/[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 396 by ph10, Fri Mar 20 11:23:20 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    16. Added comments and documentation about mis-use of no_arg in the C++
61        wrapper.
62    
63    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
64        from Martin Jerabek that uses macro names for all relevant character and
65        string constants.
66    
67    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
68        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
69        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
70        these, but not everybody uses configure.
71    
72    19. A conditional group that had only one branch was not being correctly
73        recognized as an item that could match an empty string. This meant that an
74        enclosing group might also not be so recognized, causing infinite looping
75        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
76        with the subject "ab", where knowledge that the repeated group can match
77        nothing is needed in order to break the loop.
78    
79    
80    Version 7.8 05-Sep-08
81    ---------------------
82    
83    1.  Replaced UCP searching code with optimized version as implemented for Ad
84        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
85        stage table and inline lookup instead of a function, giving speed ups of 2
86        to 5 times on some simple patterns that I tested. Permission was given to
87        distribute the MultiStage2.py script that generates the tables (it's not in
88        the tarball, but is in the Subversion repository).
89    
90    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
91        scripts.
92    
93    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
94        a group with a zero qualifier. The result of the study could be incorrect,
95        or the function might crash, depending on the pattern.
96    
97    4.  Caseless matching was not working for non-ASCII characters in back
98        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
99        It now works when Unicode Property Support is available.
100    
101    5.  In pcretest, an escape such as \x{de} in the data was always generating
102        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
103        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
104        truncation.
105    
106    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
107    
108    7.  Added two (int) casts to pcregrep when printing the difference of two
109        pointers, in case they are 64-bit values.
110    
111    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
112        test 2 if it fails.
113    
114    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
115        and a #define of that name to empty if it is not externally set. This is to
116        allow users of MSVC to set it if necessary.
117    
118    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
119        the convenience functions in the pcre_get.c source file.
120    
121    11. An option change at the start of a pattern that had top-level alternatives
122        could cause overwriting and/or a crash. This command provoked a crash in
123        some environments:
124    
125          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
126    
127        This potential security problem was recorded as CVE-2008-2371.
128    
129    12. For a pattern where the match had to start at the beginning or immediately
130        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
131        pcre_dfa_exec() could read past the end of the passed subject if there was
132        no match. To help with detecting such bugs (e.g. with valgrind), I modified
133        pcretest so that it places the subject at the end of its malloc-ed buffer.
134    
135    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
136        exec() might read past the end of the data buffer in UTF-8 mode.
137    
138    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
139        the data contained the byte 0x85 as part of a UTF-8 character within its
140        first line. This applied both to normal and DFA matching.
141    
142    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
143        /^[^d]*?$/8 failed to match "abc".
144    
145    16. Added a missing copyright notice to pcrecpp_internal.h.
146    
147    17. Make it more clear in the documentation that values returned from
148        pcre_exec() in ovector are byte offsets, not character counts.
149    
150    18. Tidied a few places to stop certain compilers from issuing warnings.
151    
152    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
153        supplied by Stefan Weber. I made a further small update for 7.8 because
154        there is a change of source arrangements: the pcre_searchfuncs.c module is
155        replaced by pcre_ucd.c.
156    
157    
158    Version 7.7 07-May-08
159    ---------------------
160    
161    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
162        a string to a long long, pretend we don't even have a long long." This is
163        done by checking for the strtoq, strtoll, and _strtoi64 functions.
164    
165    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
166        pre-7.6 versions, which defined a global no_arg variable instead of putting
167        it in the RE class. (See also #8 below.)
168    
169    3.  Remove a line of dead code, identified by coverity and reported by Nuno
170        Lopes.
171    
172    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
173    
174        (1) The include/exclude patterns were being applied to the whole pathnames
175            of files, instead of just to the final components.
176    
177        (2) If there was more than one level of directory, the subdirectories were
178            skipped unless they satisfied the include/exclude conditions. This is
179            inconsistent with GNU grep (and could even be seen as contrary to the
180            pcregrep specification - which I improved to make it absolutely clear).
181            The action now is always to scan all levels of directory, and just
182            apply the include/exclude patterns to regular files.
183    
184    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
185        --exclude_dir in the tests to avoid scanning .svn directories.
186    
187    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
188        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
189        doesn't support NULs in patterns.
190    
191    7.  Added some missing "const"s to declarations of static tables in
192        pcre_compile.c and pcre_dfa_exec.c.
193    
194    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
195        caused by fix #2  above. (Subsequently also a second patch to fix the
196        first patch. And a third patch - this was a messy problem.)
197    
198    9.  Applied Craig's patch to remove the use of push_back().
199    
200    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
201        matching function regexec().
202    
203    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
204        which, however, unlike Perl's \g{...}, are subroutine calls, not back
205        references. PCRE supports relative numbers with this syntax (I don't think
206        Oniguruma does).
207    
208    12. Previously, a group with a zero repeat such as (...){0} was completely
209        omitted from the compiled regex. However, this means that if the group
210        was called as a subroutine from elsewhere in the pattern, things went wrong
211        (an internal error was given). Such groups are now left in the compiled
212        pattern, with a new opcode that causes them to be skipped at execution
213        time.
214    
215    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
216        to the way PCRE behaves:
217    
218        (a) A lone ] character is dis-allowed (Perl treats it as data).
219    
220        (b) A back reference to an unmatched subpattern matches an empty string
221            (Perl fails the current match path).
222    
223        (c) A data ] in a character class must be notated as \] because if the
224            first data character in a class is ], it defines an empty class. (In
225            Perl it is not possible to have an empty class.) The empty class []
226            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
227            The negative empty class [^] matches any one character, independently
228            of the DOTALL setting.
229    
230    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
231        non-existent subpattern following a character class starting with ']' and
232        containing () gave an internal compiling error instead of "reference to
233        non-existent subpattern". Fortunately, when the pattern did exist, the
234        compiled code was correct. (When scanning forwards to check for the
235        existencd of the subpattern, it was treating the data ']' as terminating
236        the class, so got the count wrong. When actually compiling, the reference
237        was subsequently set up correctly.)
238    
239    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
240        it was being rejected as not supported by pcre_dfa_exec(), even though
241        other assertions are supported. I have made pcre_dfa_exec() support
242        (*FAIL).
243    
244    16. The implementation of 13c above involved the invention of a new opcode,
245        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
246        cannot be changed at match time, I realized I could make a small
247        improvement to matching performance by compiling OP_ALLANY instead of
248        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
249        on the OP_ANY path.
250    
251    17. Compiling pcretest on Windows with readline support failed without the
252        following two fixes: (1) Make the unistd.h include conditional on
253        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
254    
255    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
256        ncurses library to be included for pcretest when ReadLine support is
257        requested, but also to allow for it to be overridden. This patch came from
258        Daniel Bergström.
259    
260    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
261        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
262        any errors with the current Unicode tables. Thanks to Peter Kankowski for
263        spotting this.
264    
265    
266    Version 7.6 28-Jan-08
267    ---------------------
268    
269    1.  A character class containing a very large number of characters with
270        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
271        overflow.
272    
273    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
274        HAVE_LONG_LONG is not defined.
275    
276    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
277        bring them up to date and include new features. This patch includes:
278    
279        - Fixed PH's badly added libz and libbz2 support.
280        - Fixed a problem with static linking.
281        - Added pcredemo. [But later removed - see 7 below.]
282        - Fixed dftables problem and added an option.
283        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
284            HAVE_LONG_LONG.
285        - Added readline support for pcretest.
286        - Added an listing of the option settings after cmake has run.
287    
288    4.  A user submitted a patch to Makefile that makes it easy to create
289        "pcre.dll" under mingw when using Configure/Make. I added stuff to
290        Makefile.am that cause it to include this special target, without
291        affecting anything else. Note that the same mingw target plus all
292        the other distribution libraries and programs are now supported
293        when configuring with CMake (see 6 below) instead of with
294        Configure/Make.
295    
296    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
297        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
298        exported in the Windows port". It has not yet been confirmed that the patch
299        solves the problem, but it does no harm.
300    
301    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
302        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
303        with CMake, and also correct the comment about stack recursion.
304    
305    7.  Remove the automatic building of pcredemo from the ./configure system and
306        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
307        of a program that users should build themselves after PCRE is installed, so
308        building it automatically is not really right. What is more, it gave
309        trouble in some build environments.
310    
311    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
312    
313    
314    Version 7.5 10-Jan-08
315    ---------------------
316    
317    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
318        values in parens when parsing an RE using the C++ wrapper."
319    
320    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
321        Characters greater than 255 were excluded from the class instead of being
322        included.
323    
324    3.  The same bug as (2) above applied to negated POSIX classes such as
325        [:^space:].
326    
327    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
328        defined or documented. It seems to have been a typo for PCRE_STATIC, so
329        I have changed it.
330    
331    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
332        first named subpattern) and a construct such as (?&a) would reference the
333        first named subpattern whose name started with "a" (in other words, the
334        length check was missing). Both these problems are fixed. "Subpattern name
335        expected" is now given for (?&) (a zero-length name), and this patch also
336        makes it give the same error for \k'' (previously it complained that that
337        was a reference to a non-existent subpattern).
338    
339    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
340        this is right because (?- can be followed by option settings as well as by
341        digits. I have, however, made the messages clearer.
342    
343    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
344        than the number used in the conditional) now cause a compile-time error.
345        This is actually not compatible with Perl, which accepts such patterns, but
346        treats the conditional as always being FALSE (as PCRE used to), but it
347        seems to me that giving a diagnostic is better.
348    
349    8.  Change "alphameric" to the more common word "alphanumeric" in comments
350        and messages.
351    
352    9.  Fix two occurrences of "backslash" in comments that should have been
353        "backspace".
354    
355    10. Remove two redundant lines of code that can never be obeyed (their function
356        was moved elsewhere).
357    
358    11. The program that makes PCRE's Unicode character property table had a bug
359        which caused it to generate incorrect table entries for sequences of
360        characters that have the same character type, but are in different scripts.
361        It amalgamated them into a single range, with the script of the first of
362        them. In other words, some characters were in the wrong script. There were
363        thirteen such cases, affecting characters in the following ranges:
364    
365          U+002b0 - U+002c1
366          U+0060c - U+0060d
367          U+0061e - U+00612
368          U+0064b - U+0065e
369          U+0074d - U+0076d
370          U+01800 - U+01805
371          U+01d00 - U+01d77
372          U+01d9b - U+01dbf
373          U+0200b - U+0200f
374          U+030fc - U+030fe
375          U+03260 - U+0327f
376          U+0fb46 - U+0fbb1
377          U+10450 - U+1049d
378    
379    12. The -o option (show only the matching part of a line) for pcregrep was not
380        compatible with GNU grep in that, if there was more than one match in a
381        line, it showed only the first of them. It now behaves in the same way as
382        GNU grep.
383    
384    13. If the -o and -v options were combined for pcregrep, it printed a blank
385        line for every non-matching line. GNU grep prints nothing, and pcregrep now
386        does the same. The return code can be used to tell if there were any
387        non-matching lines.
388    
389    14. Added --file-offsets and --line-offsets to pcregrep.
390    
391    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
392        infinitely looping recursion. The bug was that positive lookaheads were not
393        being skipped when checking for a possible empty match (negative lookaheads
394        and both kinds of lookbehind were skipped).
395    
396    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
397        inclusion of <windows.h> to before rather than after the definition of
398        INVALID_FILE_ATTRIBUTES (patch from David Byron).
399    
400    17. Specifying a possessive quantifier with a specific limit for a Unicode
401        character property caused pcre_compile() to compile bad code, which led at
402        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
403        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
404        caused the error; without that there was no problem.
405    
406    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
407    
408    19. Added --enable-pcretest-libreadline.
409    
410    20. In pcrecpp.cc, the variable 'count' was incremented twice in
411        RE::GlobalReplace(). As a result, the number of replacements returned was
412        double what it should be. I removed one of the increments, but Craig sent a
413        later patch that removed the other one (the right fix) and added unit tests
414        that check the return values (which was not done before).
415    
416    21. Several CMake things:
417    
418        (1) Arranged that, when cmake is used on Unix, the libraries end up with
419            the names libpcre and libpcreposix, not just pcre and pcreposix.
420    
421        (2) The above change means that pcretest and pcregrep are now correctly
422            linked with the newly-built libraries, not previously installed ones.
423    
424        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
425    
426    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
427        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
428        UTF-8 newline character). The key issue is that the pattern starts .*;
429        this means that the match must be either at the beginning, or after a
430        newline. The bug was in the code for advancing after a failed match and
431        checking that the new position followed a newline. It was not taking
432        account of UTF-8 characters correctly.
433    
434    23. PCRE was behaving differently from Perl in the way it recognized POSIX
435        character classes. PCRE was not treating the sequence [:...:] as a
436        character class unless the ... were all letters. Perl, however, seems to
437        allow any characters between [: and :], though of course it rejects as
438        unknown any "names" that contain non-letters, because all the known class
439        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
440        for example, whereas PCRE did not - it did not recognize a POSIX character
441        class. This seemed a bit dangerous, so the code has been changed to be
442        closer to Perl. The behaviour is not identical to Perl, because PCRE will
443        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
444        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
445        Perl does, and where it didn't before.
446    
447    24. Rewrite so as to remove the single use of %n from pcregrep because in some
448        Windows environments %n is disabled by default.
449    
450    
451    Version 7.4 21-Sep-07
452    ---------------------
453    
454    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
455        means that a class such as [\s] counted as "explicit reference to CR or
456        LF". That isn't really right - the whole point of the change was to try to
457        help when there was an actual mention of one of the two characters. So now
458        the change happens only if \r or \n (or a literal CR or LF) character is
459        encountered.
460    
461    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
462        of both had grown to the point where there were only 3 bits left.
463        Fortunately, there was spare space in the data structure, and so I have
464        moved the internal flags into a new 16-bit field to free up more option
465        bits.
466    
467    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
468        but did not set the internal JCHANGED flag - either of these is enough to
469        control the way the "get" function works - but the PCRE_INFO_JCHANGED
470        facility is supposed to tell if (?J) was ever used, so now (?J) at the
471        start sets both bits.
472    
473    4.  Added options (at build time, compile time, exec time) to change \R from
474        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
475    
476    5.  doc/pcresyntax.html was missing from the distribution.
477    
478    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
479        compatibility, even though it is no longer used.
480    
481    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
482        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
483        windows.h file is present (where different names are used). [This was
484        reversed later after testing - see 16 below.]
485    
486    8.  Changed all #include <config.h> to #include "config.h". There were also
487        some further <pcre.h> cases that I changed to "pcre.h".
488    
489    9.  When pcregrep was used with the --colour option, it missed the line ending
490        sequence off the lines that it output.
491    
492    10. It was pointed out to me that arrays of string pointers cause lots of
493        relocations when a shared library is dynamically loaded. A technique of
494        using a single long string with a table of offsets can drastically reduce
495        these. I have refactored PCRE in four places to do this. The result is
496        dramatic:
497    
498          Originally:                          290
499          After changing UCP table:            187
500          After changing error message table:   43
501          After changing table of "verbs"       36
502          After changing table of Posix names   22
503    
504        Thanks to the folks working on Gregex for glib for this insight.
505    
506    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
507        unicode-properties was also set.
508    
509    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
510    
511    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
512        checked only for CRLF.
513    
514    14. Added casts to pcretest.c to avoid compiler warnings.
515    
516    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
517    
518    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
519        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
520        entirely. This removes changes made in 7 above.
521    
522    17. The CMake files have been updated, and there is now more information about
523        building with CMake in the NON-UNIX-USE document.
524    
525    
526    Version 7.3 28-Aug-07
527    ---------------------
528    
529     1. In the rejigging of the build system that eventually resulted in 7.1, the
530        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
531        brackets there is not right, since it causes compilers to look for an
532        installed pcre.h, not the version that is in the source that is being
533        compiled (which of course may be different). I have changed it back to:
534    
535          #include "pcre.h"
536    
537        I have a vague recollection that the change was concerned with compiling in
538        different directories, but in the new build system, that is taken care of
539        by the VPATH setting the Makefile.
540    
541     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
542        when the subject happened to end in the byte 0x85 (e.g. if the last
543        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
544        characters but of course it shouldn't be taken as a newline when it is part
545        of another character. The bug was that, for an unlimited repeat of . in
546        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
547        characters when looking for a newline.
548    
549     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
550    
551     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
552        in debug output.
553    
554     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
555        long printing in the pcrecpp unittest when running under MinGW.
556    
557     6. ESC_K was left out of the EBCDIC table.
558    
559     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
560        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
561        limit also applies to "virtual nesting" when a pattern is recursive, and in
562        this case 1000 isn't so big. I have been able to remove this limit at the
563        expense of backing off one optimization in certain circumstances. Normally,
564        when pcre_exec() would call its internal match() function recursively and
565        immediately return the result unconditionally, it uses a "tail recursion"
566        feature to save stack. However, when a subpattern that can match an empty
567        string has an unlimited repetition quantifier, it no longer makes this
568        optimization. That gives it a stack frame in which to save the data for
569        checking that an empty string has been matched. Previously this was taken
570        from the 1000-entry workspace that had been reserved. So now there is no
571        explicit limit, but more stack is used.
572    
573     8. Applied Daniel's patches to solve problems with the import/export magic
574        syntax that is required for Windows, and which was going wrong for the
575        pcreposix and pcrecpp parts of the library. These were overlooked when this
576        problem was solved for the main library.
577    
578     9. There were some crude static tests to avoid integer overflow when computing
579        the size of patterns that contain repeated groups with explicit upper
580        limits. As the maximum quantifier is 65535, the maximum group length was
581        set at 30,000 so that the product of these two numbers did not overflow a
582        32-bit integer. However, it turns out that people want to use groups that
583        are longer than 30,000 bytes (though not repeat them that many times).
584        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
585        made it possible to implement the integer overflow checks in a much more
586        dynamic way, which I have now done. The artificial limitation on group
587        length has been removed - we now have only the limit on the total length of
588        the compiled pattern, which depends on the LINK_SIZE setting.
589    
590    10. Fixed a bug in the documentation for get/copy named substring when
591        duplicate names are permitted. If none of the named substrings are set, the
592        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
593        empty string.
594    
595    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
596        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
597        because the ] is interpreted as the first data character and the
598        terminating ] is not found. PCRE has been made compatible with Perl in this
599        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
600        cause memory overwriting.
601    
602    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
603        string has been matched (to stop an infinite loop). It was not recognizing
604        a conditional subpattern that could match an empty string if that
605        subpattern was within another subpattern. For example, it looped when
606        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
607        condition was not nested. This bug has been fixed.
608    
609    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
610        past the start of the subject in the presence of bytes with the top bit
611        set, for example "\x8aBCD".
612    
613    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
614        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
615    
616    14. Optimized (?!) to (*FAIL).
617    
618    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
619        This restricts code points to be within the range 0 to 0x10FFFF, excluding
620        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
621        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
622        does: it's just the validity check that is more restrictive.
623    
624    16. Inserted checks for integer overflows during escape sequence (backslash)
625        processing, and also fixed erroneous offset values for syntax errors during
626        backslash processing.
627    
628    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
629        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
630    
631    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
632        caused an overrun.
633    
634    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
635        something other than just ASCII characters) inside a group that had an
636        unlimited repeat caused a loop at compile time (while checking to see
637        whether the group could match an empty string).
638    
639    20. Debugging a pattern containing \p or \P could cause a crash. For example,
640        [\P{Any}] did so. (Error in the code for printing property names.)
641    
642    21. An orphan \E inside a character class could cause a crash.
643    
644    22. A repeated capturing bracket such as (A)? could cause a wild memory
645        reference during compilation.
646    
647    23. There are several functions in pcre_compile() that scan along a compiled
648        expression for various reasons (e.g. to see if it's fixed length for look
649        behind). There were bugs in these functions when a repeated \p or \P was
650        present in the pattern. These operators have additional parameters compared
651        with \d, etc, and these were not being taken into account when moving along
652        the compiled data. Specifically:
653    
654        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
655            length.
656    
657        (b) An item such as \pL+ within a repeated group could cause crashes or
658            loops.
659    
660        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
661            "reference to non-existent subpattern" error.
662    
663        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
664    
665    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
666        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
667    
668    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
669    
670    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
671        character were causing crashes (broken optimization).
672    
673    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
674        \p or \P) caused a compile-time loop.
675    
676    28. More problems have arisen in unanchored patterns when CRLF is a valid line
677        break. For example, the unstudied pattern [\r\n]A does not match the string
678        "\r\nA" because change 7.0/46 below moves the current point on by two
679        characters after failing to match at the start. However, the pattern \nA
680        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
681        the same is true. There doesn't seem any very clean way out of this, but
682        what I have chosen to do makes the common cases work: PCRE now takes note
683        of whether there can be an explicit match for \r or \n anywhere in the
684        pattern, and if so, 7.0/46 no longer applies. As part of this change,
685        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
686        pattern has explicit CR or LF references.
687    
688    29. Added (*CR) etc for changing newline setting at start of pattern.
689    
690    
691    Version 7.2 19-Jun-07
692    ---------------------
693    
694     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
695        which is apparently normally available under Windows.
696    
697     2. Re-jig the pcregrep tests with different newline settings in an attempt
698        to make them independent of the local environment's newline setting.
699    
700     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
701    
702     4. Some of the "internals" tests were previously cut out when the link size
703        was not 2, because the output contained actual offsets. The recent new
704        "Z" feature of pcretest means that these can be cut out, making the tests
705        usable with all link sizes.
706    
707     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
708        stack recursion. This gives a massive performance boost under BSD, but just
709        a small improvement under Linux. However, it saves one field in the frame
710        in all cases.
711    
712     6. Added more features from the forthcoming Perl 5.10:
713    
714        (a) (?-n) (where n is a string of digits) is a relative subroutine or
715            recursion call. It refers to the nth most recently opened parentheses.
716    
717        (b) (?+n) is also a relative subroutine call; it refers to the nth next
718            to be opened parentheses.
719    
720        (c) Conditions that refer to capturing parentheses can be specified
721            relatively, for example, (?(-2)... or (?(+3)...
722    
723        (d) \K resets the start of the current match so that everything before
724            is not part of it.
725    
726        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
727    
728        (f) \g{name} is another synonym - part of Perl 5.10's unification of
729            reference syntax.
730    
731        (g) (?| introduces a group in which the numbering of parentheses in each
732            alternative starts with the same number.
733    
734        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
735    
736     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
737        PCRE_INFO_JCHANGED.
738    
739     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
740        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
741        for detecting groups that can match an empty string.
742    
743     9. A pattern with a very large number of alternatives (more than several
744        hundred) was running out of internal workspace during the pre-compile
745        phase, where pcre_compile() figures out how much memory will be needed. A
746        bit of new cunning has reduced the workspace needed for groups with
747        alternatives. The 1000-alternative test pattern now uses 12 bytes of
748        workspace instead of running out of the 4096 that are available.
749    
750    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
751    
752    11. Applied patch from Google to remove an optimization that didn't quite work.
753        The report of the bug said:
754    
755          pcrecpp::RE("a*").FullMatch("aaa") matches, while
756          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
757          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
758    
759    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
760        it matched the wrong number of bytes.
761    
762    
763    Version 7.1 24-Apr-07
764    ---------------------
765    
766     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
767        that is more "standard", making use of automake and other Autotools. There
768        is some re-arrangement of the files and adjustment of comments consequent
769        on this.
770    
771     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
772        for recursive directory scanning broke on some systems because the files
773        are not scanned in any specific order and on different systems the order
774        was different. A call to "sort" has been inserted into RunGrepTest for the
775        approprate test as a short-term fix. In the longer term there may be an
776        alternative.
777    
778     3. I had an email from Eric Raymond about problems translating some of PCRE's
779        man pages to HTML (despite the fact that I distribute HTML pages, some
780        people do their own conversions for various reasons). The problems
781        concerned the use of low-level troff macros .br and .in. I have therefore
782        removed all such uses from the man pages (some were redundant, some could
783        be replaced by .nf/.fi pairs). The 132html script that I use to generate
784        HTML has been updated to handle .nf/.fi and to complain if it encounters
785        .br or .in.
786    
787     4. Updated comments in configure.ac that get placed in config.h.in and also
788        arranged for config.h to be included in the distribution, with the name
789        config.h.generic, for the benefit of those who have to compile without
790        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
791    
792     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
793        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
794        makevp.bat for the current PCRE, using the additional files
795        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
796    
797     6. A Windows user reported a minor discrepancy with test 2, which turned out
798        to be caused by a trailing space on an input line that had got lost in his
799        copy. The trailing space was an accident, so I've just removed it.
800    
801     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
802        that is needed.
803    
804     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
805        as "const" (a) because they are and (b) because it helps the PHP
806        maintainers who have recently made a script to detect big data structures
807        in the php code that should be moved to the .rodata section. I remembered
808        to update Builducptable as well, so it won't revert if ucptable.h is ever
809        re-created.
810    
811     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
812        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
813        order to be able to cut out the UTF-8 tables in the latter when UTF-8
814        support is not required. This saves 1.5-2K of code, which is important in
815        some applications.
816    
817        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
818        so as not to refer to the tables, even though these functions will never be
819        called when UTF-8 support is disabled. Otherwise there are problems with a
820        shared library.
821    
822    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
823    
824        (a) It was defining its arguments as char * instead of void *.
825    
826        (b) It was assuming that all moves were upwards in memory; this was true
827            a long time ago when I wrote it, but is no longer the case.
828    
829        The emulated memove() is provided for those environments that have neither
830        memmove() nor bcopy(). I didn't think anyone used it these days, but that
831        is clearly not the case, as these two bugs were recently reported.
832    
833    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
834        and Detrail to create the HTML documentation, the .txt form of the man
835        pages, and it removes trailing spaces from listed files. It also creates
836        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
837        case, it wraps all the #defines with #ifndefs. This script should be run
838        before "make dist".
839    
840    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
841        with Unicode property support.
842    
843        (a) For a maximizing quantifier, if the two different cases of the
844            character were of different lengths in their UTF-8 codings (there are
845            some cases like this - I found 11), and the matching function had to
846            back up over a mixture of the two cases, it incorrectly assumed they
847            were both the same length.
848    
849        (b) When PCRE was configured to use the heap rather than the stack for
850            recursion during matching, it was not correctly preserving the data for
851            the other case of a UTF-8 character when checking ahead for a match
852            while processing a minimizing repeat. If the check also involved
853            matching a wide character, but failed, corruption could cause an
854            erroneous result when trying to check for a repeat of the original
855            character.
856    
857    13. Some tidying changes to the testing mechanism:
858    
859        (a) The RunTest script now detects the internal link size and whether there
860            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
861            values substituted by "configure". (The RunGrepTest script already did
862            this for UTF-8.) The configure.ac script no longer substitutes the
863            relevant variables.
864    
865        (b) The debugging options /B and /D in pcretest show the compiled bytecode
866            with length and offset values. This means that the output is different
867            for different internal link sizes. Test 2 is skipped for link sizes
868            other than 2 because of this, bypassing the problem. Unfortunately,
869            there was also a test in test 3 (the locale tests) that used /B and
870            failed for link sizes other than 2. Rather than cut the whole test out,
871            I have added a new /Z option to pcretest that replaces the length and
872            offset values with spaces. This is now used to make test 3 independent
873            of link size. (Test 2 will be tidied up later.)
874    
875    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
876        segmentation fault instead of returning the appropriate error message.
877    
878    15. In multiline mode when the newline sequence was set to "any", the pattern
879        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
880        This doesn't seem right; it now treats the CRLF combination as the line
881        ending, and so does not match in that case. It's only a pattern such as ^$
882        that would hit this one: something like ^ABC$ would have failed after \r
883        and then tried again after \r\n.
884    
885    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
886        in an attempt to make files that differ only in their line terminators
887        compare equal. This works on Linux.
888    
889    17. Under certain error circumstances pcregrep might try to free random memory
890        as it exited. This is now fixed, thanks to valgrind.
891    
892    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
893        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
894        was because its rules for how to advance for /g after matching an empty
895        string at the end of a line did not allow for this case. They now check for
896        it specially.
897    
898    20. pcretest is supposed to handle patterns and data of any length, by
899        extending its buffers when necessary. It was getting this wrong when the
900        buffer for a data line had to be extended.
901    
902    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
903        CRLF as a newline sequence.
904    
905    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
906        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
907        I have nevertheless tidied it up.
908    
909    23. Added some casts to kill warnings from HP-UX ia64 compiler.
910    
911    24. Added a man page for pcre-config.
912    
913    
914    Version 7.0 19-Dec-06
915    ---------------------
916    
917     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
918        moving to gcc 4.1.1.
919    
920     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
921        sys/time.h, which is documented as needed for this function. It doesn't
922        seem to matter on Linux, but it showed up on some releases of OS X.
923    
924     3. It seems that there are systems where bytes whose values are greater than
925        127 match isprint() in the "C" locale. The "C" locale should be the
926        default when a C program starts up. In most systems, only ASCII printing
927        characters match isprint(). This difference caused the output from pcretest
928        to vary, making some of the tests fail. I have changed pcretest so that:
929    
930        (a) When it is outputting text in the compiled version of a pattern, bytes
931            other than 32-126 are always shown as hex escapes.
932    
933        (b) When it is outputting text that is a matched part of a subject string,
934            it does the same, unless a different locale has been set for the match
935            (using the /L modifier). In this case, it uses isprint() to decide.
936    
937     4. Fixed a major bug that caused incorrect computation of the amount of memory
938        required for a compiled pattern when options that changed within the
939        pattern affected the logic of the preliminary scan that determines the
940        length. The relevant options are -x, and -i in UTF-8 mode. The result was
941        that the computed length was too small. The symptoms of this bug were
942        either the PCRE error "internal error: code overflow" from pcre_compile(),
943        or a glibc crash with a message such as "pcretest: free(): invalid next
944        size (fast)". Examples of patterns that provoked this bug (shown in
945        pcretest format) are:
946    
947          /(?-x: )/x
948          /(?x)(?-x: \s*#\s*)/
949          /((?i)[\x{c0}])/8
950          /(?i:[\x{c0}])/8
951    
952        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
953        is now done differently.
954    
955     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
956        wrapper classes; (b) implement a new function in the C++ scanner that is
957        more efficient than the old way of doing things because it avoids levels of
958        recursion in the regex matching; (c) add a paragraph to the documentation
959        for the FullMatch() function.
960    
961     6. The escape sequence \n was being treated as whatever was defined as
962        "newline". Not only was this contrary to the documentation, which states
963        that \n is character 10 (hex 0A), but it also went horribly wrong when
964        "newline" was defined as CRLF. This has been fixed.
965    
966     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
967        was being set to -1 for the "end of line" case (supposedly a value that no
968        character can have). Though this value is never used (the check for end of
969        line is "zero bytes in current character"), it caused compiler complaints.
970        I've changed it to 0xffffffff.
971    
972     8. In pcre_version.c, the version string was being built by a sequence of
973        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
974        string (as it is for production releases) called a macro with an empty
975        argument. The C standard says the result of this is undefined. The gcc
976        compiler treats it as an empty string (which was what was wanted) but it is
977        reported that Visual C gives an error. The source has been hacked around to
978        avoid this problem.
979    
980     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
981        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
982        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
983        of them did).
984    
985    10. Originally, pcretest opened its input and output without "b"; then I was
986        told that "b" was needed in some environments, so it was added for release
987        5.0 to both the input and output. (It makes no difference on Unix-like
988        systems.) Later I was told that it is wrong for the input on Windows. I've
989        now abstracted the modes into two macros, to make it easier to fiddle with
990        them, and removed "b" from the input mode under Windows.
991    
992    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
993    
994    12. Added -help and --help to pcretest as an official way of being reminded
995        of the options.
996    
997    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
998        and pcrecpp.cc because they annoy compilers at high warning levels.
999    
1000    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1001    
1002    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1003        scripts are not C programs :-) and which was not noticed because it works
1004        on Linux.
1005    
1006    16. pcretest is supposed to handle any length of pattern and data line (as one
1007        line or as a continued sequence of lines) by extending its input buffer if
1008        necessary. This feature was broken for very long pattern lines, leading to
1009        a string of junk being passed to pcre_compile() if the pattern was longer
1010        than about 50K.
1011    
1012    17. I have done a major re-factoring of the way pcre_compile() computes the
1013        amount of memory needed for a compiled pattern. Previously, there was code
1014        that made a preliminary scan of the pattern in order to do this. That was
1015        OK when PCRE was new, but as the facilities have expanded, it has become
1016        harder and harder to keep it in step with the real compile phase, and there
1017        have been a number of bugs (see for example, 4 above). I have now found a
1018        cunning way of running the real compile function in a "fake" mode that
1019        enables it to compute how much memory it would need, while actually only
1020        ever using a few hundred bytes of working memory and without too many
1021        tests of the mode. This should make future maintenance and development
1022        easier. A side effect of this work is that the limit of 200 on the nesting
1023        depth of parentheses has been removed (though this was never a serious
1024        limitation, I suspect). However, there is a downside: pcre_compile() now
1025        runs more slowly than before (30% or more, depending on the pattern). I
1026        hope this isn't a big issue. There is no effect on runtime performance.
1027    
1028    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1029        newline (only possible for the last line of a file) and it was a
1030        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1031    
1032    19. Added additional timing features to pcretest. (1) The -tm option now times
1033        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1034        separate command line item, by a number that specifies the number of
1035        repeats to use when timing. The default is 50000; this gives better
1036        precision, but takes uncomfortably long for very large patterns.
1037    
1038    20. Extended pcre_study() to be more clever in cases where a branch of a
1039        subpattern has no definite first character. For example, (a*|b*)[cd] would
1040        previously give no result from pcre_study(). Now it recognizes that the
1041        first character must be a, b, c, or d.
1042    
1043    21. There was an incorrect error "recursive call could loop indefinitely" if
1044        a subpattern (or the entire pattern) that was being tested for matching an
1045        empty string contained only one non-empty item after a nested subpattern.
1046        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1047        incorrectly, because the \d was being skipped in the check.
1048    
1049    22. The pcretest program now has a new pattern option /B and a command line
1050        option -b, which is equivalent to adding /B to every pattern. This causes
1051        it to show the compiled bytecode, without the additional information that
1052        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1053        is the same as /B/I).
1054    
1055    23. A new optimization is now able automatically to treat some sequences such
1056        as a*b as a*+b. More specifically, if something simple (such as a character
1057        or a simple class like \d) has an unlimited quantifier, and is followed by
1058        something that cannot possibly match the quantified thing, the quantifier
1059        is automatically "possessified".
1060    
1061    24. A recursive reference to a subpattern whose number was greater than 39
1062        went wrong under certain circumstances in UTF-8 mode. This bug could also
1063        have affected the operation of pcre_study().
1064    
1065    25. Realized that a little bit of performance could be had by replacing
1066        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1067    
1068    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1069    
1070    27. Possessive quantifiers such as a++ were previously implemented by turning
1071        them into atomic groups such as ($>a+). Now they have their own opcodes,
1072        which improves performance. This includes the automatically created ones
1073        from 23 above.
1074    
1075    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1076        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1077        the first matched character to be a colon. This applied both to named and
1078        numbered groups.
1079    
1080    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1081    
1082    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1083        building PCRE on Windows easier, so I have included it in the distribution.
1084    
1085    31. There is now a check in pcretest against a ridiculously large number being
1086        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1087        loop, the loop is abandoned.
1088    
1089    32. Forward references to subpatterns in conditions such as (?(2)...) where
1090        subpattern 2 is defined later cause pcre_compile() to search forwards in
1091        the pattern for the relevant set of parentheses. This search went wrong
1092        when there were unescaped parentheses in a character class, parentheses
1093        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1094    
1095    33. "Subroutine" calls and backreferences were previously restricted to
1096        referencing subpatterns earlier in the regex. This restriction has now
1097        been removed.
1098    
1099    34. Added a number of extra features that are going to be in Perl 5.10. On the
1100        whole, these are just syntactic alternatives for features that PCRE had
1101        previously implemented using the Python syntax or my own invention. The
1102        other formats are all retained for compatibility.
1103    
1104        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1105            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1106            also .NET compatible.
1107    
1108        (b) A recursion or subroutine call to a named group can now be defined as
1109            (?&name) as well as (?P>name).
1110    
1111        (c) A backreference to a named group can now be defined as \k<name> or
1112            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1113            5.10, are also .NET compatible.
1114    
1115        (d) A conditional reference to a named group can now use the syntax
1116            (?(<name>) or (?('name') as well as (?(name).
1117    
1118        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1119            groups (named and numbered) that are never evaluated inline, but can be
1120            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1121            is always false. There may be only one alternative in such a group.
1122    
1123        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1124            as the simple (?(R). The condition is true only if the most recent
1125            recursion is that of the given number or name. It does not search out
1126            through the entire recursion stack.
1127    
1128        (g) The escape \gN or \g{N} has been added, where N is a positive or
1129            negative number, specifying an absolute or relative reference.
1130    
1131    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1132        some "unreachable code" warnings.
1133    
1134    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1135        things, this adds five new scripts.
1136    
1137    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1138        There were also incompatibilities regarding the handling of \Q..\E inside
1139        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1140        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1141    
1142    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1143        matches an empty string, and forcibly breaks the loop. There were bugs in
1144        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1145        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1146        separate and independent bugs (that affected different cases) have been
1147        fixed.
1148    
1149    39. Refactored the code to abolish the use of different opcodes for small
1150        capturing bracket numbers. This is a tidy that I avoided doing when I
1151        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1152        The new approach is not only tidier, it makes it possible to reduce the
1153        memory needed to fix the previous bug (38).
1154    
1155    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1156        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1157        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1158        mode.
1159    
1160    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1161        report.
1162    
1163    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1164        copy construction and assignment in the C++ wrapper.
1165    
1166    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1167        couple of bugs that could have given wrong results in the "--newline=crlf"
1168        case.
1169    
1170    44. Added a number of casts and did some reorganization of signed/unsigned int
1171        variables following suggestions from Dair Grant. Also renamed the variable
1172        "this" as "item" because it is a C++ keyword.
1173    
1174    45. Arranged for dftables to add
1175    
1176          #include "pcre_internal.h"
1177    
1178        to pcre_chartables.c because without it, gcc 4.x may remove the array
1179        definition from the final binary if PCRE is built into a static library and
1180        dead code stripping is activated.
1181    
1182    46. For an unanchored pattern, if a match attempt fails at the start of a
1183        newline sequence, and the newline setting is CRLF or ANY, and the next two
1184        characters are CRLF, advance by two characters instead of one.
1185    
1186    
1187    Version 6.7 04-Jul-06
1188    ---------------------
1189    
1190     1. In order to handle tests when input lines are enormously long, pcretest has
1191        been re-factored so that it automatically extends its buffers when
1192        necessary. The code is crude, but this _is_ just a test program. The
1193        default size has been increased from 32K to 50K.
1194    
1195     2. The code in pcre_study() was using the value of the re argument before
1196        testing it for NULL. (Of course, in any sensible call of the function, it
1197        won't be NULL.)
1198    
1199     3. The memmove() emulation function in pcre_internal.h, which is used on
1200        systems that lack both memmove() and bcopy() - that is, hardly ever -
1201        was missing a "static" storage class specifier.
1202    
1203     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1204        containing an extended class (one that cannot be represented by a bitmap
1205        because it contains high-valued characters or Unicode property items, e.g.
1206        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1207        pattern, but PCRE should not loop if you do not (it no longer does).
1208        [Detail: two cases were found: (a) a repeated subpattern containing an
1209        extended class; (b) a recursive reference to a subpattern that followed a
1210        previous extended class. It wasn't skipping over the extended class
1211        correctly when UTF-8 mode was not set.]
1212    
1213     5. A negated single-character class was not being recognized as fixed-length
1214        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1215        compile error "lookbehind assertion is not fixed length".
1216    
1217     6. The RunPerlTest auxiliary script was showing an unexpected difference
1218        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1219        write a Perl script that can interpret lines of an input file either as
1220        byte characters or as UTF-8, which is what "perltest" was being required to
1221        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1222        can't do is switch easily at run time between having the "use utf8;" pragma
1223        or not. In the end, I fudged it by using the RunPerlTest script to insert
1224        "use utf8;" explicitly for the UTF-8 tests.
1225    
1226     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1227        the end of the subject string, contrary to the documentation and to what
1228        Perl does. This was true of both matching functions. Now it matches only at
1229        the start of the subject and immediately after *internal* newlines.
1230    
1231     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1232        a pointer to an int instead of a pointer to an unsigned long int. This
1233        caused problems on 64-bit systems.
1234    
1235     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1236        instance of the 'standard' template library not being so standard".
1237    
1238    10. There was no check on the number of named subpatterns nor the maximum
1239        length of a subpattern name. The product of these values is used to compute
1240        the size of the memory block for a compiled pattern. By supplying a very
1241        long subpattern name and a large number of named subpatterns, the size
1242        computation could be caused to overflow. This is now prevented by limiting
1243        the length of names to 32 characters, and the number of named subpatterns
1244        to 10,000.
1245    
1246    11. Subpatterns that are repeated with specific counts have to be replicated in
1247        the compiled pattern. The size of memory for this was computed from the
1248        length of the subpattern and the repeat count. The latter is limited to
1249        65535, but there was no limit on the former, meaning that integer overflow
1250        could in principle occur. The compiled length of a repeated subpattern is
1251        now limited to 30,000 bytes in order to prevent this.
1252    
1253    12. Added the optional facility to have named substrings with the same name.
1254    
1255    13. Added the ability to use a named substring as a condition, using the
1256        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1257        are numbers (not recommended). Forward references are permitted.
1258    
1259    14. Added forward references in named backreferences (if you see what I mean).
1260    
1261    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1262        pattern could run off the end of the subject. For example, the pattern
1263        "(?s)(.{1,5})"8 did this with the subject "ab".
1264    
1265    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1266        PCRE_CASELESS was set when matching characters that were quantified with ?
1267        or *.
1268    
1269    17. A character class other than a single negated character that had a minimum
1270        but no maximum quantifier - for example [ab]{6,} - was not handled
1271        correctly by pce_dfa_exec(). It would match only one character.
1272    
1273    18. A valid (though odd) pattern that looked like a POSIX character
1274        class but used an invalid character after [ (for example [[,abc,]]) caused
1275        pcre_compile() to give the error "Failed: internal error: code overflow" or
1276        in some cases to crash with a glibc free() error. This could even happen if
1277        the pattern terminated after [[ but there just happened to be a sequence of
1278        letters, a binary zero, and a closing ] in the memory that followed.
1279    
1280    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1281        over the years. Originally (before any Unicode support), just the bottom 8
1282        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1283        output from "man perlunicode" includes this:
1284    
1285          The regular expression compiler produces polymorphic opcodes.  That
1286          is, the pattern adapts to the data and automatically switches to
1287          the Unicode character scheme when presented with Unicode data--or
1288          instead uses a traditional byte scheme when presented with byte
1289          data.
1290    
1291        Sadly, a wide octal escape does not cause a switch, and in a string with
1292        no other multibyte characters, these octal escapes are treated as before.
1293        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1294        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1295        Unicode string.
1296    
1297        I have not perpetrated such confusion in PCRE. Up till now, it took just
1298        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1299        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1300        translate to the appropriate multibyte character.
1301    
1302    29. Applied some refactoring to reduce the number of warnings from Microsoft
1303        and Borland compilers. This has included removing the fudge introduced
1304        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1305        a warning about an unused variable.
1306    
1307    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1308        characters since release 4.0, because Perl (from release 5.004) does not.
1309        [Or at least, is documented not to: some releases seem to be in conflict
1310        with the documentation.] However, when a pattern was studied with
1311        pcre_study() and all its branches started with \s, PCRE still included VT
1312        as a possible starting character. Of course, this did no harm; it just
1313        caused an unnecessary match attempt.
1314    
1315    22. Removed a now-redundant internal flag bit that recorded the fact that case
1316        dependency changed within the pattern. This was once needed for "required
1317        byte" processing, but is no longer used. This recovers a now-scarce options
1318        bit. Also moved the least significant internal flag bit to the most-
1319        significant bit of the word, which was not previously used (hangover from
1320        the days when it was an int rather than a uint) to free up another bit for
1321        the future.
1322    
1323    23. Added support for CRLF line endings as well as CR and LF. As well as the
1324        default being selectable at build time, it can now be changed at runtime
1325        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1326        specify that it is scanning data with non-default line endings.
1327    
1328    24. Changed the definition of CXXLINK to make it agree with the definition of
1329        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1330    
1331    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1332        recursions. This makes a big different to stack usage for some patterns.
1333    
1334    26. If a subpattern containing a named recursion or subroutine reference such
1335        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1336        the space required for the compiled pattern went wrong and gave too small a
1337        value. Depending on the environment, this could lead to "Failed: internal
1338        error: code overflow at offset 49" or "glibc detected double free or
1339        corruption" errors.
1340    
1341    27. Applied patches from Google (a) to support the new newline modes and (b) to
1342        advance over multibyte UTF-8 characters in GlobalReplace.
1343    
1344    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1345        difference for some implementation of PCRE in some Windows version.
1346    
1347    29. Added some extra testing facilities to pcretest:
1348    
1349        \q<number>   in a data line sets the "match limit" value
1350        \Q<number>   in a data line sets the "match recursion limt" value
1351        -S <number>  sets the stack size, where <number> is in megabytes
1352    
1353        The -S option isn't available for Windows.
1354    
1355    
1356    Version 6.6 06-Feb-06
1357    ---------------------
1358    
1359     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1360        in pcreposix.h. I have copied the definition from pcre.h.
1361    
1362     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1363        because pcre.h is no longer a built file.
1364    
1365     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1366        not normally included in the compiled code.
1367    
1368    
1369  Version 6.5 01-Feb-06  Version 6.5 01-Feb-06
1370  ---------------------  ---------------------
1371    

Legend:
Removed from v.87  
changed lines
  Added in v.396

  ViewVC Help
Powered by ViewVC 1.1.5