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

Diff of /code/trunk/ChangeLog

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

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

Legend:
Removed from v.55  
changed lines
  Added in v.407

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12