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

Legend:
Removed from v.87  
changed lines
  Added in v.380

  ViewVC Help
Powered by ViewVC 1.1.5