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

Diff of /code/trunk/ChangeLog

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

revision 31 by nigel, Sat Feb 24 21:38:57 2007 UTC revision 376 by ph10, Sun Mar 1 12:00:59 2009 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.9 xx-xxx-09
5    ---------------------
6    
7    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
8        (pcretest), all targets were linked against these libraries. This included
9        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
10        libraries. This caused unwanted dependencies to be created. This problem
11        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
12        pcretest is linked with readline.
13    
14    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
15        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
16        moved outside it again, because FALSE and TRUE are already defined in AIX,
17        but BOOL is not.
18    
19    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
20        PCRE_MATCH_LIMIT_RETURSION values as ints, when they should be long ints.
21    
22    4.  The pcregrep documentation said spaces were inserted as well as colons (or
23        hyphens) following file names and line numbers when outputting matching
24        lines. This is not true; no spaces are inserted. I have also clarified the
25        wording for the --colour (or --color) option.
26    
27    
28    Version 7.8 05-Sep-08
29    ---------------------
30    
31    1.  Replaced UCP searching code with optimized version as implemented for Ad
32        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
33        stage table and inline lookup instead of a function, giving speed ups of 2
34        to 5 times on some simple patterns that I tested. Permission was given to
35        distribute the MultiStage2.py script that generates the tables (it's not in
36        the tarball, but is in the Subversion repository).
37    
38    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
39        scripts.
40    
41    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
42        a group with a zero qualifier. The result of the study could be incorrect,
43        or the function might crash, depending on the pattern.
44    
45    4.  Caseless matching was not working for non-ASCII characters in back
46        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
47        It now works when Unicode Property Support is available.
48    
49    5.  In pcretest, an escape such as \x{de} in the data was always generating
50        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
51        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
52        truncation.
53    
54    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
55    
56    7.  Added two (int) casts to pcregrep when printing the difference of two
57        pointers, in case they are 64-bit values.
58    
59    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
60        test 2 if it fails.
61    
62    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
63        and a #define of that name to empty if it is not externally set. This is to
64        allow users of MSVC to set it if necessary.
65    
66    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
67        the convenience functions in the pcre_get.c source file.
68    
69    11. An option change at the start of a pattern that had top-level alternatives
70        could cause overwriting and/or a crash. This command provoked a crash in
71        some environments:
72    
73          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
74    
75        This potential security problem was recorded as CVE-2008-2371.
76    
77    12. For a pattern where the match had to start at the beginning or immediately
78        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
79        pcre_dfa_exec() could read past the end of the passed subject if there was
80        no match. To help with detecting such bugs (e.g. with valgrind), I modified
81        pcretest so that it places the subject at the end of its malloc-ed buffer.
82    
83    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
84        exec() might read past the end of the data buffer in UTF-8 mode.
85    
86    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
87        the data contained the byte 0x85 as part of a UTF-8 character within its
88        first line. This applied both to normal and DFA matching.
89    
90    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
91        /^[^d]*?$/8 failed to match "abc".
92    
93    16. Added a missing copyright notice to pcrecpp_internal.h.
94    
95    17. Make it more clear in the documentation that values returned from
96        pcre_exec() in ovector are byte offsets, not character counts.
97    
98    18. Tidied a few places to stop certain compilers from issuing warnings.
99    
100    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
101        supplied by Stefan Weber. I made a further small update for 7.8 because
102        there is a change of source arrangements: the pcre_searchfuncs.c module is
103        replaced by pcre_ucd.c.
104    
105    
106    Version 7.7 07-May-08
107    ---------------------
108    
109    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
110        a string to a long long, pretend we don't even have a long long." This is
111        done by checking for the strtoq, strtoll, and _strtoi64 functions.
112    
113    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
114        pre-7.6 versions, which defined a global no_arg variable instead of putting
115        it in the RE class. (See also #8 below.)
116    
117    3.  Remove a line of dead code, identified by coverity and reported by Nuno
118        Lopes.
119    
120    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
121    
122        (1) The include/exclude patterns were being applied to the whole pathnames
123            of files, instead of just to the final components.
124    
125        (2) If there was more than one level of directory, the subdirectories were
126            skipped unless they satisfied the include/exclude conditions. This is
127            inconsistent with GNU grep (and could even be seen as contrary to the
128            pcregrep specification - which I improved to make it absolutely clear).
129            The action now is always to scan all levels of directory, and just
130            apply the include/exclude patterns to regular files.
131    
132    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
133        --exclude_dir in the tests to avoid scanning .svn directories.
134    
135    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
136        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
137        doesn't support NULs in patterns.
138    
139    7.  Added some missing "const"s to declarations of static tables in
140        pcre_compile.c and pcre_dfa_exec.c.
141    
142    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
143        caused by fix #2  above. (Subsequently also a second patch to fix the
144        first patch. And a third patch - this was a messy problem.)
145    
146    9.  Applied Craig's patch to remove the use of push_back().
147    
148    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
149        matching function regexec().
150    
151    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
152        which, however, unlike Perl's \g{...}, are subroutine calls, not back
153        references. PCRE supports relative numbers with this syntax (I don't think
154        Oniguruma does).
155    
156    12. Previously, a group with a zero repeat such as (...){0} was completely
157        omitted from the compiled regex. However, this means that if the group
158        was called as a subroutine from elsewhere in the pattern, things went wrong
159        (an internal error was given). Such groups are now left in the compiled
160        pattern, with a new opcode that causes them to be skipped at execution
161        time.
162    
163    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
164        to the way PCRE behaves:
165    
166        (a) A lone ] character is dis-allowed (Perl treats it as data).
167    
168        (b) A back reference to an unmatched subpattern matches an empty string
169            (Perl fails the current match path).
170    
171        (c) A data ] in a character class must be notated as \] because if the
172            first data character in a class is ], it defines an empty class. (In
173            Perl it is not possible to have an empty class.) The empty class []
174            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
175            The negative empty class [^] matches any one character, independently
176            of the DOTALL setting.
177    
178    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
179        non-existent subpattern following a character class starting with ']' and
180        containing () gave an internal compiling error instead of "reference to
181        non-existent subpattern". Fortunately, when the pattern did exist, the
182        compiled code was correct. (When scanning forwards to check for the
183        existencd of the subpattern, it was treating the data ']' as terminating
184        the class, so got the count wrong. When actually compiling, the reference
185        was subsequently set up correctly.)
186    
187    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
188        it was being rejected as not supported by pcre_dfa_exec(), even though
189        other assertions are supported. I have made pcre_dfa_exec() support
190        (*FAIL).
191    
192    16. The implementation of 13c above involved the invention of a new opcode,
193        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
194        cannot be changed at match time, I realized I could make a small
195        improvement to matching performance by compiling OP_ALLANY instead of
196        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
197        on the OP_ANY path.
198    
199    17. Compiling pcretest on Windows with readline support failed without the
200        following two fixes: (1) Make the unistd.h include conditional on
201        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
202    
203    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
204        ncurses library to be included for pcretest when ReadLine support is
205        requested, but also to allow for it to be overridden. This patch came from
206        Daniel Bergström.
207    
208    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
209        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
210        any errors with the current Unicode tables. Thanks to Peter Kankowski for
211        spotting this.
212    
213    
214    Version 7.6 28-Jan-08
215    ---------------------
216    
217    1.  A character class containing a very large number of characters with
218        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
219        overflow.
220    
221    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
222        HAVE_LONG_LONG is not defined.
223    
224    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
225        bring them up to date and include new features. This patch includes:
226    
227        - Fixed PH's badly added libz and libbz2 support.
228        - Fixed a problem with static linking.
229        - Added pcredemo. [But later removed - see 7 below.]
230        - Fixed dftables problem and added an option.
231        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
232            HAVE_LONG_LONG.
233        - Added readline support for pcretest.
234        - Added an listing of the option settings after cmake has run.
235    
236    4.  A user submitted a patch to Makefile that makes it easy to create
237        "pcre.dll" under mingw when using Configure/Make. I added stuff to
238        Makefile.am that cause it to include this special target, without
239        affecting anything else. Note that the same mingw target plus all
240        the other distribution libraries and programs are now supported
241        when configuring with CMake (see 6 below) instead of with
242        Configure/Make.
243    
244    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
245        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
246        exported in the Windows port". It has not yet been confirmed that the patch
247        solves the problem, but it does no harm.
248    
249    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
250        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
251        with CMake, and also correct the comment about stack recursion.
252    
253    7.  Remove the automatic building of pcredemo from the ./configure system and
254        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
255        of a program that users should build themselves after PCRE is installed, so
256        building it automatically is not really right. What is more, it gave
257        trouble in some build environments.
258    
259    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
260    
261    
262    Version 7.5 10-Jan-08
263    ---------------------
264    
265    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
266        values in parens when parsing an RE using the C++ wrapper."
267    
268    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
269        Characters greater than 255 were excluded from the class instead of being
270        included.
271    
272    3.  The same bug as (2) above applied to negated POSIX classes such as
273        [:^space:].
274    
275    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
276        defined or documented. It seems to have been a typo for PCRE_STATIC, so
277        I have changed it.
278    
279    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
280        first named subpattern) and a construct such as (?&a) would reference the
281        first named subpattern whose name started with "a" (in other words, the
282        length check was missing). Both these problems are fixed. "Subpattern name
283        expected" is now given for (?&) (a zero-length name), and this patch also
284        makes it give the same error for \k'' (previously it complained that that
285        was a reference to a non-existent subpattern).
286    
287    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
288        this is right because (?- can be followed by option settings as well as by
289        digits. I have, however, made the messages clearer.
290    
291    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
292        than the number used in the conditional) now cause a compile-time error.
293        This is actually not compatible with Perl, which accepts such patterns, but
294        treats the conditional as always being FALSE (as PCRE used to), but it
295        seems to me that giving a diagnostic is better.
296    
297    8.  Change "alphameric" to the more common word "alphanumeric" in comments
298        and messages.
299    
300    9.  Fix two occurrences of "backslash" in comments that should have been
301        "backspace".
302    
303    10. Remove two redundant lines of code that can never be obeyed (their function
304        was moved elsewhere).
305    
306    11. The program that makes PCRE's Unicode character property table had a bug
307        which caused it to generate incorrect table entries for sequences of
308        characters that have the same character type, but are in different scripts.
309        It amalgamated them into a single range, with the script of the first of
310        them. In other words, some characters were in the wrong script. There were
311        thirteen such cases, affecting characters in the following ranges:
312    
313          U+002b0 - U+002c1
314          U+0060c - U+0060d
315          U+0061e - U+00612
316          U+0064b - U+0065e
317          U+0074d - U+0076d
318          U+01800 - U+01805
319          U+01d00 - U+01d77
320          U+01d9b - U+01dbf
321          U+0200b - U+0200f
322          U+030fc - U+030fe
323          U+03260 - U+0327f
324          U+0fb46 - U+0fbb1
325          U+10450 - U+1049d
326    
327    12. The -o option (show only the matching part of a line) for pcregrep was not
328        compatible with GNU grep in that, if there was more than one match in a
329        line, it showed only the first of them. It now behaves in the same way as
330        GNU grep.
331    
332    13. If the -o and -v options were combined for pcregrep, it printed a blank
333        line for every non-matching line. GNU grep prints nothing, and pcregrep now
334        does the same. The return code can be used to tell if there were any
335        non-matching lines.
336    
337    14. Added --file-offsets and --line-offsets to pcregrep.
338    
339    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
340        infinitely looping recursion. The bug was that positive lookaheads were not
341        being skipped when checking for a possible empty match (negative lookaheads
342        and both kinds of lookbehind were skipped).
343    
344    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
345        inclusion of <windows.h> to before rather than after the definition of
346        INVALID_FILE_ATTRIBUTES (patch from David Byron).
347    
348    17. Specifying a possessive quantifier with a specific limit for a Unicode
349        character property caused pcre_compile() to compile bad code, which led at
350        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
351        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
352        caused the error; without that there was no problem.
353    
354    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
355    
356    19. Added --enable-pcretest-libreadline.
357    
358    20. In pcrecpp.cc, the variable 'count' was incremented twice in
359        RE::GlobalReplace(). As a result, the number of replacements returned was
360        double what it should be. I removed one of the increments, but Craig sent a
361        later patch that removed the other one (the right fix) and added unit tests
362        that check the return values (which was not done before).
363    
364    21. Several CMake things:
365    
366        (1) Arranged that, when cmake is used on Unix, the libraries end up with
367            the names libpcre and libpcreposix, not just pcre and pcreposix.
368    
369        (2) The above change means that pcretest and pcregrep are now correctly
370            linked with the newly-built libraries, not previously installed ones.
371    
372        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
373    
374    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
375        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
376        UTF-8 newline character). The key issue is that the pattern starts .*;
377        this means that the match must be either at the beginning, or after a
378        newline. The bug was in the code for advancing after a failed match and
379        checking that the new position followed a newline. It was not taking
380        account of UTF-8 characters correctly.
381    
382    23. PCRE was behaving differently from Perl in the way it recognized POSIX
383        character classes. PCRE was not treating the sequence [:...:] as a
384        character class unless the ... were all letters. Perl, however, seems to
385        allow any characters between [: and :], though of course it rejects as
386        unknown any "names" that contain non-letters, because all the known class
387        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
388        for example, whereas PCRE did not - it did not recognize a POSIX character
389        class. This seemed a bit dangerous, so the code has been changed to be
390        closer to Perl. The behaviour is not identical to Perl, because PCRE will
391        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
392        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
393        Perl does, and where it didn't before.
394    
395    24. Rewrite so as to remove the single use of %n from pcregrep because in some
396        Windows environments %n is disabled by default.
397    
398    
399    Version 7.4 21-Sep-07
400    ---------------------
401    
402    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
403        means that a class such as [\s] counted as "explicit reference to CR or
404        LF". That isn't really right - the whole point of the change was to try to
405        help when there was an actual mention of one of the two characters. So now
406        the change happens only if \r or \n (or a literal CR or LF) character is
407        encountered.
408    
409    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
410        of both had grown to the point where there were only 3 bits left.
411        Fortunately, there was spare space in the data structure, and so I have
412        moved the internal flags into a new 16-bit field to free up more option
413        bits.
414    
415    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
416        but did not set the internal JCHANGED flag - either of these is enough to
417        control the way the "get" function works - but the PCRE_INFO_JCHANGED
418        facility is supposed to tell if (?J) was ever used, so now (?J) at the
419        start sets both bits.
420    
421    4.  Added options (at build time, compile time, exec time) to change \R from
422        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
423    
424    5.  doc/pcresyntax.html was missing from the distribution.
425    
426    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
427        compatibility, even though it is no longer used.
428    
429    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
430        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
431        windows.h file is present (where different names are used). [This was
432        reversed later after testing - see 16 below.]
433    
434    8.  Changed all #include <config.h> to #include "config.h". There were also
435        some further <pcre.h> cases that I changed to "pcre.h".
436    
437    9.  When pcregrep was used with the --colour option, it missed the line ending
438        sequence off the lines that it output.
439    
440    10. It was pointed out to me that arrays of string pointers cause lots of
441        relocations when a shared library is dynamically loaded. A technique of
442        using a single long string with a table of offsets can drastically reduce
443        these. I have refactored PCRE in four places to do this. The result is
444        dramatic:
445    
446          Originally:                          290
447          After changing UCP table:            187
448          After changing error message table:   43
449          After changing table of "verbs"       36
450          After changing table of Posix names   22
451    
452        Thanks to the folks working on Gregex for glib for this insight.
453    
454    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
455        unicode-properties was also set.
456    
457    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
458    
459    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
460        checked only for CRLF.
461    
462    14. Added casts to pcretest.c to avoid compiler warnings.
463    
464    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
465    
466    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
467        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
468        entirely. This removes changes made in 7 above.
469    
470    17. The CMake files have been updated, and there is now more information about
471        building with CMake in the NON-UNIX-USE document.
472    
473    
474    Version 7.3 28-Aug-07
475    ---------------------
476    
477     1. In the rejigging of the build system that eventually resulted in 7.1, the
478        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
479        brackets there is not right, since it causes compilers to look for an
480        installed pcre.h, not the version that is in the source that is being
481        compiled (which of course may be different). I have changed it back to:
482    
483          #include "pcre.h"
484    
485        I have a vague recollection that the change was concerned with compiling in
486        different directories, but in the new build system, that is taken care of
487        by the VPATH setting the Makefile.
488    
489     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
490        when the subject happened to end in the byte 0x85 (e.g. if the last
491        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
492        characters but of course it shouldn't be taken as a newline when it is part
493        of another character. The bug was that, for an unlimited repeat of . in
494        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
495        characters when looking for a newline.
496    
497     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
498    
499     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
500        in debug output.
501    
502     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
503        long printing in the pcrecpp unittest when running under MinGW.
504    
505     6. ESC_K was left out of the EBCDIC table.
506    
507     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
508        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
509        limit also applies to "virtual nesting" when a pattern is recursive, and in
510        this case 1000 isn't so big. I have been able to remove this limit at the
511        expense of backing off one optimization in certain circumstances. Normally,
512        when pcre_exec() would call its internal match() function recursively and
513        immediately return the result unconditionally, it uses a "tail recursion"
514        feature to save stack. However, when a subpattern that can match an empty
515        string has an unlimited repetition quantifier, it no longer makes this
516        optimization. That gives it a stack frame in which to save the data for
517        checking that an empty string has been matched. Previously this was taken
518        from the 1000-entry workspace that had been reserved. So now there is no
519        explicit limit, but more stack is used.
520    
521     8. Applied Daniel's patches to solve problems with the import/export magic
522        syntax that is required for Windows, and which was going wrong for the
523        pcreposix and pcrecpp parts of the library. These were overlooked when this
524        problem was solved for the main library.
525    
526     9. There were some crude static tests to avoid integer overflow when computing
527        the size of patterns that contain repeated groups with explicit upper
528        limits. As the maximum quantifier is 65535, the maximum group length was
529        set at 30,000 so that the product of these two numbers did not overflow a
530        32-bit integer. However, it turns out that people want to use groups that
531        are longer than 30,000 bytes (though not repeat them that many times).
532        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
533        made it possible to implement the integer overflow checks in a much more
534        dynamic way, which I have now done. The artificial limitation on group
535        length has been removed - we now have only the limit on the total length of
536        the compiled pattern, which depends on the LINK_SIZE setting.
537    
538    10. Fixed a bug in the documentation for get/copy named substring when
539        duplicate names are permitted. If none of the named substrings are set, the
540        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
541        empty string.
542    
543    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
544        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
545        because the ] is interpreted as the first data character and the
546        terminating ] is not found. PCRE has been made compatible with Perl in this
547        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
548        cause memory overwriting.
549    
550    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
551        string has been matched (to stop an infinite loop). It was not recognizing
552        a conditional subpattern that could match an empty string if that
553        subpattern was within another subpattern. For example, it looped when
554        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
555        condition was not nested. This bug has been fixed.
556    
557    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
558        past the start of the subject in the presence of bytes with the top bit
559        set, for example "\x8aBCD".
560    
561    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
562        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
563    
564    14. Optimized (?!) to (*FAIL).
565    
566    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
567        This restricts code points to be within the range 0 to 0x10FFFF, excluding
568        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
569        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
570        does: it's just the validity check that is more restrictive.
571    
572    16. Inserted checks for integer overflows during escape sequence (backslash)
573        processing, and also fixed erroneous offset values for syntax errors during
574        backslash processing.
575    
576    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
577        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
578    
579    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
580        caused an overrun.
581    
582    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
583        something other than just ASCII characters) inside a group that had an
584        unlimited repeat caused a loop at compile time (while checking to see
585        whether the group could match an empty string).
586    
587    20. Debugging a pattern containing \p or \P could cause a crash. For example,
588        [\P{Any}] did so. (Error in the code for printing property names.)
589    
590    21. An orphan \E inside a character class could cause a crash.
591    
592    22. A repeated capturing bracket such as (A)? could cause a wild memory
593        reference during compilation.
594    
595    23. There are several functions in pcre_compile() that scan along a compiled
596        expression for various reasons (e.g. to see if it's fixed length for look
597        behind). There were bugs in these functions when a repeated \p or \P was
598        present in the pattern. These operators have additional parameters compared
599        with \d, etc, and these were not being taken into account when moving along
600        the compiled data. Specifically:
601    
602        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
603            length.
604    
605        (b) An item such as \pL+ within a repeated group could cause crashes or
606            loops.
607    
608        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
609            "reference to non-existent subpattern" error.
610    
611        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
612    
613    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
614        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
615    
616    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
617    
618    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
619        character were causing crashes (broken optimization).
620    
621    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
622        \p or \P) caused a compile-time loop.
623    
624    28. More problems have arisen in unanchored patterns when CRLF is a valid line
625        break. For example, the unstudied pattern [\r\n]A does not match the string
626        "\r\nA" because change 7.0/46 below moves the current point on by two
627        characters after failing to match at the start. However, the pattern \nA
628        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
629        the same is true. There doesn't seem any very clean way out of this, but
630        what I have chosen to do makes the common cases work: PCRE now takes note
631        of whether there can be an explicit match for \r or \n anywhere in the
632        pattern, and if so, 7.0/46 no longer applies. As part of this change,
633        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
634        pattern has explicit CR or LF references.
635    
636    29. Added (*CR) etc for changing newline setting at start of pattern.
637    
638    
639    Version 7.2 19-Jun-07
640    ---------------------
641    
642     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
643        which is apparently normally available under Windows.
644    
645     2. Re-jig the pcregrep tests with different newline settings in an attempt
646        to make them independent of the local environment's newline setting.
647    
648     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
649    
650     4. Some of the "internals" tests were previously cut out when the link size
651        was not 2, because the output contained actual offsets. The recent new
652        "Z" feature of pcretest means that these can be cut out, making the tests
653        usable with all link sizes.
654    
655     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
656        stack recursion. This gives a massive performance boost under BSD, but just
657        a small improvement under Linux. However, it saves one field in the frame
658        in all cases.
659    
660     6. Added more features from the forthcoming Perl 5.10:
661    
662        (a) (?-n) (where n is a string of digits) is a relative subroutine or
663            recursion call. It refers to the nth most recently opened parentheses.
664    
665        (b) (?+n) is also a relative subroutine call; it refers to the nth next
666            to be opened parentheses.
667    
668        (c) Conditions that refer to capturing parentheses can be specified
669            relatively, for example, (?(-2)... or (?(+3)...
670    
671        (d) \K resets the start of the current match so that everything before
672            is not part of it.
673    
674        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
675    
676        (f) \g{name} is another synonym - part of Perl 5.10's unification of
677            reference syntax.
678    
679        (g) (?| introduces a group in which the numbering of parentheses in each
680            alternative starts with the same number.
681    
682        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
683    
684     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
685        PCRE_INFO_JCHANGED.
686    
687     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
688        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
689        for detecting groups that can match an empty string.
690    
691     9. A pattern with a very large number of alternatives (more than several
692        hundred) was running out of internal workspace during the pre-compile
693        phase, where pcre_compile() figures out how much memory will be needed. A
694        bit of new cunning has reduced the workspace needed for groups with
695        alternatives. The 1000-alternative test pattern now uses 12 bytes of
696        workspace instead of running out of the 4096 that are available.
697    
698    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
699    
700    11. Applied patch from Google to remove an optimization that didn't quite work.
701        The report of the bug said:
702    
703          pcrecpp::RE("a*").FullMatch("aaa") matches, while
704          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
705          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
706    
707    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
708        it matched the wrong number of bytes.
709    
710    
711    Version 7.1 24-Apr-07
712    ---------------------
713    
714     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
715        that is more "standard", making use of automake and other Autotools. There
716        is some re-arrangement of the files and adjustment of comments consequent
717        on this.
718    
719     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
720        for recursive directory scanning broke on some systems because the files
721        are not scanned in any specific order and on different systems the order
722        was different. A call to "sort" has been inserted into RunGrepTest for the
723        approprate test as a short-term fix. In the longer term there may be an
724        alternative.
725    
726     3. I had an email from Eric Raymond about problems translating some of PCRE's
727        man pages to HTML (despite the fact that I distribute HTML pages, some
728        people do their own conversions for various reasons). The problems
729        concerned the use of low-level troff macros .br and .in. I have therefore
730        removed all such uses from the man pages (some were redundant, some could
731        be replaced by .nf/.fi pairs). The 132html script that I use to generate
732        HTML has been updated to handle .nf/.fi and to complain if it encounters
733        .br or .in.
734    
735     4. Updated comments in configure.ac that get placed in config.h.in and also
736        arranged for config.h to be included in the distribution, with the name
737        config.h.generic, for the benefit of those who have to compile without
738        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
739    
740     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
741        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
742        makevp.bat for the current PCRE, using the additional files
743        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
744    
745     6. A Windows user reported a minor discrepancy with test 2, which turned out
746        to be caused by a trailing space on an input line that had got lost in his
747        copy. The trailing space was an accident, so I've just removed it.
748    
749     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
750        that is needed.
751    
752     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
753        as "const" (a) because they are and (b) because it helps the PHP
754        maintainers who have recently made a script to detect big data structures
755        in the php code that should be moved to the .rodata section. I remembered
756        to update Builducptable as well, so it won't revert if ucptable.h is ever
757        re-created.
758    
759     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
760        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
761        order to be able to cut out the UTF-8 tables in the latter when UTF-8
762        support is not required. This saves 1.5-2K of code, which is important in
763        some applications.
764    
765        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
766        so as not to refer to the tables, even though these functions will never be
767        called when UTF-8 support is disabled. Otherwise there are problems with a
768        shared library.
769    
770    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
771    
772        (a) It was defining its arguments as char * instead of void *.
773    
774        (b) It was assuming that all moves were upwards in memory; this was true
775            a long time ago when I wrote it, but is no longer the case.
776    
777        The emulated memove() is provided for those environments that have neither
778        memmove() nor bcopy(). I didn't think anyone used it these days, but that
779        is clearly not the case, as these two bugs were recently reported.
780    
781    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
782        and Detrail to create the HTML documentation, the .txt form of the man
783        pages, and it removes trailing spaces from listed files. It also creates
784        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
785        case, it wraps all the #defines with #ifndefs. This script should be run
786        before "make dist".
787    
788    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
789        with Unicode property support.
790    
791        (a) For a maximizing quantifier, if the two different cases of the
792            character were of different lengths in their UTF-8 codings (there are
793            some cases like this - I found 11), and the matching function had to
794            back up over a mixture of the two cases, it incorrectly assumed they
795            were both the same length.
796    
797        (b) When PCRE was configured to use the heap rather than the stack for
798            recursion during matching, it was not correctly preserving the data for
799            the other case of a UTF-8 character when checking ahead for a match
800            while processing a minimizing repeat. If the check also involved
801            matching a wide character, but failed, corruption could cause an
802            erroneous result when trying to check for a repeat of the original
803            character.
804    
805    13. Some tidying changes to the testing mechanism:
806    
807        (a) The RunTest script now detects the internal link size and whether there
808            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
809            values substituted by "configure". (The RunGrepTest script already did
810            this for UTF-8.) The configure.ac script no longer substitutes the
811            relevant variables.
812    
813        (b) The debugging options /B and /D in pcretest show the compiled bytecode
814            with length and offset values. This means that the output is different
815            for different internal link sizes. Test 2 is skipped for link sizes
816            other than 2 because of this, bypassing the problem. Unfortunately,
817            there was also a test in test 3 (the locale tests) that used /B and
818            failed for link sizes other than 2. Rather than cut the whole test out,
819            I have added a new /Z option to pcretest that replaces the length and
820            offset values with spaces. This is now used to make test 3 independent
821            of link size. (Test 2 will be tidied up later.)
822    
823    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
824        segmentation fault instead of returning the appropriate error message.
825    
826    15. In multiline mode when the newline sequence was set to "any", the pattern
827        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
828        This doesn't seem right; it now treats the CRLF combination as the line
829        ending, and so does not match in that case. It's only a pattern such as ^$
830        that would hit this one: something like ^ABC$ would have failed after \r
831        and then tried again after \r\n.
832    
833    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
834        in an attempt to make files that differ only in their line terminators
835        compare equal. This works on Linux.
836    
837    17. Under certain error circumstances pcregrep might try to free random memory
838        as it exited. This is now fixed, thanks to valgrind.
839    
840    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
841        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
842        was because its rules for how to advance for /g after matching an empty
843        string at the end of a line did not allow for this case. They now check for
844        it specially.
845    
846    20. pcretest is supposed to handle patterns and data of any length, by
847        extending its buffers when necessary. It was getting this wrong when the
848        buffer for a data line had to be extended.
849    
850    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
851        CRLF as a newline sequence.
852    
853    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
854        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
855        I have nevertheless tidied it up.
856    
857    23. Added some casts to kill warnings from HP-UX ia64 compiler.
858    
859    24. Added a man page for pcre-config.
860    
861    
862    Version 7.0 19-Dec-06
863    ---------------------
864    
865     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
866        moving to gcc 4.1.1.
867    
868     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
869        sys/time.h, which is documented as needed for this function. It doesn't
870        seem to matter on Linux, but it showed up on some releases of OS X.
871    
872     3. It seems that there are systems where bytes whose values are greater than
873        127 match isprint() in the "C" locale. The "C" locale should be the
874        default when a C program starts up. In most systems, only ASCII printing
875        characters match isprint(). This difference caused the output from pcretest
876        to vary, making some of the tests fail. I have changed pcretest so that:
877    
878        (a) When it is outputting text in the compiled version of a pattern, bytes
879            other than 32-126 are always shown as hex escapes.
880    
881        (b) When it is outputting text that is a matched part of a subject string,
882            it does the same, unless a different locale has been set for the match
883            (using the /L modifier). In this case, it uses isprint() to decide.
884    
885     4. Fixed a major bug that caused incorrect computation of the amount of memory
886        required for a compiled pattern when options that changed within the
887        pattern affected the logic of the preliminary scan that determines the
888        length. The relevant options are -x, and -i in UTF-8 mode. The result was
889        that the computed length was too small. The symptoms of this bug were
890        either the PCRE error "internal error: code overflow" from pcre_compile(),
891        or a glibc crash with a message such as "pcretest: free(): invalid next
892        size (fast)". Examples of patterns that provoked this bug (shown in
893        pcretest format) are:
894    
895          /(?-x: )/x
896          /(?x)(?-x: \s*#\s*)/
897          /((?i)[\x{c0}])/8
898          /(?i:[\x{c0}])/8
899    
900        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
901        is now done differently.
902    
903     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
904        wrapper classes; (b) implement a new function in the C++ scanner that is
905        more efficient than the old way of doing things because it avoids levels of
906        recursion in the regex matching; (c) add a paragraph to the documentation
907        for the FullMatch() function.
908    
909     6. The escape sequence \n was being treated as whatever was defined as
910        "newline". Not only was this contrary to the documentation, which states
911        that \n is character 10 (hex 0A), but it also went horribly wrong when
912        "newline" was defined as CRLF. This has been fixed.
913    
914     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
915        was being set to -1 for the "end of line" case (supposedly a value that no
916        character can have). Though this value is never used (the check for end of
917        line is "zero bytes in current character"), it caused compiler complaints.
918        I've changed it to 0xffffffff.
919    
920     8. In pcre_version.c, the version string was being built by a sequence of
921        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
922        string (as it is for production releases) called a macro with an empty
923        argument. The C standard says the result of this is undefined. The gcc
924        compiler treats it as an empty string (which was what was wanted) but it is
925        reported that Visual C gives an error. The source has been hacked around to
926        avoid this problem.
927    
928     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
929        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
930        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
931        of them did).
932    
933    10. Originally, pcretest opened its input and output without "b"; then I was
934        told that "b" was needed in some environments, so it was added for release
935        5.0 to both the input and output. (It makes no difference on Unix-like
936        systems.) Later I was told that it is wrong for the input on Windows. I've
937        now abstracted the modes into two macros, to make it easier to fiddle with
938        them, and removed "b" from the input mode under Windows.
939    
940    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
941    
942    12. Added -help and --help to pcretest as an official way of being reminded
943        of the options.
944    
945    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
946        and pcrecpp.cc because they annoy compilers at high warning levels.
947    
948    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
949    
950    15. Fixed an occurrence of == in configure.ac that should have been = (shell
951        scripts are not C programs :-) and which was not noticed because it works
952        on Linux.
953    
954    16. pcretest is supposed to handle any length of pattern and data line (as one
955        line or as a continued sequence of lines) by extending its input buffer if
956        necessary. This feature was broken for very long pattern lines, leading to
957        a string of junk being passed to pcre_compile() if the pattern was longer
958        than about 50K.
959    
960    17. I have done a major re-factoring of the way pcre_compile() computes the
961        amount of memory needed for a compiled pattern. Previously, there was code
962        that made a preliminary scan of the pattern in order to do this. That was
963        OK when PCRE was new, but as the facilities have expanded, it has become
964        harder and harder to keep it in step with the real compile phase, and there
965        have been a number of bugs (see for example, 4 above). I have now found a
966        cunning way of running the real compile function in a "fake" mode that
967        enables it to compute how much memory it would need, while actually only
968        ever using a few hundred bytes of working memory and without too many
969        tests of the mode. This should make future maintenance and development
970        easier. A side effect of this work is that the limit of 200 on the nesting
971        depth of parentheses has been removed (though this was never a serious
972        limitation, I suspect). However, there is a downside: pcre_compile() now
973        runs more slowly than before (30% or more, depending on the pattern). I
974        hope this isn't a big issue. There is no effect on runtime performance.
975    
976    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
977        newline (only possible for the last line of a file) and it was a
978        pattern that set a locale (followed by /Lsomething), pcretest crashed.
979    
980    19. Added additional timing features to pcretest. (1) The -tm option now times
981        matching only, not compiling. (2) Both -t and -tm can be followed, as a
982        separate command line item, by a number that specifies the number of
983        repeats to use when timing. The default is 50000; this gives better
984        precision, but takes uncomfortably long for very large patterns.
985    
986    20. Extended pcre_study() to be more clever in cases where a branch of a
987        subpattern has no definite first character. For example, (a*|b*)[cd] would
988        previously give no result from pcre_study(). Now it recognizes that the
989        first character must be a, b, c, or d.
990    
991    21. There was an incorrect error "recursive call could loop indefinitely" if
992        a subpattern (or the entire pattern) that was being tested for matching an
993        empty string contained only one non-empty item after a nested subpattern.
994        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
995        incorrectly, because the \d was being skipped in the check.
996    
997    22. The pcretest program now has a new pattern option /B and a command line
998        option -b, which is equivalent to adding /B to every pattern. This causes
999        it to show the compiled bytecode, without the additional information that
1000        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1001        is the same as /B/I).
1002    
1003    23. A new optimization is now able automatically to treat some sequences such
1004        as a*b as a*+b. More specifically, if something simple (such as a character
1005        or a simple class like \d) has an unlimited quantifier, and is followed by
1006        something that cannot possibly match the quantified thing, the quantifier
1007        is automatically "possessified".
1008    
1009    24. A recursive reference to a subpattern whose number was greater than 39
1010        went wrong under certain circumstances in UTF-8 mode. This bug could also
1011        have affected the operation of pcre_study().
1012    
1013    25. Realized that a little bit of performance could be had by replacing
1014        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1015    
1016    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1017    
1018    27. Possessive quantifiers such as a++ were previously implemented by turning
1019        them into atomic groups such as ($>a+). Now they have their own opcodes,
1020        which improves performance. This includes the automatically created ones
1021        from 23 above.
1022    
1023    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1024        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1025        the first matched character to be a colon. This applied both to named and
1026        numbered groups.
1027    
1028    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1029    
1030    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1031        building PCRE on Windows easier, so I have included it in the distribution.
1032    
1033    31. There is now a check in pcretest against a ridiculously large number being
1034        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1035        loop, the loop is abandoned.
1036    
1037    32. Forward references to subpatterns in conditions such as (?(2)...) where
1038        subpattern 2 is defined later cause pcre_compile() to search forwards in
1039        the pattern for the relevant set of parentheses. This search went wrong
1040        when there were unescaped parentheses in a character class, parentheses
1041        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1042    
1043    33. "Subroutine" calls and backreferences were previously restricted to
1044        referencing subpatterns earlier in the regex. This restriction has now
1045        been removed.
1046    
1047    34. Added a number of extra features that are going to be in Perl 5.10. On the
1048        whole, these are just syntactic alternatives for features that PCRE had
1049        previously implemented using the Python syntax or my own invention. The
1050        other formats are all retained for compatibility.
1051    
1052        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1053            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1054            also .NET compatible.
1055    
1056        (b) A recursion or subroutine call to a named group can now be defined as
1057            (?&name) as well as (?P>name).
1058    
1059        (c) A backreference to a named group can now be defined as \k<name> or
1060            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1061            5.10, are also .NET compatible.
1062    
1063        (d) A conditional reference to a named group can now use the syntax
1064            (?(<name>) or (?('name') as well as (?(name).
1065    
1066        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1067            groups (named and numbered) that are never evaluated inline, but can be
1068            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1069            is always false. There may be only one alternative in such a group.
1070    
1071        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1072            as the simple (?(R). The condition is true only if the most recent
1073            recursion is that of the given number or name. It does not search out
1074            through the entire recursion stack.
1075    
1076        (g) The escape \gN or \g{N} has been added, where N is a positive or
1077            negative number, specifying an absolute or relative reference.
1078    
1079    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1080        some "unreachable code" warnings.
1081    
1082    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1083        things, this adds five new scripts.
1084    
1085    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1086        There were also incompatibilities regarding the handling of \Q..\E inside
1087        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1088        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1089    
1090    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1091        matches an empty string, and forcibly breaks the loop. There were bugs in
1092        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1093        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1094        separate and independent bugs (that affected different cases) have been
1095        fixed.
1096    
1097    39. Refactored the code to abolish the use of different opcodes for small
1098        capturing bracket numbers. This is a tidy that I avoided doing when I
1099        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1100        The new approach is not only tidier, it makes it possible to reduce the
1101        memory needed to fix the previous bug (38).
1102    
1103    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1104        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1105        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1106        mode.
1107    
1108    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1109        report.
1110    
1111    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1112        copy construction and assignment in the C++ wrapper.
1113    
1114    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1115        couple of bugs that could have given wrong results in the "--newline=crlf"
1116        case.
1117    
1118    44. Added a number of casts and did some reorganization of signed/unsigned int
1119        variables following suggestions from Dair Grant. Also renamed the variable
1120        "this" as "item" because it is a C++ keyword.
1121    
1122    45. Arranged for dftables to add
1123    
1124          #include "pcre_internal.h"
1125    
1126        to pcre_chartables.c because without it, gcc 4.x may remove the array
1127        definition from the final binary if PCRE is built into a static library and
1128        dead code stripping is activated.
1129    
1130    46. For an unanchored pattern, if a match attempt fails at the start of a
1131        newline sequence, and the newline setting is CRLF or ANY, and the next two
1132        characters are CRLF, advance by two characters instead of one.
1133    
1134    
1135    Version 6.7 04-Jul-06
1136    ---------------------
1137    
1138     1. In order to handle tests when input lines are enormously long, pcretest has
1139        been re-factored so that it automatically extends its buffers when
1140        necessary. The code is crude, but this _is_ just a test program. The
1141        default size has been increased from 32K to 50K.
1142    
1143     2. The code in pcre_study() was using the value of the re argument before
1144        testing it for NULL. (Of course, in any sensible call of the function, it
1145        won't be NULL.)
1146    
1147     3. The memmove() emulation function in pcre_internal.h, which is used on
1148        systems that lack both memmove() and bcopy() - that is, hardly ever -
1149        was missing a "static" storage class specifier.
1150    
1151     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1152        containing an extended class (one that cannot be represented by a bitmap
1153        because it contains high-valued characters or Unicode property items, e.g.
1154        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1155        pattern, but PCRE should not loop if you do not (it no longer does).
1156        [Detail: two cases were found: (a) a repeated subpattern containing an
1157        extended class; (b) a recursive reference to a subpattern that followed a
1158        previous extended class. It wasn't skipping over the extended class
1159        correctly when UTF-8 mode was not set.]
1160    
1161     5. A negated single-character class was not being recognized as fixed-length
1162        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1163        compile error "lookbehind assertion is not fixed length".
1164    
1165     6. The RunPerlTest auxiliary script was showing an unexpected difference
1166        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1167        write a Perl script that can interpret lines of an input file either as
1168        byte characters or as UTF-8, which is what "perltest" was being required to
1169        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1170        can't do is switch easily at run time between having the "use utf8;" pragma
1171        or not. In the end, I fudged it by using the RunPerlTest script to insert
1172        "use utf8;" explicitly for the UTF-8 tests.
1173    
1174     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1175        the end of the subject string, contrary to the documentation and to what
1176        Perl does. This was true of both matching functions. Now it matches only at
1177        the start of the subject and immediately after *internal* newlines.
1178    
1179     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1180        a pointer to an int instead of a pointer to an unsigned long int. This
1181        caused problems on 64-bit systems.
1182    
1183     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1184        instance of the 'standard' template library not being so standard".
1185    
1186    10. There was no check on the number of named subpatterns nor the maximum
1187        length of a subpattern name. The product of these values is used to compute
1188        the size of the memory block for a compiled pattern. By supplying a very
1189        long subpattern name and a large number of named subpatterns, the size
1190        computation could be caused to overflow. This is now prevented by limiting
1191        the length of names to 32 characters, and the number of named subpatterns
1192        to 10,000.
1193    
1194    11. Subpatterns that are repeated with specific counts have to be replicated in
1195        the compiled pattern. The size of memory for this was computed from the
1196        length of the subpattern and the repeat count. The latter is limited to
1197        65535, but there was no limit on the former, meaning that integer overflow
1198        could in principle occur. The compiled length of a repeated subpattern is
1199        now limited to 30,000 bytes in order to prevent this.
1200    
1201    12. Added the optional facility to have named substrings with the same name.
1202    
1203    13. Added the ability to use a named substring as a condition, using the
1204        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1205        are numbers (not recommended). Forward references are permitted.
1206    
1207    14. Added forward references in named backreferences (if you see what I mean).
1208    
1209    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1210        pattern could run off the end of the subject. For example, the pattern
1211        "(?s)(.{1,5})"8 did this with the subject "ab".
1212    
1213    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1214        PCRE_CASELESS was set when matching characters that were quantified with ?
1215        or *.
1216    
1217    17. A character class other than a single negated character that had a minimum
1218        but no maximum quantifier - for example [ab]{6,} - was not handled
1219        correctly by pce_dfa_exec(). It would match only one character.
1220    
1221    18. A valid (though odd) pattern that looked like a POSIX character
1222        class but used an invalid character after [ (for example [[,abc,]]) caused
1223        pcre_compile() to give the error "Failed: internal error: code overflow" or
1224        in some cases to crash with a glibc free() error. This could even happen if
1225        the pattern terminated after [[ but there just happened to be a sequence of
1226        letters, a binary zero, and a closing ] in the memory that followed.
1227    
1228    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1229        over the years. Originally (before any Unicode support), just the bottom 8
1230        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1231        output from "man perlunicode" includes this:
1232    
1233          The regular expression compiler produces polymorphic opcodes.  That
1234          is, the pattern adapts to the data and automatically switches to
1235          the Unicode character scheme when presented with Unicode data--or
1236          instead uses a traditional byte scheme when presented with byte
1237          data.
1238    
1239        Sadly, a wide octal escape does not cause a switch, and in a string with
1240        no other multibyte characters, these octal escapes are treated as before.
1241        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1242        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1243        Unicode string.
1244    
1245        I have not perpetrated such confusion in PCRE. Up till now, it took just
1246        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1247        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1248        translate to the appropriate multibyte character.
1249    
1250    29. Applied some refactoring to reduce the number of warnings from Microsoft
1251        and Borland compilers. This has included removing the fudge introduced
1252        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1253        a warning about an unused variable.
1254    
1255    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1256        characters since release 4.0, because Perl (from release 5.004) does not.
1257        [Or at least, is documented not to: some releases seem to be in conflict
1258        with the documentation.] However, when a pattern was studied with
1259        pcre_study() and all its branches started with \s, PCRE still included VT
1260        as a possible starting character. Of course, this did no harm; it just
1261        caused an unnecessary match attempt.
1262    
1263    22. Removed a now-redundant internal flag bit that recorded the fact that case
1264        dependency changed within the pattern. This was once needed for "required
1265        byte" processing, but is no longer used. This recovers a now-scarce options
1266        bit. Also moved the least significant internal flag bit to the most-
1267        significant bit of the word, which was not previously used (hangover from
1268        the days when it was an int rather than a uint) to free up another bit for
1269        the future.
1270    
1271    23. Added support for CRLF line endings as well as CR and LF. As well as the
1272        default being selectable at build time, it can now be changed at runtime
1273        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1274        specify that it is scanning data with non-default line endings.
1275    
1276    24. Changed the definition of CXXLINK to make it agree with the definition of
1277        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1278    
1279    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1280        recursions. This makes a big different to stack usage for some patterns.
1281    
1282    26. If a subpattern containing a named recursion or subroutine reference such
1283        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1284        the space required for the compiled pattern went wrong and gave too small a
1285        value. Depending on the environment, this could lead to "Failed: internal
1286        error: code overflow at offset 49" or "glibc detected double free or
1287        corruption" errors.
1288    
1289    27. Applied patches from Google (a) to support the new newline modes and (b) to
1290        advance over multibyte UTF-8 characters in GlobalReplace.
1291    
1292    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1293        difference for some implementation of PCRE in some Windows version.
1294    
1295    29. Added some extra testing facilities to pcretest:
1296    
1297        \q<number>   in a data line sets the "match limit" value
1298        \Q<number>   in a data line sets the "match recursion limt" value
1299        -S <number>  sets the stack size, where <number> is in megabytes
1300    
1301        The -S option isn't available for Windows.
1302    
1303    
1304    Version 6.6 06-Feb-06
1305    ---------------------
1306    
1307     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1308        in pcreposix.h. I have copied the definition from pcre.h.
1309    
1310     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1311        because pcre.h is no longer a built file.
1312    
1313     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1314        not normally included in the compiled code.
1315    
1316    
1317    Version 6.5 01-Feb-06
1318    ---------------------
1319    
1320     1. When using the partial match feature with pcre_dfa_exec(), it was not
1321        anchoring the second and subsequent partial matches at the new starting
1322        point. This could lead to incorrect results. For example, with the pattern
1323        /1234/, partially matching against "123" and then "a4" gave a match.
1324    
1325     2. Changes to pcregrep:
1326    
1327        (a) All non-match returns from pcre_exec() were being treated as failures
1328            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1329            error message is output. Some extra information is given for the
1330            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1331            probably the only errors that are likely to be caused by users (by
1332            specifying a regex that has nested indefinite repeats, for instance).
1333            If there are more than 20 of these errors, pcregrep is abandoned.
1334    
1335        (b) A binary zero was treated as data while matching, but terminated the
1336            output line if it was written out. This has been fixed: binary zeroes
1337            are now no different to any other data bytes.
1338    
1339        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1340            used to set a locale for matching. The --locale=xxxx long option has
1341            been added (no short equivalent) to specify a locale explicitly on the
1342            pcregrep command, overriding the environment variables.
1343    
1344        (d) When -B was used with -n, some line numbers in the output were one less
1345            than they should have been.
1346    
1347        (e) Added the -o (--only-matching) option.
1348    
1349        (f) If -A or -C was used with -c (count only), some lines of context were
1350            accidentally printed for the final match.
1351    
1352        (g) Added the -H (--with-filename) option.
1353    
1354        (h) The combination of options -rh failed to suppress file names for files
1355            that were found from directory arguments.
1356    
1357        (i) Added the -D (--devices) and -d (--directories) options.
1358    
1359        (j) Added the -F (--fixed-strings) option.
1360    
1361        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1362    
1363        (l) Added the --colo(u)r option.
1364    
1365        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1366            is not present by default.
1367    
1368     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1369        items such as (?R) or (?1), when the recursion could match a number of
1370        alternatives. If it matched one of the alternatives, but subsequently,
1371        outside the recursion, there was a failure, the code tried to back up into
1372        the recursion. However, because of the way PCRE is implemented, this is not
1373        possible, and the result was an incorrect result from the match.
1374    
1375        In order to prevent this happening, the specification of recursion has
1376        been changed so that all such subpatterns are automatically treated as
1377        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1378    
1379     4. I had overlooked the fact that, in some locales, there are characters for
1380        which isalpha() is true but neither isupper() nor islower() are true. In
1381        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1382        and ordfeminine) are like this. This affected the treatment of \w and \W
1383        when they appeared in character classes, but not when they appeared outside
1384        a character class. The bit map for "word" characters is now created
1385        separately from the results of isalnum() instead of just taking it from the
1386        upper, lower, and digit maps. (Plus the underscore character, of course.)
1387    
1388     5. The above bug also affected the handling of POSIX character classes such as
1389        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1390        permanent tables. Instead, the bit maps for such a class were previously
1391        created as the appropriate unions of the upper, lower, and digit bitmaps.
1392        Now they are created by subtraction from the [[:word:]] class, which has
1393        its own bitmap.
1394    
1395     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1396        It is created by subtracting the vertical space characters (\x09, \x0a,
1397        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1398        subtraction was done in the overall bitmap for a character class, meaning
1399        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1400        be recognized. This bug has been fixed.
1401    
1402     7. Patches from the folks at Google:
1403    
1404          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1405          real life, but is still worth protecting against".
1406    
1407          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1408          regular expressions".
1409    
1410          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1411          have it.
1412    
1413          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1414          "configure" and the latter not, in order to fix a problem somebody had
1415          with compiling the Arg class on HP-UX.
1416    
1417          (e) Improve the error-handling of the C++ wrapper a little bit.
1418    
1419          (f) New tests for checking recursion limiting.
1420    
1421     8. The pcre_memmove() function, which is used only if the environment does not
1422        have a standard memmove() function (and is therefore rarely compiled),
1423        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1424        returning a result (though PCRE never actually uses the result).
1425    
1426     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1427        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1428        returned instead of calling malloc() with an overflowing number that would
1429        most likely cause subsequent chaos.
1430    
1431    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1432    
1433    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1434        with this option is matched, the nmatch and pmatch options of regexec() are
1435        ignored.
1436    
1437    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1438        provided in case anyone wants to the the POSIX interface with UTF-8
1439        strings.
1440    
1441    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1442        C++ linking (needed for some HP-UX environments).
1443    
1444    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1445        (unused parameter) and in the pcre_printint() function (omitted "default"
1446        switch label when the default is to do nothing).
1447    
1448    15. Added some code to make it possible, when PCRE is compiled as a C++
1449        library, to replace subject pointers for pcre_exec() with a smart pointer
1450        class, thus making it possible to process discontinuous strings.
1451    
1452    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1453        much the same function. They were added by different people who were trying
1454        to make PCRE easy to compile on non-Unix systems. It has been suggested
1455        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1456        for compiling on Windows systems. I have therefore replaced it with
1457        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1458        defaults to "extern" for C or "extern C" for C++, which works fine on
1459        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1460        SCOPE with something explicit in config.h. In addition:
1461    
1462        (a) pcreposix.h still had just "extern" instead of either of these macros;
1463            I have replaced it with PCRE_DATA_SCOPE.
1464    
1465        (b) Functions such as _pcre_xclass(), which are internal to the library,
1466            but external in the C sense, all had PCRE_EXPORT in their definitions.
1467            This is apparently wrong for the Windows case, so I have removed it.
1468            (It makes no difference on Unix-like systems.)
1469    
1470    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1471        of recursive calls to match(). This is different to MATCH_LIMIT because
1472        that limits the total number of calls to match(), not all of which increase
1473        the depth of recursion. Limiting the recursion depth limits the amount of
1474        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1475        when PCRE is compiled, and changed at run time. A patch from Google adds
1476        this functionality to the C++ interface.
1477    
1478    18. Changes to the handling of Unicode character properties:
1479    
1480        (a) Updated the table to Unicode 4.1.0.
1481    
1482        (b) Recognize characters that are not in the table as "Cn" (undefined).
1483    
1484        (c) I revised the way the table is implemented to a much improved format
1485            which includes recognition of ranges. It now supports the ranges that
1486            are defined in UnicodeData.txt, and it also amalgamates other
1487            characters into ranges. This has reduced the number of entries in the
1488            table from around 16,000 to around 3,000, thus reducing its size
1489            considerably. I realized I did not need to use a tree structure after
1490            all - a binary chop search is just as efficient. Having reduced the
1491            number of entries, I extended their size from 6 bytes to 8 bytes to
1492            allow for more data.
1493    
1494        (d) Added support for Unicode script names via properties such as \p{Han}.
1495    
1496    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1497        matching that character.
1498    
1499    20. When matching a repeated Unicode property with a minimum greater than zero,
1500        (for example \pL{2,}), PCRE could look past the end of the subject if it
1501        reached it while seeking the minimum number of characters. This could
1502        happen only if some of the characters were more than one byte long, because
1503        there is a check for at least the minimum number of bytes.
1504    
1505    21. Refactored the implementation of \p and \P so as to be more general, to
1506        allow for more different types of property in future. This has changed the
1507        compiled form incompatibly. Anybody with saved compiled patterns that use
1508        \p or \P will have to recompile them.
1509    
1510    22. Added "Any" and "L&" to the supported property types.
1511    
1512    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1513        but give a compile time error if the value is greater than 0xff.
1514    
1515    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1516        accidentally not being installed or uninstalled.
1517    
1518    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1519        made were to insert the current release number. This seemed silly, because
1520        it made things harder for people building PCRE on systems that don't run
1521        "configure". I have turned pcre.h into a distributed file, no longer built
1522        by "configure", with the version identification directly included. There is
1523        no longer a pcre.h.in file.
1524    
1525        However, this change necessitated a change to the pcre-config script as
1526        well. It is built from pcre-config.in, and one of the substitutions was the
1527        release number. I have updated configure.ac so that ./configure now finds
1528        the release number by grepping pcre.h.
1529    
1530    26. Added the ability to run the tests under valgrind.
1531    
1532    
1533    Version 6.4 05-Sep-05
1534    ---------------------
1535    
1536     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1537        "--" to be printed when multiple files were scanned, even when none of the
1538        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1539        consider it to be a bug, and have restored the previous behaviour.
1540    
1541     2. A couple of code tidies to get rid of compiler warnings.
1542    
1543     3. The pcretest program used to cheat by referring to symbols in the library
1544        whose names begin with _pcre_. These are internal symbols that are not
1545        really supposed to be visible externally, and in some environments it is
1546        possible to suppress them. The cheating is now confined to including
1547        certain files from the library's source, which is a bit cleaner.
1548    
1549     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1550        file's purpose clearer.
1551    
1552     5. Reorganized pcre_ucp_findchar().
1553    
1554    
1555    Version 6.3 15-Aug-05
1556    ---------------------
1557    
1558     1. The file libpcre.pc.in did not have general read permission in the tarball.
1559    
1560     2. There were some problems when building without C++ support:
1561    
1562        (a) If C++ support was not built, "make install" and "make test" still
1563            tried to test it.
1564    
1565        (b) There were problems when the value of CXX was explicitly set. Some
1566            changes have been made to try to fix these, and ...
1567    
1568        (c) --disable-cpp can now be used to explicitly disable C++ support.
1569    
1570        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1571            backslash in a target when C++ was disabled. This confuses some
1572            versions of "make", apparently. Using an intermediate variable solves
1573            this. (Same for CPP_LOBJ.)
1574    
1575     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1576        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1577        necessary on certain architectures.
1578    
1579     4. Added a setting of -export-symbols-regex to the link command to remove
1580        those symbols that are exported in the C sense, but actually are local
1581        within the library, and not documented. Their names all begin with
1582        "_pcre_". This is not a perfect job, because (a) we have to except some
1583        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1584        available (and never for static libraries). I have made a note to try to
1585        find a way round (a) in the future.
1586    
1587    
1588    Version 6.2 01-Aug-05
1589    ---------------------
1590    
1591     1. There was no test for integer overflow of quantifier values. A construction
1592        such as {1111111111111111} would give undefined results. What is worse, if
1593        a minimum quantifier for a parenthesized subpattern overflowed and became
1594        negative, the calculation of the memory size went wrong. This could have
1595        led to memory overwriting.
1596    
1597     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1598    
1599     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1600        operating environments where this matters.
1601    
1602     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1603        PCRE options from within the C++ wrapper.
1604    
1605     5. Named capturing subpatterns were not being correctly counted when a pattern
1606        was compiled. This caused two problems: (a) If there were more than 100
1607        such subpatterns, the calculation of the memory needed for the whole
1608        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1609        back references of the form \12, where the number was greater than 9, were
1610        not recognized as back references, even though there were sufficient
1611        previous subpatterns.
1612    
1613     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1614        versions of gcc, e.g. 2.95.4.
1615    
1616    
1617    Version 6.1 21-Jun-05
1618    ---------------------
1619    
1620     1. There was one reference to the variable "posix" in pcretest.c that was not
1621        surrounded by "#if !defined NOPOSIX".
1622    
1623     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1624        the cross-check on the old pcre_info() function, for the benefit of the
1625        cut-down version of PCRE that is currently imported into Exim.
1626    
1627     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1628        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1629        patterns that start (?i) but I don't think that matters. The use of (?i) is
1630        just an example; this all applies to the other options as well.
1631    
1632     4. Since libtool seems to echo the compile commands it is issuing, the output
1633        from "make" can be reduced a bit by putting "@" in front of each libtool
1634        compile command.
1635    
1636     5. Patch from the folks at Google for configure.in to be a bit more thorough
1637        in checking for a suitable C++ installation before trying to compile the
1638        C++ stuff. This should fix a reported problem when a compiler was present,
1639        but no suitable headers.
1640    
1641     6. The man pages all had just "PCRE" as their title. I have changed them to
1642        be the relevant file name. I have also arranged that these names are
1643        retained in the file doc/pcre.txt, which is a concatenation in text format
1644        of all the man pages except the little individual ones for each function.
1645    
1646     7. The NON-UNIX-USE file had not been updated for the different set of source
1647        files that come with release 6. I also added a few comments about the C++
1648        wrapper.
1649    
1650    
1651    Version 6.0 07-Jun-05
1652    ---------------------
1653    
1654     1. Some minor internal re-organization to help with my DFA experiments.
1655    
1656     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1657        didn't matter for the library itself when fully configured, but did matter
1658        when compiling without UCP support, or within Exim, where the ucp files are
1659        not imported.
1660    
1661     3. Refactoring of the library code to split up the various functions into
1662        different source modules. The addition of the new DFA matching code (see
1663        below) to a single monolithic source would have made it really too
1664        unwieldy, quite apart from causing all the code to be include in a
1665        statically linked application, when only some functions are used. This is
1666        relevant even without the DFA addition now that patterns can be compiled in
1667        one application and matched in another.
1668    
1669        The downside of splitting up is that there have to be some external
1670        functions and data tables that are used internally in different modules of
1671        the library but which are not part of the API. These have all had their
1672        names changed to start with "_pcre_" so that they are unlikely to clash
1673        with other external names.
1674    
1675     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1676        a different (DFA) algorithm. Although it is slower than the original
1677        function, it does have some advantages for certain types of matching
1678        problem.
1679    
1680     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1681        including restarting after a partial match.
1682    
1683     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1684        defined when compiling for Windows was sent to me. I have put it into the
1685        code, though I have no means of testing or verifying it.
1686    
1687     7. Added the pcre_refcount() auxiliary function.
1688    
1689     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1690        match before or at the first newline in the subject string. In pcretest,
1691        the /f option on a pattern can be used to set this.
1692    
1693     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1694        would behave wrongly. This has been present in PCRE since release 4.0.
1695    
1696    10. A number of changes to the pcregrep command:
1697    
1698        (a) Refactored how -x works; insert ^(...)$ instead of setting
1699            PCRE_ANCHORED and checking the length, in preparation for adding
1700            something similar for -w.
1701    
1702        (b) Added the -w (match as a word) option.
1703    
1704        (c) Refactored the way lines are read and buffered so as to have more
1705            than one at a time available.
1706    
1707        (d) Implemented a pcregrep test script.
1708    
1709        (e) Added the -M (multiline match) option. This allows patterns to match
1710            over several lines of the subject. The buffering ensures that at least
1711            8K, or the rest of the document (whichever is the shorter) is available
1712            for matching (and similarly the previous 8K for lookbehind assertions).
1713    
1714        (f) Changed the --help output so that it now says
1715    
1716              -w, --word-regex(p)
1717    
1718            instead of two lines, one with "regex" and the other with "regexp"
1719            because that confused at least one person since the short forms are the
1720            same. (This required a bit of code, as the output is generated
1721            automatically from a table. It wasn't just a text change.)
1722    
1723        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1724            option but starts with a hyphen. Could be a pattern or a path name
1725            starting with a hyphen, for instance.
1726    
1727        (h) "-" can be given as a file name to represent stdin.
1728    
1729        (i) When file names are being printed, "(standard input)" is used for
1730            the standard input, for compatibility with GNU grep. Previously
1731            "<stdin>" was used.
1732    
1733        (j) The option --label=xxx can be used to supply a name to be used for
1734            stdin when file names are being printed. There is no short form.
1735    
1736        (k) Re-factored the options decoding logic because we are going to add
1737            two more options that take data. Such options can now be given in four
1738            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1739    
1740        (l) Added the -A, -B, and -C options for requesting that lines of context
1741            around matches be printed.
1742    
1743        (m) Added the -L option to print the names of files that do not contain
1744            any matching lines, that is, the complement of -l.
1745    
1746        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1747            continue to scan other files.
1748    
1749        (o) The -s option was incorrectly implemented. For compatibility with other
1750            greps, it now suppresses the error message for a non-existent or non-
1751            accessible file (but not the return code). There is a new option called
1752            -q that suppresses the output of matching lines, which was what -s was
1753            previously doing.
1754    
1755        (p) Added --include and --exclude options to specify files for inclusion
1756            and exclusion when recursing.
1757    
1758    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1759        Hopefully, it now does.
1760    
1761    12. Missing cast in pcre_study().
1762    
1763    13. Added an "uninstall" target to the makefile.
1764    
1765    14. Replaced "extern" in the function prototypes in Makefile.in with
1766        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1767        world, but is set differently for Windows.
1768    
1769    15. Added a second compiling function called pcre_compile2(). The only
1770        difference is that it has an extra argument, which is a pointer to an
1771        integer error code. When there is a compile-time failure, this is set
1772        non-zero, in addition to the error test pointer being set to point to an
1773        error message. The new argument may be NULL if no error number is required
1774        (but then you may as well call pcre_compile(), which is now just a
1775        wrapper). This facility is provided because some applications need a
1776        numeric error indication, but it has also enabled me to tidy up the way
1777        compile-time errors are handled in the POSIX wrapper.
1778    
1779    16. Added VPATH=.libs to the makefile; this should help when building with one
1780        prefix path and installing with another. (Or so I'm told by someone who
1781        knows more about this stuff than I do.)
1782    
1783    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1784        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1785        match everything, including newlines. This is not POSIX-compatible, but
1786        somebody wanted the feature. From pcretest it can be activated by using
1787        both the P and the s flags.
1788    
1789    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1790    
1791    19. libpcre.pc was being incorrectly installed as executable.
1792    
1793    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1794        it now also looks for '\r' so that it will work unmodified on Windows.
1795    
1796    21. Added Google's contributed C++ wrapper to the distribution.
1797    
1798    22. Added some untidy missing memory free() calls in pcretest, to keep
1799        Electric Fence happy when testing.
1800    
1801    
1802    
1803    Version 5.0 13-Sep-04
1804    ---------------------
1805    
1806     1. Internal change: literal characters are no longer packed up into items
1807        containing multiple characters in a single byte-string. Each character
1808        is now matched using a separate opcode. However, there may be more than one
1809        byte in the character in UTF-8 mode.
1810    
1811     2. The pcre_callout_block structure has two new fields: pattern_position and
1812        next_item_length. These contain the offset in the pattern to the next match
1813        item, and its length, respectively.
1814    
1815     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1816        insertion of callouts before each pattern item. Added the /C option to
1817        pcretest to make use of this.
1818    
1819     4. On the advice of a Windows user, the lines
1820    
1821          #if defined(_WIN32) || defined(WIN32)
1822          _setmode( _fileno( stdout ), 0x8000 );
1823          #endif  /* defined(_WIN32) || defined(WIN32) */
1824    
1825        have been added to the source of pcretest. This apparently does useful
1826        magic in relation to line terminators.
1827    
1828     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1829        for the benefit of those environments where the "b" makes a difference.
1830    
1831     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1832        to know about it. I have put a hack into configure.in that adds in code
1833        to set GCC=yes if CC=icc. This seems to end up at a point in the
1834        generated configure script that is early enough to affect the setting of
1835        compiler options, which is what is needed, but I have no means of testing
1836        whether it really works. (The user who reported this had patched the
1837        generated configure script, which of course I cannot do.)
1838    
1839        LATER: After change 22 below (new libtool files), the configure script
1840        seems to know about icc (and also ecc). Therefore, I have commented out
1841        this hack in configure.in.
1842    
1843     7. Added support for pkg-config (2 patches were sent in).
1844    
1845     8. Negated POSIX character classes that used a combination of internal tables
1846        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1847        [[:^ascii]]. Typically, they would match almost any characters. The other
1848        POSIX classes were not broken in this way.
1849    
1850     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1851        to find the match, as PCRE was deluded into thinking that the match had to
1852        start at the start point or following a newline. The same bug applied to
1853        patterns with negative forward assertions or any backward assertions
1854        preceding ".*" at the start, unless the pattern required a fixed first
1855        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1856    
1857    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1858        starting at the last subject character, bytes beyond the end of the subject
1859        string were read.
1860    
1861    11. Renamed the variable "class" as "classbits" to make life easier for C++
1862        users. (Previously there was a macro definition, but it apparently wasn't
1863        enough.)
1864    
1865    12. Added the new field "tables" to the extra data so that tables can be passed
1866        in at exec time, or the internal tables can be re-selected. This allows
1867        a compiled regex to be saved and re-used at a later time by a different
1868        program that might have everything at different addresses.
1869    
1870    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1871        -R library as well as a -L library.
1872    
1873    14. The debugging options of pcretest (-d on the command line or D on a
1874        pattern) showed incorrect output for anything following an extended class
1875        that contained multibyte characters and which was followed by a quantifier.
1876    
1877    15. Added optional support for general category Unicode character properties
1878        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1879        support. It adds about 90K to the size of the library. The meanings of the
1880        inbuilt class escapes such as \d and \s have NOT been changed.
1881    
1882    16. Updated pcredemo.c to include calls to free() to release the memory for the
1883        compiled pattern.
1884    
1885    17. The generated file chartables.c was being created in the source directory
1886        instead of in the building directory. This caused the build to fail if the
1887        source directory was different from the building directory, and was
1888        read-only.
1889    
1890    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1891        file. No doubt somebody will tell me if they don't make sense... Also added
1892        Dan Mooney's comments about building on OpenVMS.
1893    
1894    19. Added support for partial matching via the PCRE_PARTIAL option for
1895        pcre_exec() and the \P data escape in pcretest.
1896    
1897    20. Extended pcretest with 3 new pattern features:
1898    
1899        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1900              write the compiled pattern to the file whose name is "rest-of-line".
1901              This is a straight binary dump of the data, with the saved pointer to
1902              the character tables forced to be NULL. The study data, if any, is
1903              written too. After writing, pcretest reads a new pattern.
1904    
1905        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1906              compiled pattern from the given file. There must not be any
1907              occurrences of "<" in the file name (pretty unlikely); if there are,
1908              pcretest will instead treat the initial "<" as a pattern delimiter.
1909              After reading in the pattern, pcretest goes on to read data lines as
1910              usual.
1911    
1912        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1913              and 16-bit fields in a compiled pattern, to simulate a pattern that
1914              was compiled on a host of opposite endianness.
1915    
1916    21. The pcre-exec() function can now cope with patterns that were compiled on
1917        hosts of opposite endianness, with this restriction:
1918    
1919          As for any compiled expression that is saved and used later, the tables
1920          pointer field cannot be preserved; the extra_data field in the arguments
1921          to pcre_exec() should be used to pass in a tables address if a value
1922          other than the default internal tables were used at compile time.
1923    
1924    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1925        now diagnosed as an error. Previously, most of the time, a negative number
1926        would have been treated as zero, but if in addition "ovector" was passed as
1927        NULL, a crash could occur.
1928    
1929    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1930        new versions from the libtool 1.5 distribution (the last one is a copy of
1931        a file called libtool.m4). This seems to have fixed the need to patch
1932        "configure" to support Darwin 1.3 (which I used to do). However, I still
1933        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1934        workstation).
1935    
1936    24. Changed the PCRE licence to be the more standard "BSD" licence.
1937    
1938    
1939    Version 4.5 01-Dec-03
1940    ---------------------
1941    
1942     1. There has been some re-arrangement of the code for the match() function so
1943        that it can be compiled in a version that does not call itself recursively.
1944        Instead, it keeps those local variables that need separate instances for
1945        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1946        needs to "recurse". Keeping track of where control must go is done by means
1947        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1948        hide most of the details from the main code, and operates only if
1949        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1950        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1951        operating.
1952    
1953        To make it easier for callers to provide specially tailored get/free
1954        functions for this usage, two new functions, pcre_stack_malloc, and
1955        pcre_stack_free, are used. They are always called in strict stacking order,
1956        and the size of block requested is always the same.
1957    
1958        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1959        PCRE has been compiled to use the stack or the heap for recursion. The
1960        -C option of pcretest uses this to show which version is compiled.
1961    
1962        A new data escape \S, is added to pcretest; it causes the amounts of store
1963        obtained and freed by both kinds of malloc/free at match time to be added
1964        to the output.
1965    
1966     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1967        what's available on my current Linux desktop machine.
1968    
1969     3. When matching a UTF-8 string, the test for a valid string at the start has
1970        been extended. If start_offset is not zero, PCRE now checks that it points
1971        to a byte that is the start of a UTF-8 character. If not, it returns
1972        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1973        this is necessary because there may be backward assertions in the pattern.
1974        When matching the same subject several times, it may save resources to use
1975        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1976    
1977     4. The code for checking the validity of UTF-8 strings has been tightened so
1978        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1979        containing "overlong sequences".
1980    
1981     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1982        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1983        should have been "&", but it just so happened that all the cases this let
1984        through by mistake were picked up later in the function.
1985    
1986     6. I had used a variable called "isblank" - this is a C99 function, causing
1987        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1988    
1989     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1990        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1991        so the version is shown; (c) stop "make test" echoing "./RunTest".
1992    
1993     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1994    
1995     9. The prototype for memmove() for systems that don't have it was using
1996        size_t, but the inclusion of the header that defines size_t was later. I've
1997        moved the #includes for the C headers earlier to avoid this.
1998    
1999    10. Added some adjustments to the code to make it easier to compiler on certain
2000        special systems:
2001    
2002          (a) Some "const" qualifiers were missing.
2003          (b) Added the macro EXPORT before all exported functions; by default this
2004              is defined to be empty.
2005          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2006              that it reads its output file name as an argument instead of writing
2007              to the standard output and assuming this can be redirected.
2008    
2009    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2010        class containing characters with values greater than 255, PCRE compilation
2011        went into a loop.
2012    
2013    12. A recursive reference to a subpattern that was within another subpattern
2014        that had a minimum quantifier of zero caused PCRE to crash. For example,
2015        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2016        recursion. If the recursively-called subpattern itself had a zero repeat,
2017        that was OK.
2018    
2019    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2020        buffer into which it was copied (for escape processing) was still set at
2021        1024, so long lines caused crashes.
2022    
2023    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2024        "internal error: code overflow...". This applied to any character class
2025        that was followed by a possessive quantifier.
2026    
2027    15. Modified the Makefile to add libpcre.la as a prerequisite for
2028        libpcreposix.la because I was told this is needed for a parallel build to
2029        work.
2030    
2031    16. If a pattern that contained .* following optional items at the start was
2032        studied, the wrong optimizing data was generated, leading to matching
2033        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2034        matching string must start with a or b or c. The correct conclusion for
2035        this pattern is that a match can start with any character.
2036    
2037    
2038    Version 4.4 13-Aug-03
2039    ---------------------
2040    
2041     1. In UTF-8 mode, a character class containing characters with values between
2042        127 and 255 was not handled correctly if the compiled pattern was studied.
2043        In fixing this, I have also improved the studying algorithm for such
2044        classes (slightly).
2045    
2046     2. Three internal functions had redundant arguments passed to them. Removal
2047        might give a very teeny performance improvement.
2048    
2049     3. Documentation bug: the value of the capture_top field in a callout is *one
2050        more than* the number of the hightest numbered captured substring.
2051    
2052     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2053        in incorrectly linking with a previously installed version. They now link
2054        explicitly with libpcre.la.
2055    
2056     5. configure.in no longer needs to recognize Cygwin specially.
2057    
2058     6. A problem in pcre.in for Windows platforms is fixed.
2059    
2060     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2061        pcretest, it used to include the size of the study block as part of its
2062        output. Unfortunately, the structure contains a field that has a different
2063        size on different hardware architectures. This meant that the tests that
2064        showed this size failed. As the block is currently always of a fixed size,
2065        this information isn't actually particularly useful in pcretest output, so
2066        I have just removed it.
2067    
2068     8. Three pre-processor statements accidentally did not start in column 1.
2069        Sadly, there are *still* compilers around that complain, even though
2070        standard C has not required this for well over a decade. Sigh.
2071    
2072     9. In pcretest, the code for checking callouts passed small integers in the
2073        callout_data field, which is a void * field. However, some picky compilers
2074        complained about the casts involved for this on 64-bit systems. Now
2075        pcretest passes the address of the small integer instead, which should get
2076        rid of the warnings.
2077    
2078    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2079        both compile and run time, and gives an error if an invalid UTF-8 sequence
2080        is found. There is a option for disabling this check in cases where the
2081        string is known to be correct and/or the maximum performance is wanted.
2082    
2083    11. In response to a bug report, I changed one line in Makefile.in from
2084    
2085            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2086        to
2087            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2088    
2089        to look similar to other lines, but I have no way of telling whether this
2090        is the right thing to do, as I do not use Windows. No doubt I'll get told
2091        if it's wrong...
2092    
2093    
2094    Version 4.3 21-May-03
2095    ---------------------
2096    
2097    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2098       Makefile.
2099    
2100    2. Some refactoring to improve the quality of the code:
2101    
2102       (i)   The utf8_table... variables are now declared "const".
2103    
2104       (ii)  The code for \cx, which used the "case flipping" table to upper case
2105             lower case letters, now just substracts 32. This is ASCII-specific,
2106             but the whole concept of \cx is ASCII-specific, so it seems
2107             reasonable.
2108    
2109       (iii) PCRE was using its character types table to recognize decimal and
2110             hexadecimal digits in the pattern. This is silly, because it handles
2111             only 0-9, a-f, and A-F, but the character types table is locale-
2112             specific, which means strange things might happen. A private
2113             table is now used for this - though it costs 256 bytes, a table is
2114             much faster than multiple explicit tests. Of course, the standard
2115             character types table is still used for matching digits in subject
2116             strings against \d.
2117    
2118       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2119             ending in _t are). So I've renamed it as ESC_tee.
2120    
2121    3. The first argument for regexec() in the POSIX wrapper should have been
2122       defined as "const".
2123    
2124    4. Changed pcretest to use malloc() for its buffers so that they can be
2125       Electric Fenced for debugging.
2126    
2127    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2128       to read one or more bytes before the start of the subject string. Often this
2129       had no effect on PCRE's behaviour, but in some circumstances it could
2130       provoke a segmentation fault.
2131    
2132    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2133       to try to read one or more bytes before the start of the subject string.
2134    
2135    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2136       UTF-8 support could misbehave in various ways if the subject string
2137       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2138       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2139       back over UTF-8 characters.)
2140    
2141    
2142    Version 4.2 14-Apr-03
2143    ---------------------
2144    
2145    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2146    
2147    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2148         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2149         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2150         [WIN_PREFIX]: new variable, "cyg" for Cygwin
2151         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2152           and BUILD_EXEEXT
2153         Note: automatic setting of the BUILD variables is not yet working
2154         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2155           compile-time but not at link-time
2156         [LINK]: use for linking executables only
2157         make different versions for Windows and non-Windows
2158         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2159           libraries
2160         [LINK_FOR_BUILD]: new variable
2161         [OBJEXT]: use throughout
2162         [EXEEXT]: use throughout
2163         <winshared>: new target
2164         <wininstall>: new target
2165         <dftables.o>: use native compiler
2166         <dftables>: use native linker
2167         <install>: handle Windows platform correctly
2168         <clean>: ditto
2169         <check>: ditto
2170         copy DLL to top builddir before testing
2171    
2172       As part of these changes, -no-undefined was removed again. This was reported
2173       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2174       in any case.
2175    
2176    3. Some tidies to get rid of compiler warnings:
2177    
2178       . In the match_data structure, match_limit was an unsigned long int, whereas
2179         match_call_count was an int. I've made them both unsigned long ints.
2180    
2181       . In pcretest the fact that a const uschar * doesn't automatically cast to
2182         a void * provoked a warning.
2183    
2184       . Turning on some more compiler warnings threw up some "shadow" variables
2185         and a few more missing casts.
2186    
2187    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2188       option, a class that contained a single character with a value between 128
2189       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2190    
2191    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2192       option, a class that contained several characters, but with at least one
2193       whose value was between 128 and 255 caused PCRE to crash.
2194    
2195    
2196    Version 4.1 12-Mar-03
2197    ---------------------
2198    
2199    1. Compiling with gcc -pedantic found a couple of places where casts were
2200    needed, and a string in dftables.c that was longer than standard compilers are
2201    required to support.
2202    
2203    2. Compiling with Sun's compiler found a few more places where the code could
2204    be tidied up in order to avoid warnings.
2205    
2206    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2207    first of these names is deprecated in the latest Autoconf in favour of the name
2208    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2209    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2210    analogy I have changed it to CFLAGS_FOR_BUILD.
2211    
2212    4. Added -no-undefined to the linking command in the Makefile, because this is
2213    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2214    linking step for the pcreposix library.
2215    
2216    5. PCRE was failing to diagnose the case of two named groups with the same
2217    name.
2218    
2219    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2220    literal character that is needed in the subject for a match, and scans along to
2221    ensure that it is present before embarking on the full matching process. This
2222    saves time in cases of nested unlimited repeats that are never going to match.
2223    Problem: the scan can take a lot of time if the subject is very long (e.g.
2224    megabytes), thus penalizing straightforward matches. It is now done only if the
2225    amount of subject to be scanned is less than 1000 bytes.
2226    
2227    7. A lesser problem with the same optimization is that it was recording the
2228    first character of an anchored pattern as "needed", thus provoking a search
2229    right along the subject, even when the first match of the pattern was going to
2230    fail. The "needed" character is now not set for anchored patterns, unless it
2231    follows something in the pattern that is of non-fixed length. Thus, it still
2232    fulfils its original purpose of finding quick non-matches in cases of nested
2233    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2234    
2235    
2236    Version 4.0 17-Feb-03
2237    ---------------------
2238    
2239    1. If a comment in an extended regex that started immediately after a meta-item
2240    extended to the end of string, PCRE compiled incorrect data. This could lead to
2241    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2242    
2243    2. Moved to autoconf 2.53 and libtool 1.4.2.
2244    
2245    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2246    the special perltest8 script is no longer needed - all the tests can be run
2247    from a single perltest script.
2248    
2249    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2250    by \s. It has now been removed in PCRE. This means it isn't recognized as
2251    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2252    class [:space:] *does* include VT, thereby creating a mess.
2253    
2254    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2255    space and tab.
2256    
2257    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2258    its new features into the main test script, reducing the number of scripts.
2259    
2260    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2261    were backward compatible, and made the (?i) apply to the whole pattern, as if
2262    /i were given. Now it behaves more logically, and applies the option setting
2263    only to what follows. PCRE has been changed to follow suit. However, if it
2264    finds options settings right at the start of the pattern, it extracts them into
2265    the global options, as before. Thus, they show up in the info data.
2266    
2267    8. Added support for the \Q...\E escape sequence. Characters in between are
2268    treated as literals. This is slightly different from Perl in that $ and @ are
2269    also handled as literals inside the quotes. In Perl, they will cause variable
2270    interpolation. Note the following examples:
2271    
2272        Pattern            PCRE matches      Perl matches
2273    
2274        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
2275        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2276        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2277    
2278    For compatibility with Perl, \Q...\E sequences are recognized inside character
2279    classes as well as outside them.
2280    
2281    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2282    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2283    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2284    signed/unsigned warnings.
2285    
2286    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2287    option for pcretest, so I've replaced it by a simple function that does just
2288    that job.
2289    
2290    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2291    "pcregrep -".
2292    
2293    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2294    Java package. This provides some syntactic sugar for simple cases of what my
2295    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2296    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2297    item, you can use this simplified notation. Note that only makes sense with
2298    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2299    greediness, whatever the setting of the PCRE_UNGREEDY option.
2300    
2301    13. A change of greediness default within a pattern was not taking effect at
2302    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2303    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2304    was abstracted outside.
2305    
2306    14. PCRE now supports the \G assertion. It is true when the current matching
2307    position is at the start point of the match. This differs from \A when the
2308    starting offset is non-zero. Used with the /g option of pcretest (or similar
2309    code), it works in the same way as it does for Perl's /g option. If all
2310    alternatives of a regex begin with \G, the expression is anchored to the start
2311    match position, and the "anchored" flag is set in the compiled expression.
2312    
2313    15. Some bugs concerning the handling of certain option changes within patterns
2314    have been fixed. These applied to options other than (?ims). For example,
2315    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2316    been the other way round. Some of this was related to change 7 above.
2317    
2318    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2319    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2320    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2321    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2322    
2323    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2324    mode. Unlike ".", it always matches newline, whatever the setting of
2325    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2326    assertions. Perl allows it, but it doesn't (in general) work because it can't
2327    calculate the length of the lookbehind. At least, that's the case for Perl
2328    5.8.0 - I've been told they are going to document that it doesn't work in
2329    future.
2330    
2331    18. Added an error diagnosis for escapes that PCRE does not support: these are
2332    \L, \l, \N, \P, \p, \U, \u, and \X.
2333    
2334    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2335    reading past the end of the pattern in cases such as /[abcd/.
2336    
2337    20. PCRE was getting more memory than necessary for patterns with classes that
2338    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2339    
2340    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2341    compiling PCRE for use with Virtual Pascal.
2342    
2343    22. Small fix to the Makefile to make it work properly if the build is done
2344    outside the source tree.
2345    
2346    23. Added a new extension: a condition to go with recursion. If a conditional
2347    subpattern starts with (?(R) the "true" branch is used if recursion has
2348    happened, whereas the "false" branch is used only at the top level.
2349    
2350    24. When there was a very long string of literal characters (over 255 bytes
2351    without UTF support, over 250 bytes with UTF support), the computation of how
2352    much memory was required could be incorrect, leading to segfaults or other
2353    strange effects.
2354    
2355    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2356    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2357    there was a subsequent back reference to those brackets. This meant that, for
2358    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2359    possible to check for precisely this case. All we can do is abandon the
2360    optimization if .* occurs inside capturing brackets when there are any back
2361    references whatsoever. (See below for a better fix that came later.)
2362    
2363    26. The handling of the optimization for finding the first character of a
2364    non-anchored pattern, and for finding a character that is required later in the
2365    match were failing in some cases. This didn't break the matching; it just
2366    failed to optimize when it could. The way this is done has been re-implemented.
2367    
2368    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2369    
2370    28. Added a new feature that provides some of the functionality that Perl
2371    provides with (?{...}). The facility is termed a "callout". The way it is done
2372    in PCRE is for the caller to provide an optional function, by setting
2373    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2374    global variable. By default it is unset, which disables all calling out. To get
2375    the function called, the regex must include (?C) at appropriate points. This
2376    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2377    This provides a means of identifying different callout points. When PCRE
2378    reaches such a point in the regex, if pcre_callout has been set, the external
2379    function is called. It is provided with data in a structure called
2380    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2381    matching continues; if it returns a non-zero value, the match at the current
2382    point fails. However, backtracking will occur if possible. [This was changed
2383    later and other features added - see item 49 below.]
2384    
2385    29. pcretest is upgraded to test the callout functionality. It provides a
2386    callout function that displays information. By default, it shows the start of
2387    the match and the current position in the text. There are some new data escapes
2388    to vary what happens:
2389    
2390        \C+         in addition, show current contents of captured substrings
2391        \C-         do not supply a callout function
2392        \C!n        return 1 when callout number n is reached
2393        \C!n!m      return 1 when callout number n is reached for the mth time
2394    
2395    30. If pcregrep was called with the -l option and just a single file name, it
2396    output "<stdin>" if a match was found, instead of the file name.
2397    
2398    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2399    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2400    pcre_exec(). This saves a malloc/free per call. The default value of
2401    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2402    when configuring.
2403    
2404    32. The default maximum size of a compiled pattern is 64K. There have been a
2405    few cases of people hitting this limit. The code now uses macros to handle the
2406    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2407    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2408    configuring. Tests 2 and 5 work only with 2-byte links because they output
2409    debugging information about compiled patterns.
2410    
2411    33. Internal code re-arrangements:
2412    
2413    (a) Moved the debugging function for printing out a compiled regex into
2414        its own source file (printint.c) and used #include to pull it into
2415        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2416        separate copies.
2417    
2418    (b) Defined the list of op-code names for debugging as a macro in
2419        internal.h so that it is next to the definition of the opcodes.
2420    
2421    (c) Defined a table of op-code lengths for simpler skipping along compiled
2422        code. This is again a macro in internal.h so that it is next to the
2423        definition of the opcodes.
2424    
2425    34. Added support for recursive calls to individual subpatterns, along the
2426    lines of Robin Houston's patch (but implemented somewhat differently).
2427    
2428    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2429    allow it to read and process whole directories in Win32. This code was
2430    contributed by Lionel Fourquaux; it has not been tested by me.
2431    
2432    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2433    used to name a group. Names consist of alphanumerics and underscores, and must
2434    be unique. Back references use the syntax (?P=name) and recursive calls use
2435    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2436    numbers. The function pcre_fullinfo() can be used after compilation to extract
2437    a name/number map. There are three relevant calls:
2438    
2439      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2440      PCRE_INFO_NAMECOUNT            yields the number of entries
2441      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2442    
2443    The map is a vector of fixed-size entries. The size of each entry depends on
2444    the length of the longest name used. The first two bytes of each entry are the
2445    group number, most significant byte first. There follows the corresponding
2446    name, zero terminated. The names are in alphabetical order.
2447    
2448    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2449    case instead of 255. Making it the same both with and without UTF-8 support
2450    means that the same test output works with both.
2451    
2452    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2453    calling malloc() with a zero argument.
2454    
2455    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2456    optimization. I've improved things by keeping a bitmap of backreferences with
2457    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2458    fact referenced, the optimization can be applied. It is unlikely that a
2459    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2460    the match to follow \n) will appear inside brackets with a number greater than
2461    31, but if it does, any back reference > 31 suppresses the optimization.
2462    
2463    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2464    of disabling numbered capturing parentheses. Any opening parenthesis that is
2465    not followed by ? behaves as if it were followed by ?: but named parentheses
2466    can still be used for capturing (and they will acquire numbers in the usual
2467    way).
2468    
2469    41. Redesigned the return codes from the match() function into yes/no/error so
2470    that errors can be passed back from deep inside the nested calls. A malloc
2471    failure while inside a recursive subpattern call now causes the
2472    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2473    
2474    42. It is now possible to set a limit on the number of times the match()
2475    function is called in a call to pcre_exec(). This facility makes it possible to
2476    limit the amount of recursion and backtracking, though not in a directly
2477    obvious way, because the match() function is used in a number of different
2478    circumstances. The count starts from zero for each position in the subject
2479    string (for non-anchored patterns). The default limit is, for compatibility, a
2480    large number, namely 10 000 000. You can change this in two ways:
2481    
2482    (a) When configuring PCRE before making, you can use --with-match-limit=n
2483        to set a default value for the compiled library.
2484    
2485    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2486        a different value is set. See 45 below.
2487    
2488    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2489    
2490    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2491    of things that can be changed at compile time. The first argument specifies
2492    what is wanted and the second points to where the information is to be placed.
2493    The current list of available information is:
2494    
2495      PCRE_CONFIG_UTF8
2496    
2497    The output is an integer that is set to one if UTF-8 support is available;
2498    otherwise it is set to zero.
2499    
2500      PCRE_CONFIG_NEWLINE
2501    
2502    The output is an integer that it set to the value of the code that is used for
2503    newline. It is either LF (10) or CR (13).
2504    
2505      PCRE_CONFIG_LINK_SIZE
2506    
2507    The output is an integer that contains the number of bytes used for internal
2508    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2509    
2510      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2511    
2512    The output is an integer that contains the threshold above which the POSIX
2513    interface uses malloc() for output vectors. See item 31 above.
2514    
2515      PCRE_CONFIG_MATCH_LIMIT
2516    
2517    The output is an unsigned integer that contains the default limit of the number
2518    of match() calls in a pcre_exec() execution. See 42 above.
2519    
2520    44. pcretest has been upgraded by the addition of the -C option. This causes it
2521    to extract all the available output from the new pcre_config() function, and to
2522    output it. The program then exits immediately.
2523    
2524    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2525    order to support additional features. One way would have been to define
2526    pcre_exec2() (for example) with extra arguments, but this would not have been
2527    extensible, and would also have required all calls to the original function to
2528    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2529    is used for passing in "extra" data from pcre_study().
2530    
2531    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2532    contains the following fields:
2533    
2534      flags         a bitmap indicating which of the following fields are set
2535      study_data    opaque data from pcre_study()
2536      match_limit   a way of specifying a limit on match() calls for a specific
2537                      call to pcre_exec()
2538      callout_data  data for callouts (see 49 below)
2539    
2540    The flag bits are also defined in pcre.h, and are
2541    
2542      PCRE_EXTRA_STUDY_DATA
2543      PCRE_EXTRA_MATCH_LIMIT
2544      PCRE_EXTRA_CALLOUT_DATA
2545    
2546    The pcre_study() function now returns one of these new pcre_extra blocks, with
2547    the actual study data pointed to by the study_data field, and the
2548    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2549    before. That is, this change is entirely upwards-compatible and requires no
2550    change to existing code.
2551    
2552    If you want to pass in additional data to pcre_exec(), you can either place it
2553    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2554    block.
2555    
2556    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2557    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2558    times with different match limits, until it finds the minimum value needed for
2559    pcre_exec() to complete. The value is then output. This can be instructive; for
2560    most simple matches the number is quite small, but for pathological cases it
2561    gets very large very quickly.
2562    
2563    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2564    returns the size of the data block pointed to by the study_data field in a
2565    pcre_extra block, that is, the value that was passed as the argument to
2566    pcre_malloc() when PCRE was getting memory in which to place the information
2567    created by pcre_study(). The fourth argument should point to a size_t variable.
2568    pcretest has been extended so that this information is shown after a successful
2569    pcre_study() call when information about the compiled regex is being displayed.
2570    
2571    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2572    because what follows is always an absolute path. (Later: it turns out that this
2573    is more than cosmetic for MinGW, because it doesn't like empty path
2574    components.)
2575    
2576    49. Some changes have been made to the callout feature (see 28 above):
2577    
2578    (i)  A callout function now has three choices for what it returns:
2579    
2580           0  =>  success, carry on matching
2581         > 0  =>  failure at this point, but backtrack if possible
2582         < 0  =>  serious error, return this value from pcre_exec()
2583    
2584         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2585         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2586         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2587         use by callout functions. It will never be used by PCRE itself.
2588    
2589    (ii) The pcre_extra structure (see 45 above) has a void * field called
2590         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2591         pcre_callout_block structure has a field of the same name. The contents of
2592         the field passed in the pcre_extra structure are passed to the callout
2593         function in the corresponding field in the callout block. This makes it
2594         easier to use the same callout-containing regex from multiple threads. For
2595         testing, the pcretest program has a new data escape
2596    
2597           \C*n        pass the number n (may be negative) as callout_data
2598    
2599         If the callout function in pcretest receives a non-zero value as
2600         callout_data, it returns that value.
2601    
2602    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2603    there were some redundant $(CFLAGS) in commands that are now specified as
2604    $(LINK), which already includes $(CFLAGS).
2605    
2606    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2607    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2608    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2609    one-byte characters throughout. Note that case-insensitive matching applies
2610    only to characters whose values are less than 256. PCRE doesn't support the
2611    notion of cases for higher-valued characters.
2612    
2613    (i)   A character class whose characters are all within 0-255 is handled as
2614          a bit map, and the map is inverted for negative classes. Previously, a
2615          character > 255 always failed to match such a class; however it should
2616          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2617    
2618    (ii)  A negated character class with a single character < 255 is coded as
2619          "not this character" (OP_NOT). This wasn't working properly when the test
2620          character was multibyte, either singly or repeated.
2621    
2622    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2623          mode, for example: \x{100}{2,3}.
2624    
2625    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2626          singly or repeated) now correctly test multibyte characters. However,
2627          PCRE doesn't recognize any characters with values greater than 255 as
2628          digits, spaces, or word characters. Such characters always match \D, \S,
2629          and \W, and never match \d, \s, or \w.
2630    
2631    (v)   Classes may now contain characters and character ranges with values
2632          greater than 255. For example: [ab\x{100}-\x{400}].
2633    
2634    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2635          PCRE in UTF-8 mode.
2636    
2637    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2638    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2639    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2640    value.)
2641    
2642    53. The single man page has become too large. I have therefore split it up into
2643    a number of separate man pages. These also give rise to individual HTML pages;
2644    these are now put in a separate directory, and there is an index.html page that
2645    lists them all. Some hyperlinking between the pages has been installed.
2646    
2647    54. Added convenience functions for handling named capturing parentheses.
2648    
2649    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2650    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2651    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2652    are faulted.
2653    
2654    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2655    calling configure. These values are used when compiling the dftables.c program
2656    which is run to generate the source of the default character tables. They
2657    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2658    you will need to set these values.
2659    
2660    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2661    
2662    
2663    Version 3.9 02-Jan-02
2664    ---------------------
2665    
2666    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2667    
2668    2. If --disable-static was given, the building process failed when trying to
2669    build pcretest and pcregrep. (For some reason it was using libtool to compile
2670    them, which is not right, as they aren't part of the library.)
2671    
2672    
2673    Version 3.8 18-Dec-01
2674    ---------------------
2675    
2676    1. The experimental UTF-8 code was completely screwed up. It was packing the
2677    bytes in the wrong order. How dumb can you get?
2678    
2679    
2680    Version 3.7 29-Oct-01
2681    ---------------------
2682    
2683    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2684    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2685    this didn't happen under Solaris 8, where I normally test things.
2686    
2687    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2688    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2689    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2690    this fix an hour or so after the initial 3.7 release.)
2691    
2692    
2693    Version 3.6 23-Oct-01
2694    ---------------------
2695    
2696    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2697    offsets passed as NULL with zero offset count.
2698    
2699    2. The config.guess and config.sub files had not been updated when I moved to
2700    the latest autoconf.
2701    
2702    
2703    Version 3.5 15-Aug-01
2704    ---------------------
2705    
2706    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2707    had been forgotten.
2708    
2709    2. By using declared but undefined structures, we can avoid using "void"
2710    definitions in pcre.h while keeping the internal definitions of the structures
2711    private.
2712    
2713    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2714    user point of view, this means that both static and shared libraries are built
2715    by default, but this can be individually controlled. More of the work of
2716    handling this static/shared cases is now inside libtool instead of PCRE's make
2717    file.
2718    
2719    4. The pcretest utility is now installed along with pcregrep because it is
2720    useful for users (to test regexs) and by doing this, it automatically gets
2721    relinked by libtool. The documentation has been turned into a man page, so
2722    there are now .1, .txt, and .html versions in /doc.
2723    
2724    5. Upgrades to pcregrep:
2725       (i)   Added long-form option names like gnu grep.
2726       (ii)  Added --help to list all options with an explanatory phrase.
2727       (iii) Added -r, --recursive to recurse into sub-directories.
2728       (iv)  Added -f, --file to read patterns from a file.
2729    
2730    6. pcre_exec() was referring to its "code" argument before testing that
2731    argument for NULL (and giving an error if it was NULL).
2732    
2733    7. Upgraded Makefile.in to allow for compiling in a different directory from
2734    the source directory.
2735    
2736    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2737    options bits, the pointer it was passed was to an int instead of to an unsigned
2738    long int. This mattered only on 64-bit systems.
2739    
2740    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2741    generated) instead of pcre.in, which it its source. Also made the same change
2742    in several of the .c files.
2743    
2744    10. A new release of gcc defines printf() as a macro, which broke pcretest
2745    because it had an ifdef in the middle of a string argument for printf(). Fixed
2746    by using separate calls to printf().
2747    
2748    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2749    script, to force use of CR or LF instead of \n in the source. On non-Unix
2750    systems, the value can be set in config.h.
2751    
2752    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2753    absolute limit. Changed the text of the error message to make this clear, and
2754    likewise updated the man page.
2755    
2756    13. The limit of 99 on the number of capturing subpatterns has been removed.
2757    The new limit is 65535, which I hope will not be a "real" limit.
2758    
2759    
2760    Version 3.4 22-Aug-00
2761    ---------------------
2762    
2763    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2764    
2765    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2766    
2767    
2768    Version 3.3 01-Aug-00
2769    ---------------------
2770    
2771    1. If an octal character was given, but the value was greater than \377, it
2772    was not getting masked to the least significant bits, as documented. This could
2773    lead to crashes in some systems.
2774    
2775    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2776    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2777    
2778    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2779    These just pass their arguments on to (pcre_free)(), but they are provided
2780    because some uses of PCRE bind it to non-C systems that can call its functions,
2781    but cannot call free() or pcre_free() directly.
2782    
2783    4. Add "make test" as a synonym for "make check". Corrected some comments in
2784    the Makefile.
2785    
2786    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2787    Makefile.
2788    
2789    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2790    command called pgrep for grepping around the active processes.
2791    
2792    7. Added the beginnings of support for UTF-8 character strings.
2793    
2794    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2795    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2796    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2797    out for the ar command.)
2798    
2799    
2800    Version 3.2 12-May-00
2801    ---------------------
2802    
2803    This is purely a bug fixing release.
2804    
2805    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2806    of ZA. This was just one example of several cases that could provoke this bug,
2807    which was introduced by change 9 of version 2.00. The code for breaking
2808    infinite loops after an iteration that matches an empty string was't working
2809    correctly.
2810    
2811    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2812    when matched against abbab (for example). After matching an empty string, it
2813    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2814    caused it to match further down the string than it should.
2815    
2816    3. The code contained an inclusion of sys/types.h. It isn't clear why this
2817    was there because it doesn't seem to be needed, and it causes trouble on some
2818    systems, as it is not a Standard C header. It has been removed.
2819    
2820    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2821    were reported on the Macintosh. The changes were from
2822    
2823      while ((c = *(++ptr)) != 0 && c != '\n');
2824    to
2825      while ((c = *(++ptr)) != 0 && c != '\n') ;
2826    
2827    Totally extraordinary, but if that's what it takes...
2828    
2829    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2830    available. Added HAVE_BCOPY and an autoconf test for it; if neither
2831    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2832    assumes the way PCRE uses memmove() (always moving upwards).
2833    
2834    6. PCRE is being used in one environment where strchr() is not available. There
2835    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2836    faster code anyway.
2837    
2838    
2839    Version 3.1 09-Feb-00
2840    ---------------------
2841    
2842    The only change in this release is the fixing of some bugs in Makefile.in for
2843    the "install" target:
2844    
2845    (1) It was failing to install pcreposix.h.
2846    
2847    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2848    
2849    
2850    Version 3.0 01-Feb-00
2851    ---------------------
2852    
2853    1. Add support for the /+ modifier to perltest (to output $` like it does in
2854    pcretest).
2855    
2856    2. Add support for the /g modifier to perltest.
2857    
2858    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2859    matches null strings.
2860    
2861    4. Fix perltest so that it doesn't do unwanted things when fed an empty
2862    pattern. Perl treats empty patterns specially - it reuses the most recent
2863    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2864    effect.
2865    
2866    5. The POSIX interface was broken in that it was just handing over the POSIX
2867    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2868    required a bigger vector, with some working space on the end. This means that
2869    the POSIX wrapper now has to get and free some memory, and copy the results.
2870    
2871    6. Added some simple autoconf support, placing the test data and the
2872    documentation in separate directories, re-organizing some of the
2873    information files, and making it build pcre-config (a GNU standard). Also added
2874    libtool support for building PCRE as a shared library, which is now the
2875    default.
2876    
2877    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2878    09 are not valid octal constants. Single digits will be used for minor values
2879    less than 10.
2880    
2881    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2882    existing programs that set these in the POSIX interface can use PCRE without
2883    modification.
2884    
2885    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2886    return all that pcre_info() returns, plus additional data. The pcre_info()
2887    function is retained for compatibility, but is considered to be obsolete.
2888    
2889    10. Added experimental recursion feature (?R) to handle one common case that
2890    Perl 5.6 will be able to do with (?p{...}).
2891    
2892    11. Added support for POSIX character classes like [:alpha:], which Perl is
2893    adopting.
2894    
2895    
2896    Version 2.08 31-Aug-99
2897    ----------------------
2898    
2899    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2900    trying to match at the startoffset position, but instead was moving forward to
2901    the next newline as if a previous match had failed.
2902    
2903    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2904    and could get into a loop if a null string was matched other than at the start
2905    of the subject.
2906    
2907    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2908    be distinguished at compile time, and for completeness also added PCRE_DATE.
2909    
2910    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2911    in GnuWin32 environments.
2912    
2913    
2914    Version 2.07 29-Jul-99
2915    ----------------------
2916    
2917    1. The documentation is now supplied in plain text form and HTML as well as in
2918    the form of man page sources.
2919    
2920    2. C++ compilers don't like assigning (void *) values to other pointer types.
2921    In particular this affects malloc(). Although there is no problem in Standard
2922    C, I've put in casts to keep C++ compilers happy.
2923    
2924    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2925    should be (const char *).
2926    
2927    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2928    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2929    However, I haven't made this a standard facility. The documentation doesn't
2930    mention it, and the Makefile doesn't support it.
2931    
2932    5. The Makefile now contains an "install" target, with editable destinations at
2933    the top of the file. The pcretest program is not installed.
2934    
2935    6. pgrep -V now gives the PCRE version number and date.
2936    
2937    7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2938    causing the entire string to be ignored, instead of just the last character.
2939    
2940    8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2941    non-matching string, it can take a very, very long time, even for strings of
2942    quite modest length, because of the nested recursion. PCRE now does better in
2943    some of these cases. It does this by remembering the last required literal
2944    character in the pattern, and pre-searching the subject to ensure it is present
2945    before running the real match. In other words, it applies a heuristic to detect
2946    some types of certain failure quickly, and in the above example, if presented
2947    with a string that has no trailing " it gives "no match" very quickly.
2948    
2949    9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2950    other alternatives are tried instead.
2951    
2952    
2953    Version 2.06 09-Jun-99
2954    ----------------------
2955    
2956    1. Change pcretest's output for amount of store used to show just the code
2957    space, because the remainder (the data block) varies in size between 32-bit and
2958    64-bit systems.
2959    
2960    2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2961    start matching at. This allows lookbehinds to work when searching for multiple
2962    occurrences in a string.
2963    
2964    3. Added additional options to pcretest for testing multiple occurrences:
2965    
2966       /+   outputs the rest of the string that follows a match
2967       /g   loops for multiple occurrences, using the new startoffset argument
2968       /G   loops for multiple occurrences by passing an incremented pointer
2969    
2970    4. PCRE wasn't doing the "first character" optimization for patterns starting
2971    with \b or \B, though it was doing it for other lookbehind assertions. That is,
2972    it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2973    the letter 'x'. On long subject strings, this gives a significant speed-up.
2974    
2975    
2976    Version 2.05 21-Apr-99
2977    ----------------------
2978    
2979    1. Changed the type of magic_number from int to long int so that it works
2980    properly on 16-bit systems.
2981    
2982    2. Fixed a bug which caused patterns starting with .* not to work correctly
2983    when the subject string contained newline characters. PCRE was assuming
2984    anchoring for such patterns in all cases, which is not correct because .* will
2985    not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2986    DOTALL is set at top level; otherwise it knows that patterns starting with .*
2987    must be retried after every newline in the subject.
2988    
2989    
2990  Version 2.04 18-Feb-99  Version 2.04 18-Feb-99
2991  ----------------------  ----------------------

Legend:
Removed from v.31  
changed lines
  Added in v.376

  ViewVC Help
Powered by ViewVC 1.1.5