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

Diff of /code/trunk/ChangeLog

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

revision 87 by nigel, Sat Feb 24 21:41:21 2007 UTC revision 377 by ph10, Sun Mar 1 12:07:19 2009 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.9 xx-xxx-09
5    ---------------------
6    
7    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
8        (pcretest), all targets were linked against these libraries. This included
9        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
10        libraries. This caused unwanted dependencies to be created. This problem
11        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
12        pcretest is linked with readline.
13    
14    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
15        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
16        moved outside it again, because FALSE and TRUE are already defined in AIX,
17        but BOOL is not.
18    
19    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
20        PCRE_MATCH_LIMIT_RETURSION values as ints, when they should be long ints.
21    
22    4.  The pcregrep documentation said spaces were inserted as well as colons (or
23        hyphens) following file names and line numbers when outputting matching
24        lines. This is not true; no spaces are inserted. I have also clarified the
25        wording for the --colour (or --color) option.
26    
27    5.  When --colour was used with -o, the list of matching strings was not
28        coloured; this is different to GNU grep, so I have changed it to be the
29        same.
30    
31    
32    Version 7.8 05-Sep-08
33    ---------------------
34    
35    1.  Replaced UCP searching code with optimized version as implemented for Ad
36        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
37        stage table and inline lookup instead of a function, giving speed ups of 2
38        to 5 times on some simple patterns that I tested. Permission was given to
39        distribute the MultiStage2.py script that generates the tables (it's not in
40        the tarball, but is in the Subversion repository).
41    
42    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
43        scripts.
44    
45    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
46        a group with a zero qualifier. The result of the study could be incorrect,
47        or the function might crash, depending on the pattern.
48    
49    4.  Caseless matching was not working for non-ASCII characters in back
50        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
51        It now works when Unicode Property Support is available.
52    
53    5.  In pcretest, an escape such as \x{de} in the data was always generating
54        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
55        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
56        truncation.
57    
58    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
59    
60    7.  Added two (int) casts to pcregrep when printing the difference of two
61        pointers, in case they are 64-bit values.
62    
63    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
64        test 2 if it fails.
65    
66    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
67        and a #define of that name to empty if it is not externally set. This is to
68        allow users of MSVC to set it if necessary.
69    
70    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
71        the convenience functions in the pcre_get.c source file.
72    
73    11. An option change at the start of a pattern that had top-level alternatives
74        could cause overwriting and/or a crash. This command provoked a crash in
75        some environments:
76    
77          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
78    
79        This potential security problem was recorded as CVE-2008-2371.
80    
81    12. For a pattern where the match had to start at the beginning or immediately
82        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
83        pcre_dfa_exec() could read past the end of the passed subject if there was
84        no match. To help with detecting such bugs (e.g. with valgrind), I modified
85        pcretest so that it places the subject at the end of its malloc-ed buffer.
86    
87    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
88        exec() might read past the end of the data buffer in UTF-8 mode.
89    
90    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
91        the data contained the byte 0x85 as part of a UTF-8 character within its
92        first line. This applied both to normal and DFA matching.
93    
94    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
95        /^[^d]*?$/8 failed to match "abc".
96    
97    16. Added a missing copyright notice to pcrecpp_internal.h.
98    
99    17. Make it more clear in the documentation that values returned from
100        pcre_exec() in ovector are byte offsets, not character counts.
101    
102    18. Tidied a few places to stop certain compilers from issuing warnings.
103    
104    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
105        supplied by Stefan Weber. I made a further small update for 7.8 because
106        there is a change of source arrangements: the pcre_searchfuncs.c module is
107        replaced by pcre_ucd.c.
108    
109    
110    Version 7.7 07-May-08
111    ---------------------
112    
113    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
114        a string to a long long, pretend we don't even have a long long." This is
115        done by checking for the strtoq, strtoll, and _strtoi64 functions.
116    
117    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
118        pre-7.6 versions, which defined a global no_arg variable instead of putting
119        it in the RE class. (See also #8 below.)
120    
121    3.  Remove a line of dead code, identified by coverity and reported by Nuno
122        Lopes.
123    
124    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
125    
126        (1) The include/exclude patterns were being applied to the whole pathnames
127            of files, instead of just to the final components.
128    
129        (2) If there was more than one level of directory, the subdirectories were
130            skipped unless they satisfied the include/exclude conditions. This is
131            inconsistent with GNU grep (and could even be seen as contrary to the
132            pcregrep specification - which I improved to make it absolutely clear).
133            The action now is always to scan all levels of directory, and just
134            apply the include/exclude patterns to regular files.
135    
136    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
137        --exclude_dir in the tests to avoid scanning .svn directories.
138    
139    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
140        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
141        doesn't support NULs in patterns.
142    
143    7.  Added some missing "const"s to declarations of static tables in
144        pcre_compile.c and pcre_dfa_exec.c.
145    
146    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
147        caused by fix #2  above. (Subsequently also a second patch to fix the
148        first patch. And a third patch - this was a messy problem.)
149    
150    9.  Applied Craig's patch to remove the use of push_back().
151    
152    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
153        matching function regexec().
154    
155    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
156        which, however, unlike Perl's \g{...}, are subroutine calls, not back
157        references. PCRE supports relative numbers with this syntax (I don't think
158        Oniguruma does).
159    
160    12. Previously, a group with a zero repeat such as (...){0} was completely
161        omitted from the compiled regex. However, this means that if the group
162        was called as a subroutine from elsewhere in the pattern, things went wrong
163        (an internal error was given). Such groups are now left in the compiled
164        pattern, with a new opcode that causes them to be skipped at execution
165        time.
166    
167    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
168        to the way PCRE behaves:
169    
170        (a) A lone ] character is dis-allowed (Perl treats it as data).
171    
172        (b) A back reference to an unmatched subpattern matches an empty string
173            (Perl fails the current match path).
174    
175        (c) A data ] in a character class must be notated as \] because if the
176            first data character in a class is ], it defines an empty class. (In
177            Perl it is not possible to have an empty class.) The empty class []
178            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
179            The negative empty class [^] matches any one character, independently
180            of the DOTALL setting.
181    
182    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
183        non-existent subpattern following a character class starting with ']' and
184        containing () gave an internal compiling error instead of "reference to
185        non-existent subpattern". Fortunately, when the pattern did exist, the
186        compiled code was correct. (When scanning forwards to check for the
187        existencd of the subpattern, it was treating the data ']' as terminating
188        the class, so got the count wrong. When actually compiling, the reference
189        was subsequently set up correctly.)
190    
191    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
192        it was being rejected as not supported by pcre_dfa_exec(), even though
193        other assertions are supported. I have made pcre_dfa_exec() support
194        (*FAIL).
195    
196    16. The implementation of 13c above involved the invention of a new opcode,
197        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
198        cannot be changed at match time, I realized I could make a small
199        improvement to matching performance by compiling OP_ALLANY instead of
200        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
201        on the OP_ANY path.
202    
203    17. Compiling pcretest on Windows with readline support failed without the
204        following two fixes: (1) Make the unistd.h include conditional on
205        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
206    
207    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
208        ncurses library to be included for pcretest when ReadLine support is
209        requested, but also to allow for it to be overridden. This patch came from
210        Daniel Bergström.
211    
212    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
213        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
214        any errors with the current Unicode tables. Thanks to Peter Kankowski for
215        spotting this.
216    
217    
218    Version 7.6 28-Jan-08
219    ---------------------
220    
221    1.  A character class containing a very large number of characters with
222        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
223        overflow.
224    
225    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
226        HAVE_LONG_LONG is not defined.
227    
228    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
229        bring them up to date and include new features. This patch includes:
230    
231        - Fixed PH's badly added libz and libbz2 support.
232        - Fixed a problem with static linking.
233        - Added pcredemo. [But later removed - see 7 below.]
234        - Fixed dftables problem and added an option.
235        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
236            HAVE_LONG_LONG.
237        - Added readline support for pcretest.
238        - Added an listing of the option settings after cmake has run.
239    
240    4.  A user submitted a patch to Makefile that makes it easy to create
241        "pcre.dll" under mingw when using Configure/Make. I added stuff to
242        Makefile.am that cause it to include this special target, without
243        affecting anything else. Note that the same mingw target plus all
244        the other distribution libraries and programs are now supported
245        when configuring with CMake (see 6 below) instead of with
246        Configure/Make.
247    
248    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
249        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
250        exported in the Windows port". It has not yet been confirmed that the patch
251        solves the problem, but it does no harm.
252    
253    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
254        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
255        with CMake, and also correct the comment about stack recursion.
256    
257    7.  Remove the automatic building of pcredemo from the ./configure system and
258        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
259        of a program that users should build themselves after PCRE is installed, so
260        building it automatically is not really right. What is more, it gave
261        trouble in some build environments.
262    
263    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
264    
265    
266    Version 7.5 10-Jan-08
267    ---------------------
268    
269    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
270        values in parens when parsing an RE using the C++ wrapper."
271    
272    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
273        Characters greater than 255 were excluded from the class instead of being
274        included.
275    
276    3.  The same bug as (2) above applied to negated POSIX classes such as
277        [:^space:].
278    
279    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
280        defined or documented. It seems to have been a typo for PCRE_STATIC, so
281        I have changed it.
282    
283    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
284        first named subpattern) and a construct such as (?&a) would reference the
285        first named subpattern whose name started with "a" (in other words, the
286        length check was missing). Both these problems are fixed. "Subpattern name
287        expected" is now given for (?&) (a zero-length name), and this patch also
288        makes it give the same error for \k'' (previously it complained that that
289        was a reference to a non-existent subpattern).
290    
291    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
292        this is right because (?- can be followed by option settings as well as by
293        digits. I have, however, made the messages clearer.
294    
295    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
296        than the number used in the conditional) now cause a compile-time error.
297        This is actually not compatible with Perl, which accepts such patterns, but
298        treats the conditional as always being FALSE (as PCRE used to), but it
299        seems to me that giving a diagnostic is better.
300    
301    8.  Change "alphameric" to the more common word "alphanumeric" in comments
302        and messages.
303    
304    9.  Fix two occurrences of "backslash" in comments that should have been
305        "backspace".
306    
307    10. Remove two redundant lines of code that can never be obeyed (their function
308        was moved elsewhere).
309    
310    11. The program that makes PCRE's Unicode character property table had a bug
311        which caused it to generate incorrect table entries for sequences of
312        characters that have the same character type, but are in different scripts.
313        It amalgamated them into a single range, with the script of the first of
314        them. In other words, some characters were in the wrong script. There were
315        thirteen such cases, affecting characters in the following ranges:
316    
317          U+002b0 - U+002c1
318          U+0060c - U+0060d
319          U+0061e - U+00612
320          U+0064b - U+0065e
321          U+0074d - U+0076d
322          U+01800 - U+01805
323          U+01d00 - U+01d77
324          U+01d9b - U+01dbf
325          U+0200b - U+0200f
326          U+030fc - U+030fe
327          U+03260 - U+0327f
328          U+0fb46 - U+0fbb1
329          U+10450 - U+1049d
330    
331    12. The -o option (show only the matching part of a line) for pcregrep was not
332        compatible with GNU grep in that, if there was more than one match in a
333        line, it showed only the first of them. It now behaves in the same way as
334        GNU grep.
335    
336    13. If the -o and -v options were combined for pcregrep, it printed a blank
337        line for every non-matching line. GNU grep prints nothing, and pcregrep now
338        does the same. The return code can be used to tell if there were any
339        non-matching lines.
340    
341    14. Added --file-offsets and --line-offsets to pcregrep.
342    
343    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
344        infinitely looping recursion. The bug was that positive lookaheads were not
345        being skipped when checking for a possible empty match (negative lookaheads
346        and both kinds of lookbehind were skipped).
347    
348    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
349        inclusion of <windows.h> to before rather than after the definition of
350        INVALID_FILE_ATTRIBUTES (patch from David Byron).
351    
352    17. Specifying a possessive quantifier with a specific limit for a Unicode
353        character property caused pcre_compile() to compile bad code, which led at
354        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
355        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
356        caused the error; without that there was no problem.
357    
358    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
359    
360    19. Added --enable-pcretest-libreadline.
361    
362    20. In pcrecpp.cc, the variable 'count' was incremented twice in
363        RE::GlobalReplace(). As a result, the number of replacements returned was
364        double what it should be. I removed one of the increments, but Craig sent a
365        later patch that removed the other one (the right fix) and added unit tests
366        that check the return values (which was not done before).
367    
368    21. Several CMake things:
369    
370        (1) Arranged that, when cmake is used on Unix, the libraries end up with
371            the names libpcre and libpcreposix, not just pcre and pcreposix.
372    
373        (2) The above change means that pcretest and pcregrep are now correctly
374            linked with the newly-built libraries, not previously installed ones.
375    
376        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
377    
378    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
379        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
380        UTF-8 newline character). The key issue is that the pattern starts .*;
381        this means that the match must be either at the beginning, or after a
382        newline. The bug was in the code for advancing after a failed match and
383        checking that the new position followed a newline. It was not taking
384        account of UTF-8 characters correctly.
385    
386    23. PCRE was behaving differently from Perl in the way it recognized POSIX
387        character classes. PCRE was not treating the sequence [:...:] as a
388        character class unless the ... were all letters. Perl, however, seems to
389        allow any characters between [: and :], though of course it rejects as
390        unknown any "names" that contain non-letters, because all the known class
391        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
392        for example, whereas PCRE did not - it did not recognize a POSIX character
393        class. This seemed a bit dangerous, so the code has been changed to be
394        closer to Perl. The behaviour is not identical to Perl, because PCRE will
395        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
396        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
397        Perl does, and where it didn't before.
398    
399    24. Rewrite so as to remove the single use of %n from pcregrep because in some
400        Windows environments %n is disabled by default.
401    
402    
403    Version 7.4 21-Sep-07
404    ---------------------
405    
406    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
407        means that a class such as [\s] counted as "explicit reference to CR or
408        LF". That isn't really right - the whole point of the change was to try to
409        help when there was an actual mention of one of the two characters. So now
410        the change happens only if \r or \n (or a literal CR or LF) character is
411        encountered.
412    
413    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
414        of both had grown to the point where there were only 3 bits left.
415        Fortunately, there was spare space in the data structure, and so I have
416        moved the internal flags into a new 16-bit field to free up more option
417        bits.
418    
419    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
420        but did not set the internal JCHANGED flag - either of these is enough to
421        control the way the "get" function works - but the PCRE_INFO_JCHANGED
422        facility is supposed to tell if (?J) was ever used, so now (?J) at the
423        start sets both bits.
424    
425    4.  Added options (at build time, compile time, exec time) to change \R from
426        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
427    
428    5.  doc/pcresyntax.html was missing from the distribution.
429    
430    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
431        compatibility, even though it is no longer used.
432    
433    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
434        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
435        windows.h file is present (where different names are used). [This was
436        reversed later after testing - see 16 below.]
437    
438    8.  Changed all #include <config.h> to #include "config.h". There were also
439        some further <pcre.h> cases that I changed to "pcre.h".
440    
441    9.  When pcregrep was used with the --colour option, it missed the line ending
442        sequence off the lines that it output.
443    
444    10. It was pointed out to me that arrays of string pointers cause lots of
445        relocations when a shared library is dynamically loaded. A technique of
446        using a single long string with a table of offsets can drastically reduce
447        these. I have refactored PCRE in four places to do this. The result is
448        dramatic:
449    
450          Originally:                          290
451          After changing UCP table:            187
452          After changing error message table:   43
453          After changing table of "verbs"       36
454          After changing table of Posix names   22
455    
456        Thanks to the folks working on Gregex for glib for this insight.
457    
458    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
459        unicode-properties was also set.
460    
461    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
462    
463    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
464        checked only for CRLF.
465    
466    14. Added casts to pcretest.c to avoid compiler warnings.
467    
468    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
469    
470    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
471        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
472        entirely. This removes changes made in 7 above.
473    
474    17. The CMake files have been updated, and there is now more information about
475        building with CMake in the NON-UNIX-USE document.
476    
477    
478    Version 7.3 28-Aug-07
479    ---------------------
480    
481     1. In the rejigging of the build system that eventually resulted in 7.1, the
482        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
483        brackets there is not right, since it causes compilers to look for an
484        installed pcre.h, not the version that is in the source that is being
485        compiled (which of course may be different). I have changed it back to:
486    
487          #include "pcre.h"
488    
489        I have a vague recollection that the change was concerned with compiling in
490        different directories, but in the new build system, that is taken care of
491        by the VPATH setting the Makefile.
492    
493     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
494        when the subject happened to end in the byte 0x85 (e.g. if the last
495        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
496        characters but of course it shouldn't be taken as a newline when it is part
497        of another character. The bug was that, for an unlimited repeat of . in
498        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
499        characters when looking for a newline.
500    
501     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
502    
503     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
504        in debug output.
505    
506     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
507        long printing in the pcrecpp unittest when running under MinGW.
508    
509     6. ESC_K was left out of the EBCDIC table.
510    
511     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
512        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
513        limit also applies to "virtual nesting" when a pattern is recursive, and in
514        this case 1000 isn't so big. I have been able to remove this limit at the
515        expense of backing off one optimization in certain circumstances. Normally,
516        when pcre_exec() would call its internal match() function recursively and
517        immediately return the result unconditionally, it uses a "tail recursion"
518        feature to save stack. However, when a subpattern that can match an empty
519        string has an unlimited repetition quantifier, it no longer makes this
520        optimization. That gives it a stack frame in which to save the data for
521        checking that an empty string has been matched. Previously this was taken
522        from the 1000-entry workspace that had been reserved. So now there is no
523        explicit limit, but more stack is used.
524    
525     8. Applied Daniel's patches to solve problems with the import/export magic
526        syntax that is required for Windows, and which was going wrong for the
527        pcreposix and pcrecpp parts of the library. These were overlooked when this
528        problem was solved for the main library.
529    
530     9. There were some crude static tests to avoid integer overflow when computing
531        the size of patterns that contain repeated groups with explicit upper
532        limits. As the maximum quantifier is 65535, the maximum group length was
533        set at 30,000 so that the product of these two numbers did not overflow a
534        32-bit integer. However, it turns out that people want to use groups that
535        are longer than 30,000 bytes (though not repeat them that many times).
536        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
537        made it possible to implement the integer overflow checks in a much more
538        dynamic way, which I have now done. The artificial limitation on group
539        length has been removed - we now have only the limit on the total length of
540        the compiled pattern, which depends on the LINK_SIZE setting.
541    
542    10. Fixed a bug in the documentation for get/copy named substring when
543        duplicate names are permitted. If none of the named substrings are set, the
544        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
545        empty string.
546    
547    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
548        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
549        because the ] is interpreted as the first data character and the
550        terminating ] is not found. PCRE has been made compatible with Perl in this
551        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
552        cause memory overwriting.
553    
554    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
555        string has been matched (to stop an infinite loop). It was not recognizing
556        a conditional subpattern that could match an empty string if that
557        subpattern was within another subpattern. For example, it looped when
558        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
559        condition was not nested. This bug has been fixed.
560    
561    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
562        past the start of the subject in the presence of bytes with the top bit
563        set, for example "\x8aBCD".
564    
565    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
566        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
567    
568    14. Optimized (?!) to (*FAIL).
569    
570    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
571        This restricts code points to be within the range 0 to 0x10FFFF, excluding
572        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
573        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
574        does: it's just the validity check that is more restrictive.
575    
576    16. Inserted checks for integer overflows during escape sequence (backslash)
577        processing, and also fixed erroneous offset values for syntax errors during
578        backslash processing.
579    
580    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
581        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
582    
583    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
584        caused an overrun.
585    
586    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
587        something other than just ASCII characters) inside a group that had an
588        unlimited repeat caused a loop at compile time (while checking to see
589        whether the group could match an empty string).
590    
591    20. Debugging a pattern containing \p or \P could cause a crash. For example,
592        [\P{Any}] did so. (Error in the code for printing property names.)
593    
594    21. An orphan \E inside a character class could cause a crash.
595    
596    22. A repeated capturing bracket such as (A)? could cause a wild memory
597        reference during compilation.
598    
599    23. There are several functions in pcre_compile() that scan along a compiled
600        expression for various reasons (e.g. to see if it's fixed length for look
601        behind). There were bugs in these functions when a repeated \p or \P was
602        present in the pattern. These operators have additional parameters compared
603        with \d, etc, and these were not being taken into account when moving along
604        the compiled data. Specifically:
605    
606        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
607            length.
608    
609        (b) An item such as \pL+ within a repeated group could cause crashes or
610            loops.
611    
612        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
613            "reference to non-existent subpattern" error.
614    
615        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
616    
617    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
618        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
619    
620    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
621    
622    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
623        character were causing crashes (broken optimization).
624    
625    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
626        \p or \P) caused a compile-time loop.
627    
628    28. More problems have arisen in unanchored patterns when CRLF is a valid line
629        break. For example, the unstudied pattern [\r\n]A does not match the string
630        "\r\nA" because change 7.0/46 below moves the current point on by two
631        characters after failing to match at the start. However, the pattern \nA
632        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
633        the same is true. There doesn't seem any very clean way out of this, but
634        what I have chosen to do makes the common cases work: PCRE now takes note
635        of whether there can be an explicit match for \r or \n anywhere in the
636        pattern, and if so, 7.0/46 no longer applies. As part of this change,
637        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
638        pattern has explicit CR or LF references.
639    
640    29. Added (*CR) etc for changing newline setting at start of pattern.
641    
642    
643    Version 7.2 19-Jun-07
644    ---------------------
645    
646     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
647        which is apparently normally available under Windows.
648    
649     2. Re-jig the pcregrep tests with different newline settings in an attempt
650        to make them independent of the local environment's newline setting.
651    
652     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
653    
654     4. Some of the "internals" tests were previously cut out when the link size
655        was not 2, because the output contained actual offsets. The recent new
656        "Z" feature of pcretest means that these can be cut out, making the tests
657        usable with all link sizes.
658    
659     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
660        stack recursion. This gives a massive performance boost under BSD, but just
661        a small improvement under Linux. However, it saves one field in the frame
662        in all cases.
663    
664     6. Added more features from the forthcoming Perl 5.10:
665    
666        (a) (?-n) (where n is a string of digits) is a relative subroutine or
667            recursion call. It refers to the nth most recently opened parentheses.
668    
669        (b) (?+n) is also a relative subroutine call; it refers to the nth next
670            to be opened parentheses.
671    
672        (c) Conditions that refer to capturing parentheses can be specified
673            relatively, for example, (?(-2)... or (?(+3)...
674    
675        (d) \K resets the start of the current match so that everything before
676            is not part of it.
677    
678        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
679    
680        (f) \g{name} is another synonym - part of Perl 5.10's unification of
681            reference syntax.
682    
683        (g) (?| introduces a group in which the numbering of parentheses in each
684            alternative starts with the same number.
685    
686        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
687    
688     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
689        PCRE_INFO_JCHANGED.
690    
691     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
692        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
693        for detecting groups that can match an empty string.
694    
695     9. A pattern with a very large number of alternatives (more than several
696        hundred) was running out of internal workspace during the pre-compile
697        phase, where pcre_compile() figures out how much memory will be needed. A
698        bit of new cunning has reduced the workspace needed for groups with
699        alternatives. The 1000-alternative test pattern now uses 12 bytes of
700        workspace instead of running out of the 4096 that are available.
701    
702    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
703    
704    11. Applied patch from Google to remove an optimization that didn't quite work.
705        The report of the bug said:
706    
707          pcrecpp::RE("a*").FullMatch("aaa") matches, while
708          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
709          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
710    
711    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
712        it matched the wrong number of bytes.
713    
714    
715    Version 7.1 24-Apr-07
716    ---------------------
717    
718     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
719        that is more "standard", making use of automake and other Autotools. There
720        is some re-arrangement of the files and adjustment of comments consequent
721        on this.
722    
723     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
724        for recursive directory scanning broke on some systems because the files
725        are not scanned in any specific order and on different systems the order
726        was different. A call to "sort" has been inserted into RunGrepTest for the
727        approprate test as a short-term fix. In the longer term there may be an
728        alternative.
729    
730     3. I had an email from Eric Raymond about problems translating some of PCRE's
731        man pages to HTML (despite the fact that I distribute HTML pages, some
732        people do their own conversions for various reasons). The problems
733        concerned the use of low-level troff macros .br and .in. I have therefore
734        removed all such uses from the man pages (some were redundant, some could
735        be replaced by .nf/.fi pairs). The 132html script that I use to generate
736        HTML has been updated to handle .nf/.fi and to complain if it encounters
737        .br or .in.
738    
739     4. Updated comments in configure.ac that get placed in config.h.in and also
740        arranged for config.h to be included in the distribution, with the name
741        config.h.generic, for the benefit of those who have to compile without
742        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
743    
744     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
745        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
746        makevp.bat for the current PCRE, using the additional files
747        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
748    
749     6. A Windows user reported a minor discrepancy with test 2, which turned out
750        to be caused by a trailing space on an input line that had got lost in his
751        copy. The trailing space was an accident, so I've just removed it.
752    
753     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
754        that is needed.
755    
756     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
757        as "const" (a) because they are and (b) because it helps the PHP
758        maintainers who have recently made a script to detect big data structures
759        in the php code that should be moved to the .rodata section. I remembered
760        to update Builducptable as well, so it won't revert if ucptable.h is ever
761        re-created.
762    
763     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
764        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
765        order to be able to cut out the UTF-8 tables in the latter when UTF-8
766        support is not required. This saves 1.5-2K of code, which is important in
767        some applications.
768    
769        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
770        so as not to refer to the tables, even though these functions will never be
771        called when UTF-8 support is disabled. Otherwise there are problems with a
772        shared library.
773    
774    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
775    
776        (a) It was defining its arguments as char * instead of void *.
777    
778        (b) It was assuming that all moves were upwards in memory; this was true
779            a long time ago when I wrote it, but is no longer the case.
780    
781        The emulated memove() is provided for those environments that have neither
782        memmove() nor bcopy(). I didn't think anyone used it these days, but that
783        is clearly not the case, as these two bugs were recently reported.
784    
785    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
786        and Detrail to create the HTML documentation, the .txt form of the man
787        pages, and it removes trailing spaces from listed files. It also creates
788        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
789        case, it wraps all the #defines with #ifndefs. This script should be run
790        before "make dist".
791    
792    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
793        with Unicode property support.
794    
795        (a) For a maximizing quantifier, if the two different cases of the
796            character were of different lengths in their UTF-8 codings (there are
797            some cases like this - I found 11), and the matching function had to
798            back up over a mixture of the two cases, it incorrectly assumed they
799            were both the same length.
800    
801        (b) When PCRE was configured to use the heap rather than the stack for
802            recursion during matching, it was not correctly preserving the data for
803            the other case of a UTF-8 character when checking ahead for a match
804            while processing a minimizing repeat. If the check also involved
805            matching a wide character, but failed, corruption could cause an
806            erroneous result when trying to check for a repeat of the original
807            character.
808    
809    13. Some tidying changes to the testing mechanism:
810    
811        (a) The RunTest script now detects the internal link size and whether there
812            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
813            values substituted by "configure". (The RunGrepTest script already did
814            this for UTF-8.) The configure.ac script no longer substitutes the
815            relevant variables.
816    
817        (b) The debugging options /B and /D in pcretest show the compiled bytecode
818            with length and offset values. This means that the output is different
819            for different internal link sizes. Test 2 is skipped for link sizes
820            other than 2 because of this, bypassing the problem. Unfortunately,
821            there was also a test in test 3 (the locale tests) that used /B and
822            failed for link sizes other than 2. Rather than cut the whole test out,
823            I have added a new /Z option to pcretest that replaces the length and
824            offset values with spaces. This is now used to make test 3 independent
825            of link size. (Test 2 will be tidied up later.)
826    
827    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
828        segmentation fault instead of returning the appropriate error message.
829    
830    15. In multiline mode when the newline sequence was set to "any", the pattern
831        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
832        This doesn't seem right; it now treats the CRLF combination as the line
833        ending, and so does not match in that case. It's only a pattern such as ^$
834        that would hit this one: something like ^ABC$ would have failed after \r
835        and then tried again after \r\n.
836    
837    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
838        in an attempt to make files that differ only in their line terminators
839        compare equal. This works on Linux.
840    
841    17. Under certain error circumstances pcregrep might try to free random memory
842        as it exited. This is now fixed, thanks to valgrind.
843    
844    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
845        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
846        was because its rules for how to advance for /g after matching an empty
847        string at the end of a line did not allow for this case. They now check for
848        it specially.
849    
850    20. pcretest is supposed to handle patterns and data of any length, by
851        extending its buffers when necessary. It was getting this wrong when the
852        buffer for a data line had to be extended.
853    
854    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
855        CRLF as a newline sequence.
856    
857    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
858        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
859        I have nevertheless tidied it up.
860    
861    23. Added some casts to kill warnings from HP-UX ia64 compiler.
862    
863    24. Added a man page for pcre-config.
864    
865    
866    Version 7.0 19-Dec-06
867    ---------------------
868    
869     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
870        moving to gcc 4.1.1.
871    
872     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
873        sys/time.h, which is documented as needed for this function. It doesn't
874        seem to matter on Linux, but it showed up on some releases of OS X.
875    
876     3. It seems that there are systems where bytes whose values are greater than
877        127 match isprint() in the "C" locale. The "C" locale should be the
878        default when a C program starts up. In most systems, only ASCII printing
879        characters match isprint(). This difference caused the output from pcretest
880        to vary, making some of the tests fail. I have changed pcretest so that:
881    
882        (a) When it is outputting text in the compiled version of a pattern, bytes
883            other than 32-126 are always shown as hex escapes.
884    
885        (b) When it is outputting text that is a matched part of a subject string,
886            it does the same, unless a different locale has been set for the match
887            (using the /L modifier). In this case, it uses isprint() to decide.
888    
889     4. Fixed a major bug that caused incorrect computation of the amount of memory
890        required for a compiled pattern when options that changed within the
891        pattern affected the logic of the preliminary scan that determines the
892        length. The relevant options are -x, and -i in UTF-8 mode. The result was
893        that the computed length was too small. The symptoms of this bug were
894        either the PCRE error "internal error: code overflow" from pcre_compile(),
895        or a glibc crash with a message such as "pcretest: free(): invalid next
896        size (fast)". Examples of patterns that provoked this bug (shown in
897        pcretest format) are:
898    
899          /(?-x: )/x
900          /(?x)(?-x: \s*#\s*)/
901          /((?i)[\x{c0}])/8
902          /(?i:[\x{c0}])/8
903    
904        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
905        is now done differently.
906    
907     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
908        wrapper classes; (b) implement a new function in the C++ scanner that is
909        more efficient than the old way of doing things because it avoids levels of
910        recursion in the regex matching; (c) add a paragraph to the documentation
911        for the FullMatch() function.
912    
913     6. The escape sequence \n was being treated as whatever was defined as
914        "newline". Not only was this contrary to the documentation, which states
915        that \n is character 10 (hex 0A), but it also went horribly wrong when
916        "newline" was defined as CRLF. This has been fixed.
917    
918     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
919        was being set to -1 for the "end of line" case (supposedly a value that no
920        character can have). Though this value is never used (the check for end of
921        line is "zero bytes in current character"), it caused compiler complaints.
922        I've changed it to 0xffffffff.
923    
924     8. In pcre_version.c, the version string was being built by a sequence of
925        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
926        string (as it is for production releases) called a macro with an empty
927        argument. The C standard says the result of this is undefined. The gcc
928        compiler treats it as an empty string (which was what was wanted) but it is
929        reported that Visual C gives an error. The source has been hacked around to
930        avoid this problem.
931    
932     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
933        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
934        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
935        of them did).
936    
937    10. Originally, pcretest opened its input and output without "b"; then I was
938        told that "b" was needed in some environments, so it was added for release
939        5.0 to both the input and output. (It makes no difference on Unix-like
940        systems.) Later I was told that it is wrong for the input on Windows. I've
941        now abstracted the modes into two macros, to make it easier to fiddle with
942        them, and removed "b" from the input mode under Windows.
943    
944    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
945    
946    12. Added -help and --help to pcretest as an official way of being reminded
947        of the options.
948    
949    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
950        and pcrecpp.cc because they annoy compilers at high warning levels.
951    
952    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
953    
954    15. Fixed an occurrence of == in configure.ac that should have been = (shell
955        scripts are not C programs :-) and which was not noticed because it works
956        on Linux.
957    
958    16. pcretest is supposed to handle any length of pattern and data line (as one
959        line or as a continued sequence of lines) by extending its input buffer if
960        necessary. This feature was broken for very long pattern lines, leading to
961        a string of junk being passed to pcre_compile() if the pattern was longer
962        than about 50K.
963    
964    17. I have done a major re-factoring of the way pcre_compile() computes the
965        amount of memory needed for a compiled pattern. Previously, there was code
966        that made a preliminary scan of the pattern in order to do this. That was
967        OK when PCRE was new, but as the facilities have expanded, it has become
968        harder and harder to keep it in step with the real compile phase, and there
969        have been a number of bugs (see for example, 4 above). I have now found a
970        cunning way of running the real compile function in a "fake" mode that
971        enables it to compute how much memory it would need, while actually only
972        ever using a few hundred bytes of working memory and without too many
973        tests of the mode. This should make future maintenance and development
974        easier. A side effect of this work is that the limit of 200 on the nesting
975        depth of parentheses has been removed (though this was never a serious
976        limitation, I suspect). However, there is a downside: pcre_compile() now
977        runs more slowly than before (30% or more, depending on the pattern). I
978        hope this isn't a big issue. There is no effect on runtime performance.
979    
980    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
981        newline (only possible for the last line of a file) and it was a
982        pattern that set a locale (followed by /Lsomething), pcretest crashed.
983    
984    19. Added additional timing features to pcretest. (1) The -tm option now times
985        matching only, not compiling. (2) Both -t and -tm can be followed, as a
986        separate command line item, by a number that specifies the number of
987        repeats to use when timing. The default is 50000; this gives better
988        precision, but takes uncomfortably long for very large patterns.
989    
990    20. Extended pcre_study() to be more clever in cases where a branch of a
991        subpattern has no definite first character. For example, (a*|b*)[cd] would
992        previously give no result from pcre_study(). Now it recognizes that the
993        first character must be a, b, c, or d.
994    
995    21. There was an incorrect error "recursive call could loop indefinitely" if
996        a subpattern (or the entire pattern) that was being tested for matching an
997        empty string contained only one non-empty item after a nested subpattern.
998        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
999        incorrectly, because the \d was being skipped in the check.
1000    
1001    22. The pcretest program now has a new pattern option /B and a command line
1002        option -b, which is equivalent to adding /B to every pattern. This causes
1003        it to show the compiled bytecode, without the additional information that
1004        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1005        is the same as /B/I).
1006    
1007    23. A new optimization is now able automatically to treat some sequences such
1008        as a*b as a*+b. More specifically, if something simple (such as a character
1009        or a simple class like \d) has an unlimited quantifier, and is followed by
1010        something that cannot possibly match the quantified thing, the quantifier
1011        is automatically "possessified".
1012    
1013    24. A recursive reference to a subpattern whose number was greater than 39
1014        went wrong under certain circumstances in UTF-8 mode. This bug could also
1015        have affected the operation of pcre_study().
1016    
1017    25. Realized that a little bit of performance could be had by replacing
1018        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1019    
1020    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1021    
1022    27. Possessive quantifiers such as a++ were previously implemented by turning
1023        them into atomic groups such as ($>a+). Now they have their own opcodes,
1024        which improves performance. This includes the automatically created ones
1025        from 23 above.
1026    
1027    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1028        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1029        the first matched character to be a colon. This applied both to named and
1030        numbered groups.
1031    
1032    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1033    
1034    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1035        building PCRE on Windows easier, so I have included it in the distribution.
1036    
1037    31. There is now a check in pcretest against a ridiculously large number being
1038        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1039        loop, the loop is abandoned.
1040    
1041    32. Forward references to subpatterns in conditions such as (?(2)...) where
1042        subpattern 2 is defined later cause pcre_compile() to search forwards in
1043        the pattern for the relevant set of parentheses. This search went wrong
1044        when there were unescaped parentheses in a character class, parentheses
1045        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1046    
1047    33. "Subroutine" calls and backreferences were previously restricted to
1048        referencing subpatterns earlier in the regex. This restriction has now
1049        been removed.
1050    
1051    34. Added a number of extra features that are going to be in Perl 5.10. On the
1052        whole, these are just syntactic alternatives for features that PCRE had
1053        previously implemented using the Python syntax or my own invention. The
1054        other formats are all retained for compatibility.
1055    
1056        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1057            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1058            also .NET compatible.
1059    
1060        (b) A recursion or subroutine call to a named group can now be defined as
1061            (?&name) as well as (?P>name).
1062    
1063        (c) A backreference to a named group can now be defined as \k<name> or
1064            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1065            5.10, are also .NET compatible.
1066    
1067        (d) A conditional reference to a named group can now use the syntax
1068            (?(<name>) or (?('name') as well as (?(name).
1069    
1070        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1071            groups (named and numbered) that are never evaluated inline, but can be
1072            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1073            is always false. There may be only one alternative in such a group.
1074    
1075        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1076            as the simple (?(R). The condition is true only if the most recent
1077            recursion is that of the given number or name. It does not search out
1078            through the entire recursion stack.
1079    
1080        (g) The escape \gN or \g{N} has been added, where N is a positive or
1081            negative number, specifying an absolute or relative reference.
1082    
1083    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1084        some "unreachable code" warnings.
1085    
1086    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1087        things, this adds five new scripts.
1088    
1089    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1090        There were also incompatibilities regarding the handling of \Q..\E inside
1091        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1092        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1093    
1094    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1095        matches an empty string, and forcibly breaks the loop. There were bugs in
1096        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1097        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1098        separate and independent bugs (that affected different cases) have been
1099        fixed.
1100    
1101    39. Refactored the code to abolish the use of different opcodes for small
1102        capturing bracket numbers. This is a tidy that I avoided doing when I
1103        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1104        The new approach is not only tidier, it makes it possible to reduce the
1105        memory needed to fix the previous bug (38).
1106    
1107    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1108        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1109        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1110        mode.
1111    
1112    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1113        report.
1114    
1115    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1116        copy construction and assignment in the C++ wrapper.
1117    
1118    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1119        couple of bugs that could have given wrong results in the "--newline=crlf"
1120        case.
1121    
1122    44. Added a number of casts and did some reorganization of signed/unsigned int
1123        variables following suggestions from Dair Grant. Also renamed the variable
1124        "this" as "item" because it is a C++ keyword.
1125    
1126    45. Arranged for dftables to add
1127    
1128          #include "pcre_internal.h"
1129    
1130        to pcre_chartables.c because without it, gcc 4.x may remove the array
1131        definition from the final binary if PCRE is built into a static library and
1132        dead code stripping is activated.
1133    
1134    46. For an unanchored pattern, if a match attempt fails at the start of a
1135        newline sequence, and the newline setting is CRLF or ANY, and the next two
1136        characters are CRLF, advance by two characters instead of one.
1137    
1138    
1139    Version 6.7 04-Jul-06
1140    ---------------------
1141    
1142     1. In order to handle tests when input lines are enormously long, pcretest has
1143        been re-factored so that it automatically extends its buffers when
1144        necessary. The code is crude, but this _is_ just a test program. The
1145        default size has been increased from 32K to 50K.
1146    
1147     2. The code in pcre_study() was using the value of the re argument before
1148        testing it for NULL. (Of course, in any sensible call of the function, it
1149        won't be NULL.)
1150    
1151     3. The memmove() emulation function in pcre_internal.h, which is used on
1152        systems that lack both memmove() and bcopy() - that is, hardly ever -
1153        was missing a "static" storage class specifier.
1154    
1155     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1156        containing an extended class (one that cannot be represented by a bitmap
1157        because it contains high-valued characters or Unicode property items, e.g.
1158        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1159        pattern, but PCRE should not loop if you do not (it no longer does).
1160        [Detail: two cases were found: (a) a repeated subpattern containing an
1161        extended class; (b) a recursive reference to a subpattern that followed a
1162        previous extended class. It wasn't skipping over the extended class
1163        correctly when UTF-8 mode was not set.]
1164    
1165     5. A negated single-character class was not being recognized as fixed-length
1166        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1167        compile error "lookbehind assertion is not fixed length".
1168    
1169     6. The RunPerlTest auxiliary script was showing an unexpected difference
1170        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1171        write a Perl script that can interpret lines of an input file either as
1172        byte characters or as UTF-8, which is what "perltest" was being required to
1173        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1174        can't do is switch easily at run time between having the "use utf8;" pragma
1175        or not. In the end, I fudged it by using the RunPerlTest script to insert
1176        "use utf8;" explicitly for the UTF-8 tests.
1177    
1178     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1179        the end of the subject string, contrary to the documentation and to what
1180        Perl does. This was true of both matching functions. Now it matches only at
1181        the start of the subject and immediately after *internal* newlines.
1182    
1183     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1184        a pointer to an int instead of a pointer to an unsigned long int. This
1185        caused problems on 64-bit systems.
1186    
1187     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1188        instance of the 'standard' template library not being so standard".
1189    
1190    10. There was no check on the number of named subpatterns nor the maximum
1191        length of a subpattern name. The product of these values is used to compute
1192        the size of the memory block for a compiled pattern. By supplying a very
1193        long subpattern name and a large number of named subpatterns, the size
1194        computation could be caused to overflow. This is now prevented by limiting
1195        the length of names to 32 characters, and the number of named subpatterns
1196        to 10,000.
1197    
1198    11. Subpatterns that are repeated with specific counts have to be replicated in
1199        the compiled pattern. The size of memory for this was computed from the
1200        length of the subpattern and the repeat count. The latter is limited to
1201        65535, but there was no limit on the former, meaning that integer overflow
1202        could in principle occur. The compiled length of a repeated subpattern is
1203        now limited to 30,000 bytes in order to prevent this.
1204    
1205    12. Added the optional facility to have named substrings with the same name.
1206    
1207    13. Added the ability to use a named substring as a condition, using the
1208        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1209        are numbers (not recommended). Forward references are permitted.
1210    
1211    14. Added forward references in named backreferences (if you see what I mean).
1212    
1213    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1214        pattern could run off the end of the subject. For example, the pattern
1215        "(?s)(.{1,5})"8 did this with the subject "ab".
1216    
1217    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1218        PCRE_CASELESS was set when matching characters that were quantified with ?
1219        or *.
1220    
1221    17. A character class other than a single negated character that had a minimum
1222        but no maximum quantifier - for example [ab]{6,} - was not handled
1223        correctly by pce_dfa_exec(). It would match only one character.
1224    
1225    18. A valid (though odd) pattern that looked like a POSIX character
1226        class but used an invalid character after [ (for example [[,abc,]]) caused
1227        pcre_compile() to give the error "Failed: internal error: code overflow" or
1228        in some cases to crash with a glibc free() error. This could even happen if
1229        the pattern terminated after [[ but there just happened to be a sequence of
1230        letters, a binary zero, and a closing ] in the memory that followed.
1231    
1232    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1233        over the years. Originally (before any Unicode support), just the bottom 8
1234        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1235        output from "man perlunicode" includes this:
1236    
1237          The regular expression compiler produces polymorphic opcodes.  That
1238          is, the pattern adapts to the data and automatically switches to
1239          the Unicode character scheme when presented with Unicode data--or
1240          instead uses a traditional byte scheme when presented with byte
1241          data.
1242    
1243        Sadly, a wide octal escape does not cause a switch, and in a string with
1244        no other multibyte characters, these octal escapes are treated as before.
1245        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1246        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1247        Unicode string.
1248    
1249        I have not perpetrated such confusion in PCRE. Up till now, it took just
1250        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1251        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1252        translate to the appropriate multibyte character.
1253    
1254    29. Applied some refactoring to reduce the number of warnings from Microsoft
1255        and Borland compilers. This has included removing the fudge introduced
1256        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1257        a warning about an unused variable.
1258    
1259    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1260        characters since release 4.0, because Perl (from release 5.004) does not.
1261        [Or at least, is documented not to: some releases seem to be in conflict
1262        with the documentation.] However, when a pattern was studied with
1263        pcre_study() and all its branches started with \s, PCRE still included VT
1264        as a possible starting character. Of course, this did no harm; it just
1265        caused an unnecessary match attempt.
1266    
1267    22. Removed a now-redundant internal flag bit that recorded the fact that case
1268        dependency changed within the pattern. This was once needed for "required
1269        byte" processing, but is no longer used. This recovers a now-scarce options
1270        bit. Also moved the least significant internal flag bit to the most-
1271        significant bit of the word, which was not previously used (hangover from
1272        the days when it was an int rather than a uint) to free up another bit for
1273        the future.
1274    
1275    23. Added support for CRLF line endings as well as CR and LF. As well as the
1276        default being selectable at build time, it can now be changed at runtime
1277        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1278        specify that it is scanning data with non-default line endings.
1279    
1280    24. Changed the definition of CXXLINK to make it agree with the definition of
1281        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1282    
1283    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1284        recursions. This makes a big different to stack usage for some patterns.
1285    
1286    26. If a subpattern containing a named recursion or subroutine reference such
1287        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1288        the space required for the compiled pattern went wrong and gave too small a
1289        value. Depending on the environment, this could lead to "Failed: internal
1290        error: code overflow at offset 49" or "glibc detected double free or
1291        corruption" errors.
1292    
1293    27. Applied patches from Google (a) to support the new newline modes and (b) to
1294        advance over multibyte UTF-8 characters in GlobalReplace.
1295    
1296    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1297        difference for some implementation of PCRE in some Windows version.
1298    
1299    29. Added some extra testing facilities to pcretest:
1300    
1301        \q<number>   in a data line sets the "match limit" value
1302        \Q<number>   in a data line sets the "match recursion limt" value
1303        -S <number>  sets the stack size, where <number> is in megabytes
1304    
1305        The -S option isn't available for Windows.
1306    
1307    
1308    Version 6.6 06-Feb-06
1309    ---------------------
1310    
1311     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1312        in pcreposix.h. I have copied the definition from pcre.h.
1313    
1314     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1315        because pcre.h is no longer a built file.
1316    
1317     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1318        not normally included in the compiled code.
1319    
1320    
1321  Version 6.5 01-Feb-06  Version 6.5 01-Feb-06
1322  ---------------------  ---------------------
1323    

Legend:
Removed from v.87  
changed lines
  Added in v.377

  ViewVC Help
Powered by ViewVC 1.1.5