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

Diff of /code/trunk/ChangeLog

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

revision 81 by nigel, Sat Feb 24 21:40:59 2007 UTC revision 357 by ph10, Tue Jul 8 14:18:28 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    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
31    
32    7.  Added two (int) casts to pcregrep when printing the difference of two
33        pointers, in case they are 64-bit values.
34    
35    
36    Version 7.7 07-May-08
37    ---------------------
38    
39    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
40        a string to a long long, pretend we don't even have a long long." This is
41        done by checking for the strtoq, strtoll, and _strtoi64 functions.
42    
43    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
44        pre-7.6 versions, which defined a global no_arg variable instead of putting
45        it in the RE class. (See also #8 below.)
46    
47    3.  Remove a line of dead code, identified by coverity and reported by Nuno
48        Lopes.
49    
50    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
51    
52        (1) The include/exclude patterns were being applied to the whole pathnames
53            of files, instead of just to the final components.
54    
55        (2) If there was more than one level of directory, the subdirectories were
56            skipped unless they satisfied the include/exclude conditions. This is
57            inconsistent with GNU grep (and could even be seen as contrary to the
58            pcregrep specification - which I improved to make it absolutely clear).
59            The action now is always to scan all levels of directory, and just
60            apply the include/exclude patterns to regular files.
61    
62    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
63        --exclude_dir in the tests to avoid scanning .svn directories.
64    
65    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
66        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
67        doesn't support NULs in patterns.
68    
69    7.  Added some missing "const"s to declarations of static tables in
70        pcre_compile.c and pcre_dfa_exec.c.
71    
72    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
73        caused by fix #2  above. (Subsequently also a second patch to fix the
74        first patch. And a third patch - this was a messy problem.)
75    
76    9.  Applied Craig's patch to remove the use of push_back().
77    
78    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
79        matching function regexec().
80    
81    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
82        which, however, unlike Perl's \g{...}, are subroutine calls, not back
83        references. PCRE supports relative numbers with this syntax (I don't think
84        Oniguruma does).
85    
86    12. Previously, a group with a zero repeat such as (...){0} was completely
87        omitted from the compiled regex. However, this means that if the group
88        was called as a subroutine from elsewhere in the pattern, things went wrong
89        (an internal error was given). Such groups are now left in the compiled
90        pattern, with a new opcode that causes them to be skipped at execution
91        time.
92    
93    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
94        to the way PCRE behaves:
95    
96        (a) A lone ] character is dis-allowed (Perl treats it as data).
97    
98        (b) A back reference to an unmatched subpattern matches an empty string
99            (Perl fails the current match path).
100    
101        (c) A data ] in a character class must be notated as \] because if the
102            first data character in a class is ], it defines an empty class. (In
103            Perl it is not possible to have an empty class.) The empty class []
104            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
105            The negative empty class [^] matches any one character, independently
106            of the DOTALL setting.
107    
108    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
109        non-existent subpattern following a character class starting with ']' and
110        containing () gave an internal compiling error instead of "reference to
111        non-existent subpattern". Fortunately, when the pattern did exist, the
112        compiled code was correct. (When scanning forwards to check for the
113        existencd of the subpattern, it was treating the data ']' as terminating
114        the class, so got the count wrong. When actually compiling, the reference
115        was subsequently set up correctly.)
116    
117    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
118        it was being rejected as not supported by pcre_dfa_exec(), even though
119        other assertions are supported. I have made pcre_dfa_exec() support
120        (*FAIL).
121    
122    16. The implementation of 13c above involved the invention of a new opcode,
123        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
124        cannot be changed at match time, I realized I could make a small
125        improvement to matching performance by compiling OP_ALLANY instead of
126        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
127        on the OP_ANY path.
128    
129    17. Compiling pcretest on Windows with readline support failed without the
130        following two fixes: (1) Make the unistd.h include conditional on
131        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
132    
133    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
134        ncurses library to be included for pcretest when ReadLine support is
135        requested, but also to allow for it to be overridden. This patch came from
136        Daniel Bergström.
137    
138    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
139        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
140        any errors with the current Unicode tables. Thanks to Peter Kankowski for
141        spotting this.
142    
143    
144    Version 7.6 28-Jan-08
145    ---------------------
146    
147    1.  A character class containing a very large number of characters with
148        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
149        overflow.
150    
151    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
152        HAVE_LONG_LONG is not defined.
153    
154    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
155        bring them up to date and include new features. This patch includes:
156    
157        - Fixed PH's badly added libz and libbz2 support.
158        - Fixed a problem with static linking.
159        - Added pcredemo. [But later removed - see 7 below.]
160        - Fixed dftables problem and added an option.
161        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
162            HAVE_LONG_LONG.
163        - Added readline support for pcretest.
164        - Added an listing of the option settings after cmake has run.
165    
166    4.  A user submitted a patch to Makefile that makes it easy to create
167        "pcre.dll" under mingw when using Configure/Make. I added stuff to
168        Makefile.am that cause it to include this special target, without
169        affecting anything else. Note that the same mingw target plus all
170        the other distribution libraries and programs are now supported
171        when configuring with CMake (see 6 below) instead of with
172        Configure/Make.
173    
174    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
175        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
176        exported in the Windows port". It has not yet been confirmed that the patch
177        solves the problem, but it does no harm.
178    
179    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
180        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
181        with CMake, and also correct the comment about stack recursion.
182    
183    7.  Remove the automatic building of pcredemo from the ./configure system and
184        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
185        of a program that users should build themselves after PCRE is installed, so
186        building it automatically is not really right. What is more, it gave
187        trouble in some build environments.
188    
189    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
190    
191    
192    Version 7.5 10-Jan-08
193    ---------------------
194    
195    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
196        values in parens when parsing an RE using the C++ wrapper."
197    
198    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
199        Characters greater than 255 were excluded from the class instead of being
200        included.
201    
202    3.  The same bug as (2) above applied to negated POSIX classes such as
203        [:^space:].
204    
205    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
206        defined or documented. It seems to have been a typo for PCRE_STATIC, so
207        I have changed it.
208    
209    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
210        first named subpattern) and a construct such as (?&a) would reference the
211        first named subpattern whose name started with "a" (in other words, the
212        length check was missing). Both these problems are fixed. "Subpattern name
213        expected" is now given for (?&) (a zero-length name), and this patch also
214        makes it give the same error for \k'' (previously it complained that that
215        was a reference to a non-existent subpattern).
216    
217    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
218        this is right because (?- can be followed by option settings as well as by
219        digits. I have, however, made the messages clearer.
220    
221    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
222        than the number used in the conditional) now cause a compile-time error.
223        This is actually not compatible with Perl, which accepts such patterns, but
224        treats the conditional as always being FALSE (as PCRE used to), but it
225        seems to me that giving a diagnostic is better.
226    
227    8.  Change "alphameric" to the more common word "alphanumeric" in comments
228        and messages.
229    
230    9.  Fix two occurrences of "backslash" in comments that should have been
231        "backspace".
232    
233    10. Remove two redundant lines of code that can never be obeyed (their function
234        was moved elsewhere).
235    
236    11. The program that makes PCRE's Unicode character property table had a bug
237        which caused it to generate incorrect table entries for sequences of
238        characters that have the same character type, but are in different scripts.
239        It amalgamated them into a single range, with the script of the first of
240        them. In other words, some characters were in the wrong script. There were
241        thirteen such cases, affecting characters in the following ranges:
242    
243          U+002b0 - U+002c1
244          U+0060c - U+0060d
245          U+0061e - U+00612
246          U+0064b - U+0065e
247          U+0074d - U+0076d
248          U+01800 - U+01805
249          U+01d00 - U+01d77
250          U+01d9b - U+01dbf
251          U+0200b - U+0200f
252          U+030fc - U+030fe
253          U+03260 - U+0327f
254          U+0fb46 - U+0fbb1
255          U+10450 - U+1049d
256    
257    12. The -o option (show only the matching part of a line) for pcregrep was not
258        compatible with GNU grep in that, if there was more than one match in a
259        line, it showed only the first of them. It now behaves in the same way as
260        GNU grep.
261    
262    13. If the -o and -v options were combined for pcregrep, it printed a blank
263        line for every non-matching line. GNU grep prints nothing, and pcregrep now
264        does the same. The return code can be used to tell if there were any
265        non-matching lines.
266    
267    14. Added --file-offsets and --line-offsets to pcregrep.
268    
269    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
270        infinitely looping recursion. The bug was that positive lookaheads were not
271        being skipped when checking for a possible empty match (negative lookaheads
272        and both kinds of lookbehind were skipped).
273    
274    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
275        inclusion of <windows.h> to before rather than after the definition of
276        INVALID_FILE_ATTRIBUTES (patch from David Byron).
277    
278    17. Specifying a possessive quantifier with a specific limit for a Unicode
279        character property caused pcre_compile() to compile bad code, which led at
280        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
281        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
282        caused the error; without that there was no problem.
283    
284    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
285    
286    19. Added --enable-pcretest-libreadline.
287    
288    20. In pcrecpp.cc, the variable 'count' was incremented twice in
289        RE::GlobalReplace(). As a result, the number of replacements returned was
290        double what it should be. I removed one of the increments, but Craig sent a
291        later patch that removed the other one (the right fix) and added unit tests
292        that check the return values (which was not done before).
293    
294    21. Several CMake things:
295    
296        (1) Arranged that, when cmake is used on Unix, the libraries end up with
297            the names libpcre and libpcreposix, not just pcre and pcreposix.
298    
299        (2) The above change means that pcretest and pcregrep are now correctly
300            linked with the newly-built libraries, not previously installed ones.
301    
302        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
303    
304    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
305        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
306        UTF-8 newline character). The key issue is that the pattern starts .*;
307        this means that the match must be either at the beginning, or after a
308        newline. The bug was in the code for advancing after a failed match and
309        checking that the new position followed a newline. It was not taking
310        account of UTF-8 characters correctly.
311    
312    23. PCRE was behaving differently from Perl in the way it recognized POSIX
313        character classes. PCRE was not treating the sequence [:...:] as a
314        character class unless the ... were all letters. Perl, however, seems to
315        allow any characters between [: and :], though of course it rejects as
316        unknown any "names" that contain non-letters, because all the known class
317        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
318        for example, whereas PCRE did not - it did not recognize a POSIX character
319        class. This seemed a bit dangerous, so the code has been changed to be
320        closer to Perl. The behaviour is not identical to Perl, because PCRE will
321        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
322        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
323        Perl does, and where it didn't before.
324    
325    24. Rewrite so as to remove the single use of %n from pcregrep because in some
326        Windows environments %n is disabled by default.
327    
328    
329    Version 7.4 21-Sep-07
330    ---------------------
331    
332    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
333        means that a class such as [\s] counted as "explicit reference to CR or
334        LF". That isn't really right - the whole point of the change was to try to
335        help when there was an actual mention of one of the two characters. So now
336        the change happens only if \r or \n (or a literal CR or LF) character is
337        encountered.
338    
339    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
340        of both had grown to the point where there were only 3 bits left.
341        Fortunately, there was spare space in the data structure, and so I have
342        moved the internal flags into a new 16-bit field to free up more option
343        bits.
344    
345    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
346        but did not set the internal JCHANGED flag - either of these is enough to
347        control the way the "get" function works - but the PCRE_INFO_JCHANGED
348        facility is supposed to tell if (?J) was ever used, so now (?J) at the
349        start sets both bits.
350    
351    4.  Added options (at build time, compile time, exec time) to change \R from
352        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
353    
354    5.  doc/pcresyntax.html was missing from the distribution.
355    
356    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
357        compatibility, even though it is no longer used.
358    
359    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
360        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
361        windows.h file is present (where different names are used). [This was
362        reversed later after testing - see 16 below.]
363    
364    8.  Changed all #include <config.h> to #include "config.h". There were also
365        some further <pcre.h> cases that I changed to "pcre.h".
366    
367    9.  When pcregrep was used with the --colour option, it missed the line ending
368        sequence off the lines that it output.
369    
370    10. It was pointed out to me that arrays of string pointers cause lots of
371        relocations when a shared library is dynamically loaded. A technique of
372        using a single long string with a table of offsets can drastically reduce
373        these. I have refactored PCRE in four places to do this. The result is
374        dramatic:
375    
376          Originally:                          290
377          After changing UCP table:            187
378          After changing error message table:   43
379          After changing table of "verbs"       36
380          After changing table of Posix names   22
381    
382        Thanks to the folks working on Gregex for glib for this insight.
383    
384    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
385        unicode-properties was also set.
386    
387    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
388    
389    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
390        checked only for CRLF.
391    
392    14. Added casts to pcretest.c to avoid compiler warnings.
393    
394    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
395    
396    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
397        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
398        entirely. This removes changes made in 7 above.
399    
400    17. The CMake files have been updated, and there is now more information about
401        building with CMake in the NON-UNIX-USE document.
402    
403    
404    Version 7.3 28-Aug-07
405    ---------------------
406    
407     1. In the rejigging of the build system that eventually resulted in 7.1, the
408        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
409        brackets there is not right, since it causes compilers to look for an
410        installed pcre.h, not the version that is in the source that is being
411        compiled (which of course may be different). I have changed it back to:
412    
413          #include "pcre.h"
414    
415        I have a vague recollection that the change was concerned with compiling in
416        different directories, but in the new build system, that is taken care of
417        by the VPATH setting the Makefile.
418    
419     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
420        when the subject happened to end in the byte 0x85 (e.g. if the last
421        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
422        characters but of course it shouldn't be taken as a newline when it is part
423        of another character. The bug was that, for an unlimited repeat of . in
424        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
425        characters when looking for a newline.
426    
427     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
428    
429     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
430        in debug output.
431    
432     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
433        long printing in the pcrecpp unittest when running under MinGW.
434    
435     6. ESC_K was left out of the EBCDIC table.
436    
437     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
438        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
439        limit also applies to "virtual nesting" when a pattern is recursive, and in
440        this case 1000 isn't so big. I have been able to remove this limit at the
441        expense of backing off one optimization in certain circumstances. Normally,
442        when pcre_exec() would call its internal match() function recursively and
443        immediately return the result unconditionally, it uses a "tail recursion"
444        feature to save stack. However, when a subpattern that can match an empty
445        string has an unlimited repetition quantifier, it no longer makes this
446        optimization. That gives it a stack frame in which to save the data for
447        checking that an empty string has been matched. Previously this was taken
448        from the 1000-entry workspace that had been reserved. So now there is no
449        explicit limit, but more stack is used.
450    
451     8. Applied Daniel's patches to solve problems with the import/export magic
452        syntax that is required for Windows, and which was going wrong for the
453        pcreposix and pcrecpp parts of the library. These were overlooked when this
454        problem was solved for the main library.
455    
456     9. There were some crude static tests to avoid integer overflow when computing
457        the size of patterns that contain repeated groups with explicit upper
458        limits. As the maximum quantifier is 65535, the maximum group length was
459        set at 30,000 so that the product of these two numbers did not overflow a
460        32-bit integer. However, it turns out that people want to use groups that
461        are longer than 30,000 bytes (though not repeat them that many times).
462        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
463        made it possible to implement the integer overflow checks in a much more
464        dynamic way, which I have now done. The artificial limitation on group
465        length has been removed - we now have only the limit on the total length of
466        the compiled pattern, which depends on the LINK_SIZE setting.
467    
468    10. Fixed a bug in the documentation for get/copy named substring when
469        duplicate names are permitted. If none of the named substrings are set, the
470        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
471        empty string.
472    
473    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
474        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
475        because the ] is interpreted as the first data character and the
476        terminating ] is not found. PCRE has been made compatible with Perl in this
477        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
478        cause memory overwriting.
479    
480    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
481        string has been matched (to stop an infinite loop). It was not recognizing
482        a conditional subpattern that could match an empty string if that
483        subpattern was within another subpattern. For example, it looped when
484        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
485        condition was not nested. This bug has been fixed.
486    
487    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
488        past the start of the subject in the presence of bytes with the top bit
489        set, for example "\x8aBCD".
490    
491    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
492        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
493    
494    14. Optimized (?!) to (*FAIL).
495    
496    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
497        This restricts code points to be within the range 0 to 0x10FFFF, excluding
498        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
499        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
500        does: it's just the validity check that is more restrictive.
501    
502    16. Inserted checks for integer overflows during escape sequence (backslash)
503        processing, and also fixed erroneous offset values for syntax errors during
504        backslash processing.
505    
506    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
507        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
508    
509    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
510        caused an overrun.
511    
512    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
513        something other than just ASCII characters) inside a group that had an
514        unlimited repeat caused a loop at compile time (while checking to see
515        whether the group could match an empty string).
516    
517    20. Debugging a pattern containing \p or \P could cause a crash. For example,
518        [\P{Any}] did so. (Error in the code for printing property names.)
519    
520    21. An orphan \E inside a character class could cause a crash.
521    
522    22. A repeated capturing bracket such as (A)? could cause a wild memory
523        reference during compilation.
524    
525    23. There are several functions in pcre_compile() that scan along a compiled
526        expression for various reasons (e.g. to see if it's fixed length for look
527        behind). There were bugs in these functions when a repeated \p or \P was
528        present in the pattern. These operators have additional parameters compared
529        with \d, etc, and these were not being taken into account when moving along
530        the compiled data. Specifically:
531    
532        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
533            length.
534    
535        (b) An item such as \pL+ within a repeated group could cause crashes or
536            loops.
537    
538        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
539            "reference to non-existent subpattern" error.
540    
541        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
542    
543    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
544        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
545    
546    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
547    
548    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
549        character were causing crashes (broken optimization).
550    
551    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
552        \p or \P) caused a compile-time loop.
553    
554    28. More problems have arisen in unanchored patterns when CRLF is a valid line
555        break. For example, the unstudied pattern [\r\n]A does not match the string
556        "\r\nA" because change 7.0/46 below moves the current point on by two
557        characters after failing to match at the start. However, the pattern \nA
558        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
559        the same is true. There doesn't seem any very clean way out of this, but
560        what I have chosen to do makes the common cases work: PCRE now takes note
561        of whether there can be an explicit match for \r or \n anywhere in the
562        pattern, and if so, 7.0/46 no longer applies. As part of this change,
563        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
564        pattern has explicit CR or LF references.
565    
566    29. Added (*CR) etc for changing newline setting at start of pattern.
567    
568    
569    Version 7.2 19-Jun-07
570    ---------------------
571    
572     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
573        which is apparently normally available under Windows.
574    
575     2. Re-jig the pcregrep tests with different newline settings in an attempt
576        to make them independent of the local environment's newline setting.
577    
578     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
579    
580     4. Some of the "internals" tests were previously cut out when the link size
581        was not 2, because the output contained actual offsets. The recent new
582        "Z" feature of pcretest means that these can be cut out, making the tests
583        usable with all link sizes.
584    
585     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
586        stack recursion. This gives a massive performance boost under BSD, but just
587        a small improvement under Linux. However, it saves one field in the frame
588        in all cases.
589    
590     6. Added more features from the forthcoming Perl 5.10:
591    
592        (a) (?-n) (where n is a string of digits) is a relative subroutine or
593            recursion call. It refers to the nth most recently opened parentheses.
594    
595        (b) (?+n) is also a relative subroutine call; it refers to the nth next
596            to be opened parentheses.
597    
598        (c) Conditions that refer to capturing parentheses can be specified
599            relatively, for example, (?(-2)... or (?(+3)...
600    
601        (d) \K resets the start of the current match so that everything before
602            is not part of it.
603    
604        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
605    
606        (f) \g{name} is another synonym - part of Perl 5.10's unification of
607            reference syntax.
608    
609        (g) (?| introduces a group in which the numbering of parentheses in each
610            alternative starts with the same number.
611    
612        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
613    
614     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
615        PCRE_INFO_JCHANGED.
616    
617     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
618        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
619        for detecting groups that can match an empty string.
620    
621     9. A pattern with a very large number of alternatives (more than several
622        hundred) was running out of internal workspace during the pre-compile
623        phase, where pcre_compile() figures out how much memory will be needed. A
624        bit of new cunning has reduced the workspace needed for groups with
625        alternatives. The 1000-alternative test pattern now uses 12 bytes of
626        workspace instead of running out of the 4096 that are available.
627    
628    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
629    
630    11. Applied patch from Google to remove an optimization that didn't quite work.
631        The report of the bug said:
632    
633          pcrecpp::RE("a*").FullMatch("aaa") matches, while
634          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
635          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
636    
637    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
638        it matched the wrong number of bytes.
639    
640    
641    Version 7.1 24-Apr-07
642    ---------------------
643    
644     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
645        that is more "standard", making use of automake and other Autotools. There
646        is some re-arrangement of the files and adjustment of comments consequent
647        on this.
648    
649     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
650        for recursive directory scanning broke on some systems because the files
651        are not scanned in any specific order and on different systems the order
652        was different. A call to "sort" has been inserted into RunGrepTest for the
653        approprate test as a short-term fix. In the longer term there may be an
654        alternative.
655    
656     3. I had an email from Eric Raymond about problems translating some of PCRE's
657        man pages to HTML (despite the fact that I distribute HTML pages, some
658        people do their own conversions for various reasons). The problems
659        concerned the use of low-level troff macros .br and .in. I have therefore
660        removed all such uses from the man pages (some were redundant, some could
661        be replaced by .nf/.fi pairs). The 132html script that I use to generate
662        HTML has been updated to handle .nf/.fi and to complain if it encounters
663        .br or .in.
664    
665     4. Updated comments in configure.ac that get placed in config.h.in and also
666        arranged for config.h to be included in the distribution, with the name
667        config.h.generic, for the benefit of those who have to compile without
668        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
669    
670     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
671        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
672        makevp.bat for the current PCRE, using the additional files
673        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
674    
675     6. A Windows user reported a minor discrepancy with test 2, which turned out
676        to be caused by a trailing space on an input line that had got lost in his
677        copy. The trailing space was an accident, so I've just removed it.
678    
679     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
680        that is needed.
681    
682     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
683        as "const" (a) because they are and (b) because it helps the PHP
684        maintainers who have recently made a script to detect big data structures
685        in the php code that should be moved to the .rodata section. I remembered
686        to update Builducptable as well, so it won't revert if ucptable.h is ever
687        re-created.
688    
689     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
690        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
691        order to be able to cut out the UTF-8 tables in the latter when UTF-8
692        support is not required. This saves 1.5-2K of code, which is important in
693        some applications.
694    
695        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
696        so as not to refer to the tables, even though these functions will never be
697        called when UTF-8 support is disabled. Otherwise there are problems with a
698        shared library.
699    
700    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
701    
702        (a) It was defining its arguments as char * instead of void *.
703    
704        (b) It was assuming that all moves were upwards in memory; this was true
705            a long time ago when I wrote it, but is no longer the case.
706    
707        The emulated memove() is provided for those environments that have neither
708        memmove() nor bcopy(). I didn't think anyone used it these days, but that
709        is clearly not the case, as these two bugs were recently reported.
710    
711    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
712        and Detrail to create the HTML documentation, the .txt form of the man
713        pages, and it removes trailing spaces from listed files. It also creates
714        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
715        case, it wraps all the #defines with #ifndefs. This script should be run
716        before "make dist".
717    
718    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
719        with Unicode property support.
720    
721        (a) For a maximizing quantifier, if the two different cases of the
722            character were of different lengths in their UTF-8 codings (there are
723            some cases like this - I found 11), and the matching function had to
724            back up over a mixture of the two cases, it incorrectly assumed they
725            were both the same length.
726    
727        (b) When PCRE was configured to use the heap rather than the stack for
728            recursion during matching, it was not correctly preserving the data for
729            the other case of a UTF-8 character when checking ahead for a match
730            while processing a minimizing repeat. If the check also involved
731            matching a wide character, but failed, corruption could cause an
732            erroneous result when trying to check for a repeat of the original
733            character.
734    
735    13. Some tidying changes to the testing mechanism:
736    
737        (a) The RunTest script now detects the internal link size and whether there
738            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
739            values substituted by "configure". (The RunGrepTest script already did
740            this for UTF-8.) The configure.ac script no longer substitutes the
741            relevant variables.
742    
743        (b) The debugging options /B and /D in pcretest show the compiled bytecode
744            with length and offset values. This means that the output is different
745            for different internal link sizes. Test 2 is skipped for link sizes
746            other than 2 because of this, bypassing the problem. Unfortunately,
747            there was also a test in test 3 (the locale tests) that used /B and
748            failed for link sizes other than 2. Rather than cut the whole test out,
749            I have added a new /Z option to pcretest that replaces the length and
750            offset values with spaces. This is now used to make test 3 independent
751            of link size. (Test 2 will be tidied up later.)
752    
753    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
754        segmentation fault instead of returning the appropriate error message.
755    
756    15. In multiline mode when the newline sequence was set to "any", the pattern
757        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
758        This doesn't seem right; it now treats the CRLF combination as the line
759        ending, and so does not match in that case. It's only a pattern such as ^$
760        that would hit this one: something like ^ABC$ would have failed after \r
761        and then tried again after \r\n.
762    
763    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
764        in an attempt to make files that differ only in their line terminators
765        compare equal. This works on Linux.
766    
767    17. Under certain error circumstances pcregrep might try to free random memory
768        as it exited. This is now fixed, thanks to valgrind.
769    
770    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
771        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
772        was because its rules for how to advance for /g after matching an empty
773        string at the end of a line did not allow for this case. They now check for
774        it specially.
775    
776    20. pcretest is supposed to handle patterns and data of any length, by
777        extending its buffers when necessary. It was getting this wrong when the
778        buffer for a data line had to be extended.
779    
780    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
781        CRLF as a newline sequence.
782    
783    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
784        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
785        I have nevertheless tidied it up.
786    
787    23. Added some casts to kill warnings from HP-UX ia64 compiler.
788    
789    24. Added a man page for pcre-config.
790    
791    
792    Version 7.0 19-Dec-06
793    ---------------------
794    
795     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
796        moving to gcc 4.1.1.
797    
798     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
799        sys/time.h, which is documented as needed for this function. It doesn't
800        seem to matter on Linux, but it showed up on some releases of OS X.
801    
802     3. It seems that there are systems where bytes whose values are greater than
803        127 match isprint() in the "C" locale. The "C" locale should be the
804        default when a C program starts up. In most systems, only ASCII printing
805        characters match isprint(). This difference caused the output from pcretest
806        to vary, making some of the tests fail. I have changed pcretest so that:
807    
808        (a) When it is outputting text in the compiled version of a pattern, bytes
809            other than 32-126 are always shown as hex escapes.
810    
811        (b) When it is outputting text that is a matched part of a subject string,
812            it does the same, unless a different locale has been set for the match
813            (using the /L modifier). In this case, it uses isprint() to decide.
814    
815     4. Fixed a major bug that caused incorrect computation of the amount of memory
816        required for a compiled pattern when options that changed within the
817        pattern affected the logic of the preliminary scan that determines the
818        length. The relevant options are -x, and -i in UTF-8 mode. The result was
819        that the computed length was too small. The symptoms of this bug were
820        either the PCRE error "internal error: code overflow" from pcre_compile(),
821        or a glibc crash with a message such as "pcretest: free(): invalid next
822        size (fast)". Examples of patterns that provoked this bug (shown in
823        pcretest format) are:
824    
825          /(?-x: )/x
826          /(?x)(?-x: \s*#\s*)/
827          /((?i)[\x{c0}])/8
828          /(?i:[\x{c0}])/8
829    
830        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
831        is now done differently.
832    
833     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
834        wrapper classes; (b) implement a new function in the C++ scanner that is
835        more efficient than the old way of doing things because it avoids levels of
836        recursion in the regex matching; (c) add a paragraph to the documentation
837        for the FullMatch() function.
838    
839     6. The escape sequence \n was being treated as whatever was defined as
840        "newline". Not only was this contrary to the documentation, which states
841        that \n is character 10 (hex 0A), but it also went horribly wrong when
842        "newline" was defined as CRLF. This has been fixed.
843    
844     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
845        was being set to -1 for the "end of line" case (supposedly a value that no
846        character can have). Though this value is never used (the check for end of
847        line is "zero bytes in current character"), it caused compiler complaints.
848        I've changed it to 0xffffffff.
849    
850     8. In pcre_version.c, the version string was being built by a sequence of
851        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
852        string (as it is for production releases) called a macro with an empty
853        argument. The C standard says the result of this is undefined. The gcc
854        compiler treats it as an empty string (which was what was wanted) but it is
855        reported that Visual C gives an error. The source has been hacked around to
856        avoid this problem.
857    
858     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
859        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
860        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
861        of them did).
862    
863    10. Originally, pcretest opened its input and output without "b"; then I was
864        told that "b" was needed in some environments, so it was added for release
865        5.0 to both the input and output. (It makes no difference on Unix-like
866        systems.) Later I was told that it is wrong for the input on Windows. I've
867        now abstracted the modes into two macros, to make it easier to fiddle with
868        them, and removed "b" from the input mode under Windows.
869    
870    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
871    
872    12. Added -help and --help to pcretest as an official way of being reminded
873        of the options.
874    
875    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
876        and pcrecpp.cc because they annoy compilers at high warning levels.
877    
878    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
879    
880    15. Fixed an occurrence of == in configure.ac that should have been = (shell
881        scripts are not C programs :-) and which was not noticed because it works
882        on Linux.
883    
884    16. pcretest is supposed to handle any length of pattern and data line (as one
885        line or as a continued sequence of lines) by extending its input buffer if
886        necessary. This feature was broken for very long pattern lines, leading to
887        a string of junk being passed to pcre_compile() if the pattern was longer
888        than about 50K.
889    
890    17. I have done a major re-factoring of the way pcre_compile() computes the
891        amount of memory needed for a compiled pattern. Previously, there was code
892        that made a preliminary scan of the pattern in order to do this. That was
893        OK when PCRE was new, but as the facilities have expanded, it has become
894        harder and harder to keep it in step with the real compile phase, and there
895        have been a number of bugs (see for example, 4 above). I have now found a
896        cunning way of running the real compile function in a "fake" mode that
897        enables it to compute how much memory it would need, while actually only
898        ever using a few hundred bytes of working memory and without too many
899        tests of the mode. This should make future maintenance and development
900        easier. A side effect of this work is that the limit of 200 on the nesting
901        depth of parentheses has been removed (though this was never a serious
902        limitation, I suspect). However, there is a downside: pcre_compile() now
903        runs more slowly than before (30% or more, depending on the pattern). I
904        hope this isn't a big issue. There is no effect on runtime performance.
905    
906    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
907        newline (only possible for the last line of a file) and it was a
908        pattern that set a locale (followed by /Lsomething), pcretest crashed.
909    
910    19. Added additional timing features to pcretest. (1) The -tm option now times
911        matching only, not compiling. (2) Both -t and -tm can be followed, as a
912        separate command line item, by a number that specifies the number of
913        repeats to use when timing. The default is 50000; this gives better
914        precision, but takes uncomfortably long for very large patterns.
915    
916    20. Extended pcre_study() to be more clever in cases where a branch of a
917        subpattern has no definite first character. For example, (a*|b*)[cd] would
918        previously give no result from pcre_study(). Now it recognizes that the
919        first character must be a, b, c, or d.
920    
921    21. There was an incorrect error "recursive call could loop indefinitely" if
922        a subpattern (or the entire pattern) that was being tested for matching an
923        empty string contained only one non-empty item after a nested subpattern.
924        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
925        incorrectly, because the \d was being skipped in the check.
926    
927    22. The pcretest program now has a new pattern option /B and a command line
928        option -b, which is equivalent to adding /B to every pattern. This causes
929        it to show the compiled bytecode, without the additional information that
930        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
931        is the same as /B/I).
932    
933    23. A new optimization is now able automatically to treat some sequences such
934        as a*b as a*+b. More specifically, if something simple (such as a character
935        or a simple class like \d) has an unlimited quantifier, and is followed by
936        something that cannot possibly match the quantified thing, the quantifier
937        is automatically "possessified".
938    
939    24. A recursive reference to a subpattern whose number was greater than 39
940        went wrong under certain circumstances in UTF-8 mode. This bug could also
941        have affected the operation of pcre_study().
942    
943    25. Realized that a little bit of performance could be had by replacing
944        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
945    
946    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
947    
948    27. Possessive quantifiers such as a++ were previously implemented by turning
949        them into atomic groups such as ($>a+). Now they have their own opcodes,
950        which improves performance. This includes the automatically created ones
951        from 23 above.
952    
953    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
954        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
955        the first matched character to be a colon. This applied both to named and
956        numbered groups.
957    
958    29. The ucpinternal.h header file was missing its idempotency #ifdef.
959    
960    30. I was sent a "project" file called libpcre.a.dev which I understand makes
961        building PCRE on Windows easier, so I have included it in the distribution.
962    
963    31. There is now a check in pcretest against a ridiculously large number being
964        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
965        loop, the loop is abandoned.
966    
967    32. Forward references to subpatterns in conditions such as (?(2)...) where
968        subpattern 2 is defined later cause pcre_compile() to search forwards in
969        the pattern for the relevant set of parentheses. This search went wrong
970        when there were unescaped parentheses in a character class, parentheses
971        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
972    
973    33. "Subroutine" calls and backreferences were previously restricted to
974        referencing subpatterns earlier in the regex. This restriction has now
975        been removed.
976    
977    34. Added a number of extra features that are going to be in Perl 5.10. On the
978        whole, these are just syntactic alternatives for features that PCRE had
979        previously implemented using the Python syntax or my own invention. The
980        other formats are all retained for compatibility.
981    
982        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
983            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
984            also .NET compatible.
985    
986        (b) A recursion or subroutine call to a named group can now be defined as
987            (?&name) as well as (?P>name).
988    
989        (c) A backreference to a named group can now be defined as \k<name> or
990            \k'name' as well as (?P=name). The new forms, as well as being in Perl
991            5.10, are also .NET compatible.
992    
993        (d) A conditional reference to a named group can now use the syntax
994            (?(<name>) or (?('name') as well as (?(name).
995    
996        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
997            groups (named and numbered) that are never evaluated inline, but can be
998            called as "subroutines" from elsewhere. In effect, the DEFINE condition
999            is always false. There may be only one alternative in such a group.
1000    
1001        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1002            as the simple (?(R). The condition is true only if the most recent
1003            recursion is that of the given number or name. It does not search out
1004            through the entire recursion stack.
1005    
1006        (g) The escape \gN or \g{N} has been added, where N is a positive or
1007            negative number, specifying an absolute or relative reference.
1008    
1009    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1010        some "unreachable code" warnings.
1011    
1012    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1013        things, this adds five new scripts.
1014    
1015    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1016        There were also incompatibilities regarding the handling of \Q..\E inside
1017        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1018        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1019    
1020    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1021        matches an empty string, and forcibly breaks the loop. There were bugs in
1022        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1023        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1024        separate and independent bugs (that affected different cases) have been
1025        fixed.
1026    
1027    39. Refactored the code to abolish the use of different opcodes for small
1028        capturing bracket numbers. This is a tidy that I avoided doing when I
1029        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1030        The new approach is not only tidier, it makes it possible to reduce the
1031        memory needed to fix the previous bug (38).
1032    
1033    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1034        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1035        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1036        mode.
1037    
1038    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1039        report.
1040    
1041    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1042        copy construction and assignment in the C++ wrapper.
1043    
1044    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1045        couple of bugs that could have given wrong results in the "--newline=crlf"
1046        case.
1047    
1048    44. Added a number of casts and did some reorganization of signed/unsigned int
1049        variables following suggestions from Dair Grant. Also renamed the variable
1050        "this" as "item" because it is a C++ keyword.
1051    
1052    45. Arranged for dftables to add
1053    
1054          #include "pcre_internal.h"
1055    
1056        to pcre_chartables.c because without it, gcc 4.x may remove the array
1057        definition from the final binary if PCRE is built into a static library and
1058        dead code stripping is activated.
1059    
1060    46. For an unanchored pattern, if a match attempt fails at the start of a
1061        newline sequence, and the newline setting is CRLF or ANY, and the next two
1062        characters are CRLF, advance by two characters instead of one.
1063    
1064    
1065    Version 6.7 04-Jul-06
1066    ---------------------
1067    
1068     1. In order to handle tests when input lines are enormously long, pcretest has
1069        been re-factored so that it automatically extends its buffers when
1070        necessary. The code is crude, but this _is_ just a test program. The
1071        default size has been increased from 32K to 50K.
1072    
1073     2. The code in pcre_study() was using the value of the re argument before
1074        testing it for NULL. (Of course, in any sensible call of the function, it
1075        won't be NULL.)
1076    
1077     3. The memmove() emulation function in pcre_internal.h, which is used on
1078        systems that lack both memmove() and bcopy() - that is, hardly ever -
1079        was missing a "static" storage class specifier.
1080    
1081     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1082        containing an extended class (one that cannot be represented by a bitmap
1083        because it contains high-valued characters or Unicode property items, e.g.
1084        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1085        pattern, but PCRE should not loop if you do not (it no longer does).
1086        [Detail: two cases were found: (a) a repeated subpattern containing an
1087        extended class; (b) a recursive reference to a subpattern that followed a
1088        previous extended class. It wasn't skipping over the extended class
1089        correctly when UTF-8 mode was not set.]
1090    
1091     5. A negated single-character class was not being recognized as fixed-length
1092        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1093        compile error "lookbehind assertion is not fixed length".
1094    
1095     6. The RunPerlTest auxiliary script was showing an unexpected difference
1096        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1097        write a Perl script that can interpret lines of an input file either as
1098        byte characters or as UTF-8, which is what "perltest" was being required to
1099        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1100        can't do is switch easily at run time between having the "use utf8;" pragma
1101        or not. In the end, I fudged it by using the RunPerlTest script to insert
1102        "use utf8;" explicitly for the UTF-8 tests.
1103    
1104     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1105        the end of the subject string, contrary to the documentation and to what
1106        Perl does. This was true of both matching functions. Now it matches only at
1107        the start of the subject and immediately after *internal* newlines.
1108    
1109     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1110        a pointer to an int instead of a pointer to an unsigned long int. This
1111        caused problems on 64-bit systems.
1112    
1113     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1114        instance of the 'standard' template library not being so standard".
1115    
1116    10. There was no check on the number of named subpatterns nor the maximum
1117        length of a subpattern name. The product of these values is used to compute
1118        the size of the memory block for a compiled pattern. By supplying a very
1119        long subpattern name and a large number of named subpatterns, the size
1120        computation could be caused to overflow. This is now prevented by limiting
1121        the length of names to 32 characters, and the number of named subpatterns
1122        to 10,000.
1123    
1124    11. Subpatterns that are repeated with specific counts have to be replicated in
1125        the compiled pattern. The size of memory for this was computed from the
1126        length of the subpattern and the repeat count. The latter is limited to
1127        65535, but there was no limit on the former, meaning that integer overflow
1128        could in principle occur. The compiled length of a repeated subpattern is
1129        now limited to 30,000 bytes in order to prevent this.
1130    
1131    12. Added the optional facility to have named substrings with the same name.
1132    
1133    13. Added the ability to use a named substring as a condition, using the
1134        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1135        are numbers (not recommended). Forward references are permitted.
1136    
1137    14. Added forward references in named backreferences (if you see what I mean).
1138    
1139    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1140        pattern could run off the end of the subject. For example, the pattern
1141        "(?s)(.{1,5})"8 did this with the subject "ab".
1142    
1143    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1144        PCRE_CASELESS was set when matching characters that were quantified with ?
1145        or *.
1146    
1147    17. A character class other than a single negated character that had a minimum
1148        but no maximum quantifier - for example [ab]{6,} - was not handled
1149        correctly by pce_dfa_exec(). It would match only one character.
1150    
1151    18. A valid (though odd) pattern that looked like a POSIX character
1152        class but used an invalid character after [ (for example [[,abc,]]) caused
1153        pcre_compile() to give the error "Failed: internal error: code overflow" or
1154        in some cases to crash with a glibc free() error. This could even happen if
1155        the pattern terminated after [[ but there just happened to be a sequence of
1156        letters, a binary zero, and a closing ] in the memory that followed.
1157    
1158    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1159        over the years. Originally (before any Unicode support), just the bottom 8
1160        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1161        output from "man perlunicode" includes this:
1162    
1163          The regular expression compiler produces polymorphic opcodes.  That
1164          is, the pattern adapts to the data and automatically switches to
1165          the Unicode character scheme when presented with Unicode data--or
1166          instead uses a traditional byte scheme when presented with byte
1167          data.
1168    
1169        Sadly, a wide octal escape does not cause a switch, and in a string with
1170        no other multibyte characters, these octal escapes are treated as before.
1171        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1172        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1173        Unicode string.
1174    
1175        I have not perpetrated such confusion in PCRE. Up till now, it took just
1176        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1177        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1178        translate to the appropriate multibyte character.
1179    
1180    29. Applied some refactoring to reduce the number of warnings from Microsoft
1181        and Borland compilers. This has included removing the fudge introduced
1182        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1183        a warning about an unused variable.
1184    
1185    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1186        characters since release 4.0, because Perl (from release 5.004) does not.
1187        [Or at least, is documented not to: some releases seem to be in conflict
1188        with the documentation.] However, when a pattern was studied with
1189        pcre_study() and all its branches started with \s, PCRE still included VT
1190        as a possible starting character. Of course, this did no harm; it just
1191        caused an unnecessary match attempt.
1192    
1193    22. Removed a now-redundant internal flag bit that recorded the fact that case
1194        dependency changed within the pattern. This was once needed for "required
1195        byte" processing, but is no longer used. This recovers a now-scarce options
1196        bit. Also moved the least significant internal flag bit to the most-
1197        significant bit of the word, which was not previously used (hangover from
1198        the days when it was an int rather than a uint) to free up another bit for
1199        the future.
1200    
1201    23. Added support for CRLF line endings as well as CR and LF. As well as the
1202        default being selectable at build time, it can now be changed at runtime
1203        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1204        specify that it is scanning data with non-default line endings.
1205    
1206    24. Changed the definition of CXXLINK to make it agree with the definition of
1207        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1208    
1209    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1210        recursions. This makes a big different to stack usage for some patterns.
1211    
1212    26. If a subpattern containing a named recursion or subroutine reference such
1213        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1214        the space required for the compiled pattern went wrong and gave too small a
1215        value. Depending on the environment, this could lead to "Failed: internal
1216        error: code overflow at offset 49" or "glibc detected double free or
1217        corruption" errors.
1218    
1219    27. Applied patches from Google (a) to support the new newline modes and (b) to
1220        advance over multibyte UTF-8 characters in GlobalReplace.
1221    
1222    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1223        difference for some implementation of PCRE in some Windows version.
1224    
1225    29. Added some extra testing facilities to pcretest:
1226    
1227        \q<number>   in a data line sets the "match limit" value
1228        \Q<number>   in a data line sets the "match recursion limt" value
1229        -S <number>  sets the stack size, where <number> is in megabytes
1230    
1231        The -S option isn't available for Windows.
1232    
1233    
1234    Version 6.6 06-Feb-06
1235    ---------------------
1236    
1237     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1238        in pcreposix.h. I have copied the definition from pcre.h.
1239    
1240     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1241        because pcre.h is no longer a built file.
1242    
1243     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1244        not normally included in the compiled code.
1245    
1246    
1247    Version 6.5 01-Feb-06
1248    ---------------------
1249    
1250     1. When using the partial match feature with pcre_dfa_exec(), it was not
1251        anchoring the second and subsequent partial matches at the new starting
1252        point. This could lead to incorrect results. For example, with the pattern
1253        /1234/, partially matching against "123" and then "a4" gave a match.
1254    
1255     2. Changes to pcregrep:
1256    
1257        (a) All non-match returns from pcre_exec() were being treated as failures
1258            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1259            error message is output. Some extra information is given for the
1260            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1261            probably the only errors that are likely to be caused by users (by
1262            specifying a regex that has nested indefinite repeats, for instance).
1263            If there are more than 20 of these errors, pcregrep is abandoned.
1264    
1265        (b) A binary zero was treated as data while matching, but terminated the
1266            output line if it was written out. This has been fixed: binary zeroes
1267            are now no different to any other data bytes.
1268    
1269        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1270            used to set a locale for matching. The --locale=xxxx long option has
1271            been added (no short equivalent) to specify a locale explicitly on the
1272            pcregrep command, overriding the environment variables.
1273    
1274        (d) When -B was used with -n, some line numbers in the output were one less
1275            than they should have been.
1276    
1277        (e) Added the -o (--only-matching) option.
1278    
1279        (f) If -A or -C was used with -c (count only), some lines of context were
1280            accidentally printed for the final match.
1281    
1282        (g) Added the -H (--with-filename) option.
1283    
1284        (h) The combination of options -rh failed to suppress file names for files
1285            that were found from directory arguments.
1286    
1287        (i) Added the -D (--devices) and -d (--directories) options.
1288    
1289        (j) Added the -F (--fixed-strings) option.
1290    
1291        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1292    
1293        (l) Added the --colo(u)r option.
1294    
1295        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1296            is not present by default.
1297    
1298     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1299        items such as (?R) or (?1), when the recursion could match a number of
1300        alternatives. If it matched one of the alternatives, but subsequently,
1301        outside the recursion, there was a failure, the code tried to back up into
1302        the recursion. However, because of the way PCRE is implemented, this is not
1303        possible, and the result was an incorrect result from the match.
1304    
1305        In order to prevent this happening, the specification of recursion has
1306        been changed so that all such subpatterns are automatically treated as
1307        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1308    
1309     4. I had overlooked the fact that, in some locales, there are characters for
1310        which isalpha() is true but neither isupper() nor islower() are true. In
1311        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1312        and ordfeminine) are like this. This affected the treatment of \w and \W
1313        when they appeared in character classes, but not when they appeared outside
1314        a character class. The bit map for "word" characters is now created
1315        separately from the results of isalnum() instead of just taking it from the
1316        upper, lower, and digit maps. (Plus the underscore character, of course.)
1317    
1318     5. The above bug also affected the handling of POSIX character classes such as
1319        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1320        permanent tables. Instead, the bit maps for such a class were previously
1321        created as the appropriate unions of the upper, lower, and digit bitmaps.
1322        Now they are created by subtraction from the [[:word:]] class, which has
1323        its own bitmap.
1324    
1325     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1326        It is created by subtracting the vertical space characters (\x09, \x0a,
1327        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1328        subtraction was done in the overall bitmap for a character class, meaning
1329        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1330        be recognized. This bug has been fixed.
1331    
1332     7. Patches from the folks at Google:
1333    
1334          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1335          real life, but is still worth protecting against".
1336    
1337          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1338          regular expressions".
1339    
1340          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1341          have it.
1342    
1343          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1344          "configure" and the latter not, in order to fix a problem somebody had
1345          with compiling the Arg class on HP-UX.
1346    
1347          (e) Improve the error-handling of the C++ wrapper a little bit.
1348    
1349          (f) New tests for checking recursion limiting.
1350    
1351     8. The pcre_memmove() function, which is used only if the environment does not
1352        have a standard memmove() function (and is therefore rarely compiled),
1353        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1354        returning a result (though PCRE never actually uses the result).
1355    
1356     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1357        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1358        returned instead of calling malloc() with an overflowing number that would
1359        most likely cause subsequent chaos.
1360    
1361    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1362    
1363    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1364        with this option is matched, the nmatch and pmatch options of regexec() are
1365        ignored.
1366    
1367    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1368        provided in case anyone wants to the the POSIX interface with UTF-8
1369        strings.
1370    
1371    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1372        C++ linking (needed for some HP-UX environments).
1373    
1374    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1375        (unused parameter) and in the pcre_printint() function (omitted "default"
1376        switch label when the default is to do nothing).
1377    
1378    15. Added some code to make it possible, when PCRE is compiled as a C++
1379        library, to replace subject pointers for pcre_exec() with a smart pointer
1380        class, thus making it possible to process discontinuous strings.
1381    
1382    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1383        much the same function. They were added by different people who were trying
1384        to make PCRE easy to compile on non-Unix systems. It has been suggested
1385        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1386        for compiling on Windows systems. I have therefore replaced it with
1387        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1388        defaults to "extern" for C or "extern C" for C++, which works fine on
1389        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1390        SCOPE with something explicit in config.h. In addition:
1391    
1392        (a) pcreposix.h still had just "extern" instead of either of these macros;
1393            I have replaced it with PCRE_DATA_SCOPE.
1394    
1395        (b) Functions such as _pcre_xclass(), which are internal to the library,
1396            but external in the C sense, all had PCRE_EXPORT in their definitions.
1397            This is apparently wrong for the Windows case, so I have removed it.
1398            (It makes no difference on Unix-like systems.)
1399    
1400    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1401        of recursive calls to match(). This is different to MATCH_LIMIT because
1402        that limits the total number of calls to match(), not all of which increase
1403        the depth of recursion. Limiting the recursion depth limits the amount of
1404        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1405        when PCRE is compiled, and changed at run time. A patch from Google adds
1406        this functionality to the C++ interface.
1407    
1408    18. Changes to the handling of Unicode character properties:
1409    
1410        (a) Updated the table to Unicode 4.1.0.
1411    
1412        (b) Recognize characters that are not in the table as "Cn" (undefined).
1413    
1414        (c) I revised the way the table is implemented to a much improved format
1415            which includes recognition of ranges. It now supports the ranges that
1416            are defined in UnicodeData.txt, and it also amalgamates other
1417            characters into ranges. This has reduced the number of entries in the
1418            table from around 16,000 to around 3,000, thus reducing its size
1419            considerably. I realized I did not need to use a tree structure after
1420            all - a binary chop search is just as efficient. Having reduced the
1421            number of entries, I extended their size from 6 bytes to 8 bytes to
1422            allow for more data.
1423    
1424        (d) Added support for Unicode script names via properties such as \p{Han}.
1425    
1426    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1427        matching that character.
1428    
1429    20. When matching a repeated Unicode property with a minimum greater than zero,
1430        (for example \pL{2,}), PCRE could look past the end of the subject if it
1431        reached it while seeking the minimum number of characters. This could
1432        happen only if some of the characters were more than one byte long, because
1433        there is a check for at least the minimum number of bytes.
1434    
1435    21. Refactored the implementation of \p and \P so as to be more general, to
1436        allow for more different types of property in future. This has changed the
1437        compiled form incompatibly. Anybody with saved compiled patterns that use
1438        \p or \P will have to recompile them.
1439    
1440    22. Added "Any" and "L&" to the supported property types.
1441    
1442    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1443        but give a compile time error if the value is greater than 0xff.
1444    
1445    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1446        accidentally not being installed or uninstalled.
1447    
1448    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1449        made were to insert the current release number. This seemed silly, because
1450        it made things harder for people building PCRE on systems that don't run
1451        "configure". I have turned pcre.h into a distributed file, no longer built
1452        by "configure", with the version identification directly included. There is
1453        no longer a pcre.h.in file.
1454    
1455        However, this change necessitated a change to the pcre-config script as
1456        well. It is built from pcre-config.in, and one of the substitutions was the
1457        release number. I have updated configure.ac so that ./configure now finds
1458        the release number by grepping pcre.h.
1459    
1460    26. Added the ability to run the tests under valgrind.
1461    
1462    
1463    Version 6.4 05-Sep-05
1464    ---------------------
1465    
1466     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1467        "--" to be printed when multiple files were scanned, even when none of the
1468        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1469        consider it to be a bug, and have restored the previous behaviour.
1470    
1471     2. A couple of code tidies to get rid of compiler warnings.
1472    
1473     3. The pcretest program used to cheat by referring to symbols in the library
1474        whose names begin with _pcre_. These are internal symbols that are not
1475        really supposed to be visible externally, and in some environments it is
1476        possible to suppress them. The cheating is now confined to including
1477        certain files from the library's source, which is a bit cleaner.
1478    
1479     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1480        file's purpose clearer.
1481    
1482     5. Reorganized pcre_ucp_findchar().
1483    
1484    
1485    Version 6.3 15-Aug-05
1486    ---------------------
1487    
1488     1. The file libpcre.pc.in did not have general read permission in the tarball.
1489    
1490     2. There were some problems when building without C++ support:
1491    
1492        (a) If C++ support was not built, "make install" and "make test" still
1493            tried to test it.
1494    
1495        (b) There were problems when the value of CXX was explicitly set. Some
1496            changes have been made to try to fix these, and ...
1497    
1498        (c) --disable-cpp can now be used to explicitly disable C++ support.
1499    
1500        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1501            backslash in a target when C++ was disabled. This confuses some
1502            versions of "make", apparently. Using an intermediate variable solves
1503            this. (Same for CPP_LOBJ.)
1504    
1505     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1506        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1507        necessary on certain architectures.
1508    
1509     4. Added a setting of -export-symbols-regex to the link command to remove
1510        those symbols that are exported in the C sense, but actually are local
1511        within the library, and not documented. Their names all begin with
1512        "_pcre_". This is not a perfect job, because (a) we have to except some
1513        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1514        available (and never for static libraries). I have made a note to try to
1515        find a way round (a) in the future.
1516    
1517    
1518  Version 6.2 01-Aug-05  Version 6.2 01-Aug-05
1519  ---------------------  ---------------------
1520    

Legend:
Removed from v.81  
changed lines
  Added in v.357

  ViewVC Help
Powered by ViewVC 1.1.5