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

Diff of /code/trunk/ChangeLog

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

revision 83 by nigel, Sat Feb 24 21:41:06 2007 UTC revision 308 by ph10, Mon Jan 21 14:57:19 2008 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.6 19-Jan-08
5    ---------------------
6    
7    1.  A character class containing a very large number of characters with
8        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
9        overflow.
10    
11    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
12        HAVE_LONG_LONG is not defined.
13    
14    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
15        bring them up to date and include new features. This patch includes:
16    
17        - Fixed PH's badly added libz and libbz2 support.
18        - Fixed a problem with static linking.
19        - Added pcredemo.
20        - Fixed dftables problem and added an option.
21        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
22            HAVE_LONG_LONG.
23        - Added readline support for pcretest.
24        - Added an listing of the option settings after cmake has run.
25    
26    4.  A user submitted a patch to Makefile that makes it easy to created a dll
27        under mingw. I added stuff to Makefile.am that cause it to include this
28        special target, without affecting anything else.
29    
30    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
31        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
32        exported in the Windows port". It has not yet been confirmed that the patch
33        solves the problem, but it does no harm.
34    
35    
36    Version 7.5 10-Jan-08
37    ---------------------
38    
39    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
40        values in parens when parsing an RE using the C++ wrapper."
41    
42    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
43        Characters greater than 255 were excluded from the class instead of being
44        included.
45    
46    3.  The same bug as (2) above applied to negated POSIX classes such as
47        [:^space:].
48    
49    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
50        defined or documented. It seems to have been a typo for PCRE_STATIC, so
51        I have changed it.
52    
53    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
54        first named subpattern) and a construct such as (?&a) would reference the
55        first named subpattern whose name started with "a" (in other words, the
56        length check was missing). Both these problems are fixed. "Subpattern name
57        expected" is now given for (?&) (a zero-length name), and this patch also
58        makes it give the same error for \k'' (previously it complained that that
59        was a reference to a non-existent subpattern).
60    
61    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
62        this is right because (?- can be followed by option settings as well as by
63        digits. I have, however, made the messages clearer.
64    
65    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
66        than the number used in the conditional) now cause a compile-time error.
67        This is actually not compatible with Perl, which accepts such patterns, but
68        treats the conditional as always being FALSE (as PCRE used to), but it
69        seems to me that giving a diagnostic is better.
70    
71    8.  Change "alphameric" to the more common word "alphanumeric" in comments
72        and messages.
73    
74    9.  Fix two occurrences of "backslash" in comments that should have been
75        "backspace".
76    
77    10. Remove two redundant lines of code that can never be obeyed (their function
78        was moved elsewhere).
79    
80    11. The program that makes PCRE's Unicode character property table had a bug
81        which caused it to generate incorrect table entries for sequences of
82        characters that have the same character type, but are in different scripts.
83        It amalgamated them into a single range, with the script of the first of
84        them. In other words, some characters were in the wrong script. There were
85        thirteen such cases, affecting characters in the following ranges:
86    
87          U+002b0 - U+002c1
88          U+0060c - U+0060d
89          U+0061e - U+00612
90          U+0064b - U+0065e
91          U+0074d - U+0076d
92          U+01800 - U+01805
93          U+01d00 - U+01d77
94          U+01d9b - U+01dbf
95          U+0200b - U+0200f
96          U+030fc - U+030fe
97          U+03260 - U+0327f
98          U+0fb46 - U+0fbb1
99          U+10450 - U+1049d
100    
101    12. The -o option (show only the matching part of a line) for pcregrep was not
102        compatible with GNU grep in that, if there was more than one match in a
103        line, it showed only the first of them. It now behaves in the same way as
104        GNU grep.
105    
106    13. If the -o and -v options were combined for pcregrep, it printed a blank
107        line for every non-matching line. GNU grep prints nothing, and pcregrep now
108        does the same. The return code can be used to tell if there were any
109        non-matching lines.
110    
111    14. Added --file-offsets and --line-offsets to pcregrep.
112    
113    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
114        infinitely looping recursion. The bug was that positive lookaheads were not
115        being skipped when checking for a possible empty match (negative lookaheads
116        and both kinds of lookbehind were skipped).
117    
118    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
119        inclusion of <windows.h> to before rather than after the definition of
120        INVALID_FILE_ATTRIBUTES (patch from David Byron).
121    
122    17. Specifying a possessive quantifier with a specific limit for a Unicode
123        character property caused pcre_compile() to compile bad code, which led at
124        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
125        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
126        caused the error; without that there was no problem.
127    
128    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
129    
130    19. Added --enable-pcretest-libreadline.
131    
132    20. In pcrecpp.cc, the variable 'count' was incremented twice in
133        RE::GlobalReplace(). As a result, the number of replacements returned was
134        double what it should be. I removed one of the increments, but Craig sent a
135        later patch that removed the other one (the right fix) and added unit tests
136        that check the return values (which was not done before).
137    
138    21. Several CMake things:
139    
140        (1) Arranged that, when cmake is used on Unix, the libraries end up with
141            the names libpcre and libpcreposix, not just pcre and pcreposix.
142    
143        (2) The above change means that pcretest and pcregrep are now correctly
144            linked with the newly-built libraries, not previously installed ones.
145    
146        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
147    
148    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
149        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
150        UTF-8 newline character). The key issue is that the pattern starts .*;
151        this means that the match must be either at the beginning, or after a
152        newline. The bug was in the code for advancing after a failed match and
153        checking that the new position followed a newline. It was not taking
154        account of UTF-8 characters correctly.
155    
156    23. PCRE was behaving differently from Perl in the way it recognized POSIX
157        character classes. PCRE was not treating the sequence [:...:] as a
158        character class unless the ... were all letters. Perl, however, seems to
159        allow any characters between [: and :], though of course it rejects as
160        unknown any "names" that contain non-letters, because all the known class
161        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
162        for example, whereas PCRE did not - it did not recognize a POSIX character
163        class. This seemed a bit dangerous, so the code has been changed to be
164        closer to Perl. The behaviour is not identical to Perl, because PCRE will
165        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
166        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
167        Perl does, and where it didn't before.
168    
169    24. Rewrite so as to remove the single use of %n from pcregrep because in some
170        Windows environments %n is disabled by default.
171    
172    
173    Version 7.4 21-Sep-07
174    ---------------------
175    
176    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
177        means that a class such as [\s] counted as "explicit reference to CR or
178        LF". That isn't really right - the whole point of the change was to try to
179        help when there was an actual mention of one of the two characters. So now
180        the change happens only if \r or \n (or a literal CR or LF) character is
181        encountered.
182    
183    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
184        of both had grown to the point where there were only 3 bits left.
185        Fortunately, there was spare space in the data structure, and so I have
186        moved the internal flags into a new 16-bit field to free up more option
187        bits.
188    
189    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
190        but did not set the internal JCHANGED flag - either of these is enough to
191        control the way the "get" function works - but the PCRE_INFO_JCHANGED
192        facility is supposed to tell if (?J) was ever used, so now (?J) at the
193        start sets both bits.
194    
195    4.  Added options (at build time, compile time, exec time) to change \R from
196        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
197    
198    5.  doc/pcresyntax.html was missing from the distribution.
199    
200    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
201        compatibility, even though it is no longer used.
202    
203    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
204        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
205        windows.h file is present (where different names are used). [This was
206        reversed later after testing - see 16 below.]
207    
208    8.  Changed all #include <config.h> to #include "config.h". There were also
209        some further <pcre.h> cases that I changed to "pcre.h".
210    
211    9.  When pcregrep was used with the --colour option, it missed the line ending
212        sequence off the lines that it output.
213    
214    10. It was pointed out to me that arrays of string pointers cause lots of
215        relocations when a shared library is dynamically loaded. A technique of
216        using a single long string with a table of offsets can drastically reduce
217        these. I have refactored PCRE in four places to do this. The result is
218        dramatic:
219    
220          Originally:                          290
221          After changing UCP table:            187
222          After changing error message table:   43
223          After changing table of "verbs"       36
224          After changing table of Posix names   22
225    
226        Thanks to the folks working on Gregex for glib for this insight.
227    
228    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
229        unicode-properties was also set.
230    
231    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
232    
233    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
234        checked only for CRLF.
235    
236    14. Added casts to pcretest.c to avoid compiler warnings.
237    
238    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
239    
240    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
241        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
242        entirely. This removes changes made in 7 above.
243    
244    17. The CMake files have been updated, and there is now more information about
245        building with CMake in the NON-UNIX-USE document.
246    
247    
248    Version 7.3 28-Aug-07
249    ---------------------
250    
251     1. In the rejigging of the build system that eventually resulted in 7.1, the
252        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
253        brackets there is not right, since it causes compilers to look for an
254        installed pcre.h, not the version that is in the source that is being
255        compiled (which of course may be different). I have changed it back to:
256    
257          #include "pcre.h"
258    
259        I have a vague recollection that the change was concerned with compiling in
260        different directories, but in the new build system, that is taken care of
261        by the VPATH setting the Makefile.
262    
263     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
264        when the subject happened to end in the byte 0x85 (e.g. if the last
265        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
266        characters but of course it shouldn't be taken as a newline when it is part
267        of another character. The bug was that, for an unlimited repeat of . in
268        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
269        characters when looking for a newline.
270    
271     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
272    
273     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
274        in debug output.
275    
276     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
277        long printing in the pcrecpp unittest when running under MinGW.
278    
279     6. ESC_K was left out of the EBCDIC table.
280    
281     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
282        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
283        limit also applies to "virtual nesting" when a pattern is recursive, and in
284        this case 1000 isn't so big. I have been able to remove this limit at the
285        expense of backing off one optimization in certain circumstances. Normally,
286        when pcre_exec() would call its internal match() function recursively and
287        immediately return the result unconditionally, it uses a "tail recursion"
288        feature to save stack. However, when a subpattern that can match an empty
289        string has an unlimited repetition quantifier, it no longer makes this
290        optimization. That gives it a stack frame in which to save the data for
291        checking that an empty string has been matched. Previously this was taken
292        from the 1000-entry workspace that had been reserved. So now there is no
293        explicit limit, but more stack is used.
294    
295     8. Applied Daniel's patches to solve problems with the import/export magic
296        syntax that is required for Windows, and which was going wrong for the
297        pcreposix and pcrecpp parts of the library. These were overlooked when this
298        problem was solved for the main library.
299    
300     9. There were some crude static tests to avoid integer overflow when computing
301        the size of patterns that contain repeated groups with explicit upper
302        limits. As the maximum quantifier is 65535, the maximum group length was
303        set at 30,000 so that the product of these two numbers did not overflow a
304        32-bit integer. However, it turns out that people want to use groups that
305        are longer than 30,000 bytes (though not repeat them that many times).
306        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
307        made it possible to implement the integer overflow checks in a much more
308        dynamic way, which I have now done. The artificial limitation on group
309        length has been removed - we now have only the limit on the total length of
310        the compiled pattern, which depends on the LINK_SIZE setting.
311    
312    10. Fixed a bug in the documentation for get/copy named substring when
313        duplicate names are permitted. If none of the named substrings are set, the
314        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
315        empty string.
316    
317    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
318        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
319        because the ] is interpreted as the first data character and the
320        terminating ] is not found. PCRE has been made compatible with Perl in this
321        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
322        cause memory overwriting.
323    
324    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
325        string has been matched (to stop an infinite loop). It was not recognizing
326        a conditional subpattern that could match an empty string if that
327        subpattern was within another subpattern. For example, it looped when
328        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
329        condition was not nested. This bug has been fixed.
330    
331    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
332        past the start of the subject in the presence of bytes with the top bit
333        set, for example "\x8aBCD".
334    
335    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
336        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
337    
338    14. Optimized (?!) to (*FAIL).
339    
340    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
341        This restricts code points to be within the range 0 to 0x10FFFF, excluding
342        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
343        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
344        does: it's just the validity check that is more restrictive.
345    
346    16. Inserted checks for integer overflows during escape sequence (backslash)
347        processing, and also fixed erroneous offset values for syntax errors during
348        backslash processing.
349    
350    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
351        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
352    
353    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
354        caused an overrun.
355    
356    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
357        something other than just ASCII characters) inside a group that had an
358        unlimited repeat caused a loop at compile time (while checking to see
359        whether the group could match an empty string).
360    
361    20. Debugging a pattern containing \p or \P could cause a crash. For example,
362        [\P{Any}] did so. (Error in the code for printing property names.)
363    
364    21. An orphan \E inside a character class could cause a crash.
365    
366    22. A repeated capturing bracket such as (A)? could cause a wild memory
367        reference during compilation.
368    
369    23. There are several functions in pcre_compile() that scan along a compiled
370        expression for various reasons (e.g. to see if it's fixed length for look
371        behind). There were bugs in these functions when a repeated \p or \P was
372        present in the pattern. These operators have additional parameters compared
373        with \d, etc, and these were not being taken into account when moving along
374        the compiled data. Specifically:
375    
376        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
377            length.
378    
379        (b) An item such as \pL+ within a repeated group could cause crashes or
380            loops.
381    
382        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
383            "reference to non-existent subpattern" error.
384    
385        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
386    
387    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
388        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
389    
390    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
391    
392    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
393        character were causing crashes (broken optimization).
394    
395    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
396        \p or \P) caused a compile-time loop.
397    
398    28. More problems have arisen in unanchored patterns when CRLF is a valid line
399        break. For example, the unstudied pattern [\r\n]A does not match the string
400        "\r\nA" because change 7.0/46 below moves the current point on by two
401        characters after failing to match at the start. However, the pattern \nA
402        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
403        the same is true. There doesn't seem any very clean way out of this, but
404        what I have chosen to do makes the common cases work: PCRE now takes note
405        of whether there can be an explicit match for \r or \n anywhere in the
406        pattern, and if so, 7.0/46 no longer applies. As part of this change,
407        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
408        pattern has explicit CR or LF references.
409    
410    29. Added (*CR) etc for changing newline setting at start of pattern.
411    
412    
413    Version 7.2 19-Jun-07
414    ---------------------
415    
416     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
417        which is apparently normally available under Windows.
418    
419     2. Re-jig the pcregrep tests with different newline settings in an attempt
420        to make them independent of the local environment's newline setting.
421    
422     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
423    
424     4. Some of the "internals" tests were previously cut out when the link size
425        was not 2, because the output contained actual offsets. The recent new
426        "Z" feature of pcretest means that these can be cut out, making the tests
427        usable with all link sizes.
428    
429     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
430        stack recursion. This gives a massive performance boost under BSD, but just
431        a small improvement under Linux. However, it saves one field in the frame
432        in all cases.
433    
434     6. Added more features from the forthcoming Perl 5.10:
435    
436        (a) (?-n) (where n is a string of digits) is a relative subroutine or
437            recursion call. It refers to the nth most recently opened parentheses.
438    
439        (b) (?+n) is also a relative subroutine call; it refers to the nth next
440            to be opened parentheses.
441    
442        (c) Conditions that refer to capturing parentheses can be specified
443            relatively, for example, (?(-2)... or (?(+3)...
444    
445        (d) \K resets the start of the current match so that everything before
446            is not part of it.
447    
448        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
449    
450        (f) \g{name} is another synonym - part of Perl 5.10's unification of
451            reference syntax.
452    
453        (g) (?| introduces a group in which the numbering of parentheses in each
454            alternative starts with the same number.
455    
456        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
457    
458     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
459        PCRE_INFO_JCHANGED.
460    
461     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
462        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
463        for detecting groups that can match an empty string.
464    
465     9. A pattern with a very large number of alternatives (more than several
466        hundred) was running out of internal workspace during the pre-compile
467        phase, where pcre_compile() figures out how much memory will be needed. A
468        bit of new cunning has reduced the workspace needed for groups with
469        alternatives. The 1000-alternative test pattern now uses 12 bytes of
470        workspace instead of running out of the 4096 that are available.
471    
472    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
473    
474    11. Applied patch from Google to remove an optimization that didn't quite work.
475        The report of the bug said:
476    
477          pcrecpp::RE("a*").FullMatch("aaa") matches, while
478          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
479          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
480    
481    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
482        it matched the wrong number of bytes.
483    
484    
485    Version 7.1 24-Apr-07
486    ---------------------
487    
488     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
489        that is more "standard", making use of automake and other Autotools. There
490        is some re-arrangement of the files and adjustment of comments consequent
491        on this.
492    
493     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
494        for recursive directory scanning broke on some systems because the files
495        are not scanned in any specific order and on different systems the order
496        was different. A call to "sort" has been inserted into RunGrepTest for the
497        approprate test as a short-term fix. In the longer term there may be an
498        alternative.
499    
500     3. I had an email from Eric Raymond about problems translating some of PCRE's
501        man pages to HTML (despite the fact that I distribute HTML pages, some
502        people do their own conversions for various reasons). The problems
503        concerned the use of low-level troff macros .br and .in. I have therefore
504        removed all such uses from the man pages (some were redundant, some could
505        be replaced by .nf/.fi pairs). The 132html script that I use to generate
506        HTML has been updated to handle .nf/.fi and to complain if it encounters
507        .br or .in.
508    
509     4. Updated comments in configure.ac that get placed in config.h.in and also
510        arranged for config.h to be included in the distribution, with the name
511        config.h.generic, for the benefit of those who have to compile without
512        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
513    
514     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
515        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
516        makevp.bat for the current PCRE, using the additional files
517        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
518    
519     6. A Windows user reported a minor discrepancy with test 2, which turned out
520        to be caused by a trailing space on an input line that had got lost in his
521        copy. The trailing space was an accident, so I've just removed it.
522    
523     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
524        that is needed.
525    
526     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
527        as "const" (a) because they are and (b) because it helps the PHP
528        maintainers who have recently made a script to detect big data structures
529        in the php code that should be moved to the .rodata section. I remembered
530        to update Builducptable as well, so it won't revert if ucptable.h is ever
531        re-created.
532    
533     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
534        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
535        order to be able to cut out the UTF-8 tables in the latter when UTF-8
536        support is not required. This saves 1.5-2K of code, which is important in
537        some applications.
538    
539        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
540        so as not to refer to the tables, even though these functions will never be
541        called when UTF-8 support is disabled. Otherwise there are problems with a
542        shared library.
543    
544    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
545    
546        (a) It was defining its arguments as char * instead of void *.
547    
548        (b) It was assuming that all moves were upwards in memory; this was true
549            a long time ago when I wrote it, but is no longer the case.
550    
551        The emulated memove() is provided for those environments that have neither
552        memmove() nor bcopy(). I didn't think anyone used it these days, but that
553        is clearly not the case, as these two bugs were recently reported.
554    
555    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
556        and Detrail to create the HTML documentation, the .txt form of the man
557        pages, and it removes trailing spaces from listed files. It also creates
558        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
559        case, it wraps all the #defines with #ifndefs. This script should be run
560        before "make dist".
561    
562    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
563        with Unicode property support.
564    
565        (a) For a maximizing quantifier, if the two different cases of the
566            character were of different lengths in their UTF-8 codings (there are
567            some cases like this - I found 11), and the matching function had to
568            back up over a mixture of the two cases, it incorrectly assumed they
569            were both the same length.
570    
571        (b) When PCRE was configured to use the heap rather than the stack for
572            recursion during matching, it was not correctly preserving the data for
573            the other case of a UTF-8 character when checking ahead for a match
574            while processing a minimizing repeat. If the check also involved
575            matching a wide character, but failed, corruption could cause an
576            erroneous result when trying to check for a repeat of the original
577            character.
578    
579    13. Some tidying changes to the testing mechanism:
580    
581        (a) The RunTest script now detects the internal link size and whether there
582            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
583            values substituted by "configure". (The RunGrepTest script already did
584            this for UTF-8.) The configure.ac script no longer substitutes the
585            relevant variables.
586    
587        (b) The debugging options /B and /D in pcretest show the compiled bytecode
588            with length and offset values. This means that the output is different
589            for different internal link sizes. Test 2 is skipped for link sizes
590            other than 2 because of this, bypassing the problem. Unfortunately,
591            there was also a test in test 3 (the locale tests) that used /B and
592            failed for link sizes other than 2. Rather than cut the whole test out,
593            I have added a new /Z option to pcretest that replaces the length and
594            offset values with spaces. This is now used to make test 3 independent
595            of link size. (Test 2 will be tidied up later.)
596    
597    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
598        segmentation fault instead of returning the appropriate error message.
599    
600    15. In multiline mode when the newline sequence was set to "any", the pattern
601        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
602        This doesn't seem right; it now treats the CRLF combination as the line
603        ending, and so does not match in that case. It's only a pattern such as ^$
604        that would hit this one: something like ^ABC$ would have failed after \r
605        and then tried again after \r\n.
606    
607    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
608        in an attempt to make files that differ only in their line terminators
609        compare equal. This works on Linux.
610    
611    17. Under certain error circumstances pcregrep might try to free random memory
612        as it exited. This is now fixed, thanks to valgrind.
613    
614    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
615        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
616        was because its rules for how to advance for /g after matching an empty
617        string at the end of a line did not allow for this case. They now check for
618        it specially.
619    
620    20. pcretest is supposed to handle patterns and data of any length, by
621        extending its buffers when necessary. It was getting this wrong when the
622        buffer for a data line had to be extended.
623    
624    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
625        CRLF as a newline sequence.
626    
627    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
628        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
629        I have nevertheless tidied it up.
630    
631    23. Added some casts to kill warnings from HP-UX ia64 compiler.
632    
633    24. Added a man page for pcre-config.
634    
635    
636    Version 7.0 19-Dec-06
637    ---------------------
638    
639     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
640        moving to gcc 4.1.1.
641    
642     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
643        sys/time.h, which is documented as needed for this function. It doesn't
644        seem to matter on Linux, but it showed up on some releases of OS X.
645    
646     3. It seems that there are systems where bytes whose values are greater than
647        127 match isprint() in the "C" locale. The "C" locale should be the
648        default when a C program starts up. In most systems, only ASCII printing
649        characters match isprint(). This difference caused the output from pcretest
650        to vary, making some of the tests fail. I have changed pcretest so that:
651    
652        (a) When it is outputting text in the compiled version of a pattern, bytes
653            other than 32-126 are always shown as hex escapes.
654    
655        (b) When it is outputting text that is a matched part of a subject string,
656            it does the same, unless a different locale has been set for the match
657            (using the /L modifier). In this case, it uses isprint() to decide.
658    
659     4. Fixed a major bug that caused incorrect computation of the amount of memory
660        required for a compiled pattern when options that changed within the
661        pattern affected the logic of the preliminary scan that determines the
662        length. The relevant options are -x, and -i in UTF-8 mode. The result was
663        that the computed length was too small. The symptoms of this bug were
664        either the PCRE error "internal error: code overflow" from pcre_compile(),
665        or a glibc crash with a message such as "pcretest: free(): invalid next
666        size (fast)". Examples of patterns that provoked this bug (shown in
667        pcretest format) are:
668    
669          /(?-x: )/x
670          /(?x)(?-x: \s*#\s*)/
671          /((?i)[\x{c0}])/8
672          /(?i:[\x{c0}])/8
673    
674        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
675        is now done differently.
676    
677     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
678        wrapper classes; (b) implement a new function in the C++ scanner that is
679        more efficient than the old way of doing things because it avoids levels of
680        recursion in the regex matching; (c) add a paragraph to the documentation
681        for the FullMatch() function.
682    
683     6. The escape sequence \n was being treated as whatever was defined as
684        "newline". Not only was this contrary to the documentation, which states
685        that \n is character 10 (hex 0A), but it also went horribly wrong when
686        "newline" was defined as CRLF. This has been fixed.
687    
688     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
689        was being set to -1 for the "end of line" case (supposedly a value that no
690        character can have). Though this value is never used (the check for end of
691        line is "zero bytes in current character"), it caused compiler complaints.
692        I've changed it to 0xffffffff.
693    
694     8. In pcre_version.c, the version string was being built by a sequence of
695        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
696        string (as it is for production releases) called a macro with an empty
697        argument. The C standard says the result of this is undefined. The gcc
698        compiler treats it as an empty string (which was what was wanted) but it is
699        reported that Visual C gives an error. The source has been hacked around to
700        avoid this problem.
701    
702     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
703        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
704        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
705        of them did).
706    
707    10. Originally, pcretest opened its input and output without "b"; then I was
708        told that "b" was needed in some environments, so it was added for release
709        5.0 to both the input and output. (It makes no difference on Unix-like
710        systems.) Later I was told that it is wrong for the input on Windows. I've
711        now abstracted the modes into two macros, to make it easier to fiddle with
712        them, and removed "b" from the input mode under Windows.
713    
714    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
715    
716    12. Added -help and --help to pcretest as an official way of being reminded
717        of the options.
718    
719    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
720        and pcrecpp.cc because they annoy compilers at high warning levels.
721    
722    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
723    
724    15. Fixed an occurrence of == in configure.ac that should have been = (shell
725        scripts are not C programs :-) and which was not noticed because it works
726        on Linux.
727    
728    16. pcretest is supposed to handle any length of pattern and data line (as one
729        line or as a continued sequence of lines) by extending its input buffer if
730        necessary. This feature was broken for very long pattern lines, leading to
731        a string of junk being passed to pcre_compile() if the pattern was longer
732        than about 50K.
733    
734    17. I have done a major re-factoring of the way pcre_compile() computes the
735        amount of memory needed for a compiled pattern. Previously, there was code
736        that made a preliminary scan of the pattern in order to do this. That was
737        OK when PCRE was new, but as the facilities have expanded, it has become
738        harder and harder to keep it in step with the real compile phase, and there
739        have been a number of bugs (see for example, 4 above). I have now found a
740        cunning way of running the real compile function in a "fake" mode that
741        enables it to compute how much memory it would need, while actually only
742        ever using a few hundred bytes of working memory and without too many
743        tests of the mode. This should make future maintenance and development
744        easier. A side effect of this work is that the limit of 200 on the nesting
745        depth of parentheses has been removed (though this was never a serious
746        limitation, I suspect). However, there is a downside: pcre_compile() now
747        runs more slowly than before (30% or more, depending on the pattern). I
748        hope this isn't a big issue. There is no effect on runtime performance.
749    
750    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
751        newline (only possible for the last line of a file) and it was a
752        pattern that set a locale (followed by /Lsomething), pcretest crashed.
753    
754    19. Added additional timing features to pcretest. (1) The -tm option now times
755        matching only, not compiling. (2) Both -t and -tm can be followed, as a
756        separate command line item, by a number that specifies the number of
757        repeats to use when timing. The default is 50000; this gives better
758        precision, but takes uncomfortably long for very large patterns.
759    
760    20. Extended pcre_study() to be more clever in cases where a branch of a
761        subpattern has no definite first character. For example, (a*|b*)[cd] would
762        previously give no result from pcre_study(). Now it recognizes that the
763        first character must be a, b, c, or d.
764    
765    21. There was an incorrect error "recursive call could loop indefinitely" if
766        a subpattern (or the entire pattern) that was being tested for matching an
767        empty string contained only one non-empty item after a nested subpattern.
768        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
769        incorrectly, because the \d was being skipped in the check.
770    
771    22. The pcretest program now has a new pattern option /B and a command line
772        option -b, which is equivalent to adding /B to every pattern. This causes
773        it to show the compiled bytecode, without the additional information that
774        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
775        is the same as /B/I).
776    
777    23. A new optimization is now able automatically to treat some sequences such
778        as a*b as a*+b. More specifically, if something simple (such as a character
779        or a simple class like \d) has an unlimited quantifier, and is followed by
780        something that cannot possibly match the quantified thing, the quantifier
781        is automatically "possessified".
782    
783    24. A recursive reference to a subpattern whose number was greater than 39
784        went wrong under certain circumstances in UTF-8 mode. This bug could also
785        have affected the operation of pcre_study().
786    
787    25. Realized that a little bit of performance could be had by replacing
788        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
789    
790    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
791    
792    27. Possessive quantifiers such as a++ were previously implemented by turning
793        them into atomic groups such as ($>a+). Now they have their own opcodes,
794        which improves performance. This includes the automatically created ones
795        from 23 above.
796    
797    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
798        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
799        the first matched character to be a colon. This applied both to named and
800        numbered groups.
801    
802    29. The ucpinternal.h header file was missing its idempotency #ifdef.
803    
804    30. I was sent a "project" file called libpcre.a.dev which I understand makes
805        building PCRE on Windows easier, so I have included it in the distribution.
806    
807    31. There is now a check in pcretest against a ridiculously large number being
808        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
809        loop, the loop is abandoned.
810    
811    32. Forward references to subpatterns in conditions such as (?(2)...) where
812        subpattern 2 is defined later cause pcre_compile() to search forwards in
813        the pattern for the relevant set of parentheses. This search went wrong
814        when there were unescaped parentheses in a character class, parentheses
815        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
816    
817    33. "Subroutine" calls and backreferences were previously restricted to
818        referencing subpatterns earlier in the regex. This restriction has now
819        been removed.
820    
821    34. Added a number of extra features that are going to be in Perl 5.10. On the
822        whole, these are just syntactic alternatives for features that PCRE had
823        previously implemented using the Python syntax or my own invention. The
824        other formats are all retained for compatibility.
825    
826        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
827            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
828            also .NET compatible.
829    
830        (b) A recursion or subroutine call to a named group can now be defined as
831            (?&name) as well as (?P>name).
832    
833        (c) A backreference to a named group can now be defined as \k<name> or
834            \k'name' as well as (?P=name). The new forms, as well as being in Perl
835            5.10, are also .NET compatible.
836    
837        (d) A conditional reference to a named group can now use the syntax
838            (?(<name>) or (?('name') as well as (?(name).
839    
840        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
841            groups (named and numbered) that are never evaluated inline, but can be
842            called as "subroutines" from elsewhere. In effect, the DEFINE condition
843            is always false. There may be only one alternative in such a group.
844    
845        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
846            as the simple (?(R). The condition is true only if the most recent
847            recursion is that of the given number or name. It does not search out
848            through the entire recursion stack.
849    
850        (g) The escape \gN or \g{N} has been added, where N is a positive or
851            negative number, specifying an absolute or relative reference.
852    
853    35. Tidied to get rid of some further signed/unsigned compiler warnings and
854        some "unreachable code" warnings.
855    
856    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
857        things, this adds five new scripts.
858    
859    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
860        There were also incompatibilities regarding the handling of \Q..\E inside
861        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
862        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
863    
864    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
865        matches an empty string, and forcibly breaks the loop. There were bugs in
866        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
867        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
868        separate and independent bugs (that affected different cases) have been
869        fixed.
870    
871    39. Refactored the code to abolish the use of different opcodes for small
872        capturing bracket numbers. This is a tidy that I avoided doing when I
873        removed the limit on the number of capturing brackets for 3.5 back in 2001.
874        The new approach is not only tidier, it makes it possible to reduce the
875        memory needed to fix the previous bug (38).
876    
877    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
878        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
879        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
880        mode.
881    
882    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
883        report.
884    
885    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
886        copy construction and assignment in the C++ wrapper.
887    
888    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
889        couple of bugs that could have given wrong results in the "--newline=crlf"
890        case.
891    
892    44. Added a number of casts and did some reorganization of signed/unsigned int
893        variables following suggestions from Dair Grant. Also renamed the variable
894        "this" as "item" because it is a C++ keyword.
895    
896    45. Arranged for dftables to add
897    
898          #include "pcre_internal.h"
899    
900        to pcre_chartables.c because without it, gcc 4.x may remove the array
901        definition from the final binary if PCRE is built into a static library and
902        dead code stripping is activated.
903    
904    46. For an unanchored pattern, if a match attempt fails at the start of a
905        newline sequence, and the newline setting is CRLF or ANY, and the next two
906        characters are CRLF, advance by two characters instead of one.
907    
908    
909    Version 6.7 04-Jul-06
910    ---------------------
911    
912     1. In order to handle tests when input lines are enormously long, pcretest has
913        been re-factored so that it automatically extends its buffers when
914        necessary. The code is crude, but this _is_ just a test program. The
915        default size has been increased from 32K to 50K.
916    
917     2. The code in pcre_study() was using the value of the re argument before
918        testing it for NULL. (Of course, in any sensible call of the function, it
919        won't be NULL.)
920    
921     3. The memmove() emulation function in pcre_internal.h, which is used on
922        systems that lack both memmove() and bcopy() - that is, hardly ever -
923        was missing a "static" storage class specifier.
924    
925     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
926        containing an extended class (one that cannot be represented by a bitmap
927        because it contains high-valued characters or Unicode property items, e.g.
928        [\pZ]). Almost always one would set UTF-8 mode when processing such a
929        pattern, but PCRE should not loop if you do not (it no longer does).
930        [Detail: two cases were found: (a) a repeated subpattern containing an
931        extended class; (b) a recursive reference to a subpattern that followed a
932        previous extended class. It wasn't skipping over the extended class
933        correctly when UTF-8 mode was not set.]
934    
935     5. A negated single-character class was not being recognized as fixed-length
936        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
937        compile error "lookbehind assertion is not fixed length".
938    
939     6. The RunPerlTest auxiliary script was showing an unexpected difference
940        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
941        write a Perl script that can interpret lines of an input file either as
942        byte characters or as UTF-8, which is what "perltest" was being required to
943        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
944        can't do is switch easily at run time between having the "use utf8;" pragma
945        or not. In the end, I fudged it by using the RunPerlTest script to insert
946        "use utf8;" explicitly for the UTF-8 tests.
947    
948     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
949        the end of the subject string, contrary to the documentation and to what
950        Perl does. This was true of both matching functions. Now it matches only at
951        the start of the subject and immediately after *internal* newlines.
952    
953     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
954        a pointer to an int instead of a pointer to an unsigned long int. This
955        caused problems on 64-bit systems.
956    
957     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
958        instance of the 'standard' template library not being so standard".
959    
960    10. There was no check on the number of named subpatterns nor the maximum
961        length of a subpattern name. The product of these values is used to compute
962        the size of the memory block for a compiled pattern. By supplying a very
963        long subpattern name and a large number of named subpatterns, the size
964        computation could be caused to overflow. This is now prevented by limiting
965        the length of names to 32 characters, and the number of named subpatterns
966        to 10,000.
967    
968    11. Subpatterns that are repeated with specific counts have to be replicated in
969        the compiled pattern. The size of memory for this was computed from the
970        length of the subpattern and the repeat count. The latter is limited to
971        65535, but there was no limit on the former, meaning that integer overflow
972        could in principle occur. The compiled length of a repeated subpattern is
973        now limited to 30,000 bytes in order to prevent this.
974    
975    12. Added the optional facility to have named substrings with the same name.
976    
977    13. Added the ability to use a named substring as a condition, using the
978        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
979        are numbers (not recommended). Forward references are permitted.
980    
981    14. Added forward references in named backreferences (if you see what I mean).
982    
983    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
984        pattern could run off the end of the subject. For example, the pattern
985        "(?s)(.{1,5})"8 did this with the subject "ab".
986    
987    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
988        PCRE_CASELESS was set when matching characters that were quantified with ?
989        or *.
990    
991    17. A character class other than a single negated character that had a minimum
992        but no maximum quantifier - for example [ab]{6,} - was not handled
993        correctly by pce_dfa_exec(). It would match only one character.
994    
995    18. A valid (though odd) pattern that looked like a POSIX character
996        class but used an invalid character after [ (for example [[,abc,]]) caused
997        pcre_compile() to give the error "Failed: internal error: code overflow" or
998        in some cases to crash with a glibc free() error. This could even happen if
999        the pattern terminated after [[ but there just happened to be a sequence of
1000        letters, a binary zero, and a closing ] in the memory that followed.
1001    
1002    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1003        over the years. Originally (before any Unicode support), just the bottom 8
1004        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1005        output from "man perlunicode" includes this:
1006    
1007          The regular expression compiler produces polymorphic opcodes.  That
1008          is, the pattern adapts to the data and automatically switches to
1009          the Unicode character scheme when presented with Unicode data--or
1010          instead uses a traditional byte scheme when presented with byte
1011          data.
1012    
1013        Sadly, a wide octal escape does not cause a switch, and in a string with
1014        no other multibyte characters, these octal escapes are treated as before.
1015        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1016        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1017        Unicode string.
1018    
1019        I have not perpetrated such confusion in PCRE. Up till now, it took just
1020        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1021        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1022        translate to the appropriate multibyte character.
1023    
1024    29. Applied some refactoring to reduce the number of warnings from Microsoft
1025        and Borland compilers. This has included removing the fudge introduced
1026        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1027        a warning about an unused variable.
1028    
1029    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1030        characters since release 4.0, because Perl (from release 5.004) does not.
1031        [Or at least, is documented not to: some releases seem to be in conflict
1032        with the documentation.] However, when a pattern was studied with
1033        pcre_study() and all its branches started with \s, PCRE still included VT
1034        as a possible starting character. Of course, this did no harm; it just
1035        caused an unnecessary match attempt.
1036    
1037    22. Removed a now-redundant internal flag bit that recorded the fact that case
1038        dependency changed within the pattern. This was once needed for "required
1039        byte" processing, but is no longer used. This recovers a now-scarce options
1040        bit. Also moved the least significant internal flag bit to the most-
1041        significant bit of the word, which was not previously used (hangover from
1042        the days when it was an int rather than a uint) to free up another bit for
1043        the future.
1044    
1045    23. Added support for CRLF line endings as well as CR and LF. As well as the
1046        default being selectable at build time, it can now be changed at runtime
1047        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1048        specify that it is scanning data with non-default line endings.
1049    
1050    24. Changed the definition of CXXLINK to make it agree with the definition of
1051        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1052    
1053    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1054        recursions. This makes a big different to stack usage for some patterns.
1055    
1056    26. If a subpattern containing a named recursion or subroutine reference such
1057        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1058        the space required for the compiled pattern went wrong and gave too small a
1059        value. Depending on the environment, this could lead to "Failed: internal
1060        error: code overflow at offset 49" or "glibc detected double free or
1061        corruption" errors.
1062    
1063    27. Applied patches from Google (a) to support the new newline modes and (b) to
1064        advance over multibyte UTF-8 characters in GlobalReplace.
1065    
1066    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1067        difference for some implementation of PCRE in some Windows version.
1068    
1069    29. Added some extra testing facilities to pcretest:
1070    
1071        \q<number>   in a data line sets the "match limit" value
1072        \Q<number>   in a data line sets the "match recursion limt" value
1073        -S <number>  sets the stack size, where <number> is in megabytes
1074    
1075        The -S option isn't available for Windows.
1076    
1077    
1078    Version 6.6 06-Feb-06
1079    ---------------------
1080    
1081     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1082        in pcreposix.h. I have copied the definition from pcre.h.
1083    
1084     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1085        because pcre.h is no longer a built file.
1086    
1087     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1088        not normally included in the compiled code.
1089    
1090    
1091    Version 6.5 01-Feb-06
1092    ---------------------
1093    
1094     1. When using the partial match feature with pcre_dfa_exec(), it was not
1095        anchoring the second and subsequent partial matches at the new starting
1096        point. This could lead to incorrect results. For example, with the pattern
1097        /1234/, partially matching against "123" and then "a4" gave a match.
1098    
1099     2. Changes to pcregrep:
1100    
1101        (a) All non-match returns from pcre_exec() were being treated as failures
1102            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1103            error message is output. Some extra information is given for the
1104            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1105            probably the only errors that are likely to be caused by users (by
1106            specifying a regex that has nested indefinite repeats, for instance).
1107            If there are more than 20 of these errors, pcregrep is abandoned.
1108    
1109        (b) A binary zero was treated as data while matching, but terminated the
1110            output line if it was written out. This has been fixed: binary zeroes
1111            are now no different to any other data bytes.
1112    
1113        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1114            used to set a locale for matching. The --locale=xxxx long option has
1115            been added (no short equivalent) to specify a locale explicitly on the
1116            pcregrep command, overriding the environment variables.
1117    
1118        (d) When -B was used with -n, some line numbers in the output were one less
1119            than they should have been.
1120    
1121        (e) Added the -o (--only-matching) option.
1122    
1123        (f) If -A or -C was used with -c (count only), some lines of context were
1124            accidentally printed for the final match.
1125    
1126        (g) Added the -H (--with-filename) option.
1127    
1128        (h) The combination of options -rh failed to suppress file names for files
1129            that were found from directory arguments.
1130    
1131        (i) Added the -D (--devices) and -d (--directories) options.
1132    
1133        (j) Added the -F (--fixed-strings) option.
1134    
1135        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1136    
1137        (l) Added the --colo(u)r option.
1138    
1139        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1140            is not present by default.
1141    
1142     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1143        items such as (?R) or (?1), when the recursion could match a number of
1144        alternatives. If it matched one of the alternatives, but subsequently,
1145        outside the recursion, there was a failure, the code tried to back up into
1146        the recursion. However, because of the way PCRE is implemented, this is not
1147        possible, and the result was an incorrect result from the match.
1148    
1149        In order to prevent this happening, the specification of recursion has
1150        been changed so that all such subpatterns are automatically treated as
1151        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1152    
1153     4. I had overlooked the fact that, in some locales, there are characters for
1154        which isalpha() is true but neither isupper() nor islower() are true. In
1155        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1156        and ordfeminine) are like this. This affected the treatment of \w and \W
1157        when they appeared in character classes, but not when they appeared outside
1158        a character class. The bit map for "word" characters is now created
1159        separately from the results of isalnum() instead of just taking it from the
1160        upper, lower, and digit maps. (Plus the underscore character, of course.)
1161    
1162     5. The above bug also affected the handling of POSIX character classes such as
1163        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1164        permanent tables. Instead, the bit maps for such a class were previously
1165        created as the appropriate unions of the upper, lower, and digit bitmaps.
1166        Now they are created by subtraction from the [[:word:]] class, which has
1167        its own bitmap.
1168    
1169     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1170        It is created by subtracting the vertical space characters (\x09, \x0a,
1171        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1172        subtraction was done in the overall bitmap for a character class, meaning
1173        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1174        be recognized. This bug has been fixed.
1175    
1176     7. Patches from the folks at Google:
1177    
1178          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1179          real life, but is still worth protecting against".
1180    
1181          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1182          regular expressions".
1183    
1184          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1185          have it.
1186    
1187          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1188          "configure" and the latter not, in order to fix a problem somebody had
1189          with compiling the Arg class on HP-UX.
1190    
1191          (e) Improve the error-handling of the C++ wrapper a little bit.
1192    
1193          (f) New tests for checking recursion limiting.
1194    
1195     8. The pcre_memmove() function, which is used only if the environment does not
1196        have a standard memmove() function (and is therefore rarely compiled),
1197        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1198        returning a result (though PCRE never actually uses the result).
1199    
1200     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1201        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1202        returned instead of calling malloc() with an overflowing number that would
1203        most likely cause subsequent chaos.
1204    
1205    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1206    
1207    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1208        with this option is matched, the nmatch and pmatch options of regexec() are
1209        ignored.
1210    
1211    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1212        provided in case anyone wants to the the POSIX interface with UTF-8
1213        strings.
1214    
1215    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1216        C++ linking (needed for some HP-UX environments).
1217    
1218    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1219        (unused parameter) and in the pcre_printint() function (omitted "default"
1220        switch label when the default is to do nothing).
1221    
1222    15. Added some code to make it possible, when PCRE is compiled as a C++
1223        library, to replace subject pointers for pcre_exec() with a smart pointer
1224        class, thus making it possible to process discontinuous strings.
1225    
1226    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1227        much the same function. They were added by different people who were trying
1228        to make PCRE easy to compile on non-Unix systems. It has been suggested
1229        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1230        for compiling on Windows systems. I have therefore replaced it with
1231        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1232        defaults to "extern" for C or "extern C" for C++, which works fine on
1233        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1234        SCOPE with something explicit in config.h. In addition:
1235    
1236        (a) pcreposix.h still had just "extern" instead of either of these macros;
1237            I have replaced it with PCRE_DATA_SCOPE.
1238    
1239        (b) Functions such as _pcre_xclass(), which are internal to the library,
1240            but external in the C sense, all had PCRE_EXPORT in their definitions.
1241            This is apparently wrong for the Windows case, so I have removed it.
1242            (It makes no difference on Unix-like systems.)
1243    
1244    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1245        of recursive calls to match(). This is different to MATCH_LIMIT because
1246        that limits the total number of calls to match(), not all of which increase
1247        the depth of recursion. Limiting the recursion depth limits the amount of
1248        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1249        when PCRE is compiled, and changed at run time. A patch from Google adds
1250        this functionality to the C++ interface.
1251    
1252    18. Changes to the handling of Unicode character properties:
1253    
1254        (a) Updated the table to Unicode 4.1.0.
1255    
1256        (b) Recognize characters that are not in the table as "Cn" (undefined).
1257    
1258        (c) I revised the way the table is implemented to a much improved format
1259            which includes recognition of ranges. It now supports the ranges that
1260            are defined in UnicodeData.txt, and it also amalgamates other
1261            characters into ranges. This has reduced the number of entries in the
1262            table from around 16,000 to around 3,000, thus reducing its size
1263            considerably. I realized I did not need to use a tree structure after
1264            all - a binary chop search is just as efficient. Having reduced the
1265            number of entries, I extended their size from 6 bytes to 8 bytes to
1266            allow for more data.
1267    
1268        (d) Added support for Unicode script names via properties such as \p{Han}.
1269    
1270    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1271        matching that character.
1272    
1273    20. When matching a repeated Unicode property with a minimum greater than zero,
1274        (for example \pL{2,}), PCRE could look past the end of the subject if it
1275        reached it while seeking the minimum number of characters. This could
1276        happen only if some of the characters were more than one byte long, because
1277        there is a check for at least the minimum number of bytes.
1278    
1279    21. Refactored the implementation of \p and \P so as to be more general, to
1280        allow for more different types of property in future. This has changed the
1281        compiled form incompatibly. Anybody with saved compiled patterns that use
1282        \p or \P will have to recompile them.
1283    
1284    22. Added "Any" and "L&" to the supported property types.
1285    
1286    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1287        but give a compile time error if the value is greater than 0xff.
1288    
1289    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1290        accidentally not being installed or uninstalled.
1291    
1292    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1293        made were to insert the current release number. This seemed silly, because
1294        it made things harder for people building PCRE on systems that don't run
1295        "configure". I have turned pcre.h into a distributed file, no longer built
1296        by "configure", with the version identification directly included. There is
1297        no longer a pcre.h.in file.
1298    
1299        However, this change necessitated a change to the pcre-config script as
1300        well. It is built from pcre-config.in, and one of the substitutions was the
1301        release number. I have updated configure.ac so that ./configure now finds
1302        the release number by grepping pcre.h.
1303    
1304    26. Added the ability to run the tests under valgrind.
1305    
1306    
1307    Version 6.4 05-Sep-05
1308    ---------------------
1309    
1310     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1311        "--" to be printed when multiple files were scanned, even when none of the
1312        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1313        consider it to be a bug, and have restored the previous behaviour.
1314    
1315     2. A couple of code tidies to get rid of compiler warnings.
1316    
1317     3. The pcretest program used to cheat by referring to symbols in the library
1318        whose names begin with _pcre_. These are internal symbols that are not
1319        really supposed to be visible externally, and in some environments it is
1320        possible to suppress them. The cheating is now confined to including
1321        certain files from the library's source, which is a bit cleaner.
1322    
1323     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1324        file's purpose clearer.
1325    
1326     5. Reorganized pcre_ucp_findchar().
1327    
1328    
1329  Version 6.3 15-Aug-05  Version 6.3 15-Aug-05
1330  ---------------------  ---------------------
1331    

Legend:
Removed from v.83  
changed lines
  Added in v.308

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12