/[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 355 by ph10, Mon Jul 7 17:45:23 2008 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 8.0 02 Jul-08
5    ---------------------
6    
7    1.  Replaced UCP searching code with optimized version as implemented for Ad
8        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
9        stage table and inline lookup instead of a function, giving speed ups of 2
10        to 5 times on some simple patterns that I tested. Permission was given to
11        distribute the MultiStage2.py script that generates the tables (it's not in
12        the tarball, but is in the Subversion repository).
13    
14    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
15        scripts.
16    
17    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
18        a group with a zero qualifier. The result of the study could be incorrect,
19        or the function might crash, depending on the pattern.
20    
21    4.  Caseless matching was not working for non-ASCII characters in back
22        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
23        It now works when Unicode Property Support is available.
24    
25    5.  In pcretest, an escape such as \x{de} in the data was always generating
26        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
27        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
28        truncation.
29    
30    
31    Version 7.7 07-May-08
32    ---------------------
33    
34    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
35        a string to a long long, pretend we don't even have a long long." This is
36        done by checking for the strtoq, strtoll, and _strtoi64 functions.
37    
38    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
39        pre-7.6 versions, which defined a global no_arg variable instead of putting
40        it in the RE class. (See also #8 below.)
41    
42    3.  Remove a line of dead code, identified by coverity and reported by Nuno
43        Lopes.
44    
45    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
46    
47        (1) The include/exclude patterns were being applied to the whole pathnames
48            of files, instead of just to the final components.
49    
50        (2) If there was more than one level of directory, the subdirectories were
51            skipped unless they satisfied the include/exclude conditions. This is
52            inconsistent with GNU grep (and could even be seen as contrary to the
53            pcregrep specification - which I improved to make it absolutely clear).
54            The action now is always to scan all levels of directory, and just
55            apply the include/exclude patterns to regular files.
56    
57    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
58        --exclude_dir in the tests to avoid scanning .svn directories.
59    
60    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
61        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
62        doesn't support NULs in patterns.
63    
64    7.  Added some missing "const"s to declarations of static tables in
65        pcre_compile.c and pcre_dfa_exec.c.
66    
67    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
68        caused by fix #2  above. (Subsequently also a second patch to fix the
69        first patch. And a third patch - this was a messy problem.)
70    
71    9.  Applied Craig's patch to remove the use of push_back().
72    
73    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
74        matching function regexec().
75    
76    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
77        which, however, unlike Perl's \g{...}, are subroutine calls, not back
78        references. PCRE supports relative numbers with this syntax (I don't think
79        Oniguruma does).
80    
81    12. Previously, a group with a zero repeat such as (...){0} was completely
82        omitted from the compiled regex. However, this means that if the group
83        was called as a subroutine from elsewhere in the pattern, things went wrong
84        (an internal error was given). Such groups are now left in the compiled
85        pattern, with a new opcode that causes them to be skipped at execution
86        time.
87    
88    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
89        to the way PCRE behaves:
90    
91        (a) A lone ] character is dis-allowed (Perl treats it as data).
92    
93        (b) A back reference to an unmatched subpattern matches an empty string
94            (Perl fails the current match path).
95    
96        (c) A data ] in a character class must be notated as \] because if the
97            first data character in a class is ], it defines an empty class. (In
98            Perl it is not possible to have an empty class.) The empty class []
99            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
100            The negative empty class [^] matches any one character, independently
101            of the DOTALL setting.
102    
103    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
104        non-existent subpattern following a character class starting with ']' and
105        containing () gave an internal compiling error instead of "reference to
106        non-existent subpattern". Fortunately, when the pattern did exist, the
107        compiled code was correct. (When scanning forwards to check for the
108        existencd of the subpattern, it was treating the data ']' as terminating
109        the class, so got the count wrong. When actually compiling, the reference
110        was subsequently set up correctly.)
111    
112    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
113        it was being rejected as not supported by pcre_dfa_exec(), even though
114        other assertions are supported. I have made pcre_dfa_exec() support
115        (*FAIL).
116    
117    16. The implementation of 13c above involved the invention of a new opcode,
118        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
119        cannot be changed at match time, I realized I could make a small
120        improvement to matching performance by compiling OP_ALLANY instead of
121        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
122        on the OP_ANY path.
123    
124    17. Compiling pcretest on Windows with readline support failed without the
125        following two fixes: (1) Make the unistd.h include conditional on
126        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
127    
128    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
129        ncurses library to be included for pcretest when ReadLine support is
130        requested, but also to allow for it to be overridden. This patch came from
131        Daniel Bergström.
132    
133    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
134        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
135        any errors with the current Unicode tables. Thanks to Peter Kankowski for
136        spotting this.
137    
138    
139    Version 7.6 28-Jan-08
140    ---------------------
141    
142    1.  A character class containing a very large number of characters with
143        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
144        overflow.
145    
146    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
147        HAVE_LONG_LONG is not defined.
148    
149    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
150        bring them up to date and include new features. This patch includes:
151    
152        - Fixed PH's badly added libz and libbz2 support.
153        - Fixed a problem with static linking.
154        - Added pcredemo. [But later removed - see 7 below.]
155        - Fixed dftables problem and added an option.
156        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
157            HAVE_LONG_LONG.
158        - Added readline support for pcretest.
159        - Added an listing of the option settings after cmake has run.
160    
161    4.  A user submitted a patch to Makefile that makes it easy to create
162        "pcre.dll" under mingw when using Configure/Make. I added stuff to
163        Makefile.am that cause it to include this special target, without
164        affecting anything else. Note that the same mingw target plus all
165        the other distribution libraries and programs are now supported
166        when configuring with CMake (see 6 below) instead of with
167        Configure/Make.
168    
169    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
170        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
171        exported in the Windows port". It has not yet been confirmed that the patch
172        solves the problem, but it does no harm.
173    
174    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
175        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
176        with CMake, and also correct the comment about stack recursion.
177    
178    7.  Remove the automatic building of pcredemo from the ./configure system and
179        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
180        of a program that users should build themselves after PCRE is installed, so
181        building it automatically is not really right. What is more, it gave
182        trouble in some build environments.
183    
184    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
185    
186    
187    Version 7.5 10-Jan-08
188    ---------------------
189    
190    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
191        values in parens when parsing an RE using the C++ wrapper."
192    
193    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
194        Characters greater than 255 were excluded from the class instead of being
195        included.
196    
197    3.  The same bug as (2) above applied to negated POSIX classes such as
198        [:^space:].
199    
200    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
201        defined or documented. It seems to have been a typo for PCRE_STATIC, so
202        I have changed it.
203    
204    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
205        first named subpattern) and a construct such as (?&a) would reference the
206        first named subpattern whose name started with "a" (in other words, the
207        length check was missing). Both these problems are fixed. "Subpattern name
208        expected" is now given for (?&) (a zero-length name), and this patch also
209        makes it give the same error for \k'' (previously it complained that that
210        was a reference to a non-existent subpattern).
211    
212    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
213        this is right because (?- can be followed by option settings as well as by
214        digits. I have, however, made the messages clearer.
215    
216    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
217        than the number used in the conditional) now cause a compile-time error.
218        This is actually not compatible with Perl, which accepts such patterns, but
219        treats the conditional as always being FALSE (as PCRE used to), but it
220        seems to me that giving a diagnostic is better.
221    
222    8.  Change "alphameric" to the more common word "alphanumeric" in comments
223        and messages.
224    
225    9.  Fix two occurrences of "backslash" in comments that should have been
226        "backspace".
227    
228    10. Remove two redundant lines of code that can never be obeyed (their function
229        was moved elsewhere).
230    
231    11. The program that makes PCRE's Unicode character property table had a bug
232        which caused it to generate incorrect table entries for sequences of
233        characters that have the same character type, but are in different scripts.
234        It amalgamated them into a single range, with the script of the first of
235        them. In other words, some characters were in the wrong script. There were
236        thirteen such cases, affecting characters in the following ranges:
237    
238          U+002b0 - U+002c1
239          U+0060c - U+0060d
240          U+0061e - U+00612
241          U+0064b - U+0065e
242          U+0074d - U+0076d
243          U+01800 - U+01805
244          U+01d00 - U+01d77
245          U+01d9b - U+01dbf
246          U+0200b - U+0200f
247          U+030fc - U+030fe
248          U+03260 - U+0327f
249          U+0fb46 - U+0fbb1
250          U+10450 - U+1049d
251    
252    12. The -o option (show only the matching part of a line) for pcregrep was not
253        compatible with GNU grep in that, if there was more than one match in a
254        line, it showed only the first of them. It now behaves in the same way as
255        GNU grep.
256    
257    13. If the -o and -v options were combined for pcregrep, it printed a blank
258        line for every non-matching line. GNU grep prints nothing, and pcregrep now
259        does the same. The return code can be used to tell if there were any
260        non-matching lines.
261    
262    14. Added --file-offsets and --line-offsets to pcregrep.
263    
264    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
265        infinitely looping recursion. The bug was that positive lookaheads were not
266        being skipped when checking for a possible empty match (negative lookaheads
267        and both kinds of lookbehind were skipped).
268    
269    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
270        inclusion of <windows.h> to before rather than after the definition of
271        INVALID_FILE_ATTRIBUTES (patch from David Byron).
272    
273    17. Specifying a possessive quantifier with a specific limit for a Unicode
274        character property caused pcre_compile() to compile bad code, which led at
275        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
276        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
277        caused the error; without that there was no problem.
278    
279    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
280    
281    19. Added --enable-pcretest-libreadline.
282    
283    20. In pcrecpp.cc, the variable 'count' was incremented twice in
284        RE::GlobalReplace(). As a result, the number of replacements returned was
285        double what it should be. I removed one of the increments, but Craig sent a
286        later patch that removed the other one (the right fix) and added unit tests
287        that check the return values (which was not done before).
288    
289    21. Several CMake things:
290    
291        (1) Arranged that, when cmake is used on Unix, the libraries end up with
292            the names libpcre and libpcreposix, not just pcre and pcreposix.
293    
294        (2) The above change means that pcretest and pcregrep are now correctly
295            linked with the newly-built libraries, not previously installed ones.
296    
297        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
298    
299    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
300        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
301        UTF-8 newline character). The key issue is that the pattern starts .*;
302        this means that the match must be either at the beginning, or after a
303        newline. The bug was in the code for advancing after a failed match and
304        checking that the new position followed a newline. It was not taking
305        account of UTF-8 characters correctly.
306    
307    23. PCRE was behaving differently from Perl in the way it recognized POSIX
308        character classes. PCRE was not treating the sequence [:...:] as a
309        character class unless the ... were all letters. Perl, however, seems to
310        allow any characters between [: and :], though of course it rejects as
311        unknown any "names" that contain non-letters, because all the known class
312        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
313        for example, whereas PCRE did not - it did not recognize a POSIX character
314        class. This seemed a bit dangerous, so the code has been changed to be
315        closer to Perl. The behaviour is not identical to Perl, because PCRE will
316        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
317        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
318        Perl does, and where it didn't before.
319    
320    24. Rewrite so as to remove the single use of %n from pcregrep because in some
321        Windows environments %n is disabled by default.
322    
323    
324    Version 7.4 21-Sep-07
325    ---------------------
326    
327    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
328        means that a class such as [\s] counted as "explicit reference to CR or
329        LF". That isn't really right - the whole point of the change was to try to
330        help when there was an actual mention of one of the two characters. So now
331        the change happens only if \r or \n (or a literal CR or LF) character is
332        encountered.
333    
334    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
335        of both had grown to the point where there were only 3 bits left.
336        Fortunately, there was spare space in the data structure, and so I have
337        moved the internal flags into a new 16-bit field to free up more option
338        bits.
339    
340    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
341        but did not set the internal JCHANGED flag - either of these is enough to
342        control the way the "get" function works - but the PCRE_INFO_JCHANGED
343        facility is supposed to tell if (?J) was ever used, so now (?J) at the
344        start sets both bits.
345    
346    4.  Added options (at build time, compile time, exec time) to change \R from
347        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
348    
349    5.  doc/pcresyntax.html was missing from the distribution.
350    
351    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
352        compatibility, even though it is no longer used.
353    
354    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
355        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
356        windows.h file is present (where different names are used). [This was
357        reversed later after testing - see 16 below.]
358    
359    8.  Changed all #include <config.h> to #include "config.h". There were also
360        some further <pcre.h> cases that I changed to "pcre.h".
361    
362    9.  When pcregrep was used with the --colour option, it missed the line ending
363        sequence off the lines that it output.
364    
365    10. It was pointed out to me that arrays of string pointers cause lots of
366        relocations when a shared library is dynamically loaded. A technique of
367        using a single long string with a table of offsets can drastically reduce
368        these. I have refactored PCRE in four places to do this. The result is
369        dramatic:
370    
371          Originally:                          290
372          After changing UCP table:            187
373          After changing error message table:   43
374          After changing table of "verbs"       36
375          After changing table of Posix names   22
376    
377        Thanks to the folks working on Gregex for glib for this insight.
378    
379    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
380        unicode-properties was also set.
381    
382    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
383    
384    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
385        checked only for CRLF.
386    
387    14. Added casts to pcretest.c to avoid compiler warnings.
388    
389    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
390    
391    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
392        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
393        entirely. This removes changes made in 7 above.
394    
395    17. The CMake files have been updated, and there is now more information about
396        building with CMake in the NON-UNIX-USE document.
397    
398    
399    Version 7.3 28-Aug-07
400    ---------------------
401    
402     1. In the rejigging of the build system that eventually resulted in 7.1, the
403        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
404        brackets there is not right, since it causes compilers to look for an
405        installed pcre.h, not the version that is in the source that is being
406        compiled (which of course may be different). I have changed it back to:
407    
408          #include "pcre.h"
409    
410        I have a vague recollection that the change was concerned with compiling in
411        different directories, but in the new build system, that is taken care of
412        by the VPATH setting the Makefile.
413    
414     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
415        when the subject happened to end in the byte 0x85 (e.g. if the last
416        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
417        characters but of course it shouldn't be taken as a newline when it is part
418        of another character. The bug was that, for an unlimited repeat of . in
419        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
420        characters when looking for a newline.
421    
422     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
423    
424     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
425        in debug output.
426    
427     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
428        long printing in the pcrecpp unittest when running under MinGW.
429    
430     6. ESC_K was left out of the EBCDIC table.
431    
432     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
433        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
434        limit also applies to "virtual nesting" when a pattern is recursive, and in
435        this case 1000 isn't so big. I have been able to remove this limit at the
436        expense of backing off one optimization in certain circumstances. Normally,
437        when pcre_exec() would call its internal match() function recursively and
438        immediately return the result unconditionally, it uses a "tail recursion"
439        feature to save stack. However, when a subpattern that can match an empty
440        string has an unlimited repetition quantifier, it no longer makes this
441        optimization. That gives it a stack frame in which to save the data for
442        checking that an empty string has been matched. Previously this was taken
443        from the 1000-entry workspace that had been reserved. So now there is no
444        explicit limit, but more stack is used.
445    
446     8. Applied Daniel's patches to solve problems with the import/export magic
447        syntax that is required for Windows, and which was going wrong for the
448        pcreposix and pcrecpp parts of the library. These were overlooked when this
449        problem was solved for the main library.
450    
451     9. There were some crude static tests to avoid integer overflow when computing
452        the size of patterns that contain repeated groups with explicit upper
453        limits. As the maximum quantifier is 65535, the maximum group length was
454        set at 30,000 so that the product of these two numbers did not overflow a
455        32-bit integer. However, it turns out that people want to use groups that
456        are longer than 30,000 bytes (though not repeat them that many times).
457        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
458        made it possible to implement the integer overflow checks in a much more
459        dynamic way, which I have now done. The artificial limitation on group
460        length has been removed - we now have only the limit on the total length of
461        the compiled pattern, which depends on the LINK_SIZE setting.
462    
463    10. Fixed a bug in the documentation for get/copy named substring when
464        duplicate names are permitted. If none of the named substrings are set, the
465        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
466        empty string.
467    
468    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
469        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
470        because the ] is interpreted as the first data character and the
471        terminating ] is not found. PCRE has been made compatible with Perl in this
472        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
473        cause memory overwriting.
474    
475    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
476        string has been matched (to stop an infinite loop). It was not recognizing
477        a conditional subpattern that could match an empty string if that
478        subpattern was within another subpattern. For example, it looped when
479        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
480        condition was not nested. This bug has been fixed.
481    
482    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
483        past the start of the subject in the presence of bytes with the top bit
484        set, for example "\x8aBCD".
485    
486    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
487        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
488    
489    14. Optimized (?!) to (*FAIL).
490    
491    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
492        This restricts code points to be within the range 0 to 0x10FFFF, excluding
493        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
494        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
495        does: it's just the validity check that is more restrictive.
496    
497    16. Inserted checks for integer overflows during escape sequence (backslash)
498        processing, and also fixed erroneous offset values for syntax errors during
499        backslash processing.
500    
501    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
502        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
503    
504    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
505        caused an overrun.
506    
507    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
508        something other than just ASCII characters) inside a group that had an
509        unlimited repeat caused a loop at compile time (while checking to see
510        whether the group could match an empty string).
511    
512    20. Debugging a pattern containing \p or \P could cause a crash. For example,
513        [\P{Any}] did so. (Error in the code for printing property names.)
514    
515    21. An orphan \E inside a character class could cause a crash.
516    
517    22. A repeated capturing bracket such as (A)? could cause a wild memory
518        reference during compilation.
519    
520    23. There are several functions in pcre_compile() that scan along a compiled
521        expression for various reasons (e.g. to see if it's fixed length for look
522        behind). There were bugs in these functions when a repeated \p or \P was
523        present in the pattern. These operators have additional parameters compared
524        with \d, etc, and these were not being taken into account when moving along
525        the compiled data. Specifically:
526    
527        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
528            length.
529    
530        (b) An item such as \pL+ within a repeated group could cause crashes or
531            loops.
532    
533        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
534            "reference to non-existent subpattern" error.
535    
536        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
537    
538    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
539        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
540    
541    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
542    
543    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
544        character were causing crashes (broken optimization).
545    
546    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
547        \p or \P) caused a compile-time loop.
548    
549    28. More problems have arisen in unanchored patterns when CRLF is a valid line
550        break. For example, the unstudied pattern [\r\n]A does not match the string
551        "\r\nA" because change 7.0/46 below moves the current point on by two
552        characters after failing to match at the start. However, the pattern \nA
553        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
554        the same is true. There doesn't seem any very clean way out of this, but
555        what I have chosen to do makes the common cases work: PCRE now takes note
556        of whether there can be an explicit match for \r or \n anywhere in the
557        pattern, and if so, 7.0/46 no longer applies. As part of this change,
558        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
559        pattern has explicit CR or LF references.
560    
561    29. Added (*CR) etc for changing newline setting at start of pattern.
562    
563    
564    Version 7.2 19-Jun-07
565    ---------------------
566    
567     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
568        which is apparently normally available under Windows.
569    
570     2. Re-jig the pcregrep tests with different newline settings in an attempt
571        to make them independent of the local environment's newline setting.
572    
573     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
574    
575     4. Some of the "internals" tests were previously cut out when the link size
576        was not 2, because the output contained actual offsets. The recent new
577        "Z" feature of pcretest means that these can be cut out, making the tests
578        usable with all link sizes.
579    
580     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
581        stack recursion. This gives a massive performance boost under BSD, but just
582        a small improvement under Linux. However, it saves one field in the frame
583        in all cases.
584    
585     6. Added more features from the forthcoming Perl 5.10:
586    
587        (a) (?-n) (where n is a string of digits) is a relative subroutine or
588            recursion call. It refers to the nth most recently opened parentheses.
589    
590        (b) (?+n) is also a relative subroutine call; it refers to the nth next
591            to be opened parentheses.
592    
593        (c) Conditions that refer to capturing parentheses can be specified
594            relatively, for example, (?(-2)... or (?(+3)...
595    
596        (d) \K resets the start of the current match so that everything before
597            is not part of it.
598    
599        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
600    
601        (f) \g{name} is another synonym - part of Perl 5.10's unification of
602            reference syntax.
603    
604        (g) (?| introduces a group in which the numbering of parentheses in each
605            alternative starts with the same number.
606    
607        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
608    
609     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
610        PCRE_INFO_JCHANGED.
611    
612     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
613        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
614        for detecting groups that can match an empty string.
615    
616     9. A pattern with a very large number of alternatives (more than several
617        hundred) was running out of internal workspace during the pre-compile
618        phase, where pcre_compile() figures out how much memory will be needed. A
619        bit of new cunning has reduced the workspace needed for groups with
620        alternatives. The 1000-alternative test pattern now uses 12 bytes of
621        workspace instead of running out of the 4096 that are available.
622    
623    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
624    
625    11. Applied patch from Google to remove an optimization that didn't quite work.
626        The report of the bug said:
627    
628          pcrecpp::RE("a*").FullMatch("aaa") matches, while
629          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
630          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
631    
632    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
633        it matched the wrong number of bytes.
634    
635    
636    Version 7.1 24-Apr-07
637    ---------------------
638    
639     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
640        that is more "standard", making use of automake and other Autotools. There
641        is some re-arrangement of the files and adjustment of comments consequent
642        on this.
643    
644     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
645        for recursive directory scanning broke on some systems because the files
646        are not scanned in any specific order and on different systems the order
647        was different. A call to "sort" has been inserted into RunGrepTest for the
648        approprate test as a short-term fix. In the longer term there may be an
649        alternative.
650    
651     3. I had an email from Eric Raymond about problems translating some of PCRE's
652        man pages to HTML (despite the fact that I distribute HTML pages, some
653        people do their own conversions for various reasons). The problems
654        concerned the use of low-level troff macros .br and .in. I have therefore
655        removed all such uses from the man pages (some were redundant, some could
656        be replaced by .nf/.fi pairs). The 132html script that I use to generate
657        HTML has been updated to handle .nf/.fi and to complain if it encounters
658        .br or .in.
659    
660     4. Updated comments in configure.ac that get placed in config.h.in and also
661        arranged for config.h to be included in the distribution, with the name
662        config.h.generic, for the benefit of those who have to compile without
663        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
664    
665     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
666        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
667        makevp.bat for the current PCRE, using the additional files
668        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
669    
670     6. A Windows user reported a minor discrepancy with test 2, which turned out
671        to be caused by a trailing space on an input line that had got lost in his
672        copy. The trailing space was an accident, so I've just removed it.
673    
674     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
675        that is needed.
676    
677     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
678        as "const" (a) because they are and (b) because it helps the PHP
679        maintainers who have recently made a script to detect big data structures
680        in the php code that should be moved to the .rodata section. I remembered
681        to update Builducptable as well, so it won't revert if ucptable.h is ever
682        re-created.
683    
684     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
685        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
686        order to be able to cut out the UTF-8 tables in the latter when UTF-8
687        support is not required. This saves 1.5-2K of code, which is important in
688        some applications.
689    
690        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
691        so as not to refer to the tables, even though these functions will never be
692        called when UTF-8 support is disabled. Otherwise there are problems with a
693        shared library.
694    
695    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
696    
697        (a) It was defining its arguments as char * instead of void *.
698    
699        (b) It was assuming that all moves were upwards in memory; this was true
700            a long time ago when I wrote it, but is no longer the case.
701    
702        The emulated memove() is provided for those environments that have neither
703        memmove() nor bcopy(). I didn't think anyone used it these days, but that
704        is clearly not the case, as these two bugs were recently reported.
705    
706    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
707        and Detrail to create the HTML documentation, the .txt form of the man
708        pages, and it removes trailing spaces from listed files. It also creates
709        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
710        case, it wraps all the #defines with #ifndefs. This script should be run
711        before "make dist".
712    
713    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
714        with Unicode property support.
715    
716        (a) For a maximizing quantifier, if the two different cases of the
717            character were of different lengths in their UTF-8 codings (there are
718            some cases like this - I found 11), and the matching function had to
719            back up over a mixture of the two cases, it incorrectly assumed they
720            were both the same length.
721    
722        (b) When PCRE was configured to use the heap rather than the stack for
723            recursion during matching, it was not correctly preserving the data for
724            the other case of a UTF-8 character when checking ahead for a match
725            while processing a minimizing repeat. If the check also involved
726            matching a wide character, but failed, corruption could cause an
727            erroneous result when trying to check for a repeat of the original
728            character.
729    
730    13. Some tidying changes to the testing mechanism:
731    
732        (a) The RunTest script now detects the internal link size and whether there
733            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
734            values substituted by "configure". (The RunGrepTest script already did
735            this for UTF-8.) The configure.ac script no longer substitutes the
736            relevant variables.
737    
738        (b) The debugging options /B and /D in pcretest show the compiled bytecode
739            with length and offset values. This means that the output is different
740            for different internal link sizes. Test 2 is skipped for link sizes
741            other than 2 because of this, bypassing the problem. Unfortunately,
742            there was also a test in test 3 (the locale tests) that used /B and
743            failed for link sizes other than 2. Rather than cut the whole test out,
744            I have added a new /Z option to pcretest that replaces the length and
745            offset values with spaces. This is now used to make test 3 independent
746            of link size. (Test 2 will be tidied up later.)
747    
748    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
749        segmentation fault instead of returning the appropriate error message.
750    
751    15. In multiline mode when the newline sequence was set to "any", the pattern
752        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
753        This doesn't seem right; it now treats the CRLF combination as the line
754        ending, and so does not match in that case. It's only a pattern such as ^$
755        that would hit this one: something like ^ABC$ would have failed after \r
756        and then tried again after \r\n.
757    
758    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
759        in an attempt to make files that differ only in their line terminators
760        compare equal. This works on Linux.
761    
762    17. Under certain error circumstances pcregrep might try to free random memory
763        as it exited. This is now fixed, thanks to valgrind.
764    
765    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
766        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
767        was because its rules for how to advance for /g after matching an empty
768        string at the end of a line did not allow for this case. They now check for
769        it specially.
770    
771    20. pcretest is supposed to handle patterns and data of any length, by
772        extending its buffers when necessary. It was getting this wrong when the
773        buffer for a data line had to be extended.
774    
775    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
776        CRLF as a newline sequence.
777    
778    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
779        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
780        I have nevertheless tidied it up.
781    
782    23. Added some casts to kill warnings from HP-UX ia64 compiler.
783    
784    24. Added a man page for pcre-config.
785    
786    
787    Version 7.0 19-Dec-06
788    ---------------------
789    
790     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
791        moving to gcc 4.1.1.
792    
793     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
794        sys/time.h, which is documented as needed for this function. It doesn't
795        seem to matter on Linux, but it showed up on some releases of OS X.
796    
797     3. It seems that there are systems where bytes whose values are greater than
798        127 match isprint() in the "C" locale. The "C" locale should be the
799        default when a C program starts up. In most systems, only ASCII printing
800        characters match isprint(). This difference caused the output from pcretest
801        to vary, making some of the tests fail. I have changed pcretest so that:
802    
803        (a) When it is outputting text in the compiled version of a pattern, bytes
804            other than 32-126 are always shown as hex escapes.
805    
806        (b) When it is outputting text that is a matched part of a subject string,
807            it does the same, unless a different locale has been set for the match
808            (using the /L modifier). In this case, it uses isprint() to decide.
809    
810     4. Fixed a major bug that caused incorrect computation of the amount of memory
811        required for a compiled pattern when options that changed within the
812        pattern affected the logic of the preliminary scan that determines the
813        length. The relevant options are -x, and -i in UTF-8 mode. The result was
814        that the computed length was too small. The symptoms of this bug were
815        either the PCRE error "internal error: code overflow" from pcre_compile(),
816        or a glibc crash with a message such as "pcretest: free(): invalid next
817        size (fast)". Examples of patterns that provoked this bug (shown in
818        pcretest format) are:
819    
820          /(?-x: )/x
821          /(?x)(?-x: \s*#\s*)/
822          /((?i)[\x{c0}])/8
823          /(?i:[\x{c0}])/8
824    
825        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
826        is now done differently.
827    
828     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
829        wrapper classes; (b) implement a new function in the C++ scanner that is
830        more efficient than the old way of doing things because it avoids levels of
831        recursion in the regex matching; (c) add a paragraph to the documentation
832        for the FullMatch() function.
833    
834     6. The escape sequence \n was being treated as whatever was defined as
835        "newline". Not only was this contrary to the documentation, which states
836        that \n is character 10 (hex 0A), but it also went horribly wrong when
837        "newline" was defined as CRLF. This has been fixed.
838    
839     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
840        was being set to -1 for the "end of line" case (supposedly a value that no
841        character can have). Though this value is never used (the check for end of
842        line is "zero bytes in current character"), it caused compiler complaints.
843        I've changed it to 0xffffffff.
844    
845     8. In pcre_version.c, the version string was being built by a sequence of
846        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
847        string (as it is for production releases) called a macro with an empty
848        argument. The C standard says the result of this is undefined. The gcc
849        compiler treats it as an empty string (which was what was wanted) but it is
850        reported that Visual C gives an error. The source has been hacked around to
851        avoid this problem.
852    
853     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
854        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
855        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
856        of them did).
857    
858    10. Originally, pcretest opened its input and output without "b"; then I was
859        told that "b" was needed in some environments, so it was added for release
860        5.0 to both the input and output. (It makes no difference on Unix-like
861        systems.) Later I was told that it is wrong for the input on Windows. I've
862        now abstracted the modes into two macros, to make it easier to fiddle with
863        them, and removed "b" from the input mode under Windows.
864    
865    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
866    
867    12. Added -help and --help to pcretest as an official way of being reminded
868        of the options.
869    
870    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
871        and pcrecpp.cc because they annoy compilers at high warning levels.
872    
873    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
874    
875    15. Fixed an occurrence of == in configure.ac that should have been = (shell
876        scripts are not C programs :-) and which was not noticed because it works
877        on Linux.
878    
879    16. pcretest is supposed to handle any length of pattern and data line (as one
880        line or as a continued sequence of lines) by extending its input buffer if
881        necessary. This feature was broken for very long pattern lines, leading to
882        a string of junk being passed to pcre_compile() if the pattern was longer
883        than about 50K.
884    
885    17. I have done a major re-factoring of the way pcre_compile() computes the
886        amount of memory needed for a compiled pattern. Previously, there was code
887        that made a preliminary scan of the pattern in order to do this. That was
888        OK when PCRE was new, but as the facilities have expanded, it has become
889        harder and harder to keep it in step with the real compile phase, and there
890        have been a number of bugs (see for example, 4 above). I have now found a
891        cunning way of running the real compile function in a "fake" mode that
892        enables it to compute how much memory it would need, while actually only
893        ever using a few hundred bytes of working memory and without too many
894        tests of the mode. This should make future maintenance and development
895        easier. A side effect of this work is that the limit of 200 on the nesting
896        depth of parentheses has been removed (though this was never a serious
897        limitation, I suspect). However, there is a downside: pcre_compile() now
898        runs more slowly than before (30% or more, depending on the pattern). I
899        hope this isn't a big issue. There is no effect on runtime performance.
900    
901    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
902        newline (only possible for the last line of a file) and it was a
903        pattern that set a locale (followed by /Lsomething), pcretest crashed.
904    
905    19. Added additional timing features to pcretest. (1) The -tm option now times
906        matching only, not compiling. (2) Both -t and -tm can be followed, as a
907        separate command line item, by a number that specifies the number of
908        repeats to use when timing. The default is 50000; this gives better
909        precision, but takes uncomfortably long for very large patterns.
910    
911    20. Extended pcre_study() to be more clever in cases where a branch of a
912        subpattern has no definite first character. For example, (a*|b*)[cd] would
913        previously give no result from pcre_study(). Now it recognizes that the
914        first character must be a, b, c, or d.
915    
916    21. There was an incorrect error "recursive call could loop indefinitely" if
917        a subpattern (or the entire pattern) that was being tested for matching an
918        empty string contained only one non-empty item after a nested subpattern.
919        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
920        incorrectly, because the \d was being skipped in the check.
921    
922    22. The pcretest program now has a new pattern option /B and a command line
923        option -b, which is equivalent to adding /B to every pattern. This causes
924        it to show the compiled bytecode, without the additional information that
925        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
926        is the same as /B/I).
927    
928    23. A new optimization is now able automatically to treat some sequences such
929        as a*b as a*+b. More specifically, if something simple (such as a character
930        or a simple class like \d) has an unlimited quantifier, and is followed by
931        something that cannot possibly match the quantified thing, the quantifier
932        is automatically "possessified".
933    
934    24. A recursive reference to a subpattern whose number was greater than 39
935        went wrong under certain circumstances in UTF-8 mode. This bug could also
936        have affected the operation of pcre_study().
937    
938    25. Realized that a little bit of performance could be had by replacing
939        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
940    
941    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
942    
943    27. Possessive quantifiers such as a++ were previously implemented by turning
944        them into atomic groups such as ($>a+). Now they have their own opcodes,
945        which improves performance. This includes the automatically created ones
946        from 23 above.
947    
948    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
949        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
950        the first matched character to be a colon. This applied both to named and
951        numbered groups.
952    
953    29. The ucpinternal.h header file was missing its idempotency #ifdef.
954    
955    30. I was sent a "project" file called libpcre.a.dev which I understand makes
956        building PCRE on Windows easier, so I have included it in the distribution.
957    
958    31. There is now a check in pcretest against a ridiculously large number being
959        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
960        loop, the loop is abandoned.
961    
962    32. Forward references to subpatterns in conditions such as (?(2)...) where
963        subpattern 2 is defined later cause pcre_compile() to search forwards in
964        the pattern for the relevant set of parentheses. This search went wrong
965        when there were unescaped parentheses in a character class, parentheses
966        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
967    
968    33. "Subroutine" calls and backreferences were previously restricted to
969        referencing subpatterns earlier in the regex. This restriction has now
970        been removed.
971    
972    34. Added a number of extra features that are going to be in Perl 5.10. On the
973        whole, these are just syntactic alternatives for features that PCRE had
974        previously implemented using the Python syntax or my own invention. The
975        other formats are all retained for compatibility.
976    
977        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
978            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
979            also .NET compatible.
980    
981        (b) A recursion or subroutine call to a named group can now be defined as
982            (?&name) as well as (?P>name).
983    
984        (c) A backreference to a named group can now be defined as \k<name> or
985            \k'name' as well as (?P=name). The new forms, as well as being in Perl
986            5.10, are also .NET compatible.
987    
988        (d) A conditional reference to a named group can now use the syntax
989            (?(<name>) or (?('name') as well as (?(name).
990    
991        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
992            groups (named and numbered) that are never evaluated inline, but can be
993            called as "subroutines" from elsewhere. In effect, the DEFINE condition
994            is always false. There may be only one alternative in such a group.
995    
996        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
997            as the simple (?(R). The condition is true only if the most recent
998            recursion is that of the given number or name. It does not search out
999            through the entire recursion stack.
1000    
1001        (g) The escape \gN or \g{N} has been added, where N is a positive or
1002            negative number, specifying an absolute or relative reference.
1003    
1004    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1005        some "unreachable code" warnings.
1006    
1007    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1008        things, this adds five new scripts.
1009    
1010    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1011        There were also incompatibilities regarding the handling of \Q..\E inside
1012        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1013        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1014    
1015    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1016        matches an empty string, and forcibly breaks the loop. There were bugs in
1017        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1018        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1019        separate and independent bugs (that affected different cases) have been
1020        fixed.
1021    
1022    39. Refactored the code to abolish the use of different opcodes for small
1023        capturing bracket numbers. This is a tidy that I avoided doing when I
1024        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1025        The new approach is not only tidier, it makes it possible to reduce the
1026        memory needed to fix the previous bug (38).
1027    
1028    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1029        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1030        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1031        mode.
1032    
1033    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1034        report.
1035    
1036    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1037        copy construction and assignment in the C++ wrapper.
1038    
1039    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1040        couple of bugs that could have given wrong results in the "--newline=crlf"
1041        case.
1042    
1043    44. Added a number of casts and did some reorganization of signed/unsigned int
1044        variables following suggestions from Dair Grant. Also renamed the variable
1045        "this" as "item" because it is a C++ keyword.
1046    
1047    45. Arranged for dftables to add
1048    
1049          #include "pcre_internal.h"
1050    
1051        to pcre_chartables.c because without it, gcc 4.x may remove the array
1052        definition from the final binary if PCRE is built into a static library and
1053        dead code stripping is activated.
1054    
1055    46. For an unanchored pattern, if a match attempt fails at the start of a
1056        newline sequence, and the newline setting is CRLF or ANY, and the next two
1057        characters are CRLF, advance by two characters instead of one.
1058    
1059    
1060    Version 6.7 04-Jul-06
1061    ---------------------
1062    
1063     1. In order to handle tests when input lines are enormously long, pcretest has
1064        been re-factored so that it automatically extends its buffers when
1065        necessary. The code is crude, but this _is_ just a test program. The
1066        default size has been increased from 32K to 50K.
1067    
1068     2. The code in pcre_study() was using the value of the re argument before
1069        testing it for NULL. (Of course, in any sensible call of the function, it
1070        won't be NULL.)
1071    
1072     3. The memmove() emulation function in pcre_internal.h, which is used on
1073        systems that lack both memmove() and bcopy() - that is, hardly ever -
1074        was missing a "static" storage class specifier.
1075    
1076     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1077        containing an extended class (one that cannot be represented by a bitmap
1078        because it contains high-valued characters or Unicode property items, e.g.
1079        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1080        pattern, but PCRE should not loop if you do not (it no longer does).
1081        [Detail: two cases were found: (a) a repeated subpattern containing an
1082        extended class; (b) a recursive reference to a subpattern that followed a
1083        previous extended class. It wasn't skipping over the extended class
1084        correctly when UTF-8 mode was not set.]
1085    
1086     5. A negated single-character class was not being recognized as fixed-length
1087        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1088        compile error "lookbehind assertion is not fixed length".
1089    
1090     6. The RunPerlTest auxiliary script was showing an unexpected difference
1091        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1092        write a Perl script that can interpret lines of an input file either as
1093        byte characters or as UTF-8, which is what "perltest" was being required to
1094        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1095        can't do is switch easily at run time between having the "use utf8;" pragma
1096        or not. In the end, I fudged it by using the RunPerlTest script to insert
1097        "use utf8;" explicitly for the UTF-8 tests.
1098    
1099     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1100        the end of the subject string, contrary to the documentation and to what
1101        Perl does. This was true of both matching functions. Now it matches only at
1102        the start of the subject and immediately after *internal* newlines.
1103    
1104     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1105        a pointer to an int instead of a pointer to an unsigned long int. This
1106        caused problems on 64-bit systems.
1107    
1108     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1109        instance of the 'standard' template library not being so standard".
1110    
1111    10. There was no check on the number of named subpatterns nor the maximum
1112        length of a subpattern name. The product of these values is used to compute
1113        the size of the memory block for a compiled pattern. By supplying a very
1114        long subpattern name and a large number of named subpatterns, the size
1115        computation could be caused to overflow. This is now prevented by limiting
1116        the length of names to 32 characters, and the number of named subpatterns
1117        to 10,000.
1118    
1119    11. Subpatterns that are repeated with specific counts have to be replicated in
1120        the compiled pattern. The size of memory for this was computed from the
1121        length of the subpattern and the repeat count. The latter is limited to
1122        65535, but there was no limit on the former, meaning that integer overflow
1123        could in principle occur. The compiled length of a repeated subpattern is
1124        now limited to 30,000 bytes in order to prevent this.
1125    
1126    12. Added the optional facility to have named substrings with the same name.
1127    
1128    13. Added the ability to use a named substring as a condition, using the
1129        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1130        are numbers (not recommended). Forward references are permitted.
1131    
1132    14. Added forward references in named backreferences (if you see what I mean).
1133    
1134    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1135        pattern could run off the end of the subject. For example, the pattern
1136        "(?s)(.{1,5})"8 did this with the subject "ab".
1137    
1138    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1139        PCRE_CASELESS was set when matching characters that were quantified with ?
1140        or *.
1141    
1142    17. A character class other than a single negated character that had a minimum
1143        but no maximum quantifier - for example [ab]{6,} - was not handled
1144        correctly by pce_dfa_exec(). It would match only one character.
1145    
1146    18. A valid (though odd) pattern that looked like a POSIX character
1147        class but used an invalid character after [ (for example [[,abc,]]) caused
1148        pcre_compile() to give the error "Failed: internal error: code overflow" or
1149        in some cases to crash with a glibc free() error. This could even happen if
1150        the pattern terminated after [[ but there just happened to be a sequence of
1151        letters, a binary zero, and a closing ] in the memory that followed.
1152    
1153    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1154        over the years. Originally (before any Unicode support), just the bottom 8
1155        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1156        output from "man perlunicode" includes this:
1157    
1158          The regular expression compiler produces polymorphic opcodes.  That
1159          is, the pattern adapts to the data and automatically switches to
1160          the Unicode character scheme when presented with Unicode data--or
1161          instead uses a traditional byte scheme when presented with byte
1162          data.
1163    
1164        Sadly, a wide octal escape does not cause a switch, and in a string with
1165        no other multibyte characters, these octal escapes are treated as before.
1166        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1167        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1168        Unicode string.
1169    
1170        I have not perpetrated such confusion in PCRE. Up till now, it took just
1171        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1172        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1173        translate to the appropriate multibyte character.
1174    
1175    29. Applied some refactoring to reduce the number of warnings from Microsoft
1176        and Borland compilers. This has included removing the fudge introduced
1177        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1178        a warning about an unused variable.
1179    
1180    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1181        characters since release 4.0, because Perl (from release 5.004) does not.
1182        [Or at least, is documented not to: some releases seem to be in conflict
1183        with the documentation.] However, when a pattern was studied with
1184        pcre_study() and all its branches started with \s, PCRE still included VT
1185        as a possible starting character. Of course, this did no harm; it just
1186        caused an unnecessary match attempt.
1187    
1188    22. Removed a now-redundant internal flag bit that recorded the fact that case
1189        dependency changed within the pattern. This was once needed for "required
1190        byte" processing, but is no longer used. This recovers a now-scarce options
1191        bit. Also moved the least significant internal flag bit to the most-
1192        significant bit of the word, which was not previously used (hangover from
1193        the days when it was an int rather than a uint) to free up another bit for
1194        the future.
1195    
1196    23. Added support for CRLF line endings as well as CR and LF. As well as the
1197        default being selectable at build time, it can now be changed at runtime
1198        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1199        specify that it is scanning data with non-default line endings.
1200    
1201    24. Changed the definition of CXXLINK to make it agree with the definition of
1202        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1203    
1204    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1205        recursions. This makes a big different to stack usage for some patterns.
1206    
1207    26. If a subpattern containing a named recursion or subroutine reference such
1208        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1209        the space required for the compiled pattern went wrong and gave too small a
1210        value. Depending on the environment, this could lead to "Failed: internal
1211        error: code overflow at offset 49" or "glibc detected double free or
1212        corruption" errors.
1213    
1214    27. Applied patches from Google (a) to support the new newline modes and (b) to
1215        advance over multibyte UTF-8 characters in GlobalReplace.
1216    
1217    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1218        difference for some implementation of PCRE in some Windows version.
1219    
1220    29. Added some extra testing facilities to pcretest:
1221    
1222        \q<number>   in a data line sets the "match limit" value
1223        \Q<number>   in a data line sets the "match recursion limt" value
1224        -S <number>  sets the stack size, where <number> is in megabytes
1225    
1226        The -S option isn't available for Windows.
1227    
1228    
1229    Version 6.6 06-Feb-06
1230    ---------------------
1231    
1232     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1233        in pcreposix.h. I have copied the definition from pcre.h.
1234    
1235     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1236        because pcre.h is no longer a built file.
1237    
1238     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1239        not normally included in the compiled code.
1240    
1241    
1242  Version 6.5 01-Feb-06  Version 6.5 01-Feb-06
1243  ---------------------  ---------------------
1244    

Legend:
Removed from v.87  
changed lines
  Added in v.355

  ViewVC Help
Powered by ViewVC 1.1.5