ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 27 by nigel, Sat Feb 24 21:38:49 2007 UTC revision 428 by ph10, Mon Aug 31 17:10:26 2009 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Version 8.00 ??-???-??
5    ----------------------
7    1.  The table for translating pcre_compile() error codes into POSIX error codes
8        was out-of-date, and there was no check on the pcre_compile() error code
9        being within the table. This could lead to an OK return being given in
10        error.
12    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
13        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
14        in a Windows environment.
16    3.  The pcregrep --count option prints the count for each file even when it is
17        zero, as does GNU grep. However, pcregrep was also printing all files when
18        --files-with-matches was added. Now, when both options are given, it prints
19        counts only for those files that have at least one match. (GNU grep just
20        prints the file name in this circumstance, but including the count seems
21        more useful - otherwise, why use --count?) Also ensured that the
22        combination -clh just lists non-zero counts, with no names.
24    4.  The long form of the pcregrep -F option was incorrectly implemented as
25        --fixed_strings instead of --fixed-strings. This is an incompatible change,
26        but it seems right to fix it, and I didn't think it was worth preserving
27        the old behaviour.
29    5.  The command line items --regex=pattern and --regexp=pattern were not
30        recognized by pcregrep, which required --regex pattern or --regexp pattern
31        (with a space rather than an '='). The man page documented the '=' forms,
32        which are compatible with GNU grep; these now work.
34    6.  No libpcreposix.pc file was created for pkg-config; there was just
35        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
37    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
38        when UCP support is not needed, by modifying the Python script that
39        generates it from Unicode data files. This should not matter if the module
40        is correctly used as a library, but I received one complaint about 50K of
41        unwanted data. My guess is that the person linked everything into his
42        program rather than using a library. Anyway, it does no harm.
44    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
45        was a minimum greater than 1 for a wide character in a possessive
46        repetition. Chaos could result.
48    9.  The restrictions on what a pattern can contain when partial matching is
49        requested for pcre_exec() have been removed. All patterns can now be
50        partially matched by this function. In addition, if there are at least two
51        slots in the offset vector, the offsets of the first-encountered partial
52        match are set in them when PCRE_ERROR_PARTIAL is returned.
54    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
55        synonymous with PCRE_PARTIAL, for backwards compatibility, and
56        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
57        and may be more useful for multi-segment matching, especially with
58        pcre_exec().
60    11. Partial matching with pcre_exec() is now more intuitive. A partial match
61        used to be given if ever the end of the subject was reached; now it is
62        given only if matching could not proceed because another character was
63        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
64        string "Z", which now yields "no match" instead of "partial match". In the
65        case of pcre_dfa_exec(), "no match" is given if every matching path for the
66        final character ended with (*FAIL).
68    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
69        if the pattern had a "must contain" character that was already found in the
70        earlier partial match, unless partial matching was again requested. For
71        example, with the pattern /dog.(body)?/, the "must contain" character is
72        "g". If the first part-match was for the string "dog", restarting with
73        "sbody" failed.
76    Version 7.9 11-Apr-09
77    ---------------------
79    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
80        (pcretest), all targets were linked against these libraries. This included
81        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
82        libraries. This caused unwanted dependencies to be created. This problem
83        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
84        pcretest is linked with readline.
86    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
87        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
88        moved outside it again, because FALSE and TRUE are already defined in AIX,
89        but BOOL is not.
91    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
92        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
94    4.  The pcregrep documentation said spaces were inserted as well as colons (or
95        hyphens) following file names and line numbers when outputting matching
96        lines. This is not true; no spaces are inserted. I have also clarified the
97        wording for the --colour (or --color) option.
99    5.  In pcregrep, when --colour was used with -o, the list of matching strings
100        was not coloured; this is different to GNU grep, so I have changed it to be
101        the same.
103    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
104        each matching line was coloured. Now it goes on to look for further matches
105        of any of the test patterns, which is the same behaviour as GNU grep.
107    7.  A pattern that could match an empty string could cause pcregrep to loop; it
108        doesn't make sense to accept an empty string match in pcregrep, so I have
109        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
110        seems to be how GNU grep behaves.
112    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
113        start or after a newline", because the conditional assertion was not being
114        correctly handled. The rule now is that both the assertion and what follows
115        in the first alternative must satisfy the test.
117    9.  If auto-callout was enabled in a pattern with a conditional group whose
118        condition was an assertion, PCRE could crash during matching, both with
119        pcre_exec() and pcre_dfa_exec().
121    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
122        used for matching.
124    11. Unicode property support in character classes was not working for
125        characters (bytes) greater than 127 when not in UTF-8 mode.
127    12. Added the -M command line option to pcretest.
129    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
131    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
133    16. Added comments and documentation about mis-use of no_arg in the C++
134        wrapper.
136    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
137        from Martin Jerabek that uses macro names for all relevant character and
138        string constants.
140    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
141        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
142        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
143        these, but not everybody uses configure.
145    19. A conditional group that had only one branch was not being correctly
146        recognized as an item that could match an empty string. This meant that an
147        enclosing group might also not be so recognized, causing infinite looping
148        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
149        with the subject "ab", where knowledge that the repeated group can match
150        nothing is needed in order to break the loop.
152    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
153        exec(), but without supplying a callout function, matching went wrong.
155    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
156        leak if the size of the offset vector was greater than 30. When the vector
157        is smaller, the saved offsets during recursion go onto a local stack
158        vector, but for larger vectors malloc() is used. It was failing to free
159        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
160        error, in fact).
162    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
163        heapframe that is used only when UTF-8 support is enabled. This caused no
164        problem, but was untidy.
166    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
167        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
168        included within another project.
170    24. Steven Van Ingelgem's patches to add more options to the CMake support,
171        slightly modified by me:
173          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
174              not building pcregrep.
176          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
177              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
179    25. Forward references, both numeric and by name, in patterns that made use of
180        duplicate group numbers, could behave incorrectly or give incorrect errors,
181        because when scanning forward to find the reference group, PCRE was not
182        taking into account the duplicate group numbers. A pattern such as
183        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
185    26. Changed a few more instances of "const unsigned char *" to USPTR, making
186        the feature of a custom pointer more persuasive (as requested by a user).
188    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
189        pcretest.c, inside #ifndefs, because it seems they are sometimes already
190        pre-defined.
192    28. Added support for (*UTF8) at the start of a pattern.
194    29. Arrange for flags added by the "release type" setting in CMake to be shown
195        in the configuration summary.
198    Version 7.8 05-Sep-08
199    ---------------------
201    1.  Replaced UCP searching code with optimized version as implemented for Ad
202        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
203        stage table and inline lookup instead of a function, giving speed ups of 2
204        to 5 times on some simple patterns that I tested. Permission was given to
205        distribute the MultiStage2.py script that generates the tables (it's not in
206        the tarball, but is in the Subversion repository).
208    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
209        scripts.
211    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
212        a group with a zero qualifier. The result of the study could be incorrect,
213        or the function might crash, depending on the pattern.
215    4.  Caseless matching was not working for non-ASCII characters in back
216        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
217        It now works when Unicode Property Support is available.
219    5.  In pcretest, an escape such as \x{de} in the data was always generating
220        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
221        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
222        truncation.
224    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
226    7.  Added two (int) casts to pcregrep when printing the difference of two
227        pointers, in case they are 64-bit values.
229    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
230        test 2 if it fails.
232    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
233        and a #define of that name to empty if it is not externally set. This is to
234        allow users of MSVC to set it if necessary.
236    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
237        the convenience functions in the pcre_get.c source file.
239    11. An option change at the start of a pattern that had top-level alternatives
240        could cause overwriting and/or a crash. This command provoked a crash in
241        some environments:
243          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
245        This potential security problem was recorded as CVE-2008-2371.
247    12. For a pattern where the match had to start at the beginning or immediately
248        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
249        pcre_dfa_exec() could read past the end of the passed subject if there was
250        no match. To help with detecting such bugs (e.g. with valgrind), I modified
251        pcretest so that it places the subject at the end of its malloc-ed buffer.
253    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
254        exec() might read past the end of the data buffer in UTF-8 mode.
256    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
257        the data contained the byte 0x85 as part of a UTF-8 character within its
258        first line. This applied both to normal and DFA matching.
260    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
261        /^[^d]*?$/8 failed to match "abc".
263    16. Added a missing copyright notice to pcrecpp_internal.h.
265    17. Make it more clear in the documentation that values returned from
266        pcre_exec() in ovector are byte offsets, not character counts.
268    18. Tidied a few places to stop certain compilers from issuing warnings.
270    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
271        supplied by Stefan Weber. I made a further small update for 7.8 because
272        there is a change of source arrangements: the pcre_searchfuncs.c module is
273        replaced by pcre_ucd.c.
276    Version 7.7 07-May-08
277    ---------------------
279    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
280        a string to a long long, pretend we don't even have a long long." This is
281        done by checking for the strtoq, strtoll, and _strtoi64 functions.
283    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
284        pre-7.6 versions, which defined a global no_arg variable instead of putting
285        it in the RE class. (See also #8 below.)
287    3.  Remove a line of dead code, identified by coverity and reported by Nuno
288        Lopes.
290    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
292        (1) The include/exclude patterns were being applied to the whole pathnames
293            of files, instead of just to the final components.
295        (2) If there was more than one level of directory, the subdirectories were
296            skipped unless they satisfied the include/exclude conditions. This is
297            inconsistent with GNU grep (and could even be seen as contrary to the
298            pcregrep specification - which I improved to make it absolutely clear).
299            The action now is always to scan all levels of directory, and just
300            apply the include/exclude patterns to regular files.
302    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
303        --exclude_dir in the tests to avoid scanning .svn directories.
305    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
306        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
307        doesn't support NULs in patterns.
309    7.  Added some missing "const"s to declarations of static tables in
310        pcre_compile.c and pcre_dfa_exec.c.
312    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
313        caused by fix #2  above. (Subsequently also a second patch to fix the
314        first patch. And a third patch - this was a messy problem.)
316    9.  Applied Craig's patch to remove the use of push_back().
318    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
319        matching function regexec().
321    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
322        which, however, unlike Perl's \g{...}, are subroutine calls, not back
323        references. PCRE supports relative numbers with this syntax (I don't think
324        Oniguruma does).
326    12. Previously, a group with a zero repeat such as (...){0} was completely
327        omitted from the compiled regex. However, this means that if the group
328        was called as a subroutine from elsewhere in the pattern, things went wrong
329        (an internal error was given). Such groups are now left in the compiled
330        pattern, with a new opcode that causes them to be skipped at execution
331        time.
333    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
334        to the way PCRE behaves:
336        (a) A lone ] character is dis-allowed (Perl treats it as data).
338        (b) A back reference to an unmatched subpattern matches an empty string
339            (Perl fails the current match path).
341        (c) A data ] in a character class must be notated as \] because if the
342            first data character in a class is ], it defines an empty class. (In
343            Perl it is not possible to have an empty class.) The empty class []
344            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
345            The negative empty class [^] matches any one character, independently
346            of the DOTALL setting.
348    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
349        non-existent subpattern following a character class starting with ']' and
350        containing () gave an internal compiling error instead of "reference to
351        non-existent subpattern". Fortunately, when the pattern did exist, the
352        compiled code was correct. (When scanning forwards to check for the
353        existencd of the subpattern, it was treating the data ']' as terminating
354        the class, so got the count wrong. When actually compiling, the reference
355        was subsequently set up correctly.)
357    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
358        it was being rejected as not supported by pcre_dfa_exec(), even though
359        other assertions are supported. I have made pcre_dfa_exec() support
360        (*FAIL).
362    16. The implementation of 13c above involved the invention of a new opcode,
363        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
364        cannot be changed at match time, I realized I could make a small
365        improvement to matching performance by compiling OP_ALLANY instead of
366        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
367        on the OP_ANY path.
369    17. Compiling pcretest on Windows with readline support failed without the
370        following two fixes: (1) Make the unistd.h include conditional on
371        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
373    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
374        ncurses library to be included for pcretest when ReadLine support is
375        requested, but also to allow for it to be overridden. This patch came from
376        Daniel Bergström.
378    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
379        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
380        any errors with the current Unicode tables. Thanks to Peter Kankowski for
381        spotting this.
384    Version 7.6 28-Jan-08
385    ---------------------
387    1.  A character class containing a very large number of characters with
388        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
389        overflow.
391    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
392        HAVE_LONG_LONG is not defined.
394    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
395        bring them up to date and include new features. This patch includes:
397        - Fixed PH's badly added libz and libbz2 support.
398        - Fixed a problem with static linking.
399        - Added pcredemo. [But later removed - see 7 below.]
400        - Fixed dftables problem and added an option.
401        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
402            HAVE_LONG_LONG.
403        - Added readline support for pcretest.
404        - Added an listing of the option settings after cmake has run.
406    4.  A user submitted a patch to Makefile that makes it easy to create
407        "pcre.dll" under mingw when using Configure/Make. I added stuff to
408        Makefile.am that cause it to include this special target, without
409        affecting anything else. Note that the same mingw target plus all
410        the other distribution libraries and programs are now supported
411        when configuring with CMake (see 6 below) instead of with
412        Configure/Make.
414    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
415        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
416        exported in the Windows port". It has not yet been confirmed that the patch
417        solves the problem, but it does no harm.
419    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
420        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
421        with CMake, and also correct the comment about stack recursion.
423    7.  Remove the automatic building of pcredemo from the ./configure system and
424        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
425        of a program that users should build themselves after PCRE is installed, so
426        building it automatically is not really right. What is more, it gave
427        trouble in some build environments.
429    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
432    Version 7.5 10-Jan-08
433    ---------------------
435    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
436        values in parens when parsing an RE using the C++ wrapper."
438    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
439        Characters greater than 255 were excluded from the class instead of being
440        included.
442    3.  The same bug as (2) above applied to negated POSIX classes such as
443        [:^space:].
445    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
446        defined or documented. It seems to have been a typo for PCRE_STATIC, so
447        I have changed it.
449    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
450        first named subpattern) and a construct such as (?&a) would reference the
451        first named subpattern whose name started with "a" (in other words, the
452        length check was missing). Both these problems are fixed. "Subpattern name
453        expected" is now given for (?&) (a zero-length name), and this patch also
454        makes it give the same error for \k'' (previously it complained that that
455        was a reference to a non-existent subpattern).
457    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
458        this is right because (?- can be followed by option settings as well as by
459        digits. I have, however, made the messages clearer.
461    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
462        than the number used in the conditional) now cause a compile-time error.
463        This is actually not compatible with Perl, which accepts such patterns, but
464        treats the conditional as always being FALSE (as PCRE used to), but it
465        seems to me that giving a diagnostic is better.
467    8.  Change "alphameric" to the more common word "alphanumeric" in comments
468        and messages.
470    9.  Fix two occurrences of "backslash" in comments that should have been
471        "backspace".
473    10. Remove two redundant lines of code that can never be obeyed (their function
474        was moved elsewhere).
476    11. The program that makes PCRE's Unicode character property table had a bug
477        which caused it to generate incorrect table entries for sequences of
478        characters that have the same character type, but are in different scripts.
479        It amalgamated them into a single range, with the script of the first of
480        them. In other words, some characters were in the wrong script. There were
481        thirteen such cases, affecting characters in the following ranges:
483          U+002b0 - U+002c1
484          U+0060c - U+0060d
485          U+0061e - U+00612
486          U+0064b - U+0065e
487          U+0074d - U+0076d
488          U+01800 - U+01805
489          U+01d00 - U+01d77
490          U+01d9b - U+01dbf
491          U+0200b - U+0200f
492          U+030fc - U+030fe
493          U+03260 - U+0327f
494          U+0fb46 - U+0fbb1
495          U+10450 - U+1049d
497    12. The -o option (show only the matching part of a line) for pcregrep was not
498        compatible with GNU grep in that, if there was more than one match in a
499        line, it showed only the first of them. It now behaves in the same way as
500        GNU grep.
502    13. If the -o and -v options were combined for pcregrep, it printed a blank
503        line for every non-matching line. GNU grep prints nothing, and pcregrep now
504        does the same. The return code can be used to tell if there were any
505        non-matching lines.
507    14. Added --file-offsets and --line-offsets to pcregrep.
509    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
510        infinitely looping recursion. The bug was that positive lookaheads were not
511        being skipped when checking for a possible empty match (negative lookaheads
512        and both kinds of lookbehind were skipped).
514    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
515        inclusion of <windows.h> to before rather than after the definition of
516        INVALID_FILE_ATTRIBUTES (patch from David Byron).
518    17. Specifying a possessive quantifier with a specific limit for a Unicode
519        character property caused pcre_compile() to compile bad code, which led at
520        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
521        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
522        caused the error; without that there was no problem.
524    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
526    19. Added --enable-pcretest-libreadline.
528    20. In pcrecpp.cc, the variable 'count' was incremented twice in
529        RE::GlobalReplace(). As a result, the number of replacements returned was
530        double what it should be. I removed one of the increments, but Craig sent a
531        later patch that removed the other one (the right fix) and added unit tests
532        that check the return values (which was not done before).
534    21. Several CMake things:
536        (1) Arranged that, when cmake is used on Unix, the libraries end up with
537            the names libpcre and libpcreposix, not just pcre and pcreposix.
539        (2) The above change means that pcretest and pcregrep are now correctly
540            linked with the newly-built libraries, not previously installed ones.
544    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
545        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
546        UTF-8 newline character). The key issue is that the pattern starts .*;
547        this means that the match must be either at the beginning, or after a
548        newline. The bug was in the code for advancing after a failed match and
549        checking that the new position followed a newline. It was not taking
550        account of UTF-8 characters correctly.
552    23. PCRE was behaving differently from Perl in the way it recognized POSIX
553        character classes. PCRE was not treating the sequence [:...:] as a
554        character class unless the ... were all letters. Perl, however, seems to
555        allow any characters between [: and :], though of course it rejects as
556        unknown any "names" that contain non-letters, because all the known class
557        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
558        for example, whereas PCRE did not - it did not recognize a POSIX character
559        class. This seemed a bit dangerous, so the code has been changed to be
560        closer to Perl. The behaviour is not identical to Perl, because PCRE will
561        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
562        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
563        Perl does, and where it didn't before.
565    24. Rewrite so as to remove the single use of %n from pcregrep because in some
566        Windows environments %n is disabled by default.
569    Version 7.4 21-Sep-07
570    ---------------------
572    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
573        means that a class such as [\s] counted as "explicit reference to CR or
574        LF". That isn't really right - the whole point of the change was to try to
575        help when there was an actual mention of one of the two characters. So now
576        the change happens only if \r or \n (or a literal CR or LF) character is
577        encountered.
579    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
580        of both had grown to the point where there were only 3 bits left.
581        Fortunately, there was spare space in the data structure, and so I have
582        moved the internal flags into a new 16-bit field to free up more option
583        bits.
585    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
586        but did not set the internal JCHANGED flag - either of these is enough to
587        control the way the "get" function works - but the PCRE_INFO_JCHANGED
588        facility is supposed to tell if (?J) was ever used, so now (?J) at the
589        start sets both bits.
591    4.  Added options (at build time, compile time, exec time) to change \R from
592        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
594    5.  doc/pcresyntax.html was missing from the distribution.
596    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
597        compatibility, even though it is no longer used.
599    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
600        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
601        windows.h file is present (where different names are used). [This was
602        reversed later after testing - see 16 below.]
604    8.  Changed all #include <config.h> to #include "config.h". There were also
605        some further <pcre.h> cases that I changed to "pcre.h".
607    9.  When pcregrep was used with the --colour option, it missed the line ending
608        sequence off the lines that it output.
610    10. It was pointed out to me that arrays of string pointers cause lots of
611        relocations when a shared library is dynamically loaded. A technique of
612        using a single long string with a table of offsets can drastically reduce
613        these. I have refactored PCRE in four places to do this. The result is
614        dramatic:
616          Originally:                          290
617          After changing UCP table:            187
618          After changing error message table:   43
619          After changing table of "verbs"       36
620          After changing table of Posix names   22
622        Thanks to the folks working on Gregex for glib for this insight.
624    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
625        unicode-properties was also set.
627    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
629    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
630        checked only for CRLF.
632    14. Added casts to pcretest.c to avoid compiler warnings.
634    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
636    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
637        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
638        entirely. This removes changes made in 7 above.
640    17. The CMake files have been updated, and there is now more information about
641        building with CMake in the NON-UNIX-USE document.
644    Version 7.3 28-Aug-07
645    ---------------------
647     1. In the rejigging of the build system that eventually resulted in 7.1, the
648        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
649        brackets there is not right, since it causes compilers to look for an
650        installed pcre.h, not the version that is in the source that is being
651        compiled (which of course may be different). I have changed it back to:
653          #include "pcre.h"
655        I have a vague recollection that the change was concerned with compiling in
656        different directories, but in the new build system, that is taken care of
657        by the VPATH setting the Makefile.
659     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
660        when the subject happened to end in the byte 0x85 (e.g. if the last
661        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
662        characters but of course it shouldn't be taken as a newline when it is part
663        of another character. The bug was that, for an unlimited repeat of . in
664        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
665        characters when looking for a newline.
667     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
669     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
670        in debug output.
672     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
673        long printing in the pcrecpp unittest when running under MinGW.
675     6. ESC_K was left out of the EBCDIC table.
677     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
678        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
679        limit also applies to "virtual nesting" when a pattern is recursive, and in
680        this case 1000 isn't so big. I have been able to remove this limit at the
681        expense of backing off one optimization in certain circumstances. Normally,
682        when pcre_exec() would call its internal match() function recursively and
683        immediately return the result unconditionally, it uses a "tail recursion"
684        feature to save stack. However, when a subpattern that can match an empty
685        string has an unlimited repetition quantifier, it no longer makes this
686        optimization. That gives it a stack frame in which to save the data for
687        checking that an empty string has been matched. Previously this was taken
688        from the 1000-entry workspace that had been reserved. So now there is no
689        explicit limit, but more stack is used.
691     8. Applied Daniel's patches to solve problems with the import/export magic
692        syntax that is required for Windows, and which was going wrong for the
693        pcreposix and pcrecpp parts of the library. These were overlooked when this
694        problem was solved for the main library.
696     9. There were some crude static tests to avoid integer overflow when computing
697        the size of patterns that contain repeated groups with explicit upper
698        limits. As the maximum quantifier is 65535, the maximum group length was
699        set at 30,000 so that the product of these two numbers did not overflow a
700        32-bit integer. However, it turns out that people want to use groups that
701        are longer than 30,000 bytes (though not repeat them that many times).
702        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
703        made it possible to implement the integer overflow checks in a much more
704        dynamic way, which I have now done. The artificial limitation on group
705        length has been removed - we now have only the limit on the total length of
706        the compiled pattern, which depends on the LINK_SIZE setting.
708    10. Fixed a bug in the documentation for get/copy named substring when
709        duplicate names are permitted. If none of the named substrings are set, the
710        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
711        empty string.
713    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
714        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
715        because the ] is interpreted as the first data character and the
716        terminating ] is not found. PCRE has been made compatible with Perl in this
717        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
718        cause memory overwriting.
720    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
721        string has been matched (to stop an infinite loop). It was not recognizing
722        a conditional subpattern that could match an empty string if that
723        subpattern was within another subpattern. For example, it looped when
724        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
725        condition was not nested. This bug has been fixed.
727    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
728        past the start of the subject in the presence of bytes with the top bit
729        set, for example "\x8aBCD".
731    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
732        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
734    14. Optimized (?!) to (*FAIL).
736    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
737        This restricts code points to be within the range 0 to 0x10FFFF, excluding
738        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
739        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
740        does: it's just the validity check that is more restrictive.
742    16. Inserted checks for integer overflows during escape sequence (backslash)
743        processing, and also fixed erroneous offset values for syntax errors during
744        backslash processing.
746    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
747        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
749    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
750        caused an overrun.
752    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
753        something other than just ASCII characters) inside a group that had an
754        unlimited repeat caused a loop at compile time (while checking to see
755        whether the group could match an empty string).
757    20. Debugging a pattern containing \p or \P could cause a crash. For example,
758        [\P{Any}] did so. (Error in the code for printing property names.)
760    21. An orphan \E inside a character class could cause a crash.
762    22. A repeated capturing bracket such as (A)? could cause a wild memory
763        reference during compilation.
765    23. There are several functions in pcre_compile() that scan along a compiled
766        expression for various reasons (e.g. to see if it's fixed length for look
767        behind). There were bugs in these functions when a repeated \p or \P was
768        present in the pattern. These operators have additional parameters compared
769        with \d, etc, and these were not being taken into account when moving along
770        the compiled data. Specifically:
772        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
773            length.
775        (b) An item such as \pL+ within a repeated group could cause crashes or
776            loops.
778        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
779            "reference to non-existent subpattern" error.
781        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
783    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
784        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
786    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
788    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
789        character were causing crashes (broken optimization).
791    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
792        \p or \P) caused a compile-time loop.
794    28. More problems have arisen in unanchored patterns when CRLF is a valid line
795        break. For example, the unstudied pattern [\r\n]A does not match the string
796        "\r\nA" because change 7.0/46 below moves the current point on by two
797        characters after failing to match at the start. However, the pattern \nA
798        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
799        the same is true. There doesn't seem any very clean way out of this, but
800        what I have chosen to do makes the common cases work: PCRE now takes note
801        of whether there can be an explicit match for \r or \n anywhere in the
802        pattern, and if so, 7.0/46 no longer applies. As part of this change,
803        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
804        pattern has explicit CR or LF references.
806    29. Added (*CR) etc for changing newline setting at start of pattern.
809    Version 7.2 19-Jun-07
810    ---------------------
812     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
813        which is apparently normally available under Windows.
815     2. Re-jig the pcregrep tests with different newline settings in an attempt
816        to make them independent of the local environment's newline setting.
818     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
820     4. Some of the "internals" tests were previously cut out when the link size
821        was not 2, because the output contained actual offsets. The recent new
822        "Z" feature of pcretest means that these can be cut out, making the tests
823        usable with all link sizes.
825     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
826        stack recursion. This gives a massive performance boost under BSD, but just
827        a small improvement under Linux. However, it saves one field in the frame
828        in all cases.
830     6. Added more features from the forthcoming Perl 5.10:
832        (a) (?-n) (where n is a string of digits) is a relative subroutine or
833            recursion call. It refers to the nth most recently opened parentheses.
835        (b) (?+n) is also a relative subroutine call; it refers to the nth next
836            to be opened parentheses.
838        (c) Conditions that refer to capturing parentheses can be specified
839            relatively, for example, (?(-2)... or (?(+3)...
841        (d) \K resets the start of the current match so that everything before
842            is not part of it.
844        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
846        (f) \g{name} is another synonym - part of Perl 5.10's unification of
847            reference syntax.
849        (g) (?| introduces a group in which the numbering of parentheses in each
850            alternative starts with the same number.
852        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
854     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
857     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
858        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
859        for detecting groups that can match an empty string.
861     9. A pattern with a very large number of alternatives (more than several
862        hundred) was running out of internal workspace during the pre-compile
863        phase, where pcre_compile() figures out how much memory will be needed. A
864        bit of new cunning has reduced the workspace needed for groups with
865        alternatives. The 1000-alternative test pattern now uses 12 bytes of
866        workspace instead of running out of the 4096 that are available.
868    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
870    11. Applied patch from Google to remove an optimization that didn't quite work.
871        The report of the bug said:
873          pcrecpp::RE("a*").FullMatch("aaa") matches, while
874          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
875          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
877    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
878        it matched the wrong number of bytes.
881    Version 7.1 24-Apr-07
882    ---------------------
884     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
885        that is more "standard", making use of automake and other Autotools. There
886        is some re-arrangement of the files and adjustment of comments consequent
887        on this.
889     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
890        for recursive directory scanning broke on some systems because the files
891        are not scanned in any specific order and on different systems the order
892        was different. A call to "sort" has been inserted into RunGrepTest for the
893        approprate test as a short-term fix. In the longer term there may be an
894        alternative.
896     3. I had an email from Eric Raymond about problems translating some of PCRE's
897        man pages to HTML (despite the fact that I distribute HTML pages, some
898        people do their own conversions for various reasons). The problems
899        concerned the use of low-level troff macros .br and .in. I have therefore
900        removed all such uses from the man pages (some were redundant, some could
901        be replaced by .nf/.fi pairs). The 132html script that I use to generate
902        HTML has been updated to handle .nf/.fi and to complain if it encounters
903        .br or .in.
905     4. Updated comments in configure.ac that get placed in config.h.in and also
906        arranged for config.h to be included in the distribution, with the name
907        config.h.generic, for the benefit of those who have to compile without
908        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
910     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
911        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
912        makevp.bat for the current PCRE, using the additional files
913        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
915     6. A Windows user reported a minor discrepancy with test 2, which turned out
916        to be caused by a trailing space on an input line that had got lost in his
917        copy. The trailing space was an accident, so I've just removed it.
919     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
920        that is needed.
922     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
923        as "const" (a) because they are and (b) because it helps the PHP
924        maintainers who have recently made a script to detect big data structures
925        in the php code that should be moved to the .rodata section. I remembered
926        to update Builducptable as well, so it won't revert if ucptable.h is ever
927        re-created.
929     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
930        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
931        order to be able to cut out the UTF-8 tables in the latter when UTF-8
932        support is not required. This saves 1.5-2K of code, which is important in
933        some applications.
935        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
936        so as not to refer to the tables, even though these functions will never be
937        called when UTF-8 support is disabled. Otherwise there are problems with a
938        shared library.
940    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
942        (a) It was defining its arguments as char * instead of void *.
944        (b) It was assuming that all moves were upwards in memory; this was true
945            a long time ago when I wrote it, but is no longer the case.
947        The emulated memove() is provided for those environments that have neither
948        memmove() nor bcopy(). I didn't think anyone used it these days, but that
949        is clearly not the case, as these two bugs were recently reported.
951    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
952        and Detrail to create the HTML documentation, the .txt form of the man
953        pages, and it removes trailing spaces from listed files. It also creates
954        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
955        case, it wraps all the #defines with #ifndefs. This script should be run
956        before "make dist".
958    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
959        with Unicode property support.
961        (a) For a maximizing quantifier, if the two different cases of the
962            character were of different lengths in their UTF-8 codings (there are
963            some cases like this - I found 11), and the matching function had to
964            back up over a mixture of the two cases, it incorrectly assumed they
965            were both the same length.
967        (b) When PCRE was configured to use the heap rather than the stack for
968            recursion during matching, it was not correctly preserving the data for
969            the other case of a UTF-8 character when checking ahead for a match
970            while processing a minimizing repeat. If the check also involved
971            matching a wide character, but failed, corruption could cause an
972            erroneous result when trying to check for a repeat of the original
973            character.
975    13. Some tidying changes to the testing mechanism:
977        (a) The RunTest script now detects the internal link size and whether there
978            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
979            values substituted by "configure". (The RunGrepTest script already did
980            this for UTF-8.) The configure.ac script no longer substitutes the
981            relevant variables.
983        (b) The debugging options /B and /D in pcretest show the compiled bytecode
984            with length and offset values. This means that the output is different
985            for different internal link sizes. Test 2 is skipped for link sizes
986            other than 2 because of this, bypassing the problem. Unfortunately,
987            there was also a test in test 3 (the locale tests) that used /B and
988            failed for link sizes other than 2. Rather than cut the whole test out,
989            I have added a new /Z option to pcretest that replaces the length and
990            offset values with spaces. This is now used to make test 3 independent
991            of link size. (Test 2 will be tidied up later.)
993    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
994        segmentation fault instead of returning the appropriate error message.
996    15. In multiline mode when the newline sequence was set to "any", the pattern
997        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
998        This doesn't seem right; it now treats the CRLF combination as the line
999        ending, and so does not match in that case. It's only a pattern such as ^$
1000        that would hit this one: something like ^ABC$ would have failed after \r
1001        and then tried again after \r\n.
1003    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1004        in an attempt to make files that differ only in their line terminators
1005        compare equal. This works on Linux.
1007    17. Under certain error circumstances pcregrep might try to free random memory
1008        as it exited. This is now fixed, thanks to valgrind.
1010    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1011        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1012        was because its rules for how to advance for /g after matching an empty
1013        string at the end of a line did not allow for this case. They now check for
1014        it specially.
1016    20. pcretest is supposed to handle patterns and data of any length, by
1017        extending its buffers when necessary. It was getting this wrong when the
1018        buffer for a data line had to be extended.
1020    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1021        CRLF as a newline sequence.
1023    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1024        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1025        I have nevertheless tidied it up.
1027    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1029    24. Added a man page for pcre-config.
1032    Version 7.0 19-Dec-06
1033    ---------------------
1035     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1036        moving to gcc 4.1.1.
1038     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1039        sys/time.h, which is documented as needed for this function. It doesn't
1040        seem to matter on Linux, but it showed up on some releases of OS X.
1042     3. It seems that there are systems where bytes whose values are greater than
1043        127 match isprint() in the "C" locale. The "C" locale should be the
1044        default when a C program starts up. In most systems, only ASCII printing
1045        characters match isprint(). This difference caused the output from pcretest
1046        to vary, making some of the tests fail. I have changed pcretest so that:
1048        (a) When it is outputting text in the compiled version of a pattern, bytes
1049            other than 32-126 are always shown as hex escapes.
1051        (b) When it is outputting text that is a matched part of a subject string,
1052            it does the same, unless a different locale has been set for the match
1053            (using the /L modifier). In this case, it uses isprint() to decide.
1055     4. Fixed a major bug that caused incorrect computation of the amount of memory
1056        required for a compiled pattern when options that changed within the
1057        pattern affected the logic of the preliminary scan that determines the
1058        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1059        that the computed length was too small. The symptoms of this bug were
1060        either the PCRE error "internal error: code overflow" from pcre_compile(),
1061        or a glibc crash with a message such as "pcretest: free(): invalid next
1062        size (fast)". Examples of patterns that provoked this bug (shown in
1063        pcretest format) are:
1065          /(?-x: )/x
1066          /(?x)(?-x: \s*#\s*)/
1067          /((?i)[\x{c0}])/8
1068          /(?i:[\x{c0}])/8
1070        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1071        is now done differently.
1073     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1074        wrapper classes; (b) implement a new function in the C++ scanner that is
1075        more efficient than the old way of doing things because it avoids levels of
1076        recursion in the regex matching; (c) add a paragraph to the documentation
1077        for the FullMatch() function.
1079     6. The escape sequence \n was being treated as whatever was defined as
1080        "newline". Not only was this contrary to the documentation, which states
1081        that \n is character 10 (hex 0A), but it also went horribly wrong when
1082        "newline" was defined as CRLF. This has been fixed.
1084     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1085        was being set to -1 for the "end of line" case (supposedly a value that no
1086        character can have). Though this value is never used (the check for end of
1087        line is "zero bytes in current character"), it caused compiler complaints.
1088        I've changed it to 0xffffffff.
1090     8. In pcre_version.c, the version string was being built by a sequence of
1091        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1092        string (as it is for production releases) called a macro with an empty
1093        argument. The C standard says the result of this is undefined. The gcc
1094        compiler treats it as an empty string (which was what was wanted) but it is
1095        reported that Visual C gives an error. The source has been hacked around to
1096        avoid this problem.
1098     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1099        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1100        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1101        of them did).
1103    10. Originally, pcretest opened its input and output without "b"; then I was
1104        told that "b" was needed in some environments, so it was added for release
1105        5.0 to both the input and output. (It makes no difference on Unix-like
1106        systems.) Later I was told that it is wrong for the input on Windows. I've
1107        now abstracted the modes into two macros, to make it easier to fiddle with
1108        them, and removed "b" from the input mode under Windows.
1110    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1112    12. Added -help and --help to pcretest as an official way of being reminded
1113        of the options.
1115    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1116        and pcrecpp.cc because they annoy compilers at high warning levels.
1118    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1120    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1121        scripts are not C programs :-) and which was not noticed because it works
1122        on Linux.
1124    16. pcretest is supposed to handle any length of pattern and data line (as one
1125        line or as a continued sequence of lines) by extending its input buffer if
1126        necessary. This feature was broken for very long pattern lines, leading to
1127        a string of junk being passed to pcre_compile() if the pattern was longer
1128        than about 50K.
1130    17. I have done a major re-factoring of the way pcre_compile() computes the
1131        amount of memory needed for a compiled pattern. Previously, there was code
1132        that made a preliminary scan of the pattern in order to do this. That was
1133        OK when PCRE was new, but as the facilities have expanded, it has become
1134        harder and harder to keep it in step with the real compile phase, and there
1135        have been a number of bugs (see for example, 4 above). I have now found a
1136        cunning way of running the real compile function in a "fake" mode that
1137        enables it to compute how much memory it would need, while actually only
1138        ever using a few hundred bytes of working memory and without too many
1139        tests of the mode. This should make future maintenance and development
1140        easier. A side effect of this work is that the limit of 200 on the nesting
1141        depth of parentheses has been removed (though this was never a serious
1142        limitation, I suspect). However, there is a downside: pcre_compile() now
1143        runs more slowly than before (30% or more, depending on the pattern). I
1144        hope this isn't a big issue. There is no effect on runtime performance.
1146    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1147        newline (only possible for the last line of a file) and it was a
1148        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1150    19. Added additional timing features to pcretest. (1) The -tm option now times
1151        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1152        separate command line item, by a number that specifies the number of
1153        repeats to use when timing. The default is 50000; this gives better
1154        precision, but takes uncomfortably long for very large patterns.
1156    20. Extended pcre_study() to be more clever in cases where a branch of a
1157        subpattern has no definite first character. For example, (a*|b*)[cd] would
1158        previously give no result from pcre_study(). Now it recognizes that the
1159        first character must be a, b, c, or d.
1161    21. There was an incorrect error "recursive call could loop indefinitely" if
1162        a subpattern (or the entire pattern) that was being tested for matching an
1163        empty string contained only one non-empty item after a nested subpattern.
1164        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1165        incorrectly, because the \d was being skipped in the check.
1167    22. The pcretest program now has a new pattern option /B and a command line
1168        option -b, which is equivalent to adding /B to every pattern. This causes
1169        it to show the compiled bytecode, without the additional information that
1170        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1171        is the same as /B/I).
1173    23. A new optimization is now able automatically to treat some sequences such
1174        as a*b as a*+b. More specifically, if something simple (such as a character
1175        or a simple class like \d) has an unlimited quantifier, and is followed by
1176        something that cannot possibly match the quantified thing, the quantifier
1177        is automatically "possessified".
1179    24. A recursive reference to a subpattern whose number was greater than 39
1180        went wrong under certain circumstances in UTF-8 mode. This bug could also
1181        have affected the operation of pcre_study().
1183    25. Realized that a little bit of performance could be had by replacing
1184        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1186    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1188    27. Possessive quantifiers such as a++ were previously implemented by turning
1189        them into atomic groups such as ($>a+). Now they have their own opcodes,
1190        which improves performance. This includes the automatically created ones
1191        from 23 above.
1193    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1194        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1195        the first matched character to be a colon. This applied both to named and
1196        numbered groups.
1198    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1200    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1201        building PCRE on Windows easier, so I have included it in the distribution.
1203    31. There is now a check in pcretest against a ridiculously large number being
1204        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1205        loop, the loop is abandoned.
1207    32. Forward references to subpatterns in conditions such as (?(2)...) where
1208        subpattern 2 is defined later cause pcre_compile() to search forwards in
1209        the pattern for the relevant set of parentheses. This search went wrong
1210        when there were unescaped parentheses in a character class, parentheses
1211        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1213    33. "Subroutine" calls and backreferences were previously restricted to
1214        referencing subpatterns earlier in the regex. This restriction has now
1215        been removed.
1217    34. Added a number of extra features that are going to be in Perl 5.10. On the
1218        whole, these are just syntactic alternatives for features that PCRE had
1219        previously implemented using the Python syntax or my own invention. The
1220        other formats are all retained for compatibility.
1222        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1223            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1224            also .NET compatible.
1226        (b) A recursion or subroutine call to a named group can now be defined as
1227            (?&name) as well as (?P>name).
1229        (c) A backreference to a named group can now be defined as \k<name> or
1230            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1231            5.10, are also .NET compatible.
1233        (d) A conditional reference to a named group can now use the syntax
1234            (?(<name>) or (?('name') as well as (?(name).
1236        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1237            groups (named and numbered) that are never evaluated inline, but can be
1238            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1239            is always false. There may be only one alternative in such a group.
1241        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1242            as the simple (?(R). The condition is true only if the most recent
1243            recursion is that of the given number or name. It does not search out
1244            through the entire recursion stack.
1246        (g) The escape \gN or \g{N} has been added, where N is a positive or
1247            negative number, specifying an absolute or relative reference.
1249    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1250        some "unreachable code" warnings.
1252    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1253        things, this adds five new scripts.
1255    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1256        There were also incompatibilities regarding the handling of \Q..\E inside
1257        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1258        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1260    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1261        matches an empty string, and forcibly breaks the loop. There were bugs in
1262        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1263        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1264        separate and independent bugs (that affected different cases) have been
1265        fixed.
1267    39. Refactored the code to abolish the use of different opcodes for small
1268        capturing bracket numbers. This is a tidy that I avoided doing when I
1269        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1270        The new approach is not only tidier, it makes it possible to reduce the
1271        memory needed to fix the previous bug (38).
1273    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1274        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1275        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1276        mode.
1278    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1279        report.
1281    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1282        copy construction and assignment in the C++ wrapper.
1284    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1285        couple of bugs that could have given wrong results in the "--newline=crlf"
1286        case.
1288    44. Added a number of casts and did some reorganization of signed/unsigned int
1289        variables following suggestions from Dair Grant. Also renamed the variable
1290        "this" as "item" because it is a C++ keyword.
1292    45. Arranged for dftables to add
1294          #include "pcre_internal.h"
1296        to pcre_chartables.c because without it, gcc 4.x may remove the array
1297        definition from the final binary if PCRE is built into a static library and
1298        dead code stripping is activated.
1300    46. For an unanchored pattern, if a match attempt fails at the start of a
1301        newline sequence, and the newline setting is CRLF or ANY, and the next two
1302        characters are CRLF, advance by two characters instead of one.
1305    Version 6.7 04-Jul-06
1306    ---------------------
1308     1. In order to handle tests when input lines are enormously long, pcretest has
1309        been re-factored so that it automatically extends its buffers when
1310        necessary. The code is crude, but this _is_ just a test program. The
1311        default size has been increased from 32K to 50K.
1313     2. The code in pcre_study() was using the value of the re argument before
1314        testing it for NULL. (Of course, in any sensible call of the function, it
1315        won't be NULL.)
1317     3. The memmove() emulation function in pcre_internal.h, which is used on
1318        systems that lack both memmove() and bcopy() - that is, hardly ever -
1319        was missing a "static" storage class specifier.
1321     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1322        containing an extended class (one that cannot be represented by a bitmap
1323        because it contains high-valued characters or Unicode property items, e.g.
1324        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1325        pattern, but PCRE should not loop if you do not (it no longer does).
1326        [Detail: two cases were found: (a) a repeated subpattern containing an
1327        extended class; (b) a recursive reference to a subpattern that followed a
1328        previous extended class. It wasn't skipping over the extended class
1329        correctly when UTF-8 mode was not set.]
1331     5. A negated single-character class was not being recognized as fixed-length
1332        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1333        compile error "lookbehind assertion is not fixed length".
1335     6. The RunPerlTest auxiliary script was showing an unexpected difference
1336        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1337        write a Perl script that can interpret lines of an input file either as
1338        byte characters or as UTF-8, which is what "perltest" was being required to
1339        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1340        can't do is switch easily at run time between having the "use utf8;" pragma
1341        or not. In the end, I fudged it by using the RunPerlTest script to insert
1342        "use utf8;" explicitly for the UTF-8 tests.
1344     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1345        the end of the subject string, contrary to the documentation and to what
1346        Perl does. This was true of both matching functions. Now it matches only at
1347        the start of the subject and immediately after *internal* newlines.
1349     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1350        a pointer to an int instead of a pointer to an unsigned long int. This
1351        caused problems on 64-bit systems.
1353     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1354        instance of the 'standard' template library not being so standard".
1356    10. There was no check on the number of named subpatterns nor the maximum
1357        length of a subpattern name. The product of these values is used to compute
1358        the size of the memory block for a compiled pattern. By supplying a very
1359        long subpattern name and a large number of named subpatterns, the size
1360        computation could be caused to overflow. This is now prevented by limiting
1361        the length of names to 32 characters, and the number of named subpatterns
1362        to 10,000.
1364    11. Subpatterns that are repeated with specific counts have to be replicated in
1365        the compiled pattern. The size of memory for this was computed from the
1366        length of the subpattern and the repeat count. The latter is limited to
1367        65535, but there was no limit on the former, meaning that integer overflow
1368        could in principle occur. The compiled length of a repeated subpattern is
1369        now limited to 30,000 bytes in order to prevent this.
1371    12. Added the optional facility to have named substrings with the same name.
1373    13. Added the ability to use a named substring as a condition, using the
1374        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1375        are numbers (not recommended). Forward references are permitted.
1377    14. Added forward references in named backreferences (if you see what I mean).
1379    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1380        pattern could run off the end of the subject. For example, the pattern
1381        "(?s)(.{1,5})"8 did this with the subject "ab".
1383    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1384        PCRE_CASELESS was set when matching characters that were quantified with ?
1385        or *.
1387    17. A character class other than a single negated character that had a minimum
1388        but no maximum quantifier - for example [ab]{6,} - was not handled
1389        correctly by pce_dfa_exec(). It would match only one character.
1391    18. A valid (though odd) pattern that looked like a POSIX character
1392        class but used an invalid character after [ (for example [[,abc,]]) caused
1393        pcre_compile() to give the error "Failed: internal error: code overflow" or
1394        in some cases to crash with a glibc free() error. This could even happen if
1395        the pattern terminated after [[ but there just happened to be a sequence of
1396        letters, a binary zero, and a closing ] in the memory that followed.
1398    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1399        over the years. Originally (before any Unicode support), just the bottom 8
1400        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1401        output from "man perlunicode" includes this:
1403          The regular expression compiler produces polymorphic opcodes.  That
1404          is, the pattern adapts to the data and automatically switches to
1405          the Unicode character scheme when presented with Unicode data--or
1406          instead uses a traditional byte scheme when presented with byte
1407          data.
1409        Sadly, a wide octal escape does not cause a switch, and in a string with
1410        no other multibyte characters, these octal escapes are treated as before.
1411        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1412        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1413        Unicode string.
1415        I have not perpetrated such confusion in PCRE. Up till now, it took just
1416        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1417        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1418        translate to the appropriate multibyte character.
1420    29. Applied some refactoring to reduce the number of warnings from Microsoft
1421        and Borland compilers. This has included removing the fudge introduced
1422        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1423        a warning about an unused variable.
1425    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1426        characters since release 4.0, because Perl (from release 5.004) does not.
1427        [Or at least, is documented not to: some releases seem to be in conflict
1428        with the documentation.] However, when a pattern was studied with
1429        pcre_study() and all its branches started with \s, PCRE still included VT
1430        as a possible starting character. Of course, this did no harm; it just
1431        caused an unnecessary match attempt.
1433    22. Removed a now-redundant internal flag bit that recorded the fact that case
1434        dependency changed within the pattern. This was once needed for "required
1435        byte" processing, but is no longer used. This recovers a now-scarce options
1436        bit. Also moved the least significant internal flag bit to the most-
1437        significant bit of the word, which was not previously used (hangover from
1438        the days when it was an int rather than a uint) to free up another bit for
1439        the future.
1441    23. Added support for CRLF line endings as well as CR and LF. As well as the
1442        default being selectable at build time, it can now be changed at runtime
1443        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1444        specify that it is scanning data with non-default line endings.
1446    24. Changed the definition of CXXLINK to make it agree with the definition of
1447        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1449    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1450        recursions. This makes a big different to stack usage for some patterns.
1452    26. If a subpattern containing a named recursion or subroutine reference such
1453        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1454        the space required for the compiled pattern went wrong and gave too small a
1455        value. Depending on the environment, this could lead to "Failed: internal
1456        error: code overflow at offset 49" or "glibc detected double free or
1457        corruption" errors.
1459    27. Applied patches from Google (a) to support the new newline modes and (b) to
1460        advance over multibyte UTF-8 characters in GlobalReplace.
1462    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1463        difference for some implementation of PCRE in some Windows version.
1465    29. Added some extra testing facilities to pcretest:
1467        \q<number>   in a data line sets the "match limit" value
1468        \Q<number>   in a data line sets the "match recursion limt" value
1469        -S <number>  sets the stack size, where <number> is in megabytes
1471        The -S option isn't available for Windows.
1474    Version 6.6 06-Feb-06
1475    ---------------------
1477     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1478        in pcreposix.h. I have copied the definition from pcre.h.
1480     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1481        because pcre.h is no longer a built file.
1483     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1484        not normally included in the compiled code.
1487    Version 6.5 01-Feb-06
1488    ---------------------
1490     1. When using the partial match feature with pcre_dfa_exec(), it was not
1491        anchoring the second and subsequent partial matches at the new starting
1492        point. This could lead to incorrect results. For example, with the pattern
1493        /1234/, partially matching against "123" and then "a4" gave a match.
1495     2. Changes to pcregrep:
1497        (a) All non-match returns from pcre_exec() were being treated as failures
1498            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1499            error message is output. Some extra information is given for the
1500            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1501            probably the only errors that are likely to be caused by users (by
1502            specifying a regex that has nested indefinite repeats, for instance).
1503            If there are more than 20 of these errors, pcregrep is abandoned.
1505        (b) A binary zero was treated as data while matching, but terminated the
1506            output line if it was written out. This has been fixed: binary zeroes
1507            are now no different to any other data bytes.
1509        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1510            used to set a locale for matching. The --locale=xxxx long option has
1511            been added (no short equivalent) to specify a locale explicitly on the
1512            pcregrep command, overriding the environment variables.
1514        (d) When -B was used with -n, some line numbers in the output were one less
1515            than they should have been.
1517        (e) Added the -o (--only-matching) option.
1519        (f) If -A or -C was used with -c (count only), some lines of context were
1520            accidentally printed for the final match.
1522        (g) Added the -H (--with-filename) option.
1524        (h) The combination of options -rh failed to suppress file names for files
1525            that were found from directory arguments.
1527        (i) Added the -D (--devices) and -d (--directories) options.
1529        (j) Added the -F (--fixed-strings) option.
1531        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1533        (l) Added the --colo(u)r option.
1535        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1536            is not present by default.
1538     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1539        items such as (?R) or (?1), when the recursion could match a number of
1540        alternatives. If it matched one of the alternatives, but subsequently,
1541        outside the recursion, there was a failure, the code tried to back up into
1542        the recursion. However, because of the way PCRE is implemented, this is not
1543        possible, and the result was an incorrect result from the match.
1545        In order to prevent this happening, the specification of recursion has
1546        been changed so that all such subpatterns are automatically treated as
1547        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1549     4. I had overlooked the fact that, in some locales, there are characters for
1550        which isalpha() is true but neither isupper() nor islower() are true. In
1551        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1552        and ordfeminine) are like this. This affected the treatment of \w and \W
1553        when they appeared in character classes, but not when they appeared outside
1554        a character class. The bit map for "word" characters is now created
1555        separately from the results of isalnum() instead of just taking it from the
1556        upper, lower, and digit maps. (Plus the underscore character, of course.)
1558     5. The above bug also affected the handling of POSIX character classes such as
1559        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1560        permanent tables. Instead, the bit maps for such a class were previously
1561        created as the appropriate unions of the upper, lower, and digit bitmaps.
1562        Now they are created by subtraction from the [[:word:]] class, which has
1563        its own bitmap.
1565     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1566        It is created by subtracting the vertical space characters (\x09, \x0a,
1567        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1568        subtraction was done in the overall bitmap for a character class, meaning
1569        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1570        be recognized. This bug has been fixed.
1572     7. Patches from the folks at Google:
1574          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1575          real life, but is still worth protecting against".
1577          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1578          regular expressions".
1580          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1581          have it.
1583          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1584          "configure" and the latter not, in order to fix a problem somebody had
1585          with compiling the Arg class on HP-UX.
1587          (e) Improve the error-handling of the C++ wrapper a little bit.
1589          (f) New tests for checking recursion limiting.
1591     8. The pcre_memmove() function, which is used only if the environment does not
1592        have a standard memmove() function (and is therefore rarely compiled),
1593        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1594        returning a result (though PCRE never actually uses the result).
1596     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1597        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1598        returned instead of calling malloc() with an overflowing number that would
1599        most likely cause subsequent chaos.
1601    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1603    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1604        with this option is matched, the nmatch and pmatch options of regexec() are
1605        ignored.
1607    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1608        provided in case anyone wants to the the POSIX interface with UTF-8
1609        strings.
1611    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1612        C++ linking (needed for some HP-UX environments).
1614    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1615        (unused parameter) and in the pcre_printint() function (omitted "default"
1616        switch label when the default is to do nothing).
1618    15. Added some code to make it possible, when PCRE is compiled as a C++
1619        library, to replace subject pointers for pcre_exec() with a smart pointer
1620        class, thus making it possible to process discontinuous strings.
1622    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1623        much the same function. They were added by different people who were trying
1624        to make PCRE easy to compile on non-Unix systems. It has been suggested
1625        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1626        for compiling on Windows systems. I have therefore replaced it with
1627        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1628        defaults to "extern" for C or "extern C" for C++, which works fine on
1629        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1630        SCOPE with something explicit in config.h. In addition:
1632        (a) pcreposix.h still had just "extern" instead of either of these macros;
1633            I have replaced it with PCRE_DATA_SCOPE.
1635        (b) Functions such as _pcre_xclass(), which are internal to the library,
1636            but external in the C sense, all had PCRE_EXPORT in their definitions.
1637            This is apparently wrong for the Windows case, so I have removed it.
1638            (It makes no difference on Unix-like systems.)
1640    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1641        of recursive calls to match(). This is different to MATCH_LIMIT because
1642        that limits the total number of calls to match(), not all of which increase
1643        the depth of recursion. Limiting the recursion depth limits the amount of
1644        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1645        when PCRE is compiled, and changed at run time. A patch from Google adds
1646        this functionality to the C++ interface.
1648    18. Changes to the handling of Unicode character properties:
1650        (a) Updated the table to Unicode 4.1.0.
1652        (b) Recognize characters that are not in the table as "Cn" (undefined).
1654        (c) I revised the way the table is implemented to a much improved format
1655            which includes recognition of ranges. It now supports the ranges that
1656            are defined in UnicodeData.txt, and it also amalgamates other
1657            characters into ranges. This has reduced the number of entries in the
1658            table from around 16,000 to around 3,000, thus reducing its size
1659            considerably. I realized I did not need to use a tree structure after
1660            all - a binary chop search is just as efficient. Having reduced the
1661            number of entries, I extended their size from 6 bytes to 8 bytes to
1662            allow for more data.
1664        (d) Added support for Unicode script names via properties such as \p{Han}.
1666    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1667        matching that character.
1669    20. When matching a repeated Unicode property with a minimum greater than zero,
1670        (for example \pL{2,}), PCRE could look past the end of the subject if it
1671        reached it while seeking the minimum number of characters. This could
1672        happen only if some of the characters were more than one byte long, because
1673        there is a check for at least the minimum number of bytes.
1675    21. Refactored the implementation of \p and \P so as to be more general, to
1676        allow for more different types of property in future. This has changed the
1677        compiled form incompatibly. Anybody with saved compiled patterns that use
1678        \p or \P will have to recompile them.
1680    22. Added "Any" and "L&" to the supported property types.
1682    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1683        but give a compile time error if the value is greater than 0xff.
1685    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1686        accidentally not being installed or uninstalled.
1688    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1689        made were to insert the current release number. This seemed silly, because
1690        it made things harder for people building PCRE on systems that don't run
1691        "configure". I have turned pcre.h into a distributed file, no longer built
1692        by "configure", with the version identification directly included. There is
1693        no longer a pcre.h.in file.
1695        However, this change necessitated a change to the pcre-config script as
1696        well. It is built from pcre-config.in, and one of the substitutions was the
1697        release number. I have updated configure.ac so that ./configure now finds
1698        the release number by grepping pcre.h.
1700    26. Added the ability to run the tests under valgrind.
1703    Version 6.4 05-Sep-05
1704    ---------------------
1706     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1707        "--" to be printed when multiple files were scanned, even when none of the
1708        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1709        consider it to be a bug, and have restored the previous behaviour.
1711     2. A couple of code tidies to get rid of compiler warnings.
1713     3. The pcretest program used to cheat by referring to symbols in the library
1714        whose names begin with _pcre_. These are internal symbols that are not
1715        really supposed to be visible externally, and in some environments it is
1716        possible to suppress them. The cheating is now confined to including
1717        certain files from the library's source, which is a bit cleaner.
1719     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1720        file's purpose clearer.
1722     5. Reorganized pcre_ucp_findchar().
1725    Version 6.3 15-Aug-05
1726    ---------------------
1728     1. The file libpcre.pc.in did not have general read permission in the tarball.
1730     2. There were some problems when building without C++ support:
1732        (a) If C++ support was not built, "make install" and "make test" still
1733            tried to test it.
1735        (b) There were problems when the value of CXX was explicitly set. Some
1736            changes have been made to try to fix these, and ...
1738        (c) --disable-cpp can now be used to explicitly disable C++ support.
1740        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1741            backslash in a target when C++ was disabled. This confuses some
1742            versions of "make", apparently. Using an intermediate variable solves
1743            this. (Same for CPP_LOBJ.)
1745     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1746        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1747        necessary on certain architectures.
1749     4. Added a setting of -export-symbols-regex to the link command to remove
1750        those symbols that are exported in the C sense, but actually are local
1751        within the library, and not documented. Their names all begin with
1752        "_pcre_". This is not a perfect job, because (a) we have to except some
1753        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1754        available (and never for static libraries). I have made a note to try to
1755        find a way round (a) in the future.
1758    Version 6.2 01-Aug-05
1759    ---------------------
1761     1. There was no test for integer overflow of quantifier values. A construction
1762        such as {1111111111111111} would give undefined results. What is worse, if
1763        a minimum quantifier for a parenthesized subpattern overflowed and became
1764        negative, the calculation of the memory size went wrong. This could have
1765        led to memory overwriting.
1767     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1769     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1770        operating environments where this matters.
1772     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1773        PCRE options from within the C++ wrapper.
1775     5. Named capturing subpatterns were not being correctly counted when a pattern
1776        was compiled. This caused two problems: (a) If there were more than 100
1777        such subpatterns, the calculation of the memory needed for the whole
1778        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1779        back references of the form \12, where the number was greater than 9, were
1780        not recognized as back references, even though there were sufficient
1781        previous subpatterns.
1783     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1784        versions of gcc, e.g. 2.95.4.
1787    Version 6.1 21-Jun-05
1788    ---------------------
1790     1. There was one reference to the variable "posix" in pcretest.c that was not
1791        surrounded by "#if !defined NOPOSIX".
1793     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1794        the cross-check on the old pcre_info() function, for the benefit of the
1795        cut-down version of PCRE that is currently imported into Exim.
1797     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1798        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1799        patterns that start (?i) but I don't think that matters. The use of (?i) is
1800        just an example; this all applies to the other options as well.
1802     4. Since libtool seems to echo the compile commands it is issuing, the output
1803        from "make" can be reduced a bit by putting "@" in front of each libtool
1804        compile command.
1806     5. Patch from the folks at Google for configure.in to be a bit more thorough
1807        in checking for a suitable C++ installation before trying to compile the
1808        C++ stuff. This should fix a reported problem when a compiler was present,
1809        but no suitable headers.
1811     6. The man pages all had just "PCRE" as their title. I have changed them to
1812        be the relevant file name. I have also arranged that these names are
1813        retained in the file doc/pcre.txt, which is a concatenation in text format
1814        of all the man pages except the little individual ones for each function.
1816     7. The NON-UNIX-USE file had not been updated for the different set of source
1817        files that come with release 6. I also added a few comments about the C++
1818        wrapper.
1821    Version 6.0 07-Jun-05
1822    ---------------------
1824     1. Some minor internal re-organization to help with my DFA experiments.
1826     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1827        didn't matter for the library itself when fully configured, but did matter
1828        when compiling without UCP support, or within Exim, where the ucp files are
1829        not imported.
1831     3. Refactoring of the library code to split up the various functions into
1832        different source modules. The addition of the new DFA matching code (see
1833        below) to a single monolithic source would have made it really too
1834        unwieldy, quite apart from causing all the code to be include in a
1835        statically linked application, when only some functions are used. This is
1836        relevant even without the DFA addition now that patterns can be compiled in
1837        one application and matched in another.
1839        The downside of splitting up is that there have to be some external
1840        functions and data tables that are used internally in different modules of
1841        the library but which are not part of the API. These have all had their
1842        names changed to start with "_pcre_" so that they are unlikely to clash
1843        with other external names.
1845     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1846        a different (DFA) algorithm. Although it is slower than the original
1847        function, it does have some advantages for certain types of matching
1848        problem.
1850     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1851        including restarting after a partial match.
1853     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1854        defined when compiling for Windows was sent to me. I have put it into the
1855        code, though I have no means of testing or verifying it.
1857     7. Added the pcre_refcount() auxiliary function.
1859     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1860        match before or at the first newline in the subject string. In pcretest,
1861        the /f option on a pattern can be used to set this.
1863     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1864        would behave wrongly. This has been present in PCRE since release 4.0.
1866    10. A number of changes to the pcregrep command:
1868        (a) Refactored how -x works; insert ^(...)$ instead of setting
1869            PCRE_ANCHORED and checking the length, in preparation for adding
1870            something similar for -w.
1872        (b) Added the -w (match as a word) option.
1874        (c) Refactored the way lines are read and buffered so as to have more
1875            than one at a time available.
1877        (d) Implemented a pcregrep test script.
1879        (e) Added the -M (multiline match) option. This allows patterns to match
1880            over several lines of the subject. The buffering ensures that at least
1881            8K, or the rest of the document (whichever is the shorter) is available
1882            for matching (and similarly the previous 8K for lookbehind assertions).
1884        (f) Changed the --help output so that it now says
1886              -w, --word-regex(p)
1888            instead of two lines, one with "regex" and the other with "regexp"
1889            because that confused at least one person since the short forms are the
1890            same. (This required a bit of code, as the output is generated
1891            automatically from a table. It wasn't just a text change.)
1893        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1894            option but starts with a hyphen. Could be a pattern or a path name
1895            starting with a hyphen, for instance.
1897        (h) "-" can be given as a file name to represent stdin.
1899        (i) When file names are being printed, "(standard input)" is used for
1900            the standard input, for compatibility with GNU grep. Previously
1901            "<stdin>" was used.
1903        (j) The option --label=xxx can be used to supply a name to be used for
1904            stdin when file names are being printed. There is no short form.
1906        (k) Re-factored the options decoding logic because we are going to add
1907            two more options that take data. Such options can now be given in four
1908            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1910        (l) Added the -A, -B, and -C options for requesting that lines of context
1911            around matches be printed.
1913        (m) Added the -L option to print the names of files that do not contain
1914            any matching lines, that is, the complement of -l.
1916        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1917            continue to scan other files.
1919        (o) The -s option was incorrectly implemented. For compatibility with other
1920            greps, it now suppresses the error message for a non-existent or non-
1921            accessible file (but not the return code). There is a new option called
1922            -q that suppresses the output of matching lines, which was what -s was
1923            previously doing.
1925        (p) Added --include and --exclude options to specify files for inclusion
1926            and exclusion when recursing.
1928    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1929        Hopefully, it now does.
1931    12. Missing cast in pcre_study().
1933    13. Added an "uninstall" target to the makefile.
1935    14. Replaced "extern" in the function prototypes in Makefile.in with
1936        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1937        world, but is set differently for Windows.
1939    15. Added a second compiling function called pcre_compile2(). The only
1940        difference is that it has an extra argument, which is a pointer to an
1941        integer error code. When there is a compile-time failure, this is set
1942        non-zero, in addition to the error test pointer being set to point to an
1943        error message. The new argument may be NULL if no error number is required
1944        (but then you may as well call pcre_compile(), which is now just a
1945        wrapper). This facility is provided because some applications need a
1946        numeric error indication, but it has also enabled me to tidy up the way
1947        compile-time errors are handled in the POSIX wrapper.
1949    16. Added VPATH=.libs to the makefile; this should help when building with one
1950        prefix path and installing with another. (Or so I'm told by someone who
1951        knows more about this stuff than I do.)
1953    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1954        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1955        match everything, including newlines. This is not POSIX-compatible, but
1956        somebody wanted the feature. From pcretest it can be activated by using
1957        both the P and the s flags.
1959    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1961    19. libpcre.pc was being incorrectly installed as executable.
1963    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1964        it now also looks for '\r' so that it will work unmodified on Windows.
1966    21. Added Google's contributed C++ wrapper to the distribution.
1968    22. Added some untidy missing memory free() calls in pcretest, to keep
1969        Electric Fence happy when testing.
1973    Version 5.0 13-Sep-04
1974    ---------------------
1976     1. Internal change: literal characters are no longer packed up into items
1977        containing multiple characters in a single byte-string. Each character
1978        is now matched using a separate opcode. However, there may be more than one
1979        byte in the character in UTF-8 mode.
1981     2. The pcre_callout_block structure has two new fields: pattern_position and
1982        next_item_length. These contain the offset in the pattern to the next match
1983        item, and its length, respectively.
1985     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1986        insertion of callouts before each pattern item. Added the /C option to
1987        pcretest to make use of this.
1989     4. On the advice of a Windows user, the lines
1991          #if defined(_WIN32) || defined(WIN32)
1992          _setmode( _fileno( stdout ), 0x8000 );
1993          #endif  /* defined(_WIN32) || defined(WIN32) */
1995        have been added to the source of pcretest. This apparently does useful
1996        magic in relation to line terminators.
1998     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1999        for the benefit of those environments where the "b" makes a difference.
2001     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2002        to know about it. I have put a hack into configure.in that adds in code
2003        to set GCC=yes if CC=icc. This seems to end up at a point in the
2004        generated configure script that is early enough to affect the setting of
2005        compiler options, which is what is needed, but I have no means of testing
2006        whether it really works. (The user who reported this had patched the
2007        generated configure script, which of course I cannot do.)
2009        LATER: After change 22 below (new libtool files), the configure script
2010        seems to know about icc (and also ecc). Therefore, I have commented out
2011        this hack in configure.in.
2013     7. Added support for pkg-config (2 patches were sent in).
2015     8. Negated POSIX character classes that used a combination of internal tables
2016        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2017        [[:^ascii]]. Typically, they would match almost any characters. The other
2018        POSIX classes were not broken in this way.
2020     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2021        to find the match, as PCRE was deluded into thinking that the match had to
2022        start at the start point or following a newline. The same bug applied to
2023        patterns with negative forward assertions or any backward assertions
2024        preceding ".*" at the start, unless the pattern required a fixed first
2025        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2027    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2028        starting at the last subject character, bytes beyond the end of the subject
2029        string were read.
2031    11. Renamed the variable "class" as "classbits" to make life easier for C++
2032        users. (Previously there was a macro definition, but it apparently wasn't
2033        enough.)
2035    12. Added the new field "tables" to the extra data so that tables can be passed
2036        in at exec time, or the internal tables can be re-selected. This allows
2037        a compiled regex to be saved and re-used at a later time by a different
2038        program that might have everything at different addresses.
2040    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2041        -R library as well as a -L library.
2043    14. The debugging options of pcretest (-d on the command line or D on a
2044        pattern) showed incorrect output for anything following an extended class
2045        that contained multibyte characters and which was followed by a quantifier.
2047    15. Added optional support for general category Unicode character properties
2048        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2049        support. It adds about 90K to the size of the library. The meanings of the
2050        inbuilt class escapes such as \d and \s have NOT been changed.
2052    16. Updated pcredemo.c to include calls to free() to release the memory for the
2053        compiled pattern.
2055    17. The generated file chartables.c was being created in the source directory
2056        instead of in the building directory. This caused the build to fail if the
2057        source directory was different from the building directory, and was
2058        read-only.
2060    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2061        file. No doubt somebody will tell me if they don't make sense... Also added
2062        Dan Mooney's comments about building on OpenVMS.
2064    19. Added support for partial matching via the PCRE_PARTIAL option for
2065        pcre_exec() and the \P data escape in pcretest.
2067    20. Extended pcretest with 3 new pattern features:
2069        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2070              write the compiled pattern to the file whose name is "rest-of-line".
2071              This is a straight binary dump of the data, with the saved pointer to
2072              the character tables forced to be NULL. The study data, if any, is
2073              written too. After writing, pcretest reads a new pattern.
2075        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2076              compiled pattern from the given file. There must not be any
2077              occurrences of "<" in the file name (pretty unlikely); if there are,
2078              pcretest will instead treat the initial "<" as a pattern delimiter.
2079              After reading in the pattern, pcretest goes on to read data lines as
2080              usual.
2082        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2083              and 16-bit fields in a compiled pattern, to simulate a pattern that
2084              was compiled on a host of opposite endianness.
2086    21. The pcre-exec() function can now cope with patterns that were compiled on
2087        hosts of opposite endianness, with this restriction:
2089          As for any compiled expression that is saved and used later, the tables
2090          pointer field cannot be preserved; the extra_data field in the arguments
2091          to pcre_exec() should be used to pass in a tables address if a value
2092          other than the default internal tables were used at compile time.
2094    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2095        now diagnosed as an error. Previously, most of the time, a negative number
2096        would have been treated as zero, but if in addition "ovector" was passed as
2097        NULL, a crash could occur.
2099    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2100        new versions from the libtool 1.5 distribution (the last one is a copy of
2101        a file called libtool.m4). This seems to have fixed the need to patch
2102        "configure" to support Darwin 1.3 (which I used to do). However, I still
2103        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2104        workstation).
2106    24. Changed the PCRE licence to be the more standard "BSD" licence.
2109    Version 4.5 01-Dec-03
2110    ---------------------
2112     1. There has been some re-arrangement of the code for the match() function so
2113        that it can be compiled in a version that does not call itself recursively.
2114        Instead, it keeps those local variables that need separate instances for
2115        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2116        needs to "recurse". Keeping track of where control must go is done by means
2117        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2118        hide most of the details from the main code, and operates only if
2119        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2120        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2121        operating.
2123        To make it easier for callers to provide specially tailored get/free
2124        functions for this usage, two new functions, pcre_stack_malloc, and
2125        pcre_stack_free, are used. They are always called in strict stacking order,
2126        and the size of block requested is always the same.
2128        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2129        PCRE has been compiled to use the stack or the heap for recursion. The
2130        -C option of pcretest uses this to show which version is compiled.
2132        A new data escape \S, is added to pcretest; it causes the amounts of store
2133        obtained and freed by both kinds of malloc/free at match time to be added
2134        to the output.
2136     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2137        what's available on my current Linux desktop machine.
2139     3. When matching a UTF-8 string, the test for a valid string at the start has
2140        been extended. If start_offset is not zero, PCRE now checks that it points
2141        to a byte that is the start of a UTF-8 character. If not, it returns
2142        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2143        this is necessary because there may be backward assertions in the pattern.
2144        When matching the same subject several times, it may save resources to use
2145        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2147     4. The code for checking the validity of UTF-8 strings has been tightened so
2148        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2149        containing "overlong sequences".
2151     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2152        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2153        should have been "&", but it just so happened that all the cases this let
2154        through by mistake were picked up later in the function.
2156     6. I had used a variable called "isblank" - this is a C99 function, causing
2157        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2159     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2160        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2161        so the version is shown; (c) stop "make test" echoing "./RunTest".
2163     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2165     9. The prototype for memmove() for systems that don't have it was using
2166        size_t, but the inclusion of the header that defines size_t was later. I've
2167        moved the #includes for the C headers earlier to avoid this.
2169    10. Added some adjustments to the code to make it easier to compiler on certain
2170        special systems:
2172          (a) Some "const" qualifiers were missing.
2173          (b) Added the macro EXPORT before all exported functions; by default this
2174              is defined to be empty.
2175          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2176              that it reads its output file name as an argument instead of writing
2177              to the standard output and assuming this can be redirected.
2179    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2180        class containing characters with values greater than 255, PCRE compilation
2181        went into a loop.
2183    12. A recursive reference to a subpattern that was within another subpattern
2184        that had a minimum quantifier of zero caused PCRE to crash. For example,
2185        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2186        recursion. If the recursively-called subpattern itself had a zero repeat,
2187        that was OK.
2189    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2190        buffer into which it was copied (for escape processing) was still set at
2191        1024, so long lines caused crashes.
2193    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2194        "internal error: code overflow...". This applied to any character class
2195        that was followed by a possessive quantifier.
2197    15. Modified the Makefile to add libpcre.la as a prerequisite for
2198        libpcreposix.la because I was told this is needed for a parallel build to
2199        work.
2201    16. If a pattern that contained .* following optional items at the start was
2202        studied, the wrong optimizing data was generated, leading to matching
2203        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2204        matching string must start with a or b or c. The correct conclusion for
2205        this pattern is that a match can start with any character.
2208    Version 4.4 13-Aug-03
2209    ---------------------
2211     1. In UTF-8 mode, a character class containing characters with values between
2212        127 and 255 was not handled correctly if the compiled pattern was studied.
2213        In fixing this, I have also improved the studying algorithm for such
2214        classes (slightly).
2216     2. Three internal functions had redundant arguments passed to them. Removal
2217        might give a very teeny performance improvement.
2219     3. Documentation bug: the value of the capture_top field in a callout is *one
2220        more than* the number of the hightest numbered captured substring.
2222     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2223        in incorrectly linking with a previously installed version. They now link
2224        explicitly with libpcre.la.
2226     5. configure.in no longer needs to recognize Cygwin specially.
2228     6. A problem in pcre.in for Windows platforms is fixed.
2230     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2231        pcretest, it used to include the size of the study block as part of its
2232        output. Unfortunately, the structure contains a field that has a different
2233        size on different hardware architectures. This meant that the tests that
2234        showed this size failed. As the block is currently always of a fixed size,
2235        this information isn't actually particularly useful in pcretest output, so
2236        I have just removed it.
2238     8. Three pre-processor statements accidentally did not start in column 1.
2239        Sadly, there are *still* compilers around that complain, even though
2240        standard C has not required this for well over a decade. Sigh.
2242     9. In pcretest, the code for checking callouts passed small integers in the
2243        callout_data field, which is a void * field. However, some picky compilers
2244        complained about the casts involved for this on 64-bit systems. Now
2245        pcretest passes the address of the small integer instead, which should get
2246        rid of the warnings.
2248    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2249        both compile and run time, and gives an error if an invalid UTF-8 sequence
2250        is found. There is a option for disabling this check in cases where the
2251        string is known to be correct and/or the maximum performance is wanted.
2253    11. In response to a bug report, I changed one line in Makefile.in from
2255            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2256        to
2257            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2259        to look similar to other lines, but I have no way of telling whether this
2260        is the right thing to do, as I do not use Windows. No doubt I'll get told
2261        if it's wrong...
2264    Version 4.3 21-May-03
2265    ---------------------
2267    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2268       Makefile.
2270    2. Some refactoring to improve the quality of the code:
2272       (i)   The utf8_table... variables are now declared "const".
2274       (ii)  The code for \cx, which used the "case flipping" table to upper case
2275             lower case letters, now just substracts 32. This is ASCII-specific,
2276             but the whole concept of \cx is ASCII-specific, so it seems
2277             reasonable.
2279       (iii) PCRE was using its character types table to recognize decimal and
2280             hexadecimal digits in the pattern. This is silly, because it handles
2281             only 0-9, a-f, and A-F, but the character types table is locale-
2282             specific, which means strange things might happen. A private
2283             table is now used for this - though it costs 256 bytes, a table is
2284             much faster than multiple explicit tests. Of course, the standard
2285             character types table is still used for matching digits in subject
2286             strings against \d.
2288       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2289             ending in _t are). So I've renamed it as ESC_tee.
2291    3. The first argument for regexec() in the POSIX wrapper should have been
2292       defined as "const".
2294    4. Changed pcretest to use malloc() for its buffers so that they can be
2295       Electric Fenced for debugging.
2297    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2298       to read one or more bytes before the start of the subject string. Often this
2299       had no effect on PCRE's behaviour, but in some circumstances it could
2300       provoke a segmentation fault.
2302    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2303       to try to read one or more bytes before the start of the subject string.
2305    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2306       UTF-8 support could misbehave in various ways if the subject string
2307       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2308       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2309       back over UTF-8 characters.)
2312    Version 4.2 14-Apr-03
2313    ---------------------
2315    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2317    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2318         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2319         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2320         [WIN_PREFIX]: new variable, "cyg" for Cygwin
2321         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2322           and BUILD_EXEEXT
2323         Note: automatic setting of the BUILD variables is not yet working
2324         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2325           compile-time but not at link-time
2326         [LINK]: use for linking executables only
2327         make different versions for Windows and non-Windows
2328         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2329           libraries
2330         [LINK_FOR_BUILD]: new variable
2331         [OBJEXT]: use throughout
2332         [EXEEXT]: use throughout
2333         <winshared>: new target
2334         <wininstall>: new target
2335         <dftables.o>: use native compiler
2336         <dftables>: use native linker
2337         <install>: handle Windows platform correctly
2338         <clean>: ditto
2339         <check>: ditto
2340         copy DLL to top builddir before testing
2342       As part of these changes, -no-undefined was removed again. This was reported
2343       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2344       in any case.
2346    3. Some tidies to get rid of compiler warnings:
2348       . In the match_data structure, match_limit was an unsigned long int, whereas
2349         match_call_count was an int. I've made them both unsigned long ints.
2351       . In pcretest the fact that a const uschar * doesn't automatically cast to
2352         a void * provoked a warning.
2354       . Turning on some more compiler warnings threw up some "shadow" variables
2355         and a few more missing casts.
2357    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2358       option, a class that contained a single character with a value between 128
2359       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2361    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2362       option, a class that contained several characters, but with at least one
2363       whose value was between 128 and 255 caused PCRE to crash.
2366    Version 4.1 12-Mar-03
2367    ---------------------
2369    1. Compiling with gcc -pedantic found a couple of places where casts were
2370    needed, and a string in dftables.c that was longer than standard compilers are
2371    required to support.
2373    2. Compiling with Sun's compiler found a few more places where the code could
2374    be tidied up in order to avoid warnings.
2376    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2377    first of these names is deprecated in the latest Autoconf in favour of the name
2378    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2379    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2380    analogy I have changed it to CFLAGS_FOR_BUILD.
2382    4. Added -no-undefined to the linking command in the Makefile, because this is
2383    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2384    linking step for the pcreposix library.
2386    5. PCRE was failing to diagnose the case of two named groups with the same
2387    name.
2389    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2390    literal character that is needed in the subject for a match, and scans along to
2391    ensure that it is present before embarking on the full matching process. This
2392    saves time in cases of nested unlimited repeats that are never going to match.
2393    Problem: the scan can take a lot of time if the subject is very long (e.g.
2394    megabytes), thus penalizing straightforward matches. It is now done only if the
2395    amount of subject to be scanned is less than 1000 bytes.
2397    7. A lesser problem with the same optimization is that it was recording the
2398    first character of an anchored pattern as "needed", thus provoking a search
2399    right along the subject, even when the first match of the pattern was going to
2400    fail. The "needed" character is now not set for anchored patterns, unless it
2401    follows something in the pattern that is of non-fixed length. Thus, it still
2402    fulfils its original purpose of finding quick non-matches in cases of nested
2403    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2406    Version 4.0 17-Feb-03
2407    ---------------------
2409    1. If a comment in an extended regex that started immediately after a meta-item
2410    extended to the end of string, PCRE compiled incorrect data. This could lead to
2411    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2413    2. Moved to autoconf 2.53 and libtool 1.4.2.
2415    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2416    the special perltest8 script is no longer needed - all the tests can be run
2417    from a single perltest script.
2419    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2420    by \s. It has now been removed in PCRE. This means it isn't recognized as
2421    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2422    class [:space:] *does* include VT, thereby creating a mess.
2424    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2425    space and tab.
2427    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2428    its new features into the main test script, reducing the number of scripts.
2430    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2431    were backward compatible, and made the (?i) apply to the whole pattern, as if
2432    /i were given. Now it behaves more logically, and applies the option setting
2433    only to what follows. PCRE has been changed to follow suit. However, if it
2434    finds options settings right at the start of the pattern, it extracts them into
2435    the global options, as before. Thus, they show up in the info data.
2437    8. Added support for the \Q...\E escape sequence. Characters in between are
2438    treated as literals. This is slightly different from Perl in that $ and @ are
2439    also handled as literals inside the quotes. In Perl, they will cause variable
2440    interpolation. Note the following examples:
2442        Pattern            PCRE matches      Perl matches
2444        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
2445        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2446        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2448    For compatibility with Perl, \Q...\E sequences are recognized inside character
2449    classes as well as outside them.
2451    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2452    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2453    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2454    signed/unsigned warnings.
2456    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2457    option for pcretest, so I've replaced it by a simple function that does just
2458    that job.
2460    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2461    "pcregrep -".
2463    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2464    Java package. This provides some syntactic sugar for simple cases of what my
2465    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2466    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2467    item, you can use this simplified notation. Note that only makes sense with
2468    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2469    greediness, whatever the setting of the PCRE_UNGREEDY option.
2471    13. A change of greediness default within a pattern was not taking effect at
2472    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2473    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2474    was abstracted outside.
2476    14. PCRE now supports the \G assertion. It is true when the current matching
2477    position is at the start point of the match. This differs from \A when the
2478    starting offset is non-zero. Used with the /g option of pcretest (or similar
2479    code), it works in the same way as it does for Perl's /g option. If all
2480    alternatives of a regex begin with \G, the expression is anchored to the start
2481    match position, and the "anchored" flag is set in the compiled expression.
2483    15. Some bugs concerning the handling of certain option changes within patterns
2484    have been fixed. These applied to options other than (?ims). For example,
2485    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2486    been the other way round. Some of this was related to change 7 above.
2488    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2489    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2490    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2491    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2493    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2494    mode. Unlike ".", it always matches newline, whatever the setting of
2495    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2496    assertions. Perl allows it, but it doesn't (in general) work because it can't
2497    calculate the length of the lookbehind. At least, that's the case for Perl
2498    5.8.0 - I've been told they are going to document that it doesn't work in
2499    future.
2501    18. Added an error diagnosis for escapes that PCRE does not support: these are
2502    \L, \l, \N, \P, \p, \U, \u, and \X.
2504    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2505    reading past the end of the pattern in cases such as /[abcd/.
2507    20. PCRE was getting more memory than necessary for patterns with classes that
2508    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2510    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2511    compiling PCRE for use with Virtual Pascal.
2513    22. Small fix to the Makefile to make it work properly if the build is done
2514    outside the source tree.
2516    23. Added a new extension: a condition to go with recursion. If a conditional
2517    subpattern starts with (?(R) the "true" branch is used if recursion has
2518    happened, whereas the "false" branch is used only at the top level.
2520    24. When there was a very long string of literal characters (over 255 bytes
2521    without UTF support, over 250 bytes with UTF support), the computation of how
2522    much memory was required could be incorrect, leading to segfaults or other
2523    strange effects.
2525    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2526    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2527    there was a subsequent back reference to those brackets. This meant that, for
2528    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2529    possible to check for precisely this case. All we can do is abandon the
2530    optimization if .* occurs inside capturing brackets when there are any back
2531    references whatsoever. (See below for a better fix that came later.)
2533    26. The handling of the optimization for finding the first character of a
2534    non-anchored pattern, and for finding a character that is required later in the
2535    match were failing in some cases. This didn't break the matching; it just
2536    failed to optimize when it could. The way this is done has been re-implemented.
2538    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2540    28. Added a new feature that provides some of the functionality that Perl
2541    provides with (?{...}). The facility is termed a "callout". The way it is done
2542    in PCRE is for the caller to provide an optional function, by setting
2543    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2544    global variable. By default it is unset, which disables all calling out. To get
2545    the function called, the regex must include (?C) at appropriate points. This
2546    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2547    This provides a means of identifying different callout points. When PCRE
2548    reaches such a point in the regex, if pcre_callout has been set, the external
2549    function is called. It is provided with data in a structure called
2550    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2551    matching continues; if it returns a non-zero value, the match at the current
2552    point fails. However, backtracking will occur if possible. [This was changed
2553    later and other features added - see item 49 below.]
2555    29. pcretest is upgraded to test the callout functionality. It provides a
2556    callout function that displays information. By default, it shows the start of
2557    the match and the current position in the text. There are some new data escapes
2558    to vary what happens:
2560        \C+         in addition, show current contents of captured substrings
2561        \C-         do not supply a callout function
2562        \C!n        return 1 when callout number n is reached
2563        \C!n!m      return 1 when callout number n is reached for the mth time
2565    30. If pcregrep was called with the -l option and just a single file name, it
2566    output "<stdin>" if a match was found, instead of the file name.
2568    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2569    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2570    pcre_exec(). This saves a malloc/free per call. The default value of
2571    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2572    when configuring.
2574    32. The default maximum size of a compiled pattern is 64K. There have been a
2575    few cases of people hitting this limit. The code now uses macros to handle the
2576    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2577    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2578    configuring. Tests 2 and 5 work only with 2-byte links because they output
2579    debugging information about compiled patterns.
2581    33. Internal code re-arrangements:
2583    (a) Moved the debugging function for printing out a compiled regex into
2584        its own source file (printint.c) and used #include to pull it into
2585        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2586        separate copies.
2588    (b) Defined the list of op-code names for debugging as a macro in
2589        internal.h so that it is next to the definition of the opcodes.
2591    (c) Defined a table of op-code lengths for simpler skipping along compiled
2592        code. This is again a macro in internal.h so that it is next to the
2593        definition of the opcodes.
2595    34. Added support for recursive calls to individual subpatterns, along the
2596    lines of Robin Houston's patch (but implemented somewhat differently).
2598    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2599    allow it to read and process whole directories in Win32. This code was
2600    contributed by Lionel Fourquaux; it has not been tested by me.
2602    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2603    used to name a group. Names consist of alphanumerics and underscores, and must
2604    be unique. Back references use the syntax (?P=name) and recursive calls use
2605    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2606    numbers. The function pcre_fullinfo() can be used after compilation to extract
2607    a name/number map. There are three relevant calls:
2609      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2610      PCRE_INFO_NAMECOUNT            yields the number of entries
2611      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2613    The map is a vector of fixed-size entries. The size of each entry depends on
2614    the length of the longest name used. The first two bytes of each entry are the
2615    group number, most significant byte first. There follows the corresponding
2616    name, zero terminated. The names are in alphabetical order.
2618    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2619    case instead of 255. Making it the same both with and without UTF-8 support
2620    means that the same test output works with both.
2622    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2623    calling malloc() with a zero argument.
2625    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2626    optimization. I've improved things by keeping a bitmap of backreferences with
2627    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2628    fact referenced, the optimization can be applied. It is unlikely that a
2629    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2630    the match to follow \n) will appear inside brackets with a number greater than
2631    31, but if it does, any back reference > 31 suppresses the optimization.
2633    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2634    of disabling numbered capturing parentheses. Any opening parenthesis that is
2635    not followed by ? behaves as if it were followed by ?: but named parentheses
2636    can still be used for capturing (and they will acquire numbers in the usual
2637    way).
2639    41. Redesigned the return codes from the match() function into yes/no/error so
2640    that errors can be passed back from deep inside the nested calls. A malloc
2641    failure while inside a recursive subpattern call now causes the
2642    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2644    42. It is now possible to set a limit on the number of times the match()
2645    function is called in a call to pcre_exec(). This facility makes it possible to
2646    limit the amount of recursion and backtracking, though not in a directly
2647    obvious way, because the match() function is used in a number of different
2648    circumstances. The count starts from zero for each position in the subject
2649    string (for non-anchored patterns). The default limit is, for compatibility, a
2650    large number, namely 10 000 000. You can change this in two ways:
2652    (a) When configuring PCRE before making, you can use --with-match-limit=n
2653        to set a default value for the compiled library.
2655    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2656        a different value is set. See 45 below.
2658    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2660    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2661    of things that can be changed at compile time. The first argument specifies
2662    what is wanted and the second points to where the information is to be placed.
2663    The current list of available information is:
2665      PCRE_CONFIG_UTF8
2667    The output is an integer that is set to one if UTF-8 support is available;
2668    otherwise it is set to zero.
2672    The output is an integer that it set to the value of the code that is used for
2673    newline. It is either LF (10) or CR (13).
2677    The output is an integer that contains the number of bytes used for internal
2678    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2682    The output is an integer that contains the threshold above which the POSIX
2683    interface uses malloc() for output vectors. See item 31 above.
2687    The output is an unsigned integer that contains the default limit of the number
2688    of match() calls in a pcre_exec() execution. See 42 above.
2690    44. pcretest has been upgraded by the addition of the -C option. This causes it
2691    to extract all the available output from the new pcre_config() function, and to
2692    output it. The program then exits immediately.
2694    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2695    order to support additional features. One way would have been to define
2696    pcre_exec2() (for example) with extra arguments, but this would not have been
2697    extensible, and would also have required all calls to the original function to
2698    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2699    is used for passing in "extra" data from pcre_study().
2701    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2702    contains the following fields:
2704      flags         a bitmap indicating which of the following fields are set
2705      study_data    opaque data from pcre_study()
2706      match_limit   a way of specifying a limit on match() calls for a specific
2707                      call to pcre_exec()
2708      callout_data  data for callouts (see 49 below)
2710    The flag bits are also defined in pcre.h, and are
2716    The pcre_study() function now returns one of these new pcre_extra blocks, with
2717    the actual study data pointed to by the study_data field, and the
2718    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2719    before. That is, this change is entirely upwards-compatible and requires no
2720    change to existing code.
2722    If you want to pass in additional data to pcre_exec(), you can either place it
2723    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2724    block.
2726    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2727    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2728    times with different match limits, until it finds the minimum value needed for
2729    pcre_exec() to complete. The value is then output. This can be instructive; for
2730    most simple matches the number is quite small, but for pathological cases it
2731    gets very large very quickly.
2733    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2734    returns the size of the data block pointed to by the study_data field in a
2735    pcre_extra block, that is, the value that was passed as the argument to
2736    pcre_malloc() when PCRE was getting memory in which to place the information
2737    created by pcre_study(). The fourth argument should point to a size_t variable.
2738    pcretest has been extended so that this information is shown after a successful
2739    pcre_study() call when information about the compiled regex is being displayed.
2741    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2742    because what follows is always an absolute path. (Later: it turns out that this
2743    is more than cosmetic for MinGW, because it doesn't like empty path
2744    components.)
2746    49. Some changes have been made to the callout feature (see 28 above):
2748    (i)  A callout function now has three choices for what it returns:
2750           0  =>  success, carry on matching
2751         > 0  =>  failure at this point, but backtrack if possible
2752         < 0  =>  serious error, return this value from pcre_exec()
2754         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2755         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2756         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2757         use by callout functions. It will never be used by PCRE itself.
2759    (ii) The pcre_extra structure (see 45 above) has a void * field called
2760         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2761         pcre_callout_block structure has a field of the same name. The contents of
2762         the field passed in the pcre_extra structure are passed to the callout
2763         function in the corresponding field in the callout block. This makes it
2764         easier to use the same callout-containing regex from multiple threads. For
2765         testing, the pcretest program has a new data escape
2767           \C*n        pass the number n (may be negative) as callout_data
2769         If the callout function in pcretest receives a non-zero value as
2770         callout_data, it returns that value.
2772    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2773    there were some redundant $(CFLAGS) in commands that are now specified as
2774    $(LINK), which already includes $(CFLAGS).
2776    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2777    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2778    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2779    one-byte characters throughout. Note that case-insensitive matching applies
2780    only to characters whose values are less than 256. PCRE doesn't support the
2781    notion of cases for higher-valued characters.
2783    (i)   A character class whose characters are all within 0-255 is handled as
2784          a bit map, and the map is inverted for negative classes. Previously, a
2785          character > 255 always failed to match such a class; however it should
2786          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2788    (ii)  A negated character class with a single character < 255 is coded as
2789          "not this character" (OP_NOT). This wasn't working properly when the test
2790          character was multibyte, either singly or repeated.
2792    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2793          mode, for example: \x{100}{2,3}.
2795    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2796          singly or repeated) now correctly test multibyte characters. However,
2797          PCRE doesn't recognize any characters with values greater than 255 as
2798          digits, spaces, or word characters. Such characters always match \D, \S,
2799          and \W, and never match \d, \s, or \w.
2801    (v)   Classes may now contain characters and character ranges with values
2802          greater than 255. For example: [ab\x{100}-\x{400}].
2804    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2805          PCRE in UTF-8 mode.
2807    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2808    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2809    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2810    value.)
2812    53. The single man page has become too large. I have therefore split it up into
2813    a number of separate man pages. These also give rise to individual HTML pages;
2814    these are now put in a separate directory, and there is an index.html page that
2815    lists them all. Some hyperlinking between the pages has been installed.
2817    54. Added convenience functions for handling named capturing parentheses.
2819    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2820    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2821    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2822    are faulted.
2824    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2825    calling configure. These values are used when compiling the dftables.c program
2826    which is run to generate the source of the default character tables. They
2827    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2828    you will need to set these values.
2830    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2833    Version 3.9 02-Jan-02
2834    ---------------------
2836    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2838    2. If --disable-static was given, the building process failed when trying to
2839    build pcretest and pcregrep. (For some reason it was using libtool to compile
2840    them, which is not right, as they aren't part of the library.)
2843    Version 3.8 18-Dec-01
2844    ---------------------
2846    1. The experimental UTF-8 code was completely screwed up. It was packing the
2847    bytes in the wrong order. How dumb can you get?
2850    Version 3.7 29-Oct-01
2851    ---------------------
2853    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2854    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2855    this didn't happen under Solaris 8, where I normally test things.
2857    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2858    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2859    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2860    this fix an hour or so after the initial 3.7 release.)
2863    Version 3.6 23-Oct-01
2864    ---------------------
2866    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2867    offsets passed as NULL with zero offset count.
2869    2. The config.guess and config.sub files had not been updated when I moved to
2870    the latest autoconf.
2873    Version 3.5 15-Aug-01
2874    ---------------------
2876    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2877    had been forgotten.
2879    2. By using declared but undefined structures, we can avoid using "void"
2880    definitions in pcre.h while keeping the internal definitions of the structures
2881    private.
2883    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2884    user point of view, this means that both static and shared libraries are built
2885    by default, but this can be individually controlled. More of the work of
2886    handling this static/shared cases is now inside libtool instead of PCRE's make
2887    file.
2889    4. The pcretest utility is now installed along with pcregrep because it is
2890    useful for users (to test regexs) and by doing this, it automatically gets
2891    relinked by libtool. The documentation has been turned into a man page, so
2892    there are now .1, .txt, and .html versions in /doc.
2894    5. Upgrades to pcregrep:
2895       (i)   Added long-form option names like gnu grep.
2896       (ii)  Added --help to list all options with an explanatory phrase.
2897       (iii) Added -r, --recursive to recurse into sub-directories.
2898       (iv)  Added -f, --file to read patterns from a file.
2900    6. pcre_exec() was referring to its "code" argument before testing that
2901    argument for NULL (and giving an error if it was NULL).
2903    7. Upgraded Makefile.in to allow for compiling in a different directory from
2904    the source directory.
2906    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2907    options bits, the pointer it was passed was to an int instead of to an unsigned
2908    long int. This mattered only on 64-bit systems.
2910    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2911    generated) instead of pcre.in, which it its source. Also made the same change
2912    in several of the .c files.
2914    10. A new release of gcc defines printf() as a macro, which broke pcretest
2915    because it had an ifdef in the middle of a string argument for printf(). Fixed
2916    by using separate calls to printf().
2918    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2919    script, to force use of CR or LF instead of \n in the source. On non-Unix
2920    systems, the value can be set in config.h.
2922    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2923    absolute limit. Changed the text of the error message to make this clear, and
2924    likewise updated the man page.
2926    13. The limit of 99 on the number of capturing subpatterns has been removed.
2927    The new limit is 65535, which I hope will not be a "real" limit.
2930    Version 3.4 22-Aug-00
2931    ---------------------
2933    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2935    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2938    Version 3.3 01-Aug-00
2939    ---------------------
2941    1. If an octal character was given, but the value was greater than \377, it
2942    was not getting masked to the least significant bits, as documented. This could
2943    lead to crashes in some systems.
2945    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2946    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2948    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2949    These just pass their arguments on to (pcre_free)(), but they are provided
2950    because some uses of PCRE bind it to non-C systems that can call its functions,
2951    but cannot call free() or pcre_free() directly.
2953    4. Add "make test" as a synonym for "make check". Corrected some comments in
2954    the Makefile.
2956    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2957    Makefile.
2959    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2960    command called pgrep for grepping around the active processes.
2962    7. Added the beginnings of support for UTF-8 character strings.
2964    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2965    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2966    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2967    out for the ar command.)
2970    Version 3.2 12-May-00
2971    ---------------------
2973    This is purely a bug fixing release.
2975    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2976    of ZA. This was just one example of several cases that could provoke this bug,
2977    which was introduced by change 9 of version 2.00. The code for breaking
2978    infinite loops after an iteration that matches an empty string was't working
2979    correctly.
2981    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2982    when matched against abbab (for example). After matching an empty string, it
2983    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2984    caused it to match further down the string than it should.
2986    3. The code contained an inclusion of sys/types.h. It isn't clear why this
2987    was there because it doesn't seem to be needed, and it causes trouble on some
2988    systems, as it is not a Standard C header. It has been removed.
2990    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2991    were reported on the Macintosh. The changes were from
2993      while ((c = *(++ptr)) != 0 && c != '\n');
2994    to
2995      while ((c = *(++ptr)) != 0 && c != '\n') ;
2997    Totally extraordinary, but if that's what it takes...
2999    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3000    available. Added HAVE_BCOPY and an autoconf test for it; if neither
3001    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3002    assumes the way PCRE uses memmove() (always moving upwards).
3004    6. PCRE is being used in one environment where strchr() is not available. There
3005    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3006    faster code anyway.
3009    Version 3.1 09-Feb-00
3010    ---------------------
3012    The only change in this release is the fixing of some bugs in Makefile.in for
3013    the "install" target:
3015    (1) It was failing to install pcreposix.h.
3017    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3020    Version 3.0 01-Feb-00
3021    ---------------------
3023    1. Add support for the /+ modifier to perltest (to output $` like it does in
3024    pcretest).
3026    2. Add support for the /g modifier to perltest.
3028    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3029    matches null strings.
3031    4. Fix perltest so that it doesn't do unwanted things when fed an empty
3032    pattern. Perl treats empty patterns specially - it reuses the most recent
3033    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3034    effect.
3036    5. The POSIX interface was broken in that it was just handing over the POSIX
3037    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3038    required a bigger vector, with some working space on the end. This means that
3039    the POSIX wrapper now has to get and free some memory, and copy the results.
3041    6. Added some simple autoconf support, placing the test data and the
3042    documentation in separate directories, re-organizing some of the
3043    information files, and making it build pcre-config (a GNU standard). Also added
3044    libtool support for building PCRE as a shared library, which is now the
3045    default.
3047    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3048    09 are not valid octal constants. Single digits will be used for minor values
3049    less than 10.
3051    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3052    existing programs that set these in the POSIX interface can use PCRE without
3053    modification.
3055    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3056    return all that pcre_info() returns, plus additional data. The pcre_info()
3057    function is retained for compatibility, but is considered to be obsolete.
3059    10. Added experimental recursion feature (?R) to handle one common case that
3060    Perl 5.6 will be able to do with (?p{...}).
3062    11. Added support for POSIX character classes like [:alpha:], which Perl is
3063    adopting.
3066    Version 2.08 31-Aug-99
3067    ----------------------
3069    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3070    trying to match at the startoffset position, but instead was moving forward to
3071    the next newline as if a previous match had failed.
3073    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3074    and could get into a loop if a null string was matched other than at the start
3075    of the subject.
3077    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3078    be distinguished at compile time, and for completeness also added PCRE_DATE.
3080    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3081    in GnuWin32 environments.
3084    Version 2.07 29-Jul-99
3085    ----------------------
3087    1. The documentation is now supplied in plain text form and HTML as well as in
3088    the form of man page sources.
3090    2. C++ compilers don't like assigning (void *) values to other pointer types.
3091    In particular this affects malloc(). Although there is no problem in Standard
3092    C, I've put in casts to keep C++ compilers happy.
3094    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3095    should be (const char *).
3097    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3098    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3099    However, I haven't made this a standard facility. The documentation doesn't
3100    mention it, and the Makefile doesn't support it.
3102    5. The Makefile now contains an "install" target, with editable destinations at
3103    the top of the file. The pcretest program is not installed.
3105    6. pgrep -V now gives the PCRE version number and date.
3107    7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3108    causing the entire string to be ignored, instead of just the last character.
3110    8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3111    non-matching string, it can take a very, very long time, even for strings of
3112    quite modest length, because of the nested recursion. PCRE now does better in
3113    some of these cases. It does this by remembering the last required literal
3114    character in the pattern, and pre-searching the subject to ensure it is present
3115    before running the real match. In other words, it applies a heuristic to detect
3116    some types of certain failure quickly, and in the above example, if presented
3117    with a string that has no trailing " it gives "no match" very quickly.
3119    9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3120    other alternatives are tried instead.
3123    Version 2.06 09-Jun-99
3124    ----------------------
3126    1. Change pcretest's output for amount of store used to show just the code
3127    space, because the remainder (the data block) varies in size between 32-bit and
3128    64-bit systems.
3130    2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3131    start matching at. This allows lookbehinds to work when searching for multiple
3132    occurrences in a string.
3134    3. Added additional options to pcretest for testing multiple occurrences:
3136       /+   outputs the rest of the string that follows a match
3137       /g   loops for multiple occurrences, using the new startoffset argument
3138       /G   loops for multiple occurrences by passing an incremented pointer
3140    4. PCRE wasn't doing the "first character" optimization for patterns starting
3141    with \b or \B, though it was doing it for other lookbehind assertions. That is,
3142    it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3143    the letter 'x'. On long subject strings, this gives a significant speed-up.
3146    Version 2.05 21-Apr-99
3147    ----------------------
3149    1. Changed the type of magic_number from int to long int so that it works
3150    properly on 16-bit systems.
3152    2. Fixed a bug which caused patterns starting with .* not to work correctly
3153    when the subject string contained newline characters. PCRE was assuming
3154    anchoring for such patterns in all cases, which is not correct because .* will
3155    not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3156    DOTALL is set at top level; otherwise it knows that patterns starting with .*
3157    must be retried after every newline in the subject.
3160    Version 2.04 18-Feb-99
3161    ----------------------
3163    1. For parenthesized subpatterns with repeats whose minimum was zero, the
3164    computation of the store needed to hold the pattern was incorrect (too large).
3165    If such patterns were nested a few deep, this could multiply and become a real
3166    problem.
3168    2. Added /M option to pcretest to show the memory requirement of a specific
3169    pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3171    3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3172    compiled in such a way that the backtracking after subsequent failure was
3173    pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3174    ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3177    Version 2.03 02-Feb-99
3178    ----------------------
3180    1. Fixed typo and small mistake in man page.
3182    2. Added 4th condition (GPL supersedes if conflict) and created separate
3183    LICENCE file containing the conditions.
3185    3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3186    Perl, that is the internal \ allows the delimiter to be included in the
3187    pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3188    the final delimiter, add \ to the end of the pattern (to test the error).
3190    4. Added the convenience functions for extracting substrings after a successful
3191    match. Updated pcretest to make it able to test these functions.
3194  Version 2.02 14-Jan-99  Version 2.02 14-Jan-99
3195  ----------------------  ----------------------

Removed from v.27  
changed lines
  Added in v.428

  ViewVC Help
Powered by ViewVC 1.1.5