/[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 389 by ph10, Sun Mar 15 18:24:05 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.  In pcregrep, when --colour was used with -o, the list of matching strings
28        was not coloured; this is different to GNU grep, so I have changed it to be
29        the same.
30    
31    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
32        each matching line was coloured. Now it goes on to look for further matches
33        of any of the test patterns, which is the same behaviour as GNU grep.
34    
35    7.  A pattern that could match an empty string could cause pcregrep to loop; it
36        doesn't make sense to accept an empty string match in pcregrep, so I have
37        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
38        seems to be how GNU grep behaves.
39    
40    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
41        start or after a newline", because the conditional assertion was not being
42        correctly handled. The rule now is that both the assertion and what follows
43        in the first alternative must satisfy the test.
44    
45    9.  If auto-callout was enabled in a pattern with a conditional group, PCRE
46        could crash during matching.
47    
48    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
49        used for matching.
50    
51    11. Unicode property support in character classes was not working for
52        characters (bytes) greater than 127 when not in UTF-8 mode.
53    
54    12. Added the -M command line option to pcretest.
55    
56    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
57    
58    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
59    
60    
61    Version 7.8 05-Sep-08
62    ---------------------
63    
64    1.  Replaced UCP searching code with optimized version as implemented for Ad
65        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
66        stage table and inline lookup instead of a function, giving speed ups of 2
67        to 5 times on some simple patterns that I tested. Permission was given to
68        distribute the MultiStage2.py script that generates the tables (it's not in
69        the tarball, but is in the Subversion repository).
70    
71    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
72        scripts.
73    
74    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
75        a group with a zero qualifier. The result of the study could be incorrect,
76        or the function might crash, depending on the pattern.
77    
78    4.  Caseless matching was not working for non-ASCII characters in back
79        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
80        It now works when Unicode Property Support is available.
81    
82    5.  In pcretest, an escape such as \x{de} in the data was always generating
83        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
84        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
85        truncation.
86    
87    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
88    
89    7.  Added two (int) casts to pcregrep when printing the difference of two
90        pointers, in case they are 64-bit values.
91    
92    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
93        test 2 if it fails.
94    
95    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
96        and a #define of that name to empty if it is not externally set. This is to
97        allow users of MSVC to set it if necessary.
98    
99    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
100        the convenience functions in the pcre_get.c source file.
101    
102    11. An option change at the start of a pattern that had top-level alternatives
103        could cause overwriting and/or a crash. This command provoked a crash in
104        some environments:
105    
106          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
107    
108        This potential security problem was recorded as CVE-2008-2371.
109    
110    12. For a pattern where the match had to start at the beginning or immediately
111        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
112        pcre_dfa_exec() could read past the end of the passed subject if there was
113        no match. To help with detecting such bugs (e.g. with valgrind), I modified
114        pcretest so that it places the subject at the end of its malloc-ed buffer.
115    
116    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
117        exec() might read past the end of the data buffer in UTF-8 mode.
118    
119    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
120        the data contained the byte 0x85 as part of a UTF-8 character within its
121        first line. This applied both to normal and DFA matching.
122    
123    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
124        /^[^d]*?$/8 failed to match "abc".
125    
126    16. Added a missing copyright notice to pcrecpp_internal.h.
127    
128    17. Make it more clear in the documentation that values returned from
129        pcre_exec() in ovector are byte offsets, not character counts.
130    
131    18. Tidied a few places to stop certain compilers from issuing warnings.
132    
133    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
134        supplied by Stefan Weber. I made a further small update for 7.8 because
135        there is a change of source arrangements: the pcre_searchfuncs.c module is
136        replaced by pcre_ucd.c.
137    
138    
139    Version 7.7 07-May-08
140    ---------------------
141    
142    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
143        a string to a long long, pretend we don't even have a long long." This is
144        done by checking for the strtoq, strtoll, and _strtoi64 functions.
145    
146    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
147        pre-7.6 versions, which defined a global no_arg variable instead of putting
148        it in the RE class. (See also #8 below.)
149    
150    3.  Remove a line of dead code, identified by coverity and reported by Nuno
151        Lopes.
152    
153    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
154    
155        (1) The include/exclude patterns were being applied to the whole pathnames
156            of files, instead of just to the final components.
157    
158        (2) If there was more than one level of directory, the subdirectories were
159            skipped unless they satisfied the include/exclude conditions. This is
160            inconsistent with GNU grep (and could even be seen as contrary to the
161            pcregrep specification - which I improved to make it absolutely clear).
162            The action now is always to scan all levels of directory, and just
163            apply the include/exclude patterns to regular files.
164    
165    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
166        --exclude_dir in the tests to avoid scanning .svn directories.
167    
168    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
169        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
170        doesn't support NULs in patterns.
171    
172    7.  Added some missing "const"s to declarations of static tables in
173        pcre_compile.c and pcre_dfa_exec.c.
174    
175    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
176        caused by fix #2  above. (Subsequently also a second patch to fix the
177        first patch. And a third patch - this was a messy problem.)
178    
179    9.  Applied Craig's patch to remove the use of push_back().
180    
181    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
182        matching function regexec().
183    
184    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
185        which, however, unlike Perl's \g{...}, are subroutine calls, not back
186        references. PCRE supports relative numbers with this syntax (I don't think
187        Oniguruma does).
188    
189    12. Previously, a group with a zero repeat such as (...){0} was completely
190        omitted from the compiled regex. However, this means that if the group
191        was called as a subroutine from elsewhere in the pattern, things went wrong
192        (an internal error was given). Such groups are now left in the compiled
193        pattern, with a new opcode that causes them to be skipped at execution
194        time.
195    
196    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
197        to the way PCRE behaves:
198    
199        (a) A lone ] character is dis-allowed (Perl treats it as data).
200    
201        (b) A back reference to an unmatched subpattern matches an empty string
202            (Perl fails the current match path).
203    
204        (c) A data ] in a character class must be notated as \] because if the
205            first data character in a class is ], it defines an empty class. (In
206            Perl it is not possible to have an empty class.) The empty class []
207            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
208            The negative empty class [^] matches any one character, independently
209            of the DOTALL setting.
210    
211    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
212        non-existent subpattern following a character class starting with ']' and
213        containing () gave an internal compiling error instead of "reference to
214        non-existent subpattern". Fortunately, when the pattern did exist, the
215        compiled code was correct. (When scanning forwards to check for the
216        existencd of the subpattern, it was treating the data ']' as terminating
217        the class, so got the count wrong. When actually compiling, the reference
218        was subsequently set up correctly.)
219    
220    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
221        it was being rejected as not supported by pcre_dfa_exec(), even though
222        other assertions are supported. I have made pcre_dfa_exec() support
223        (*FAIL).
224    
225    16. The implementation of 13c above involved the invention of a new opcode,
226        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
227        cannot be changed at match time, I realized I could make a small
228        improvement to matching performance by compiling OP_ALLANY instead of
229        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
230        on the OP_ANY path.
231    
232    17. Compiling pcretest on Windows with readline support failed without the
233        following two fixes: (1) Make the unistd.h include conditional on
234        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
235    
236    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
237        ncurses library to be included for pcretest when ReadLine support is
238        requested, but also to allow for it to be overridden. This patch came from
239        Daniel Bergström.
240    
241    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
242        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
243        any errors with the current Unicode tables. Thanks to Peter Kankowski for
244        spotting this.
245    
246    
247    Version 7.6 28-Jan-08
248    ---------------------
249    
250    1.  A character class containing a very large number of characters with
251        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
252        overflow.
253    
254    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
255        HAVE_LONG_LONG is not defined.
256    
257    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
258        bring them up to date and include new features. This patch includes:
259    
260        - Fixed PH's badly added libz and libbz2 support.
261        - Fixed a problem with static linking.
262        - Added pcredemo. [But later removed - see 7 below.]
263        - Fixed dftables problem and added an option.
264        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
265            HAVE_LONG_LONG.
266        - Added readline support for pcretest.
267        - Added an listing of the option settings after cmake has run.
268    
269    4.  A user submitted a patch to Makefile that makes it easy to create
270        "pcre.dll" under mingw when using Configure/Make. I added stuff to
271        Makefile.am that cause it to include this special target, without
272        affecting anything else. Note that the same mingw target plus all
273        the other distribution libraries and programs are now supported
274        when configuring with CMake (see 6 below) instead of with
275        Configure/Make.
276    
277    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
278        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
279        exported in the Windows port". It has not yet been confirmed that the patch
280        solves the problem, but it does no harm.
281    
282    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
283        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
284        with CMake, and also correct the comment about stack recursion.
285    
286    7.  Remove the automatic building of pcredemo from the ./configure system and
287        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
288        of a program that users should build themselves after PCRE is installed, so
289        building it automatically is not really right. What is more, it gave
290        trouble in some build environments.
291    
292    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
293    
294    
295    Version 7.5 10-Jan-08
296    ---------------------
297    
298    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
299        values in parens when parsing an RE using the C++ wrapper."
300    
301    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
302        Characters greater than 255 were excluded from the class instead of being
303        included.
304    
305    3.  The same bug as (2) above applied to negated POSIX classes such as
306        [:^space:].
307    
308    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
309        defined or documented. It seems to have been a typo for PCRE_STATIC, so
310        I have changed it.
311    
312    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
313        first named subpattern) and a construct such as (?&a) would reference the
314        first named subpattern whose name started with "a" (in other words, the
315        length check was missing). Both these problems are fixed. "Subpattern name
316        expected" is now given for (?&) (a zero-length name), and this patch also
317        makes it give the same error for \k'' (previously it complained that that
318        was a reference to a non-existent subpattern).
319    
320    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
321        this is right because (?- can be followed by option settings as well as by
322        digits. I have, however, made the messages clearer.
323    
324    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
325        than the number used in the conditional) now cause a compile-time error.
326        This is actually not compatible with Perl, which accepts such patterns, but
327        treats the conditional as always being FALSE (as PCRE used to), but it
328        seems to me that giving a diagnostic is better.
329    
330    8.  Change "alphameric" to the more common word "alphanumeric" in comments
331        and messages.
332    
333    9.  Fix two occurrences of "backslash" in comments that should have been
334        "backspace".
335    
336    10. Remove two redundant lines of code that can never be obeyed (their function
337        was moved elsewhere).
338    
339    11. The program that makes PCRE's Unicode character property table had a bug
340        which caused it to generate incorrect table entries for sequences of
341        characters that have the same character type, but are in different scripts.
342        It amalgamated them into a single range, with the script of the first of
343        them. In other words, some characters were in the wrong script. There were
344        thirteen such cases, affecting characters in the following ranges:
345    
346          U+002b0 - U+002c1
347          U+0060c - U+0060d
348          U+0061e - U+00612
349          U+0064b - U+0065e
350          U+0074d - U+0076d
351          U+01800 - U+01805
352          U+01d00 - U+01d77
353          U+01d9b - U+01dbf
354          U+0200b - U+0200f
355          U+030fc - U+030fe
356          U+03260 - U+0327f
357          U+0fb46 - U+0fbb1
358          U+10450 - U+1049d
359    
360    12. The -o option (show only the matching part of a line) for pcregrep was not
361        compatible with GNU grep in that, if there was more than one match in a
362        line, it showed only the first of them. It now behaves in the same way as
363        GNU grep.
364    
365    13. If the -o and -v options were combined for pcregrep, it printed a blank
366        line for every non-matching line. GNU grep prints nothing, and pcregrep now
367        does the same. The return code can be used to tell if there were any
368        non-matching lines.
369    
370    14. Added --file-offsets and --line-offsets to pcregrep.
371    
372    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
373        infinitely looping recursion. The bug was that positive lookaheads were not
374        being skipped when checking for a possible empty match (negative lookaheads
375        and both kinds of lookbehind were skipped).
376    
377    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
378        inclusion of <windows.h> to before rather than after the definition of
379        INVALID_FILE_ATTRIBUTES (patch from David Byron).
380    
381    17. Specifying a possessive quantifier with a specific limit for a Unicode
382        character property caused pcre_compile() to compile bad code, which led at
383        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
384        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
385        caused the error; without that there was no problem.
386    
387    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
388    
389    19. Added --enable-pcretest-libreadline.
390    
391    20. In pcrecpp.cc, the variable 'count' was incremented twice in
392        RE::GlobalReplace(). As a result, the number of replacements returned was
393        double what it should be. I removed one of the increments, but Craig sent a
394        later patch that removed the other one (the right fix) and added unit tests
395        that check the return values (which was not done before).
396    
397    21. Several CMake things:
398    
399        (1) Arranged that, when cmake is used on Unix, the libraries end up with
400            the names libpcre and libpcreposix, not just pcre and pcreposix.
401    
402        (2) The above change means that pcretest and pcregrep are now correctly
403            linked with the newly-built libraries, not previously installed ones.
404    
405        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
406    
407    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
408        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
409        UTF-8 newline character). The key issue is that the pattern starts .*;
410        this means that the match must be either at the beginning, or after a
411        newline. The bug was in the code for advancing after a failed match and
412        checking that the new position followed a newline. It was not taking
413        account of UTF-8 characters correctly.
414    
415    23. PCRE was behaving differently from Perl in the way it recognized POSIX
416        character classes. PCRE was not treating the sequence [:...:] as a
417        character class unless the ... were all letters. Perl, however, seems to
418        allow any characters between [: and :], though of course it rejects as
419        unknown any "names" that contain non-letters, because all the known class
420        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
421        for example, whereas PCRE did not - it did not recognize a POSIX character
422        class. This seemed a bit dangerous, so the code has been changed to be
423        closer to Perl. The behaviour is not identical to Perl, because PCRE will
424        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
425        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
426        Perl does, and where it didn't before.
427    
428    24. Rewrite so as to remove the single use of %n from pcregrep because in some
429        Windows environments %n is disabled by default.
430    
431    
432    Version 7.4 21-Sep-07
433    ---------------------
434    
435    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
436        means that a class such as [\s] counted as "explicit reference to CR or
437        LF". That isn't really right - the whole point of the change was to try to
438        help when there was an actual mention of one of the two characters. So now
439        the change happens only if \r or \n (or a literal CR or LF) character is
440        encountered.
441    
442    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
443        of both had grown to the point where there were only 3 bits left.
444        Fortunately, there was spare space in the data structure, and so I have
445        moved the internal flags into a new 16-bit field to free up more option
446        bits.
447    
448    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
449        but did not set the internal JCHANGED flag - either of these is enough to
450        control the way the "get" function works - but the PCRE_INFO_JCHANGED
451        facility is supposed to tell if (?J) was ever used, so now (?J) at the
452        start sets both bits.
453    
454    4.  Added options (at build time, compile time, exec time) to change \R from
455        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
456    
457    5.  doc/pcresyntax.html was missing from the distribution.
458    
459    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
460        compatibility, even though it is no longer used.
461    
462    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
463        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
464        windows.h file is present (where different names are used). [This was
465        reversed later after testing - see 16 below.]
466    
467    8.  Changed all #include <config.h> to #include "config.h". There were also
468        some further <pcre.h> cases that I changed to "pcre.h".
469    
470    9.  When pcregrep was used with the --colour option, it missed the line ending
471        sequence off the lines that it output.
472    
473    10. It was pointed out to me that arrays of string pointers cause lots of
474        relocations when a shared library is dynamically loaded. A technique of
475        using a single long string with a table of offsets can drastically reduce
476        these. I have refactored PCRE in four places to do this. The result is
477        dramatic:
478    
479          Originally:                          290
480          After changing UCP table:            187
481          After changing error message table:   43
482          After changing table of "verbs"       36
483          After changing table of Posix names   22
484    
485        Thanks to the folks working on Gregex for glib for this insight.
486    
487    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
488        unicode-properties was also set.
489    
490    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
491    
492    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
493        checked only for CRLF.
494    
495    14. Added casts to pcretest.c to avoid compiler warnings.
496    
497    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
498    
499    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
500        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
501        entirely. This removes changes made in 7 above.
502    
503    17. The CMake files have been updated, and there is now more information about
504        building with CMake in the NON-UNIX-USE document.
505    
506    
507    Version 7.3 28-Aug-07
508    ---------------------
509    
510     1. In the rejigging of the build system that eventually resulted in 7.1, the
511        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
512        brackets there is not right, since it causes compilers to look for an
513        installed pcre.h, not the version that is in the source that is being
514        compiled (which of course may be different). I have changed it back to:
515    
516          #include "pcre.h"
517    
518        I have a vague recollection that the change was concerned with compiling in
519        different directories, but in the new build system, that is taken care of
520        by the VPATH setting the Makefile.
521    
522     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
523        when the subject happened to end in the byte 0x85 (e.g. if the last
524        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
525        characters but of course it shouldn't be taken as a newline when it is part
526        of another character. The bug was that, for an unlimited repeat of . in
527        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
528        characters when looking for a newline.
529    
530     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
531    
532     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
533        in debug output.
534    
535     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
536        long printing in the pcrecpp unittest when running under MinGW.
537    
538     6. ESC_K was left out of the EBCDIC table.
539    
540     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
541        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
542        limit also applies to "virtual nesting" when a pattern is recursive, and in
543        this case 1000 isn't so big. I have been able to remove this limit at the
544        expense of backing off one optimization in certain circumstances. Normally,
545        when pcre_exec() would call its internal match() function recursively and
546        immediately return the result unconditionally, it uses a "tail recursion"
547        feature to save stack. However, when a subpattern that can match an empty
548        string has an unlimited repetition quantifier, it no longer makes this
549        optimization. That gives it a stack frame in which to save the data for
550        checking that an empty string has been matched. Previously this was taken
551        from the 1000-entry workspace that had been reserved. So now there is no
552        explicit limit, but more stack is used.
553    
554     8. Applied Daniel's patches to solve problems with the import/export magic
555        syntax that is required for Windows, and which was going wrong for the
556        pcreposix and pcrecpp parts of the library. These were overlooked when this
557        problem was solved for the main library.
558    
559     9. There were some crude static tests to avoid integer overflow when computing
560        the size of patterns that contain repeated groups with explicit upper
561        limits. As the maximum quantifier is 65535, the maximum group length was
562        set at 30,000 so that the product of these two numbers did not overflow a
563        32-bit integer. However, it turns out that people want to use groups that
564        are longer than 30,000 bytes (though not repeat them that many times).
565        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
566        made it possible to implement the integer overflow checks in a much more
567        dynamic way, which I have now done. The artificial limitation on group
568        length has been removed - we now have only the limit on the total length of
569        the compiled pattern, which depends on the LINK_SIZE setting.
570    
571    10. Fixed a bug in the documentation for get/copy named substring when
572        duplicate names are permitted. If none of the named substrings are set, the
573        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
574        empty string.
575    
576    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
577        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
578        because the ] is interpreted as the first data character and the
579        terminating ] is not found. PCRE has been made compatible with Perl in this
580        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
581        cause memory overwriting.
582    
583    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
584        string has been matched (to stop an infinite loop). It was not recognizing
585        a conditional subpattern that could match an empty string if that
586        subpattern was within another subpattern. For example, it looped when
587        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
588        condition was not nested. This bug has been fixed.
589    
590    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
591        past the start of the subject in the presence of bytes with the top bit
592        set, for example "\x8aBCD".
593    
594    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
595        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
596    
597    14. Optimized (?!) to (*FAIL).
598    
599    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
600        This restricts code points to be within the range 0 to 0x10FFFF, excluding
601        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
602        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
603        does: it's just the validity check that is more restrictive.
604    
605    16. Inserted checks for integer overflows during escape sequence (backslash)
606        processing, and also fixed erroneous offset values for syntax errors during
607        backslash processing.
608    
609    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
610        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
611    
612    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
613        caused an overrun.
614    
615    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
616        something other than just ASCII characters) inside a group that had an
617        unlimited repeat caused a loop at compile time (while checking to see
618        whether the group could match an empty string).
619    
620    20. Debugging a pattern containing \p or \P could cause a crash. For example,
621        [\P{Any}] did so. (Error in the code for printing property names.)
622    
623    21. An orphan \E inside a character class could cause a crash.
624    
625    22. A repeated capturing bracket such as (A)? could cause a wild memory
626        reference during compilation.
627    
628    23. There are several functions in pcre_compile() that scan along a compiled
629        expression for various reasons (e.g. to see if it's fixed length for look
630        behind). There were bugs in these functions when a repeated \p or \P was
631        present in the pattern. These operators have additional parameters compared
632        with \d, etc, and these were not being taken into account when moving along
633        the compiled data. Specifically:
634    
635        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
636            length.
637    
638        (b) An item such as \pL+ within a repeated group could cause crashes or
639            loops.
640    
641        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
642            "reference to non-existent subpattern" error.
643    
644        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
645    
646    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
647        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
648    
649    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
650    
651    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
652        character were causing crashes (broken optimization).
653    
654    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
655        \p or \P) caused a compile-time loop.
656    
657    28. More problems have arisen in unanchored patterns when CRLF is a valid line
658        break. For example, the unstudied pattern [\r\n]A does not match the string
659        "\r\nA" because change 7.0/46 below moves the current point on by two
660        characters after failing to match at the start. However, the pattern \nA
661        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
662        the same is true. There doesn't seem any very clean way out of this, but
663        what I have chosen to do makes the common cases work: PCRE now takes note
664        of whether there can be an explicit match for \r or \n anywhere in the
665        pattern, and if so, 7.0/46 no longer applies. As part of this change,
666        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
667        pattern has explicit CR or LF references.
668    
669    29. Added (*CR) etc for changing newline setting at start of pattern.
670    
671    
672    Version 7.2 19-Jun-07
673    ---------------------
674    
675     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
676        which is apparently normally available under Windows.
677    
678     2. Re-jig the pcregrep tests with different newline settings in an attempt
679        to make them independent of the local environment's newline setting.
680    
681     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
682    
683     4. Some of the "internals" tests were previously cut out when the link size
684        was not 2, because the output contained actual offsets. The recent new
685        "Z" feature of pcretest means that these can be cut out, making the tests
686        usable with all link sizes.
687    
688     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
689        stack recursion. This gives a massive performance boost under BSD, but just
690        a small improvement under Linux. However, it saves one field in the frame
691        in all cases.
692    
693     6. Added more features from the forthcoming Perl 5.10:
694    
695        (a) (?-n) (where n is a string of digits) is a relative subroutine or
696            recursion call. It refers to the nth most recently opened parentheses.
697    
698        (b) (?+n) is also a relative subroutine call; it refers to the nth next
699            to be opened parentheses.
700    
701        (c) Conditions that refer to capturing parentheses can be specified
702            relatively, for example, (?(-2)... or (?(+3)...
703    
704        (d) \K resets the start of the current match so that everything before
705            is not part of it.
706    
707        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
708    
709        (f) \g{name} is another synonym - part of Perl 5.10's unification of
710            reference syntax.
711    
712        (g) (?| introduces a group in which the numbering of parentheses in each
713            alternative starts with the same number.
714    
715        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
716    
717     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
718        PCRE_INFO_JCHANGED.
719    
720     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
721        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
722        for detecting groups that can match an empty string.
723    
724     9. A pattern with a very large number of alternatives (more than several
725        hundred) was running out of internal workspace during the pre-compile
726        phase, where pcre_compile() figures out how much memory will be needed. A
727        bit of new cunning has reduced the workspace needed for groups with
728        alternatives. The 1000-alternative test pattern now uses 12 bytes of
729        workspace instead of running out of the 4096 that are available.
730    
731    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
732    
733    11. Applied patch from Google to remove an optimization that didn't quite work.
734        The report of the bug said:
735    
736          pcrecpp::RE("a*").FullMatch("aaa") matches, while
737          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
738          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
739    
740    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
741        it matched the wrong number of bytes.
742    
743    
744    Version 7.1 24-Apr-07
745    ---------------------
746    
747     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
748        that is more "standard", making use of automake and other Autotools. There
749        is some re-arrangement of the files and adjustment of comments consequent
750        on this.
751    
752     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
753        for recursive directory scanning broke on some systems because the files
754        are not scanned in any specific order and on different systems the order
755        was different. A call to "sort" has been inserted into RunGrepTest for the
756        approprate test as a short-term fix. In the longer term there may be an
757        alternative.
758    
759     3. I had an email from Eric Raymond about problems translating some of PCRE's
760        man pages to HTML (despite the fact that I distribute HTML pages, some
761        people do their own conversions for various reasons). The problems
762        concerned the use of low-level troff macros .br and .in. I have therefore
763        removed all such uses from the man pages (some were redundant, some could
764        be replaced by .nf/.fi pairs). The 132html script that I use to generate
765        HTML has been updated to handle .nf/.fi and to complain if it encounters
766        .br or .in.
767    
768     4. Updated comments in configure.ac that get placed in config.h.in and also
769        arranged for config.h to be included in the distribution, with the name
770        config.h.generic, for the benefit of those who have to compile without
771        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
772    
773     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
774        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
775        makevp.bat for the current PCRE, using the additional files
776        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
777    
778     6. A Windows user reported a minor discrepancy with test 2, which turned out
779        to be caused by a trailing space on an input line that had got lost in his
780        copy. The trailing space was an accident, so I've just removed it.
781    
782     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
783        that is needed.
784    
785     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
786        as "const" (a) because they are and (b) because it helps the PHP
787        maintainers who have recently made a script to detect big data structures
788        in the php code that should be moved to the .rodata section. I remembered
789        to update Builducptable as well, so it won't revert if ucptable.h is ever
790        re-created.
791    
792     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
793        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
794        order to be able to cut out the UTF-8 tables in the latter when UTF-8
795        support is not required. This saves 1.5-2K of code, which is important in
796        some applications.
797    
798        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
799        so as not to refer to the tables, even though these functions will never be
800        called when UTF-8 support is disabled. Otherwise there are problems with a
801        shared library.
802    
803    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
804    
805        (a) It was defining its arguments as char * instead of void *.
806    
807        (b) It was assuming that all moves were upwards in memory; this was true
808            a long time ago when I wrote it, but is no longer the case.
809    
810        The emulated memove() is provided for those environments that have neither
811        memmove() nor bcopy(). I didn't think anyone used it these days, but that
812        is clearly not the case, as these two bugs were recently reported.
813    
814    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
815        and Detrail to create the HTML documentation, the .txt form of the man
816        pages, and it removes trailing spaces from listed files. It also creates
817        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
818        case, it wraps all the #defines with #ifndefs. This script should be run
819        before "make dist".
820    
821    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
822        with Unicode property support.
823    
824        (a) For a maximizing quantifier, if the two different cases of the
825            character were of different lengths in their UTF-8 codings (there are
826            some cases like this - I found 11), and the matching function had to
827            back up over a mixture of the two cases, it incorrectly assumed they
828            were both the same length.
829    
830        (b) When PCRE was configured to use the heap rather than the stack for
831            recursion during matching, it was not correctly preserving the data for
832            the other case of a UTF-8 character when checking ahead for a match
833            while processing a minimizing repeat. If the check also involved
834            matching a wide character, but failed, corruption could cause an
835            erroneous result when trying to check for a repeat of the original
836            character.
837    
838    13. Some tidying changes to the testing mechanism:
839    
840        (a) The RunTest script now detects the internal link size and whether there
841            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
842            values substituted by "configure". (The RunGrepTest script already did
843            this for UTF-8.) The configure.ac script no longer substitutes the
844            relevant variables.
845    
846        (b) The debugging options /B and /D in pcretest show the compiled bytecode
847            with length and offset values. This means that the output is different
848            for different internal link sizes. Test 2 is skipped for link sizes
849            other than 2 because of this, bypassing the problem. Unfortunately,
850            there was also a test in test 3 (the locale tests) that used /B and
851            failed for link sizes other than 2. Rather than cut the whole test out,
852            I have added a new /Z option to pcretest that replaces the length and
853            offset values with spaces. This is now used to make test 3 independent
854            of link size. (Test 2 will be tidied up later.)
855    
856    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
857        segmentation fault instead of returning the appropriate error message.
858    
859    15. In multiline mode when the newline sequence was set to "any", the pattern
860        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
861        This doesn't seem right; it now treats the CRLF combination as the line
862        ending, and so does not match in that case. It's only a pattern such as ^$
863        that would hit this one: something like ^ABC$ would have failed after \r
864        and then tried again after \r\n.
865    
866    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
867        in an attempt to make files that differ only in their line terminators
868        compare equal. This works on Linux.
869    
870    17. Under certain error circumstances pcregrep might try to free random memory
871        as it exited. This is now fixed, thanks to valgrind.
872    
873    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
874        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
875        was because its rules for how to advance for /g after matching an empty
876        string at the end of a line did not allow for this case. They now check for
877        it specially.
878    
879    20. pcretest is supposed to handle patterns and data of any length, by
880        extending its buffers when necessary. It was getting this wrong when the
881        buffer for a data line had to be extended.
882    
883    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
884        CRLF as a newline sequence.
885    
886    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
887        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
888        I have nevertheless tidied it up.
889    
890    23. Added some casts to kill warnings from HP-UX ia64 compiler.
891    
892    24. Added a man page for pcre-config.
893    
894    
895    Version 7.0 19-Dec-06
896    ---------------------
897    
898     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
899        moving to gcc 4.1.1.
900    
901     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
902        sys/time.h, which is documented as needed for this function. It doesn't
903        seem to matter on Linux, but it showed up on some releases of OS X.
904    
905     3. It seems that there are systems where bytes whose values are greater than
906        127 match isprint() in the "C" locale. The "C" locale should be the
907        default when a C program starts up. In most systems, only ASCII printing
908        characters match isprint(). This difference caused the output from pcretest
909        to vary, making some of the tests fail. I have changed pcretest so that:
910    
911        (a) When it is outputting text in the compiled version of a pattern, bytes
912            other than 32-126 are always shown as hex escapes.
913    
914        (b) When it is outputting text that is a matched part of a subject string,
915            it does the same, unless a different locale has been set for the match
916            (using the /L modifier). In this case, it uses isprint() to decide.
917    
918     4. Fixed a major bug that caused incorrect computation of the amount of memory
919        required for a compiled pattern when options that changed within the
920        pattern affected the logic of the preliminary scan that determines the
921        length. The relevant options are -x, and -i in UTF-8 mode. The result was
922        that the computed length was too small. The symptoms of this bug were
923        either the PCRE error "internal error: code overflow" from pcre_compile(),
924        or a glibc crash with a message such as "pcretest: free(): invalid next
925        size (fast)". Examples of patterns that provoked this bug (shown in
926        pcretest format) are:
927    
928          /(?-x: )/x
929          /(?x)(?-x: \s*#\s*)/
930          /((?i)[\x{c0}])/8
931          /(?i:[\x{c0}])/8
932    
933        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
934        is now done differently.
935    
936     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
937        wrapper classes; (b) implement a new function in the C++ scanner that is
938        more efficient than the old way of doing things because it avoids levels of
939        recursion in the regex matching; (c) add a paragraph to the documentation
940        for the FullMatch() function.
941    
942     6. The escape sequence \n was being treated as whatever was defined as
943        "newline". Not only was this contrary to the documentation, which states
944        that \n is character 10 (hex 0A), but it also went horribly wrong when
945        "newline" was defined as CRLF. This has been fixed.
946    
947     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
948        was being set to -1 for the "end of line" case (supposedly a value that no
949        character can have). Though this value is never used (the check for end of
950        line is "zero bytes in current character"), it caused compiler complaints.
951        I've changed it to 0xffffffff.
952    
953     8. In pcre_version.c, the version string was being built by a sequence of
954        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
955        string (as it is for production releases) called a macro with an empty
956        argument. The C standard says the result of this is undefined. The gcc
957        compiler treats it as an empty string (which was what was wanted) but it is
958        reported that Visual C gives an error. The source has been hacked around to
959        avoid this problem.
960    
961     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
962        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
963        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
964        of them did).
965    
966    10. Originally, pcretest opened its input and output without "b"; then I was
967        told that "b" was needed in some environments, so it was added for release
968        5.0 to both the input and output. (It makes no difference on Unix-like
969        systems.) Later I was told that it is wrong for the input on Windows. I've
970        now abstracted the modes into two macros, to make it easier to fiddle with
971        them, and removed "b" from the input mode under Windows.
972    
973    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
974    
975    12. Added -help and --help to pcretest as an official way of being reminded
976        of the options.
977    
978    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
979        and pcrecpp.cc because they annoy compilers at high warning levels.
980    
981    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
982    
983    15. Fixed an occurrence of == in configure.ac that should have been = (shell
984        scripts are not C programs :-) and which was not noticed because it works
985        on Linux.
986    
987    16. pcretest is supposed to handle any length of pattern and data line (as one
988        line or as a continued sequence of lines) by extending its input buffer if
989        necessary. This feature was broken for very long pattern lines, leading to
990        a string of junk being passed to pcre_compile() if the pattern was longer
991        than about 50K.
992    
993    17. I have done a major re-factoring of the way pcre_compile() computes the
994        amount of memory needed for a compiled pattern. Previously, there was code
995        that made a preliminary scan of the pattern in order to do this. That was
996        OK when PCRE was new, but as the facilities have expanded, it has become
997        harder and harder to keep it in step with the real compile phase, and there
998        have been a number of bugs (see for example, 4 above). I have now found a
999        cunning way of running the real compile function in a "fake" mode that
1000        enables it to compute how much memory it would need, while actually only
1001        ever using a few hundred bytes of working memory and without too many
1002        tests of the mode. This should make future maintenance and development
1003        easier. A side effect of this work is that the limit of 200 on the nesting
1004        depth of parentheses has been removed (though this was never a serious
1005        limitation, I suspect). However, there is a downside: pcre_compile() now
1006        runs more slowly than before (30% or more, depending on the pattern). I
1007        hope this isn't a big issue. There is no effect on runtime performance.
1008    
1009    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1010        newline (only possible for the last line of a file) and it was a
1011        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1012    
1013    19. Added additional timing features to pcretest. (1) The -tm option now times
1014        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1015        separate command line item, by a number that specifies the number of
1016        repeats to use when timing. The default is 50000; this gives better
1017        precision, but takes uncomfortably long for very large patterns.
1018    
1019    20. Extended pcre_study() to be more clever in cases where a branch of a
1020        subpattern has no definite first character. For example, (a*|b*)[cd] would
1021        previously give no result from pcre_study(). Now it recognizes that the
1022        first character must be a, b, c, or d.
1023    
1024    21. There was an incorrect error "recursive call could loop indefinitely" if
1025        a subpattern (or the entire pattern) that was being tested for matching an
1026        empty string contained only one non-empty item after a nested subpattern.
1027        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1028        incorrectly, because the \d was being skipped in the check.
1029    
1030    22. The pcretest program now has a new pattern option /B and a command line
1031        option -b, which is equivalent to adding /B to every pattern. This causes
1032        it to show the compiled bytecode, without the additional information that
1033        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1034        is the same as /B/I).
1035    
1036    23. A new optimization is now able automatically to treat some sequences such
1037        as a*b as a*+b. More specifically, if something simple (such as a character
1038        or a simple class like \d) has an unlimited quantifier, and is followed by
1039        something that cannot possibly match the quantified thing, the quantifier
1040        is automatically "possessified".
1041    
1042    24. A recursive reference to a subpattern whose number was greater than 39
1043        went wrong under certain circumstances in UTF-8 mode. This bug could also
1044        have affected the operation of pcre_study().
1045    
1046    25. Realized that a little bit of performance could be had by replacing
1047        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1048    
1049    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1050    
1051    27. Possessive quantifiers such as a++ were previously implemented by turning
1052        them into atomic groups such as ($>a+). Now they have their own opcodes,
1053        which improves performance. This includes the automatically created ones
1054        from 23 above.
1055    
1056    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1057        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1058        the first matched character to be a colon. This applied both to named and
1059        numbered groups.
1060    
1061    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1062    
1063    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1064        building PCRE on Windows easier, so I have included it in the distribution.
1065    
1066    31. There is now a check in pcretest against a ridiculously large number being
1067        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1068        loop, the loop is abandoned.
1069    
1070    32. Forward references to subpatterns in conditions such as (?(2)...) where
1071        subpattern 2 is defined later cause pcre_compile() to search forwards in
1072        the pattern for the relevant set of parentheses. This search went wrong
1073        when there were unescaped parentheses in a character class, parentheses
1074        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1075    
1076    33. "Subroutine" calls and backreferences were previously restricted to
1077        referencing subpatterns earlier in the regex. This restriction has now
1078        been removed.
1079    
1080    34. Added a number of extra features that are going to be in Perl 5.10. On the
1081        whole, these are just syntactic alternatives for features that PCRE had
1082        previously implemented using the Python syntax or my own invention. The
1083        other formats are all retained for compatibility.
1084    
1085        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1086            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1087            also .NET compatible.
1088    
1089        (b) A recursion or subroutine call to a named group can now be defined as
1090            (?&name) as well as (?P>name).
1091    
1092        (c) A backreference to a named group can now be defined as \k<name> or
1093            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1094            5.10, are also .NET compatible.
1095    
1096        (d) A conditional reference to a named group can now use the syntax
1097            (?(<name>) or (?('name') as well as (?(name).
1098    
1099        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1100            groups (named and numbered) that are never evaluated inline, but can be
1101            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1102            is always false. There may be only one alternative in such a group.
1103    
1104        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1105            as the simple (?(R). The condition is true only if the most recent
1106            recursion is that of the given number or name. It does not search out
1107            through the entire recursion stack.
1108    
1109        (g) The escape \gN or \g{N} has been added, where N is a positive or
1110            negative number, specifying an absolute or relative reference.
1111    
1112    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1113        some "unreachable code" warnings.
1114    
1115    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1116        things, this adds five new scripts.
1117    
1118    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1119        There were also incompatibilities regarding the handling of \Q..\E inside
1120        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1121        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1122    
1123    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1124        matches an empty string, and forcibly breaks the loop. There were bugs in
1125        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1126        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1127        separate and independent bugs (that affected different cases) have been
1128        fixed.
1129    
1130    39. Refactored the code to abolish the use of different opcodes for small
1131        capturing bracket numbers. This is a tidy that I avoided doing when I
1132        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1133        The new approach is not only tidier, it makes it possible to reduce the
1134        memory needed to fix the previous bug (38).
1135    
1136    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1137        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1138        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1139        mode.
1140    
1141    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1142        report.
1143    
1144    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1145        copy construction and assignment in the C++ wrapper.
1146    
1147    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1148        couple of bugs that could have given wrong results in the "--newline=crlf"
1149        case.
1150    
1151    44. Added a number of casts and did some reorganization of signed/unsigned int
1152        variables following suggestions from Dair Grant. Also renamed the variable
1153        "this" as "item" because it is a C++ keyword.
1154    
1155    45. Arranged for dftables to add
1156    
1157          #include "pcre_internal.h"
1158    
1159        to pcre_chartables.c because without it, gcc 4.x may remove the array
1160        definition from the final binary if PCRE is built into a static library and
1161        dead code stripping is activated.
1162    
1163    46. For an unanchored pattern, if a match attempt fails at the start of a
1164        newline sequence, and the newline setting is CRLF or ANY, and the next two
1165        characters are CRLF, advance by two characters instead of one.
1166    
1167    
1168    Version 6.7 04-Jul-06
1169    ---------------------
1170    
1171     1. In order to handle tests when input lines are enormously long, pcretest has
1172        been re-factored so that it automatically extends its buffers when
1173        necessary. The code is crude, but this _is_ just a test program. The
1174        default size has been increased from 32K to 50K.
1175    
1176     2. The code in pcre_study() was using the value of the re argument before
1177        testing it for NULL. (Of course, in any sensible call of the function, it
1178        won't be NULL.)
1179    
1180     3. The memmove() emulation function in pcre_internal.h, which is used on
1181        systems that lack both memmove() and bcopy() - that is, hardly ever -
1182        was missing a "static" storage class specifier.
1183    
1184     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1185        containing an extended class (one that cannot be represented by a bitmap
1186        because it contains high-valued characters or Unicode property items, e.g.
1187        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1188        pattern, but PCRE should not loop if you do not (it no longer does).
1189        [Detail: two cases were found: (a) a repeated subpattern containing an
1190        extended class; (b) a recursive reference to a subpattern that followed a
1191        previous extended class. It wasn't skipping over the extended class
1192        correctly when UTF-8 mode was not set.]
1193    
1194     5. A negated single-character class was not being recognized as fixed-length
1195        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1196        compile error "lookbehind assertion is not fixed length".
1197    
1198     6. The RunPerlTest auxiliary script was showing an unexpected difference
1199        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1200        write a Perl script that can interpret lines of an input file either as
1201        byte characters or as UTF-8, which is what "perltest" was being required to
1202        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1203        can't do is switch easily at run time between having the "use utf8;" pragma
1204        or not. In the end, I fudged it by using the RunPerlTest script to insert
1205        "use utf8;" explicitly for the UTF-8 tests.
1206    
1207     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1208        the end of the subject string, contrary to the documentation and to what
1209        Perl does. This was true of both matching functions. Now it matches only at
1210        the start of the subject and immediately after *internal* newlines.
1211    
1212     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1213        a pointer to an int instead of a pointer to an unsigned long int. This
1214        caused problems on 64-bit systems.
1215    
1216     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1217        instance of the 'standard' template library not being so standard".
1218    
1219    10. There was no check on the number of named subpatterns nor the maximum
1220        length of a subpattern name. The product of these values is used to compute
1221        the size of the memory block for a compiled pattern. By supplying a very
1222        long subpattern name and a large number of named subpatterns, the size
1223        computation could be caused to overflow. This is now prevented by limiting
1224        the length of names to 32 characters, and the number of named subpatterns
1225        to 10,000.
1226    
1227    11. Subpatterns that are repeated with specific counts have to be replicated in
1228        the compiled pattern. The size of memory for this was computed from the
1229        length of the subpattern and the repeat count. The latter is limited to
1230        65535, but there was no limit on the former, meaning that integer overflow
1231        could in principle occur. The compiled length of a repeated subpattern is
1232        now limited to 30,000 bytes in order to prevent this.
1233    
1234    12. Added the optional facility to have named substrings with the same name.
1235    
1236    13. Added the ability to use a named substring as a condition, using the
1237        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1238        are numbers (not recommended). Forward references are permitted.
1239    
1240    14. Added forward references in named backreferences (if you see what I mean).
1241    
1242    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1243        pattern could run off the end of the subject. For example, the pattern
1244        "(?s)(.{1,5})"8 did this with the subject "ab".
1245    
1246    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1247        PCRE_CASELESS was set when matching characters that were quantified with ?
1248        or *.
1249    
1250    17. A character class other than a single negated character that had a minimum
1251        but no maximum quantifier - for example [ab]{6,} - was not handled
1252        correctly by pce_dfa_exec(). It would match only one character.
1253    
1254    18. A valid (though odd) pattern that looked like a POSIX character
1255        class but used an invalid character after [ (for example [[,abc,]]) caused
1256        pcre_compile() to give the error "Failed: internal error: code overflow" or
1257        in some cases to crash with a glibc free() error. This could even happen if
1258        the pattern terminated after [[ but there just happened to be a sequence of
1259        letters, a binary zero, and a closing ] in the memory that followed.
1260    
1261    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1262        over the years. Originally (before any Unicode support), just the bottom 8
1263        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1264        output from "man perlunicode" includes this:
1265    
1266          The regular expression compiler produces polymorphic opcodes.  That
1267          is, the pattern adapts to the data and automatically switches to
1268          the Unicode character scheme when presented with Unicode data--or
1269          instead uses a traditional byte scheme when presented with byte
1270          data.
1271    
1272        Sadly, a wide octal escape does not cause a switch, and in a string with
1273        no other multibyte characters, these octal escapes are treated as before.
1274        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1275        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1276        Unicode string.
1277    
1278        I have not perpetrated such confusion in PCRE. Up till now, it took just
1279        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1280        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1281        translate to the appropriate multibyte character.
1282    
1283    29. Applied some refactoring to reduce the number of warnings from Microsoft
1284        and Borland compilers. This has included removing the fudge introduced
1285        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1286        a warning about an unused variable.
1287    
1288    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1289        characters since release 4.0, because Perl (from release 5.004) does not.
1290        [Or at least, is documented not to: some releases seem to be in conflict
1291        with the documentation.] However, when a pattern was studied with
1292        pcre_study() and all its branches started with \s, PCRE still included VT
1293        as a possible starting character. Of course, this did no harm; it just
1294        caused an unnecessary match attempt.
1295    
1296    22. Removed a now-redundant internal flag bit that recorded the fact that case
1297        dependency changed within the pattern. This was once needed for "required
1298        byte" processing, but is no longer used. This recovers a now-scarce options
1299        bit. Also moved the least significant internal flag bit to the most-
1300        significant bit of the word, which was not previously used (hangover from
1301        the days when it was an int rather than a uint) to free up another bit for
1302        the future.
1303    
1304    23. Added support for CRLF line endings as well as CR and LF. As well as the
1305        default being selectable at build time, it can now be changed at runtime
1306        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1307        specify that it is scanning data with non-default line endings.
1308    
1309    24. Changed the definition of CXXLINK to make it agree with the definition of
1310        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1311    
1312    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1313        recursions. This makes a big different to stack usage for some patterns.
1314    
1315    26. If a subpattern containing a named recursion or subroutine reference such
1316        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1317        the space required for the compiled pattern went wrong and gave too small a
1318        value. Depending on the environment, this could lead to "Failed: internal
1319        error: code overflow at offset 49" or "glibc detected double free or
1320        corruption" errors.
1321    
1322    27. Applied patches from Google (a) to support the new newline modes and (b) to
1323        advance over multibyte UTF-8 characters in GlobalReplace.
1324    
1325    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1326        difference for some implementation of PCRE in some Windows version.
1327    
1328    29. Added some extra testing facilities to pcretest:
1329    
1330        \q<number>   in a data line sets the "match limit" value
1331        \Q<number>   in a data line sets the "match recursion limt" value
1332        -S <number>  sets the stack size, where <number> is in megabytes
1333    
1334        The -S option isn't available for Windows.
1335    
1336    
1337    Version 6.6 06-Feb-06
1338    ---------------------
1339    
1340     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1341        in pcreposix.h. I have copied the definition from pcre.h.
1342    
1343     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1344        because pcre.h is no longer a built file.
1345    
1346     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1347        not normally included in the compiled code.
1348    
1349    
1350  Version 6.5 01-Feb-06  Version 6.5 01-Feb-06
1351  ---------------------  ---------------------
1352    

Legend:
Removed from v.87  
changed lines
  Added in v.389

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12