTF-8 mode.
2606    
2607     2. The pcre_callout_block structure has two new fields: pattern_position and
2608        next_item_length. These contain the offset in the pattern to the next match
2609        item, and its length, respectively.
2610    
2611     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2612        insertion of callouts before each pattern item. Added the /C option to
2613        pcretest to make use of this.
2614    
2615     4. On the advice of a Windows user, the lines
2616    
2617          #if defined(_WIN32) || defined(WIN32)
2618          _setmode( _fileno( stdout ), 0x8000 );
2619          #endif  /* defined(_WIN32) || defined(WIN32) */
2620    
2621        have been added to the source of pcretest. This apparently does useful
2622        magic in relation to line terminators.
2623    
2624     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2625        for the benefit of those environments where the "b" makes a difference.
2626    
2627     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2628        to know about it. I have put a hack into configure.in that adds in code
2629        to set GCC=yes if CC=icc. This seems to end up at a point in the
2630        generated configure script that is early enough to affect the setting of
2631        compiler options, which is what is needed, but I have no means of testing
2632        whether it really works. (The user who reported this had patched the
2633        generated configure script, which of course I cannot do.)
2634    
2635        LATER: After change 22 below (new libtool files), the configure script
2636        seems to know about icc (and also ecc). Therefore, I have commented out
2637        this hack in configure.in.
2638    
2639     7. Added support for pkg-config (2 patches were sent in).
2640    
2641     8. Negated POSIX character classes that used a combination of internal tables
2642        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2643        [[:^ascii]]. Typically, they would match almost any characters. The other
2644        POSIX classes were not broken in this way.
2645    
2646     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2647        to find the match, as PCRE was deluded into thinking that the match had to
2648        start at the start point or following a newline. The same bug applied to
2649        patterns with negative forward assertions or any backward assertions
2650        preceding ".*" at the start, unless the pattern required a fixed first
2651        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2652    
2653    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2654        starting at the last subject character, bytes beyond the end of the subject
2655        string were read.
2656    
2657    11. Renamed the variable "class" as "classbits" to make life easier for C++
2658        users. (Previously there was a macro definition, but it apparently wasn't
2659        enough.)
2660    
2661    12. Added the new field "tables" to the extra data so that tables can be passed
2662        in at exec time, or the internal tables can be re-selected. This allows
2663        a compiled regex to be saved and re-used at a later time by a different
2664        program that might have everything at different addresses.
2665    
2666    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2667        -R library as well as a -L library.
2668    
2669    14. The debugging options of pcretest (-d on the command line or D on a
2670        pattern) showed incorrect output for anything following an extended class
2671        that contained multibyte characters and which was followed by a quantifier.
2672    
2673    15. Added optional support for general category Unicode character properties
2674        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2675        support. It adds about 90K to the size of the library. The meanings of the
2676        inbuilt class escapes such as \d and \s have NOT been changed.
2677    
2678    16. Updated pcredemo.c to include calls to free() to release the memory for the
2679        compiled pattern.
2680    
2681    17. The generated file chartables.c was being created in the source directory
2682        instead of in the building directory. This caused the build to fail if the
2683        source directory was different from the building directory, and was
2684        read-only.
2685    
2686    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2687        file. No doubt somebody will tell me if they don't make sense... Also added
2688        Dan Mooney's comments about building on OpenVMS.
2689    
2690    19. Added support for partial matching via the PCRE_PARTIAL option for
2691        pcre_exec() and the \P data escape in pcretest.
2692    
2693    20. Extended pcretest with 3 new pattern features:
2694    
2695        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2696              write the compiled pattern to the file whose name is "rest-of-line".
2697              This is a straight binary dump of the data, with the saved pointer to
2698              the character tables forced to be NULL. The study data, if any, is
2699              written too. After writing, pcretest reads a new pattern.
2700    
2701        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2702              compiled pattern from the given file. There must not be any
2703              occurrences of "<" in the file name (pretty unlikely); if there are,
2704              pcretest will instead treat the initial "<" as a pattern delimiter.
2705              After reading in the pattern, pcretest goes on to read data lines as
2706              usual.
2707    
2708        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2709              and 16-bit fields in a compiled pattern, to simulate a pattern that
2710              was compiled on a host of opposite endianness.
2711    
2712    21. The pcre-exec() function can now cope with patterns that were compiled on
2713        hosts of opposite endianness, with this restriction:
2714    
2715          As for any compiled expression that is saved and used later, the tables
2716          pointer field cannot be preserved; the extra_data field in the arguments
2717          to pcre_exec() should be used to pass in a tables address if a value
2718          other than the default internal tables were used at compile time.
2719    
2720    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2721        now diagnosed as an error. Previously, most of the time, a negative number
2722        would have been treated as zero, but if in addition "ovector" was passed as
2723        NULL, a crash could occur.
2724    
2725    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2726        new versions from the libtool 1.5 distribution (the last one is a copy of
2727        a file called libtool.m4). This seems to have fixed the need to patch
2728        "configure" to support Darwin 1.3 (which I used to do). However, I still
2729        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2730        workstation).
2731    
2732    24. Changed the PCRE licence to be the more standard "BSD" licence.
2733    
2734    
2735    Version 4.5 01-Dec-03
2736    ---------------------
2737    
2738     1. There has been some re-arrangement of the code for the match() function so
2739        that it can be compiled in a version that does not call itself recursively.
2740        Instead, it keeps those local variables that need separate instances for
2741        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2742        needs to "recurse". Keeping track of where control must go is done by means
2743        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2744        hide most of the details from the main code, and operates only if
2745        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2746        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2747        operating.
2748    
2749        To make it easier for callers to provide specially tailored get/free
2750        functions for this usage, two new functions, pcre_stack_malloc, and
2751        pcre_stack_free, are used. They are always called in strict stacking order,
2752        and the size of block requested is always the same.
2753    
2754        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2755        PCRE has been compiled to use the stack or the heap for recursion. The
2756        -C option of pcretest uses this to show which version is compiled.
2757    
2758        A new data escape \S, is added to pcretest; it causes the amounts of store
2759        obtained and freed by both kinds of malloc/free at match time to be added
2760        to the output.
2761    
2762     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2763        what's available on my current Linux desktop machine.
2764    
2765     3. When matching a UTF-8 string, the test for a valid string at the start has
2766        been extended. If start_offset is not zero, PCRE now checks that it points
2767        to a byte that is the start of a UTF-8 character. If not, it returns
2768        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2769        this is necessary because there may be backward assertions in the pattern.
2770        When matching the same subject several times, it may save resources to use
2771        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2772    
2773     4. The code for checking the validity of UTF-8 strings has been tightened so
2774        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2775        containing "overlong sequences".
2776    
2777     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2778        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2779        should have been "&", but it just so happened that all the cases this let
2780        through by mistake were picked up later in the function.
2781    
2782     6. I had used a variable called "isblank" - this is a C99 function, causing
2783        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2784    
2785     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2786        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2787        so the version is shown; (c) stop "make test" echoing "./RunTest".
2788    
2789     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2790    
2791     9. The prototype for memmove() for systems that don't have it was using
2792        size_t, but the inclusion of the header that defines size_t was later. I've
2793        moved the #includes for the C headers earlier to avoid this.
2794    
2795    10. Added some adjustments to the code to make it easier to compiler on certain
2796        special systems:
2797    
2798          (a) Some "const" qualifiers were missing.
2799          (b) Added the macro EXPORT before all exported functions; by default this
2800              is defined to be empty.
2801          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2802              that it reads its output file name as an argument instead of writing
2803              to the standard output and assuming this can be redirected.
2804    
2805    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2806        class containing characters with values greater than 255, PCRE compilation
2807        went into a loop.
2808    
2809    12. A recursive reference to a subpattern that was within another subpattern
2810        that had a minimum quantifier of zero caused PCRE to crash. For example,
2811        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2812        recursion. If the recursively-called subpattern itself had a zero repeat,
2813        that was OK.
2814    
2815    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2816        buffer into which it was copied (for escape processing) was still set at
2817        1024, so long lines caused crashes.
2818    
2819    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2820        "internal error: code overflow...". This applied to any character class
2821        that was followed by a possessive quantifier.
2822    
2823    15. Modified the Makefile to add libpcre.la as a prerequisite for
2824        libpcreposix.la because I was told this is needed for a parallel build to
2825        work.
2826    
2827    16. If a pattern that contained .* following optional items at the start was
2828        studied, the wrong optimizing data was generated, leading to matching
2829        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2830        matching string must start with a or b or c. The correct conclusion for
2831        this pattern is that a match can start with any character.
2832    
2833    
2834    Version 4.4 13-Aug-03
2835    ---------------------
2836    
2837     1. In UTF-8 mode, a character class containing characters with values between
2838        127 and 255 was not handled correctly if the compiled pattern was studied.
2839        In fixing this, I have also improved the studying algorithm for such
2840        classes (slightly).
2841    
2842     2. Three internal functions had redundant arguments passed to them. Removal
2843        might give a very teeny performance improvement.
2844    
2845     3. Documentation bug: the value of the capture_top field in a callout is *one
2846        more than* the number of the hightest numbered captured substring.
2847    
2848     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2849        in incorrectly linking with a previously installed version. They now link
2850        explicitly with libpcre.la.
2851    
2852     5. configure.in no longer needs to recognize Cygwin specially.
2853    
2854     6. A problem in pcre.in for Windows platforms is fixed.
2855    
2856     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2857        pcretest, it used to include the size of the study block as part of its
2858        output. Unfortunately, the structure contains a field that has a different
2859        size on different hardware architectures. This meant that the tests that
2860        showed this size failed. As the block is currently always of a fixed size,
2861        this information isn't actually particularly useful in pcretest output, so
2862        I have just removed it.
2863    
2864     8. Three pre-processor statements accidentally did not start in column 1.
2865        Sadly, there are *still* compilers around that complain, even though
2866        standard C has not required this for well over a decade. Sigh.
2867    
2868     9. In pcretest, the code for checking callouts passed small integers in the
2869        callout_data field, which is a void * field. However, some picky compilers
2870        complained about the casts involved for this on 64-bit systems. Now
2871        pcretest passes the address of the small integer instead, which should get
2872        rid of the warnings.
2873    
2874    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2875        both compile and run time, and gives an error if an invalid UTF-8 sequence
2876        is found. There is a option for disabling this check in cases where the
2877        string is known to be correct and/or the maximum performance is wanted.
2878    
2879    11. In response to a bug report, I changed one line in Makefile.in from
2880    
2881            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2882        to
2883            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2884    
2885        to look similar to other lines, but I have no way of telling whether this
2886        is the right thing to do, as I do not use Windows. No doubt I'll get told
2887        if it's wrong...
2888    
2889    
2890  Version 4.3 21-May-03  Version 4.3 21-May-03
2891  ---------------------  ---------------------
2892    

Legend:
Removed from v.69  
changed lines
  Added in v.606

  ViewVC Help
Powered by ViewVC 1.1.5