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

Diff of /code/trunk/ChangeLog

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

revision 21 by nigel, Sat Feb 24 21:38:37 2007 UTC revision 427 by ph10, Fri Aug 28 09:55:54 2009 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 8.00 ??-???-??
5    ----------------------
6    
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8        was out-of-date, and there was no check on the pcre_compile() error code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.
11    
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.
15    
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.
23    
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.
28    
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.
33    
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.
36    
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.
43    
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.
47    
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.
53    
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 longer partial match to supersede a
57        shorter full match, and may be more useful for multi-segment matching,
58        especially with pcre_exec().
59    
60    
61    Version 7.9 11-Apr-09
62    ---------------------
63    
64    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
65        (pcretest), all targets were linked against these libraries. This included
66        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
67        libraries. This caused unwanted dependencies to be created. This problem
68        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
69        pcretest is linked with readline.
70    
71    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
72        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
73        moved outside it again, because FALSE and TRUE are already defined in AIX,
74        but BOOL is not.
75    
76    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
77        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
78    
79    4.  The pcregrep documentation said spaces were inserted as well as colons (or
80        hyphens) following file names and line numbers when outputting matching
81        lines. This is not true; no spaces are inserted. I have also clarified the
82        wording for the --colour (or --color) option.
83    
84    5.  In pcregrep, when --colour was used with -o, the list of matching strings
85        was not coloured; this is different to GNU grep, so I have changed it to be
86        the same.
87    
88    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
89        each matching line was coloured. Now it goes on to look for further matches
90        of any of the test patterns, which is the same behaviour as GNU grep.
91    
92    7.  A pattern that could match an empty string could cause pcregrep to loop; it
93        doesn't make sense to accept an empty string match in pcregrep, so I have
94        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
95        seems to be how GNU grep behaves.
96    
97    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
98        start or after a newline", because the conditional assertion was not being
99        correctly handled. The rule now is that both the assertion and what follows
100        in the first alternative must satisfy the test.
101    
102    9.  If auto-callout was enabled in a pattern with a conditional group whose
103        condition was an assertion, PCRE could crash during matching, both with
104        pcre_exec() and pcre_dfa_exec().
105    
106    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
107        used for matching.
108    
109    11. Unicode property support in character classes was not working for
110        characters (bytes) greater than 127 when not in UTF-8 mode.
111    
112    12. Added the -M command line option to pcretest.
113    
114    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
115    
116    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
117    
118    16. Added comments and documentation about mis-use of no_arg in the C++
119        wrapper.
120    
121    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
122        from Martin Jerabek that uses macro names for all relevant character and
123        string constants.
124    
125    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
126        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
127        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
128        these, but not everybody uses configure.
129    
130    19. A conditional group that had only one branch was not being correctly
131        recognized as an item that could match an empty string. This meant that an
132        enclosing group might also not be so recognized, causing infinite looping
133        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
134        with the subject "ab", where knowledge that the repeated group can match
135        nothing is needed in order to break the loop.
136    
137    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
138        exec(), but without supplying a callout function, matching went wrong.
139    
140    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
141        leak if the size of the offset vector was greater than 30. When the vector
142        is smaller, the saved offsets during recursion go onto a local stack
143        vector, but for larger vectors malloc() is used. It was failing to free
144        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
145        error, in fact).
146    
147    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
148        heapframe that is used only when UTF-8 support is enabled. This caused no
149        problem, but was untidy.
150    
151    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
152        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
153        included within another project.
154    
155    24. Steven Van Ingelgem's patches to add more options to the CMake support,
156        slightly modified by me:
157    
158          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
159              not building pcregrep.
160    
161          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
162              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
163    
164    25. Forward references, both numeric and by name, in patterns that made use of
165        duplicate group numbers, could behave incorrectly or give incorrect errors,
166        because when scanning forward to find the reference group, PCRE was not
167        taking into account the duplicate group numbers. A pattern such as
168        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
169    
170    26. Changed a few more instances of "const unsigned char *" to USPTR, making
171        the feature of a custom pointer more persuasive (as requested by a user).
172    
173    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
174        pcretest.c, inside #ifndefs, because it seems they are sometimes already
175        pre-defined.
176    
177    28. Added support for (*UTF8) at the start of a pattern.
178    
179    29. Arrange for flags added by the "release type" setting in CMake to be shown
180        in the configuration summary.
181    
182    
183    Version 7.8 05-Sep-08
184    ---------------------
185    
186    1.  Replaced UCP searching code with optimized version as implemented for Ad
187        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
188        stage table and inline lookup instead of a function, giving speed ups of 2
189        to 5 times on some simple patterns that I tested. Permission was given to
190        distribute the MultiStage2.py script that generates the tables (it's not in
191        the tarball, but is in the Subversion repository).
192    
193    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
194        scripts.
195    
196    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
197        a group with a zero qualifier. The result of the study could be incorrect,
198        or the function might crash, depending on the pattern.
199    
200    4.  Caseless matching was not working for non-ASCII characters in back
201        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
202        It now works when Unicode Property Support is available.
203    
204    5.  In pcretest, an escape such as \x{de} in the data was always generating
205        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
206        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
207        truncation.
208    
209    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
210    
211    7.  Added two (int) casts to pcregrep when printing the difference of two
212        pointers, in case they are 64-bit values.
213    
214    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
215        test 2 if it fails.
216    
217    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
218        and a #define of that name to empty if it is not externally set. This is to
219        allow users of MSVC to set it if necessary.
220    
221    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
222        the convenience functions in the pcre_get.c source file.
223    
224    11. An option change at the start of a pattern that had top-level alternatives
225        could cause overwriting and/or a crash. This command provoked a crash in
226        some environments:
227    
228          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
229    
230        This potential security problem was recorded as CVE-2008-2371.
231    
232    12. For a pattern where the match had to start at the beginning or immediately
233        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
234        pcre_dfa_exec() could read past the end of the passed subject if there was
235        no match. To help with detecting such bugs (e.g. with valgrind), I modified
236        pcretest so that it places the subject at the end of its malloc-ed buffer.
237    
238    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
239        exec() might read past the end of the data buffer in UTF-8 mode.
240    
241    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
242        the data contained the byte 0x85 as part of a UTF-8 character within its
243        first line. This applied both to normal and DFA matching.
244    
245    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
246        /^[^d]*?$/8 failed to match "abc".
247    
248    16. Added a missing copyright notice to pcrecpp_internal.h.
249    
250    17. Make it more clear in the documentation that values returned from
251        pcre_exec() in ovector are byte offsets, not character counts.
252    
253    18. Tidied a few places to stop certain compilers from issuing warnings.
254    
255    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
256        supplied by Stefan Weber. I made a further small update for 7.8 because
257        there is a change of source arrangements: the pcre_searchfuncs.c module is
258        replaced by pcre_ucd.c.
259    
260    
261    Version 7.7 07-May-08
262    ---------------------
263    
264    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
265        a string to a long long, pretend we don't even have a long long." This is
266        done by checking for the strtoq, strtoll, and _strtoi64 functions.
267    
268    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
269        pre-7.6 versions, which defined a global no_arg variable instead of putting
270        it in the RE class. (See also #8 below.)
271    
272    3.  Remove a line of dead code, identified by coverity and reported by Nuno
273        Lopes.
274    
275    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
276    
277        (1) The include/exclude patterns were being applied to the whole pathnames
278            of files, instead of just to the final components.
279    
280        (2) If there was more than one level of directory, the subdirectories were
281            skipped unless they satisfied the include/exclude conditions. This is
282            inconsistent with GNU grep (and could even be seen as contrary to the
283            pcregrep specification - which I improved to make it absolutely clear).
284            The action now is always to scan all levels of directory, and just
285            apply the include/exclude patterns to regular files.
286    
287    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
288        --exclude_dir in the tests to avoid scanning .svn directories.
289    
290    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
291        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
292        doesn't support NULs in patterns.
293    
294    7.  Added some missing "const"s to declarations of static tables in
295        pcre_compile.c and pcre_dfa_exec.c.
296    
297    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
298        caused by fix #2  above. (Subsequently also a second patch to fix the
299        first patch. And a third patch - this was a messy problem.)
300    
301    9.  Applied Craig's patch to remove the use of push_back().
302    
303    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
304        matching function regexec().
305    
306    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
307        which, however, unlike Perl's \g{...}, are subroutine calls, not back
308        references. PCRE supports relative numbers with this syntax (I don't think
309        Oniguruma does).
310    
311    12. Previously, a group with a zero repeat such as (...){0} was completely
312        omitted from the compiled regex. However, this means that if the group
313        was called as a subroutine from elsewhere in the pattern, things went wrong
314        (an internal error was given). Such groups are now left in the compiled
315        pattern, with a new opcode that causes them to be skipped at execution
316        time.
317    
318    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
319        to the way PCRE behaves:
320    
321        (a) A lone ] character is dis-allowed (Perl treats it as data).
322    
323        (b) A back reference to an unmatched subpattern matches an empty string
324            (Perl fails the current match path).
325    
326        (c) A data ] in a character class must be notated as \] because if the
327            first data character in a class is ], it defines an empty class. (In
328            Perl it is not possible to have an empty class.) The empty class []
329            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
330            The negative empty class [^] matches any one character, independently
331            of the DOTALL setting.
332    
333    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
334        non-existent subpattern following a character class starting with ']' and
335        containing () gave an internal compiling error instead of "reference to
336        non-existent subpattern". Fortunately, when the pattern did exist, the
337        compiled code was correct. (When scanning forwards to check for the
338        existencd of the subpattern, it was treating the data ']' as terminating
339        the class, so got the count wrong. When actually compiling, the reference
340        was subsequently set up correctly.)
341    
342    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
343        it was being rejected as not supported by pcre_dfa_exec(), even though
344        other assertions are supported. I have made pcre_dfa_exec() support
345        (*FAIL).
346    
347    16. The implementation of 13c above involved the invention of a new opcode,
348        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
349        cannot be changed at match time, I realized I could make a small
350        improvement to matching performance by compiling OP_ALLANY instead of
351        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
352        on the OP_ANY path.
353    
354    17. Compiling pcretest on Windows with readline support failed without the
355        following two fixes: (1) Make the unistd.h include conditional on
356        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
357    
358    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
359        ncurses library to be included for pcretest when ReadLine support is
360        requested, but also to allow for it to be overridden. This patch came from
361        Daniel Bergström.
362    
363    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
364        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
365        any errors with the current Unicode tables. Thanks to Peter Kankowski for
366        spotting this.
367    
368    
369    Version 7.6 28-Jan-08
370    ---------------------
371    
372    1.  A character class containing a very large number of characters with
373        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
374        overflow.
375    
376    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
377        HAVE_LONG_LONG is not defined.
378    
379    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
380        bring them up to date and include new features. This patch includes:
381    
382        - Fixed PH's badly added libz and libbz2 support.
383        - Fixed a problem with static linking.
384        - Added pcredemo. [But later removed - see 7 below.]
385        - Fixed dftables problem and added an option.
386        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
387            HAVE_LONG_LONG.
388        - Added readline support for pcretest.
389        - Added an listing of the option settings after cmake has run.
390    
391    4.  A user submitted a patch to Makefile that makes it easy to create
392        "pcre.dll" under mingw when using Configure/Make. I added stuff to
393        Makefile.am that cause it to include this special target, without
394        affecting anything else. Note that the same mingw target plus all
395        the other distribution libraries and programs are now supported
396        when configuring with CMake (see 6 below) instead of with
397        Configure/Make.
398    
399    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
400        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
401        exported in the Windows port". It has not yet been confirmed that the patch
402        solves the problem, but it does no harm.
403    
404    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
405        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
406        with CMake, and also correct the comment about stack recursion.
407    
408    7.  Remove the automatic building of pcredemo from the ./configure system and
409        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
410        of a program that users should build themselves after PCRE is installed, so
411        building it automatically is not really right. What is more, it gave
412        trouble in some build environments.
413    
414    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
415    
416    
417    Version 7.5 10-Jan-08
418    ---------------------
419    
420    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
421        values in parens when parsing an RE using the C++ wrapper."
422    
423    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
424        Characters greater than 255 were excluded from the class instead of being
425        included.
426    
427    3.  The same bug as (2) above applied to negated POSIX classes such as
428        [:^space:].
429    
430    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
431        defined or documented. It seems to have been a typo for PCRE_STATIC, so
432        I have changed it.
433    
434    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
435        first named subpattern) and a construct such as (?&a) would reference the
436        first named subpattern whose name started with "a" (in other words, the
437        length check was missing). Both these problems are fixed. "Subpattern name
438        expected" is now given for (?&) (a zero-length name), and this patch also
439        makes it give the same error for \k'' (previously it complained that that
440        was a reference to a non-existent subpattern).
441    
442    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
443        this is right because (?- can be followed by option settings as well as by
444        digits. I have, however, made the messages clearer.
445    
446    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
447        than the number used in the conditional) now cause a compile-time error.
448        This is actually not compatible with Perl, which accepts such patterns, but
449        treats the conditional as always being FALSE (as PCRE used to), but it
450        seems to me that giving a diagnostic is better.
451    
452    8.  Change "alphameric" to the more common word "alphanumeric" in comments
453        and messages.
454    
455    9.  Fix two occurrences of "backslash" in comments that should have been
456        "backspace".
457    
458    10. Remove two redundant lines of code that can never be obeyed (their function
459        was moved elsewhere).
460    
461    11. The program that makes PCRE's Unicode character property table had a bug
462        which caused it to generate incorrect table entries for sequences of
463        characters that have the same character type, but are in different scripts.
464        It amalgamated them into a single range, with the script of the first of
465        them. In other words, some characters were in the wrong script. There were
466        thirteen such cases, affecting characters in the following ranges:
467    
468          U+002b0 - U+002c1
469          U+0060c - U+0060d
470          U+0061e - U+00612
471          U+0064b - U+0065e
472          U+0074d - U+0076d
473          U+01800 - U+01805
474          U+01d00 - U+01d77
475          U+01d9b - U+01dbf
476          U+0200b - U+0200f
477          U+030fc - U+030fe
478          U+03260 - U+0327f
479          U+0fb46 - U+0fbb1
480          U+10450 - U+1049d
481    
482    12. The -o option (show only the matching part of a line) for pcregrep was not
483        compatible with GNU grep in that, if there was more than one match in a
484        line, it showed only the first of them. It now behaves in the same way as
485        GNU grep.
486    
487    13. If the -o and -v options were combined for pcregrep, it printed a blank
488        line for every non-matching line. GNU grep prints nothing, and pcregrep now
489        does the same. The return code can be used to tell if there were any
490        non-matching lines.
491    
492    14. Added --file-offsets and --line-offsets to pcregrep.
493    
494    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
495        infinitely looping recursion. The bug was that positive lookaheads were not
496        being skipped when checking for a possible empty match (negative lookaheads
497        and both kinds of lookbehind were skipped).
498    
499    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
500        inclusion of <windows.h> to before rather than after the definition of
501        INVALID_FILE_ATTRIBUTES (patch from David Byron).
502    
503    17. Specifying a possessive quantifier with a specific limit for a Unicode
504        character property caused pcre_compile() to compile bad code, which led at
505        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
506        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
507        caused the error; without that there was no problem.
508    
509    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
510    
511    19. Added --enable-pcretest-libreadline.
512    
513    20. In pcrecpp.cc, the variable 'count' was incremented twice in
514        RE::GlobalReplace(). As a result, the number of replacements returned was
515        double what it should be. I removed one of the increments, but Craig sent a
516        later patch that removed the other one (the right fix) and added unit tests
517        that check the return values (which was not done before).
518    
519    21. Several CMake things:
520    
521        (1) Arranged that, when cmake is used on Unix, the libraries end up with
522            the names libpcre and libpcreposix, not just pcre and pcreposix.
523    
524        (2) The above change means that pcretest and pcregrep are now correctly
525            linked with the newly-built libraries, not previously installed ones.
526    
527        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
528    
529    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
530        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
531        UTF-8 newline character). The key issue is that the pattern starts .*;
532        this means that the match must be either at the beginning, or after a
533        newline. The bug was in the code for advancing after a failed match and
534        checking that the new position followed a newline. It was not taking
535        account of UTF-8 characters correctly.
536    
537    23. PCRE was behaving differently from Perl in the way it recognized POSIX
538        character classes. PCRE was not treating the sequence [:...:] as a
539        character class unless the ... were all letters. Perl, however, seems to
540        allow any characters between [: and :], though of course it rejects as
541        unknown any "names" that contain non-letters, because all the known class
542        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
543        for example, whereas PCRE did not - it did not recognize a POSIX character
544        class. This seemed a bit dangerous, so the code has been changed to be
545        closer to Perl. The behaviour is not identical to Perl, because PCRE will
546        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
547        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
548        Perl does, and where it didn't before.
549    
550    24. Rewrite so as to remove the single use of %n from pcregrep because in some
551        Windows environments %n is disabled by default.
552    
553    
554    Version 7.4 21-Sep-07
555    ---------------------
556    
557    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
558        means that a class such as [\s] counted as "explicit reference to CR or
559        LF". That isn't really right - the whole point of the change was to try to
560        help when there was an actual mention of one of the two characters. So now
561        the change happens only if \r or \n (or a literal CR or LF) character is
562        encountered.
563    
564    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
565        of both had grown to the point where there were only 3 bits left.
566        Fortunately, there was spare space in the data structure, and so I have
567        moved the internal flags into a new 16-bit field to free up more option
568        bits.
569    
570    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
571        but did not set the internal JCHANGED flag - either of these is enough to
572        control the way the "get" function works - but the PCRE_INFO_JCHANGED
573        facility is supposed to tell if (?J) was ever used, so now (?J) at the
574        start sets both bits.
575    
576    4.  Added options (at build time, compile time, exec time) to change \R from
577        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
578    
579    5.  doc/pcresyntax.html was missing from the distribution.
580    
581    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
582        compatibility, even though it is no longer used.
583    
584    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
585        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
586        windows.h file is present (where different names are used). [This was
587        reversed later after testing - see 16 below.]
588    
589    8.  Changed all #include <config.h> to #include "config.h". There were also
590        some further <pcre.h> cases that I changed to "pcre.h".
591    
592    9.  When pcregrep was used with the --colour option, it missed the line ending
593        sequence off the lines that it output.
594    
595    10. It was pointed out to me that arrays of string pointers cause lots of
596        relocations when a shared library is dynamically loaded. A technique of
597        using a single long string with a table of offsets can drastically reduce
598        these. I have refactored PCRE in four places to do this. The result is
599        dramatic:
600    
601          Originally:                          290
602          After changing UCP table:            187
603          After changing error message table:   43
604          After changing table of "verbs"       36
605          After changing table of Posix names   22
606    
607        Thanks to the folks working on Gregex for glib for this insight.
608    
609    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
610        unicode-properties was also set.
611    
612    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
613    
614    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
615        checked only for CRLF.
616    
617    14. Added casts to pcretest.c to avoid compiler warnings.
618    
619    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
620    
621    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
622        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
623        entirely. This removes changes made in 7 above.
624    
625    17. The CMake files have been updated, and there is now more information about
626        building with CMake in the NON-UNIX-USE document.
627    
628    
629    Version 7.3 28-Aug-07
630    ---------------------
631    
632     1. In the rejigging of the build system that eventually resulted in 7.1, the
633        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
634        brackets there is not right, since it causes compilers to look for an
635        installed pcre.h, not the version that is in the source that is being
636        compiled (which of course may be different). I have changed it back to:
637    
638          #include "pcre.h"
639    
640        I have a vague recollection that the change was concerned with compiling in
641        different directories, but in the new build system, that is taken care of
642        by the VPATH setting the Makefile.
643    
644     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
645        when the subject happened to end in the byte 0x85 (e.g. if the last
646        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
647        characters but of course it shouldn't be taken as a newline when it is part
648        of another character. The bug was that, for an unlimited repeat of . in
649        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
650        characters when looking for a newline.
651    
652     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
653    
654     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
655        in debug output.
656    
657     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
658        long printing in the pcrecpp unittest when running under MinGW.
659    
660     6. ESC_K was left out of the EBCDIC table.
661    
662     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
663        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
664        limit also applies to "virtual nesting" when a pattern is recursive, and in
665        this case 1000 isn't so big. I have been able to remove this limit at the
666        expense of backing off one optimization in certain circumstances. Normally,
667        when pcre_exec() would call its internal match() function recursively and
668        immediately return the result unconditionally, it uses a "tail recursion"
669        feature to save stack. However, when a subpattern that can match an empty
670        string has an unlimited repetition quantifier, it no longer makes this
671        optimization. That gives it a stack frame in which to save the data for
672        checking that an empty string has been matched. Previously this was taken
673        from the 1000-entry workspace that had been reserved. So now there is no
674        explicit limit, but more stack is used.
675    
676     8. Applied Daniel's patches to solve problems with the import/export magic
677        syntax that is required for Windows, and which was going wrong for the
678        pcreposix and pcrecpp parts of the library. These were overlooked when this
679        problem was solved for the main library.
680    
681     9. There were some crude static tests to avoid integer overflow when computing
682        the size of patterns that contain repeated groups with explicit upper
683        limits. As the maximum quantifier is 65535, the maximum group length was
684        set at 30,000 so that the product of these two numbers did not overflow a
685        32-bit integer. However, it turns out that people want to use groups that
686        are longer than 30,000 bytes (though not repeat them that many times).
687        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
688        made it possible to implement the integer overflow checks in a much more
689        dynamic way, which I have now done. The artificial limitation on group
690        length has been removed - we now have only the limit on the total length of
691        the compiled pattern, which depends on the LINK_SIZE setting.
692    
693    10. Fixed a bug in the documentation for get/copy named substring when
694        duplicate names are permitted. If none of the named substrings are set, the
695        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
696        empty string.
697    
698    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
699        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
700        because the ] is interpreted as the first data character and the
701        terminating ] is not found. PCRE has been made compatible with Perl in this
702        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
703        cause memory overwriting.
704    
705    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
706        string has been matched (to stop an infinite loop). It was not recognizing
707        a conditional subpattern that could match an empty string if that
708        subpattern was within another subpattern. For example, it looped when
709        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
710        condition was not nested. This bug has been fixed.
711    
712    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
713        past the start of the subject in the presence of bytes with the top bit
714        set, for example "\x8aBCD".
715    
716    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
717        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
718    
719    14. Optimized (?!) to (*FAIL).
720    
721    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
722        This restricts code points to be within the range 0 to 0x10FFFF, excluding
723        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
724        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
725        does: it's just the validity check that is more restrictive.
726    
727    16. Inserted checks for integer overflows during escape sequence (backslash)
728        processing, and also fixed erroneous offset values for syntax errors during
729        backslash processing.
730    
731    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
732        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
733    
734    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
735        caused an overrun.
736    
737    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
738        something other than just ASCII characters) inside a group that had an
739        unlimited repeat caused a loop at compile time (while checking to see
740        whether the group could match an empty string).
741    
742    20. Debugging a pattern containing \p or \P could cause a crash. For example,
743        [\P{Any}] did so. (Error in the code for printing property names.)
744    
745    21. An orphan \E inside a character class could cause a crash.
746    
747    22. A repeated capturing bracket such as (A)? could cause a wild memory
748        reference during compilation.
749    
750    23. There are several functions in pcre_compile() that scan along a compiled
751        expression for various reasons (e.g. to see if it's fixed length for look
752        behind). There were bugs in these functions when a repeated \p or \P was
753        present in the pattern. These operators have additional parameters compared
754        with \d, etc, and these were not being taken into account when moving along
755        the compiled data. Specifically:
756    
757        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
758            length.
759    
760        (b) An item such as \pL+ within a repeated group could cause crashes or
761            loops.
762    
763        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
764            "reference to non-existent subpattern" error.
765    
766        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
767    
768    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
769        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
770    
771    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
772    
773    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
774        character were causing crashes (broken optimization).
775    
776    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
777        \p or \P) caused a compile-time loop.
778    
779    28. More problems have arisen in unanchored patterns when CRLF is a valid line
780        break. For example, the unstudied pattern [\r\n]A does not match the string
781        "\r\nA" because change 7.0/46 below moves the current point on by two
782        characters after failing to match at the start. However, the pattern \nA
783        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
784        the same is true. There doesn't seem any very clean way out of this, but
785        what I have chosen to do makes the common cases work: PCRE now takes note
786        of whether there can be an explicit match for \r or \n anywhere in the
787        pattern, and if so, 7.0/46 no longer applies. As part of this change,
788        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
789        pattern has explicit CR or LF references.
790    
791    29. Added (*CR) etc for changing newline setting at start of pattern.
792    
793    
794    Version 7.2 19-Jun-07
795    ---------------------
796    
797     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
798        which is apparently normally available under Windows.
799    
800     2. Re-jig the pcregrep tests with different newline settings in an attempt
801        to make them independent of the local environment's newline setting.
802    
803     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
804    
805     4. Some of the "internals" tests were previously cut out when the link size
806        was not 2, because the output contained actual offsets. The recent new
807        "Z" feature of pcretest means that these can be cut out, making the tests
808        usable with all link sizes.
809    
810     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
811        stack recursion. This gives a massive performance boost under BSD, but just
812        a small improvement under Linux. However, it saves one field in the frame
813        in all cases.
814    
815     6. Added more features from the forthcoming Perl 5.10:
816    
817        (a) (?-n) (where n is a string of digits) is a relative subroutine or
818            recursion call. It refers to the nth most recently opened parentheses.
819    
820        (b) (?+n) is also a relative subroutine call; it refers to the nth next
821            to be opened parentheses.
822    
823        (c) Conditions that refer to capturing parentheses can be specified
824            relatively, for example, (?(-2)... or (?(+3)...
825    
826        (d) \K resets the start of the current match so that everything before
827            is not part of it.
828    
829        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
830    
831        (f) \g{name} is another synonym - part of Perl 5.10's unification of
832            reference syntax.
833    
834        (g) (?| introduces a group in which the numbering of parentheses in each
835            alternative starts with the same number.
836    
837        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
838    
839     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
840        PCRE_INFO_JCHANGED.
841    
842     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
843        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
844        for detecting groups that can match an empty string.
845    
846     9. A pattern with a very large number of alternatives (more than several
847        hundred) was running out of internal workspace during the pre-compile
848        phase, where pcre_compile() figures out how much memory will be needed. A
849        bit of new cunning has reduced the workspace needed for groups with
850        alternatives. The 1000-alternative test pattern now uses 12 bytes of
851        workspace instead of running out of the 4096 that are available.
852    
853    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
854    
855    11. Applied patch from Google to remove an optimization that didn't quite work.
856        The report of the bug said:
857    
858          pcrecpp::RE("a*").FullMatch("aaa") matches, while
859          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
860          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
861    
862    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
863        it matched the wrong number of bytes.
864    
865    
866    Version 7.1 24-Apr-07
867    ---------------------
868    
869     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
870        that is more "standard", making use of automake and other Autotools. There
871        is some re-arrangement of the files and adjustment of comments consequent
872        on this.
873    
874     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
875        for recursive directory scanning broke on some systems because the files
876        are not scanned in any specific order and on different systems the order
877        was different. A call to "sort" has been inserted into RunGrepTest for the
878        approprate test as a short-term fix. In the longer term there may be an
879        alternative.
880    
881     3. I had an email from Eric Raymond about problems translating some of PCRE's
882        man pages to HTML (despite the fact that I distribute HTML pages, some
883        people do their own conversions for various reasons). The problems
884        concerned the use of low-level troff macros .br and .in. I have therefore
885        removed all such uses from the man pages (some were redundant, some could
886        be replaced by .nf/.fi pairs). The 132html script that I use to generate
887        HTML has been updated to handle .nf/.fi and to complain if it encounters
888        .br or .in.
889    
890     4. Updated comments in configure.ac that get placed in config.h.in and also
891        arranged for config.h to be included in the distribution, with the name
892        config.h.generic, for the benefit of those who have to compile without
893        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
894    
895     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
896        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
897        makevp.bat for the current PCRE, using the additional files
898        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
899    
900     6. A Windows user reported a minor discrepancy with test 2, which turned out
901        to be caused by a trailing space on an input line that had got lost in his
902        copy. The trailing space was an accident, so I've just removed it.
903    
904     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
905        that is needed.
906    
907     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
908        as "const" (a) because they are and (b) because it helps the PHP
909        maintainers who have recently made a script to detect big data structures
910        in the php code that should be moved to the .rodata section. I remembered
911        to update Builducptable as well, so it won't revert if ucptable.h is ever
912        re-created.
913    
914     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
915        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
916        order to be able to cut out the UTF-8 tables in the latter when UTF-8
917        support is not required. This saves 1.5-2K of code, which is important in
918        some applications.
919    
920        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
921        so as not to refer to the tables, even though these functions will never be
922        called when UTF-8 support is disabled. Otherwise there are problems with a
923        shared library.
924    
925    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
926    
927        (a) It was defining its arguments as char * instead of void *.
928    
929        (b) It was assuming that all moves were upwards in memory; this was true
930            a long time ago when I wrote it, but is no longer the case.
931    
932        The emulated memove() is provided for those environments that have neither
933        memmove() nor bcopy(). I didn't think anyone used it these days, but that
934        is clearly not the case, as these two bugs were recently reported.
935    
936    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
937        and Detrail to create the HTML documentation, the .txt form of the man
938        pages, and it removes trailing spaces from listed files. It also creates
939        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
940        case, it wraps all the #defines with #ifndefs. This script should be run
941        before "make dist".
942    
943    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
944        with Unicode property support.
945    
946        (a) For a maximizing quantifier, if the two different cases of the
947            character were of different lengths in their UTF-8 codings (there are
948            some cases like this - I found 11), and the matching function had to
949            back up over a mixture of the two cases, it incorrectly assumed they
950            were both the same length.
951    
952        (b) When PCRE was configured to use the heap rather than the stack for
953            recursion during matching, it was not correctly preserving the data for
954            the other case of a UTF-8 character when checking ahead for a match
955            while processing a minimizing repeat. If the check also involved
956            matching a wide character, but failed, corruption could cause an
957            erroneous result when trying to check for a repeat of the original
958            character.
959    
960    13. Some tidying changes to the testing mechanism:
961    
962        (a) The RunTest script now detects the internal link size and whether there
963            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
964            values substituted by "configure". (The RunGrepTest script already did
965            this for UTF-8.) The configure.ac script no longer substitutes the
966            relevant variables.
967    
968        (b) The debugging options /B and /D in pcretest show the compiled bytecode
969            with length and offset values. This means that the output is different
970            for different internal link sizes. Test 2 is skipped for link sizes
971            other than 2 because of this, bypassing the problem. Unfortunately,
972            there was also a test in test 3 (the locale tests) that used /B and
973            failed for link sizes other than 2. Rather than cut the whole test out,
974            I have added a new /Z option to pcretest that replaces the length and
975            offset values with spaces. This is now used to make test 3 independent
976            of link size. (Test 2 will be tidied up later.)
977    
978    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
979        segmentation fault instead of returning the appropriate error message.
980    
981    15. In multiline mode when the newline sequence was set to "any", the pattern
982        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
983        This doesn't seem right; it now treats the CRLF combination as the line
984        ending, and so does not match in that case. It's only a pattern such as ^$
985        that would hit this one: something like ^ABC$ would have failed after \r
986        and then tried again after \r\n.
987    
988    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
989        in an attempt to make files that differ only in their line terminators
990        compare equal. This works on Linux.
991    
992    17. Under certain error circumstances pcregrep might try to free random memory
993        as it exited. This is now fixed, thanks to valgrind.
994    
995    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
996        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
997        was because its rules for how to advance for /g after matching an empty
998        string at the end of a line did not allow for this case. They now check for
999        it specially.
1000    
1001    20. pcretest is supposed to handle patterns and data of any length, by
1002        extending its buffers when necessary. It was getting this wrong when the
1003        buffer for a data line had to be extended.
1004    
1005    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1006        CRLF as a newline sequence.
1007    
1008    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1009        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1010        I have nevertheless tidied it up.
1011    
1012    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1013    
1014    24. Added a man page for pcre-config.
1015    
1016    
1017    Version 7.0 19-Dec-06
1018    ---------------------
1019    
1020     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1021        moving to gcc 4.1.1.
1022    
1023     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1024        sys/time.h, which is documented as needed for this function. It doesn't
1025        seem to matter on Linux, but it showed up on some releases of OS X.
1026    
1027     3. It seems that there are systems where bytes whose values are greater than
1028        127 match isprint() in the "C" locale. The "C" locale should be the
1029        default when a C program starts up. In most systems, only ASCII printing
1030        characters match isprint(). This difference caused the output from pcretest
1031        to vary, making some of the tests fail. I have changed pcretest so that:
1032    
1033        (a) When it is outputting text in the compiled version of a pattern, bytes
1034            other than 32-126 are always shown as hex escapes.
1035    
1036        (b) When it is outputting text that is a matched part of a subject string,
1037            it does the same, unless a different locale has been set for the match
1038            (using the /L modifier). In this case, it uses isprint() to decide.
1039    
1040     4. Fixed a major bug that caused incorrect computation of the amount of memory
1041        required for a compiled pattern when options that changed within the
1042        pattern affected the logic of the preliminary scan that determines the
1043        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1044        that the computed length was too small. The symptoms of this bug were
1045        either the PCRE error "internal error: code overflow" from pcre_compile(),
1046        or a glibc crash with a message such as "pcretest: free(): invalid next
1047        size (fast)". Examples of patterns that provoked this bug (shown in
1048        pcretest format) are:
1049    
1050          /(?-x: )/x
1051          /(?x)(?-x: \s*#\s*)/
1052          /((?i)[\x{c0}])/8
1053          /(?i:[\x{c0}])/8
1054    
1055        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1056        is now done differently.
1057    
1058     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1059        wrapper classes; (b) implement a new function in the C++ scanner that is
1060        more efficient than the old way of doing things because it avoids levels of
1061        recursion in the regex matching; (c) add a paragraph to the documentation
1062        for the FullMatch() function.
1063    
1064     6. The escape sequence \n was being treated as whatever was defined as
1065        "newline". Not only was this contrary to the documentation, which states
1066        that \n is character 10 (hex 0A), but it also went horribly wrong when
1067        "newline" was defined as CRLF. This has been fixed.
1068    
1069     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1070        was being set to -1 for the "end of line" case (supposedly a value that no
1071        character can have). Though this value is never used (the check for end of
1072        line is "zero bytes in current character"), it caused compiler complaints.
1073        I've changed it to 0xffffffff.
1074    
1075     8. In pcre_version.c, the version string was being built by a sequence of
1076        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1077        string (as it is for production releases) called a macro with an empty
1078        argument. The C standard says the result of this is undefined. The gcc
1079        compiler treats it as an empty string (which was what was wanted) but it is
1080        reported that Visual C gives an error. The source has been hacked around to
1081        avoid this problem.
1082    
1083     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1084        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1085        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1086        of them did).
1087    
1088    10. Originally, pcretest opened its input and output without "b"; then I was
1089        told that "b" was needed in some environments, so it was added for release
1090        5.0 to both the input and output. (It makes no difference on Unix-like
1091        systems.) Later I was told that it is wrong for the input on Windows. I've
1092        now abstracted the modes into two macros, to make it easier to fiddle with
1093        them, and removed "b" from the input mode under Windows.
1094    
1095    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1096    
1097    12. Added -help and --help to pcretest as an official way of being reminded
1098        of the options.
1099    
1100    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1101        and pcrecpp.cc because they annoy compilers at high warning levels.
1102    
1103    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1104    
1105    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1106        scripts are not C programs :-) and which was not noticed because it works
1107        on Linux.
1108    
1109    16. pcretest is supposed to handle any length of pattern and data line (as one
1110        line or as a continued sequence of lines) by extending its input buffer if
1111        necessary. This feature was broken for very long pattern lines, leading to
1112        a string of junk being passed to pcre_compile() if the pattern was longer
1113        than about 50K.
1114    
1115    17. I have done a major re-factoring of the way pcre_compile() computes the
1116        amount of memory needed for a compiled pattern. Previously, there was code
1117        that made a preliminary scan of the pattern in order to do this. That was
1118        OK when PCRE was new, but as the facilities have expanded, it has become
1119        harder and harder to keep it in step with the real compile phase, and there
1120        have been a number of bugs (see for example, 4 above). I have now found a
1121        cunning way of running the real compile function in a "fake" mode that
1122        enables it to compute how much memory it would need, while actually only
1123        ever using a few hundred bytes of working memory and without too many
1124        tests of the mode. This should make future maintenance and development
1125        easier. A side effect of this work is that the limit of 200 on the nesting
1126        depth of parentheses has been removed (though this was never a serious
1127        limitation, I suspect). However, there is a downside: pcre_compile() now
1128        runs more slowly than before (30% or more, depending on the pattern). I
1129        hope this isn't a big issue. There is no effect on runtime performance.
1130    
1131    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1132        newline (only possible for the last line of a file) and it was a
1133        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1134    
1135    19. Added additional timing features to pcretest. (1) The -tm option now times
1136        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1137        separate command line item, by a number that specifies the number of
1138        repeats to use when timing. The default is 50000; this gives better
1139        precision, but takes uncomfortably long for very large patterns.
1140    
1141    20. Extended pcre_study() to be more clever in cases where a branch of a
1142        subpattern has no definite first character. For example, (a*|b*)[cd] would
1143        previously give no result from pcre_study(). Now it recognizes that the
1144        first character must be a, b, c, or d.
1145    
1146    21. There was an incorrect error "recursive call could loop indefinitely" if
1147        a subpattern (or the entire pattern) that was being tested for matching an
1148        empty string contained only one non-empty item after a nested subpattern.
1149        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1150        incorrectly, because the \d was being skipped in the check.
1151    
1152    22. The pcretest program now has a new pattern option /B and a command line
1153        option -b, which is equivalent to adding /B to every pattern. This causes
1154        it to show the compiled bytecode, without the additional information that
1155        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1156        is the same as /B/I).
1157    
1158    23. A new optimization is now able automatically to treat some sequences such
1159        as a*b as a*+b. More specifically, if something simple (such as a character
1160        or a simple class like \d) has an unlimited quantifier, and is followed by
1161        something that cannot possibly match the quantified thing, the quantifier
1162        is automatically "possessified".
1163    
1164    24. A recursive reference to a subpattern whose number was greater than 39
1165        went wrong under certain circumstances in UTF-8 mode. This bug could also
1166        have affected the operation of pcre_study().
1167    
1168    25. Realized that a little bit of performance could be had by replacing
1169        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1170    
1171    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1172    
1173    27. Possessive quantifiers such as a++ were previously implemented by turning
1174        them into atomic groups such as ($>a+). Now they have their own opcodes,
1175        which improves performance. This includes the automatically created ones
1176        from 23 above.
1177    
1178    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1179        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1180        the first matched character to be a colon. This applied both to named and
1181        numbered groups.
1182    
1183    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1184    
1185    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1186        building PCRE on Windows easier, so I have included it in the distribution.
1187    
1188    31. There is now a check in pcretest against a ridiculously large number being
1189        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1190        loop, the loop is abandoned.
1191    
1192    32. Forward references to subpatterns in conditions such as (?(2)...) where
1193        subpattern 2 is defined later cause pcre_compile() to search forwards in
1194        the pattern for the relevant set of parentheses. This search went wrong
1195        when there were unescaped parentheses in a character class, parentheses
1196        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1197    
1198    33. "Subroutine" calls and backreferences were previously restricted to
1199        referencing subpatterns earlier in the regex. This restriction has now
1200        been removed.
1201    
1202    34. Added a number of extra features that are going to be in Perl 5.10. On the
1203        whole, these are just syntactic alternatives for features that PCRE had
1204        previously implemented using the Python syntax or my own invention. The
1205        other formats are all retained for compatibility.
1206    
1207        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1208            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1209            also .NET compatible.
1210    
1211        (b) A recursion or subroutine call to a named group can now be defined as
1212            (?&name) as well as (?P>name).
1213    
1214        (c) A backreference to a named group can now be defined as \k<name> or
1215            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1216            5.10, are also .NET compatible.
1217    
1218        (d) A conditional reference to a named group can now use the syntax
1219            (?(<name>) or (?('name') as well as (?(name).
1220    
1221        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1222            groups (named and numbered) that are never evaluated inline, but can be
1223            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1224            is always false. There may be only one alternative in such a group.
1225    
1226        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1227            as the simple (?(R). The condition is true only if the most recent
1228            recursion is that of the given number or name. It does not search out
1229            through the entire recursion stack.
1230    
1231        (g) The escape \gN or \g{N} has been added, where N is a positive or
1232            negative number, specifying an absolute or relative reference.
1233    
1234    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1235        some "unreachable code" warnings.
1236    
1237    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1238        things, this adds five new scripts.
1239    
1240    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1241        There were also incompatibilities regarding the handling of \Q..\E inside
1242        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1243        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1244    
1245    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1246        matches an empty string, and forcibly breaks the loop. There were bugs in
1247        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1248        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1249        separate and independent bugs (that affected different cases) have been
1250        fixed.
1251    
1252    39. Refactored the code to abolish the use of different opcodes for small
1253        capturing bracket numbers. This is a tidy that I avoided doing when I
1254        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1255        The new approach is not only tidier, it makes it possible to reduce the
1256        memory needed to fix the previous bug (38).
1257    
1258    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1259        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1260        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1261        mode.
1262    
1263    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1264        report.
1265    
1266    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1267        copy construction and assignment in the C++ wrapper.
1268    
1269    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1270        couple of bugs that could have given wrong results in the "--newline=crlf"
1271        case.
1272    
1273    44. Added a number of casts and did some reorganization of signed/unsigned int
1274        variables following suggestions from Dair Grant. Also renamed the variable
1275        "this" as "item" because it is a C++ keyword.
1276    
1277    45. Arranged for dftables to add
1278    
1279          #include "pcre_internal.h"
1280    
1281        to pcre_chartables.c because without it, gcc 4.x may remove the array
1282        definition from the final binary if PCRE is built into a static library and
1283        dead code stripping is activated.
1284    
1285    46. For an unanchored pattern, if a match attempt fails at the start of a
1286        newline sequence, and the newline setting is CRLF or ANY, and the next two
1287        characters are CRLF, advance by two characters instead of one.
1288    
1289    
1290    Version 6.7 04-Jul-06
1291    ---------------------
1292    
1293     1. In order to handle tests when input lines are enormously long, pcretest has
1294        been re-factored so that it automatically extends its buffers when
1295        necessary. The code is crude, but this _is_ just a test program. The
1296        default size has been increased from 32K to 50K.
1297    
1298     2. The code in pcre_study() was using the value of the re argument before
1299        testing it for NULL. (Of course, in any sensible call of the function, it
1300        won't be NULL.)
1301    
1302     3. The memmove() emulation function in pcre_internal.h, which is used on
1303        systems that lack both memmove() and bcopy() - that is, hardly ever -
1304        was missing a "static" storage class specifier.
1305    
1306     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1307        containing an extended class (one that cannot be represented by a bitmap
1308        because it contains high-valued characters or Unicode property items, e.g.
1309        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1310        pattern, but PCRE should not loop if you do not (it no longer does).
1311        [Detail: two cases were found: (a) a repeated subpattern containing an
1312        extended class; (b) a recursive reference to a subpattern that followed a
1313        previous extended class. It wasn't skipping over the extended class
1314        correctly when UTF-8 mode was not set.]
1315    
1316     5. A negated single-character class was not being recognized as fixed-length
1317        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1318        compile error "lookbehind assertion is not fixed length".
1319    
1320     6. The RunPerlTest auxiliary script was showing an unexpected difference
1321        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1322        write a Perl script that can interpret lines of an input file either as
1323        byte characters or as UTF-8, which is what "perltest" was being required to
1324        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1325        can't do is switch easily at run time between having the "use utf8;" pragma
1326        or not. In the end, I fudged it by using the RunPerlTest script to insert
1327        "use utf8;" explicitly for the UTF-8 tests.
1328    
1329     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1330        the end of the subject string, contrary to the documentation and to what
1331        Perl does. This was true of both matching functions. Now it matches only at
1332        the start of the subject and immediately after *internal* newlines.
1333    
1334     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1335        a pointer to an int instead of a pointer to an unsigned long int. This
1336        caused problems on 64-bit systems.
1337    
1338     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1339        instance of the 'standard' template library not being so standard".
1340    
1341    10. There was no check on the number of named subpatterns nor the maximum
1342        length of a subpattern name. The product of these values is used to compute
1343        the size of the memory block for a compiled pattern. By supplying a very
1344        long subpattern name and a large number of named subpatterns, the size
1345        computation could be caused to overflow. This is now prevented by limiting
1346        the length of names to 32 characters, and the number of named subpatterns
1347        to 10,000.
1348    
1349    11. Subpatterns that are repeated with specific counts have to be replicated in
1350        the compiled pattern. The size of memory for this was computed from the
1351        length of the subpattern and the repeat count. The latter is limited to
1352        65535, but there was no limit on the former, meaning that integer overflow
1353        could in principle occur. The compiled length of a repeated subpattern is
1354        now limited to 30,000 bytes in order to prevent this.
1355    
1356    12. Added the optional facility to have named substrings with the same name.
1357    
1358    13. Added the ability to use a named substring as a condition, using the
1359        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1360        are numbers (not recommended). Forward references are permitted.
1361    
1362    14. Added forward references in named backreferences (if you see what I mean).
1363    
1364    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1365        pattern could run off the end of the subject. For example, the pattern
1366        "(?s)(.{1,5})"8 did this with the subject "ab".
1367    
1368    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1369        PCRE_CASELESS was set when matching characters that were quantified with ?
1370        or *.
1371    
1372    17. A character class other than a single negated character that had a minimum
1373        but no maximum quantifier - for example [ab]{6,} - was not handled
1374        correctly by pce_dfa_exec(). It would match only one character.
1375    
1376    18. A valid (though odd) pattern that looked like a POSIX character
1377        class but used an invalid character after [ (for example [[,abc,]]) caused
1378        pcre_compile() to give the error "Failed: internal error: code overflow" or
1379        in some cases to crash with a glibc free() error. This could even happen if
1380        the pattern terminated after [[ but there just happened to be a sequence of
1381        letters, a binary zero, and a closing ] in the memory that followed.
1382    
1383    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1384        over the years. Originally (before any Unicode support), just the bottom 8
1385        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1386        output from "man perlunicode" includes this:
1387    
1388          The regular expression compiler produces polymorphic opcodes.  That
1389          is, the pattern adapts to the data and automatically switches to
1390          the Unicode character scheme when presented with Unicode data--or
1391          instead uses a traditional byte scheme when presented with byte
1392          data.
1393    
1394        Sadly, a wide octal escape does not cause a switch, and in a string with
1395        no other multibyte characters, these octal escapes are treated as before.
1396        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1397        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1398        Unicode string.
1399    
1400        I have not perpetrated such confusion in PCRE. Up till now, it took just
1401        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1402        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1403        translate to the appropriate multibyte character.
1404    
1405    29. Applied some refactoring to reduce the number of warnings from Microsoft
1406        and Borland compilers. This has included removing the fudge introduced
1407        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1408        a warning about an unused variable.
1409    
1410    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1411        characters since release 4.0, because Perl (from release 5.004) does not.
1412        [Or at least, is documented not to: some releases seem to be in conflict
1413        with the documentation.] However, when a pattern was studied with
1414        pcre_study() and all its branches started with \s, PCRE still included VT
1415        as a possible starting character. Of course, this did no harm; it just
1416        caused an unnecessary match attempt.
1417    
1418    22. Removed a now-redundant internal flag bit that recorded the fact that case
1419        dependency changed within the pattern. This was once needed for "required
1420        byte" processing, but is no longer used. This recovers a now-scarce options
1421        bit. Also moved the least significant internal flag bit to the most-
1422        significant bit of the word, which was not previously used (hangover from
1423        the days when it was an int rather than a uint) to free up another bit for
1424        the future.
1425    
1426    23. Added support for CRLF line endings as well as CR and LF. As well as the
1427        default being selectable at build time, it can now be changed at runtime
1428        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1429        specify that it is scanning data with non-default line endings.
1430    
1431    24. Changed the definition of CXXLINK to make it agree with the definition of
1432        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1433    
1434    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1435        recursions. This makes a big different to stack usage for some patterns.
1436    
1437    26. If a subpattern containing a named recursion or subroutine reference such
1438        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1439        the space required for the compiled pattern went wrong and gave too small a
1440        value. Depending on the environment, this could lead to "Failed: internal
1441        error: code overflow at offset 49" or "glibc detected double free or
1442        corruption" errors.
1443    
1444    27. Applied patches from Google (a) to support the new newline modes and (b) to
1445        advance over multibyte UTF-8 characters in GlobalReplace.
1446    
1447    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1448        difference for some implementation of PCRE in some Windows version.
1449    
1450    29. Added some extra testing facilities to pcretest:
1451    
1452        \q<number>   in a data line sets the "match limit" value
1453        \Q<number>   in a data line sets the "match recursion limt" value
1454        -S <number>  sets the stack size, where <number> is in megabytes
1455    
1456        The -S option isn't available for Windows.
1457    
1458    
1459    Version 6.6 06-Feb-06
1460    ---------------------
1461    
1462     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1463        in pcreposix.h. I have copied the definition from pcre.h.
1464    
1465     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1466        because pcre.h is no longer a built file.
1467    
1468     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1469        not normally included in the compiled code.
1470    
1471    
1472    Version 6.5 01-Feb-06
1473    ---------------------
1474    
1475     1. When using the partial match feature with pcre_dfa_exec(), it was not
1476        anchoring the second and subsequent partial matches at the new starting
1477        point. This could lead to incorrect results. For example, with the pattern
1478        /1234/, partially matching against "123" and then "a4" gave a match.
1479    
1480     2. Changes to pcregrep:
1481    
1482        (a) All non-match returns from pcre_exec() were being treated as failures
1483            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1484            error message is output. Some extra information is given for the
1485            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1486            probably the only errors that are likely to be caused by users (by
1487            specifying a regex that has nested indefinite repeats, for instance).
1488            If there are more than 20 of these errors, pcregrep is abandoned.
1489    
1490        (b) A binary zero was treated as data while matching, but terminated the
1491            output line if it was written out. This has been fixed: binary zeroes
1492            are now no different to any other data bytes.
1493    
1494        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1495            used to set a locale for matching. The --locale=xxxx long option has
1496            been added (no short equivalent) to specify a locale explicitly on the
1497            pcregrep command, overriding the environment variables.
1498    
1499        (d) When -B was used with -n, some line numbers in the output were one less
1500            than they should have been.
1501    
1502        (e) Added the -o (--only-matching) option.
1503    
1504        (f) If -A or -C was used with -c (count only), some lines of context were
1505            accidentally printed for the final match.
1506    
1507        (g) Added the -H (--with-filename) option.
1508    
1509        (h) The combination of options -rh failed to suppress file names for files
1510            that were found from directory arguments.
1511    
1512        (i) Added the -D (--devices) and -d (--directories) options.
1513    
1514        (j) Added the -F (--fixed-strings) option.
1515    
1516        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1517    
1518        (l) Added the --colo(u)r option.
1519    
1520        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1521            is not present by default.
1522    
1523     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1524        items such as (?R) or (?1), when the recursion could match a number of
1525        alternatives. If it matched one of the alternatives, but subsequently,
1526        outside the recursion, there was a failure, the code tried to back up into
1527        the recursion. However, because of the way PCRE is implemented, this is not
1528        possible, and the result was an incorrect result from the match.
1529    
1530        In order to prevent this happening, the specification of recursion has
1531        been changed so that all such subpatterns are automatically treated as
1532        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1533    
1534     4. I had overlooked the fact that, in some locales, there are characters for
1535        which isalpha() is true but neither isupper() nor islower() are true. In
1536        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1537        and ordfeminine) are like this. This affected the treatment of \w and \W
1538        when they appeared in character classes, but not when they appeared outside
1539        a character class. The bit map for "word" characters is now created
1540        separately from the results of isalnum() instead of just taking it from the
1541        upper, lower, and digit maps. (Plus the underscore character, of course.)
1542    
1543     5. The above bug also affected the handling of POSIX character classes such as
1544        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1545        permanent tables. Instead, the bit maps for such a class were previously
1546        created as the appropriate unions of the upper, lower, and digit bitmaps.
1547        Now they are created by subtraction from the [[:word:]] class, which has
1548        its own bitmap.
1549    
1550     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1551        It is created by subtracting the vertical space characters (\x09, \x0a,
1552        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1553        subtraction was done in the overall bitmap for a character class, meaning
1554        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1555        be recognized. This bug has been fixed.
1556    
1557     7. Patches from the folks at Google:
1558    
1559          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1560          real life, but is still worth protecting against".
1561    
1562          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1563          regular expressions".
1564    
1565          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1566          have it.
1567    
1568          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1569          "configure" and the latter not, in order to fix a problem somebody had
1570          with compiling the Arg class on HP-UX.
1571    
1572          (e) Improve the error-handling of the C++ wrapper a little bit.
1573    
1574          (f) New tests for checking recursion limiting.
1575    
1576     8. The pcre_memmove() function, which is used only if the environment does not
1577        have a standard memmove() function (and is therefore rarely compiled),
1578        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1579        returning a result (though PCRE never actually uses the result).
1580    
1581     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1582        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1583        returned instead of calling malloc() with an overflowing number that would
1584        most likely cause subsequent chaos.
1585    
1586    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1587    
1588    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1589        with this option is matched, the nmatch and pmatch options of regexec() are
1590        ignored.
1591    
1592    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1593        provided in case anyone wants to the the POSIX interface with UTF-8
1594        strings.
1595    
1596    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1597        C++ linking (needed for some HP-UX environments).
1598    
1599    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1600        (unused parameter) and in the pcre_printint() function (omitted "default"
1601        switch label when the default is to do nothing).
1602    
1603    15. Added some code to make it possible, when PCRE is compiled as a C++
1604        library, to replace subject pointers for pcre_exec() with a smart pointer
1605        class, thus making it possible to process discontinuous strings.
1606    
1607    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1608        much the same function. They were added by different people who were trying
1609        to make PCRE easy to compile on non-Unix systems. It has been suggested
1610        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1611        for compiling on Windows systems. I have therefore replaced it with
1612        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1613        defaults to "extern" for C or "extern C" for C++, which works fine on
1614        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1615        SCOPE with something explicit in config.h. In addition:
1616    
1617        (a) pcreposix.h still had just "extern" instead of either of these macros;
1618            I have replaced it with PCRE_DATA_SCOPE.
1619    
1620        (b) Functions such as _pcre_xclass(), which are internal to the library,
1621            but external in the C sense, all had PCRE_EXPORT in their definitions.
1622            This is apparently wrong for the Windows case, so I have removed it.
1623            (It makes no difference on Unix-like systems.)
1624    
1625    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1626        of recursive calls to match(). This is different to MATCH_LIMIT because
1627        that limits the total number of calls to match(), not all of which increase
1628        the depth of recursion. Limiting the recursion depth limits the amount of
1629        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1630        when PCRE is compiled, and changed at run time. A patch from Google adds
1631        this functionality to the C++ interface.
1632    
1633    18. Changes to the handling of Unicode character properties:
1634    
1635        (a) Updated the table to Unicode 4.1.0.
1636    
1637        (b) Recognize characters that are not in the table as "Cn" (undefined).
1638    
1639        (c) I revised the way the table is implemented to a much improved format
1640            which includes recognition of ranges. It now supports the ranges that
1641            are defined in UnicodeData.txt, and it also amalgamates other
1642            characters into ranges. This has reduced the number of entries in the
1643            table from around 16,000 to around 3,000, thus reducing its size
1644            considerably. I realized I did not need to use a tree structure after
1645            all - a binary chop search is just as efficient. Having reduced the
1646            number of entries, I extended their size from 6 bytes to 8 bytes to
1647            allow for more data.
1648    
1649        (d) Added support for Unicode script names via properties such as \p{Han}.
1650    
1651    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1652        matching that character.
1653    
1654    20. When matching a repeated Unicode property with a minimum greater than zero,
1655        (for example \pL{2,}), PCRE could look past the end of the subject if it
1656        reached it while seeking the minimum number of characters. This could
1657        happen only if some of the characters were more than one byte long, because
1658        there is a check for at least the minimum number of bytes.
1659    
1660    21. Refactored the implementation of \p and \P so as to be more general, to
1661        allow for more different types of property in future. This has changed the
1662        compiled form incompatibly. Anybody with saved compiled patterns that use
1663        \p or \P will have to recompile them.
1664    
1665    22. Added "Any" and "L&" to the supported property types.
1666    
1667    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1668        but give a compile time error if the value is greater than 0xff.
1669    
1670    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1671        accidentally not being installed or uninstalled.
1672    
1673    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1674        made were to insert the current release number. This seemed silly, because
1675        it made things harder for people building PCRE on systems that don't run
1676        "configure". I have turned pcre.h into a distributed file, no longer built
1677        by "configure", with the version identification directly included. There is
1678        no longer a pcre.h.in file.
1679    
1680        However, this change necessitated a change to the pcre-config script as
1681        well. It is built from pcre-config.in, and one of the substitutions was the
1682        release number. I have updated configure.ac so that ./configure now finds
1683        the release number by grepping pcre.h.
1684    
1685    26. Added the ability to run the tests under valgrind.
1686    
1687    
1688    Version 6.4 05-Sep-05
1689    ---------------------
1690    
1691     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1692        "--" to be printed when multiple files were scanned, even when none of the
1693        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1694        consider it to be a bug, and have restored the previous behaviour.
1695    
1696     2. A couple of code tidies to get rid of compiler warnings.
1697    
1698     3. The pcretest program used to cheat by referring to symbols in the library
1699        whose names begin with _pcre_. These are internal symbols that are not
1700        really supposed to be visible externally, and in some environments it is
1701        possible to suppress them. The cheating is now confined to including
1702        certain files from the library's source, which is a bit cleaner.
1703    
1704     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1705        file's purpose clearer.
1706    
1707     5. Reorganized pcre_ucp_findchar().
1708    
1709    
1710    Version 6.3 15-Aug-05
1711    ---------------------
1712    
1713     1. The file libpcre.pc.in did not have general read permission in the tarball.
1714    
1715     2. There were some problems when building without C++ support:
1716    
1717        (a) If C++ support was not built, "make install" and "make test" still
1718            tried to test it.
1719    
1720        (b) There were problems when the value of CXX was explicitly set. Some
1721            changes have been made to try to fix these, and ...
1722    
1723        (c) --disable-cpp can now be used to explicitly disable C++ support.
1724    
1725        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1726            backslash in a target when C++ was disabled. This confuses some
1727            versions of "make", apparently. Using an intermediate variable solves
1728            this. (Same for CPP_LOBJ.)
1729    
1730     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1731        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1732        necessary on certain architectures.
1733    
1734     4. Added a setting of -export-symbols-regex to the link command to remove
1735        those symbols that are exported in the C sense, but actually are local
1736        within the library, and not documented. Their names all begin with
1737        "_pcre_". This is not a perfect job, because (a) we have to except some
1738        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1739        available (and never for static libraries). I have made a note to try to
1740        find a way round (a) in the future.
1741    
1742    
1743    Version 6.2 01-Aug-05
1744    ---------------------
1745    
1746     1. There was no test for integer overflow of quantifier values. A construction
1747        such as {1111111111111111} would give undefined results. What is worse, if
1748        a minimum quantifier for a parenthesized subpattern overflowed and became
1749        negative, the calculation of the memory size went wrong. This could have
1750        led to memory overwriting.
1751    
1752     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1753    
1754     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1755        operating environments where this matters.
1756    
1757     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1758        PCRE options from within the C++ wrapper.
1759    
1760     5. Named capturing subpatterns were not being correctly counted when a pattern
1761        was compiled. This caused two problems: (a) If there were more than 100
1762        such subpatterns, the calculation of the memory needed for the whole
1763        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1764        back references of the form \12, where the number was greater than 9, were
1765        not recognized as back references, even though there were sufficient
1766        previous subpatterns.
1767    
1768     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1769        versions of gcc, e.g. 2.95.4.
1770    
1771    
1772    Version 6.1 21-Jun-05
1773    ---------------------
1774    
1775     1. There was one reference to the variable "posix" in pcretest.c that was not
1776        surrounded by "#if !defined NOPOSIX".
1777    
1778     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1779        the cross-check on the old pcre_info() function, for the benefit of the
1780        cut-down version of PCRE that is currently imported into Exim.
1781    
1782     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1783        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1784        patterns that start (?i) but I don't think that matters. The use of (?i) is
1785        just an example; this all applies to the other options as well.
1786    
1787     4. Since libtool seems to echo the compile commands it is issuing, the output
1788        from "make" can be reduced a bit by putting "@" in front of each libtool
1789        compile command.
1790    
1791     5. Patch from the folks at Google for configure.in to be a bit more thorough
1792        in checking for a suitable C++ installation before trying to compile the
1793        C++ stuff. This should fix a reported problem when a compiler was present,
1794        but no suitable headers.
1795    
1796     6. The man pages all had just "PCRE" as their title. I have changed them to
1797        be the relevant file name. I have also arranged that these names are
1798        retained in the file doc/pcre.txt, which is a concatenation in text format
1799        of all the man pages except the little individual ones for each function.
1800    
1801     7. The NON-UNIX-USE file had not been updated for the different set of source
1802        files that come with release 6. I also added a few comments about the C++
1803        wrapper.
1804    
1805    
1806    Version 6.0 07-Jun-05
1807    ---------------------
1808    
1809     1. Some minor internal re-organization to help with my DFA experiments.
1810    
1811     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1812        didn't matter for the library itself when fully configured, but did matter
1813        when compiling without UCP support, or within Exim, where the ucp files are
1814        not imported.
1815    
1816     3. Refactoring of the library code to split up the various functions into
1817        different source modules. The addition of the new DFA matching code (see
1818        below) to a single monolithic source would have made it really too
1819        unwieldy, quite apart from causing all the code to be include in a
1820        statically linked application, when only some functions are used. This is
1821        relevant even without the DFA addition now that patterns can be compiled in
1822        one application and matched in another.
1823    
1824        The downside of splitting up is that there have to be some external
1825        functions and data tables that are used internally in different modules of
1826        the library but which are not part of the API. These have all had their
1827        names changed to start with "_pcre_" so that they are unlikely to clash
1828        with other external names.
1829    
1830     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1831        a different (DFA) algorithm. Although it is slower than the original
1832        function, it does have some advantages for certain types of matching
1833        problem.
1834    
1835     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1836        including restarting after a partial match.
1837    
1838     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1839        defined when compiling for Windows was sent to me. I have put it into the
1840        code, though I have no means of testing or verifying it.
1841    
1842     7. Added the pcre_refcount() auxiliary function.
1843    
1844     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1845        match before or at the first newline in the subject string. In pcretest,
1846        the /f option on a pattern can be used to set this.
1847    
1848     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1849        would behave wrongly. This has been present in PCRE since release 4.0.
1850    
1851    10. A number of changes to the pcregrep command:
1852    
1853        (a) Refactored how -x works; insert ^(...)$ instead of setting
1854            PCRE_ANCHORED and checking the length, in preparation for adding
1855            something similar for -w.
1856    
1857        (b) Added the -w (match as a word) option.
1858    
1859        (c) Refactored the way lines are read and buffered so as to have more
1860            than one at a time available.
1861    
1862        (d) Implemented a pcregrep test script.
1863    
1864        (e) Added the -M (multiline match) option. This allows patterns to match
1865            over several lines of the subject. The buffering ensures that at least
1866            8K, or the rest of the document (whichever is the shorter) is available
1867            for matching (and similarly the previous 8K for lookbehind assertions).
1868    
1869        (f) Changed the --help output so that it now says
1870    
1871              -w, --word-regex(p)
1872    
1873            instead of two lines, one with "regex" and the other with "regexp"
1874            because that confused at least one person since the short forms are the
1875            same. (This required a bit of code, as the output is generated
1876            automatically from a table. It wasn't just a text change.)
1877    
1878        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1879            option but starts with a hyphen. Could be a pattern or a path name
1880            starting with a hyphen, for instance.
1881    
1882        (h) "-" can be given as a file name to represent stdin.
1883    
1884        (i) When file names are being printed, "(standard input)" is used for
1885            the standard input, for compatibility with GNU grep. Previously
1886            "<stdin>" was used.
1887    
1888        (j) The option --label=xxx can be used to supply a name to be used for
1889            stdin when file names are being printed. There is no short form.
1890    
1891        (k) Re-factored the options decoding logic because we are going to add
1892            two more options that take data. Such options can now be given in four
1893            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1894    
1895        (l) Added the -A, -B, and -C options for requesting that lines of context
1896            around matches be printed.
1897    
1898        (m) Added the -L option to print the names of files that do not contain
1899            any matching lines, that is, the complement of -l.
1900    
1901        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1902            continue to scan other files.
1903    
1904        (o) The -s option was incorrectly implemented. For compatibility with other
1905            greps, it now suppresses the error message for a non-existent or non-
1906            accessible file (but not the return code). There is a new option called
1907            -q that suppresses the output of matching lines, which was what -s was
1908            previously doing.
1909    
1910        (p) Added --include and --exclude options to specify files for inclusion
1911            and exclusion when recursing.
1912    
1913    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1914        Hopefully, it now does.
1915    
1916    12. Missing cast in pcre_study().
1917    
1918    13. Added an "uninstall" target to the makefile.
1919    
1920    14. Replaced "extern" in the function prototypes in Makefile.in with
1921        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1922        world, but is set differently for Windows.
1923    
1924    15. Added a second compiling function called pcre_compile2(). The only
1925        difference is that it has an extra argument, which is a pointer to an
1926        integer error code. When there is a compile-time failure, this is set
1927        non-zero, in addition to the error test pointer being set to point to an
1928        error message. The new argument may be NULL if no error number is required
1929        (but then you may as well call pcre_compile(), which is now just a
1930        wrapper). This facility is provided because some applications need a
1931        numeric error indication, but it has also enabled me to tidy up the way
1932        compile-time errors are handled in the POSIX wrapper.
1933    
1934    16. Added VPATH=.libs to the makefile; this should help when building with one
1935        prefix path and installing with another. (Or so I'm told by someone who
1936        knows more about this stuff than I do.)
1937    
1938    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1939        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1940        match everything, including newlines. This is not POSIX-compatible, but
1941        somebody wanted the feature. From pcretest it can be activated by using
1942        both the P and the s flags.
1943    
1944    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1945    
1946    19. libpcre.pc was being incorrectly installed as executable.
1947    
1948    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1949        it now also looks for '\r' so that it will work unmodified on Windows.
1950    
1951    21. Added Google's contributed C++ wrapper to the distribution.
1952    
1953    22. Added some untidy missing memory free() calls in pcretest, to keep
1954        Electric Fence happy when testing.
1955    
1956    
1957    
1958    Version 5.0 13-Sep-04
1959    ---------------------
1960    
1961     1. Internal change: literal characters are no longer packed up into items
1962        containing multiple characters in a single byte-string. Each character
1963        is now matched using a separate opcode. However, there may be more than one
1964        byte in the character in UTF-8 mode.
1965    
1966     2. The pcre_callout_block structure has two new fields: pattern_position and
1967        next_item_length. These contain the offset in the pattern to the next match
1968        item, and its length, respectively.
1969    
1970     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1971        insertion of callouts before each pattern item. Added the /C option to
1972        pcretest to make use of this.
1973    
1974     4. On the advice of a Windows user, the lines
1975    
1976          #if defined(_WIN32) || defined(WIN32)
1977          _setmode( _fileno( stdout ), 0x8000 );
1978          #endif  /* defined(_WIN32) || defined(WIN32) */
1979    
1980        have been added to the source of pcretest. This apparently does useful
1981        magic in relation to line terminators.
1982    
1983     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1984        for the benefit of those environments where the "b" makes a difference.
1985    
1986     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1987        to know about it. I have put a hack into configure.in that adds in code
1988        to set GCC=yes if CC=icc. This seems to end up at a point in the
1989        generated configure script that is early enough to affect the setting of
1990        compiler options, which is what is needed, but I have no means of testing
1991        whether it really works. (The user who reported this had patched the
1992        generated configure script, which of course I cannot do.)
1993    
1994        LATER: After change 22 below (new libtool files), the configure script
1995        seems to know about icc (and also ecc). Therefore, I have commented out
1996        this hack in configure.in.
1997    
1998     7. Added support for pkg-config (2 patches were sent in).
1999    
2000     8. Negated POSIX character classes that used a combination of internal tables
2001        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2002        [[:^ascii]]. Typically, they would match almost any characters. The other
2003        POSIX classes were not broken in this way.
2004    
2005     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2006        to find the match, as PCRE was deluded into thinking that the match had to
2007        start at the start point or following a newline. The same bug applied to
2008        patterns with negative forward assertions or any backward assertions
2009        preceding ".*" at the start, unless the pattern required a fixed first
2010        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2011    
2012    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2013        starting at the last subject character, bytes beyond the end of the subject
2014        string were read.
2015    
2016    11. Renamed the variable "class" as "classbits" to make life easier for C++
2017        users. (Previously there was a macro definition, but it apparently wasn't
2018        enough.)
2019    
2020    12. Added the new field "tables" to the extra data so that tables can be passed
2021        in at exec time, or the internal tables can be re-selected. This allows
2022        a compiled regex to be saved and re-used at a later time by a different
2023        program that might have everything at different addresses.
2024    
2025    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2026        -R library as well as a -L library.
2027    
2028    14. The debugging options of pcretest (-d on the command line or D on a
2029        pattern) showed incorrect output for anything following an extended class
2030        that contained multibyte characters and which was followed by a quantifier.
2031    
2032    15. Added optional support for general category Unicode character properties
2033        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2034        support. It adds about 90K to the size of the library. The meanings of the
2035        inbuilt class escapes such as \d and \s have NOT been changed.
2036    
2037    16. Updated pcredemo.c to include calls to free() to release the memory for the
2038        compiled pattern.
2039    
2040    17. The generated file chartables.c was being created in the source directory
2041        instead of in the building directory. This caused the build to fail if the
2042        source directory was different from the building directory, and was
2043        read-only.
2044    
2045    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2046        file. No doubt somebody will tell me if they don't make sense... Also added
2047        Dan Mooney's comments about building on OpenVMS.
2048    
2049    19. Added support for partial matching via the PCRE_PARTIAL option for
2050        pcre_exec() and the \P data escape in pcretest.
2051    
2052    20. Extended pcretest with 3 new pattern features:
2053    
2054        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2055              write the compiled pattern to the file whose name is "rest-of-line".
2056              This is a straight binary dump of the data, with the saved pointer to
2057              the character tables forced to be NULL. The study data, if any, is
2058              written too. After writing, pcretest reads a new pattern.
2059    
2060        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2061              compiled pattern from the given file. There must not be any
2062              occurrences of "<" in the file name (pretty unlikely); if there are,
2063              pcretest will instead treat the initial "<" as a pattern delimiter.
2064              After reading in the pattern, pcretest goes on to read data lines as
2065              usual.
2066    
2067        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2068              and 16-bit fields in a compiled pattern, to simulate a pattern that
2069              was compiled on a host of opposite endianness.
2070    
2071    21. The pcre-exec() function can now cope with patterns that were compiled on
2072        hosts of opposite endianness, with this restriction:
2073    
2074          As for any compiled expression that is saved and used later, the tables
2075          pointer field cannot be preserved; the extra_data field in the arguments
2076          to pcre_exec() should be used to pass in a tables address if a value
2077          other than the default internal tables were used at compile time.
2078    
2079    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2080        now diagnosed as an error. Previously, most of the time, a negative number
2081        would have been treated as zero, but if in addition "ovector" was passed as
2082        NULL, a crash could occur.
2083    
2084    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2085        new versions from the libtool 1.5 distribution (the last one is a copy of
2086        a file called libtool.m4). This seems to have fixed the need to patch
2087        "configure" to support Darwin 1.3 (which I used to do). However, I still
2088        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2089        workstation).
2090    
2091    24. Changed the PCRE licence to be the more standard "BSD" licence.
2092    
2093    
2094    Version 4.5 01-Dec-03
2095    ---------------------
2096    
2097     1. There has been some re-arrangement of the code for the match() function so
2098        that it can be compiled in a version that does not call itself recursively.
2099        Instead, it keeps those local variables that need separate instances for
2100        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2101        needs to "recurse". Keeping track of where control must go is done by means
2102        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2103        hide most of the details from the main code, and operates only if
2104        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2105        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2106        operating.
2107    
2108        To make it easier for callers to provide specially tailored get/free
2109        functions for this usage, two new functions, pcre_stack_malloc, and
2110        pcre_stack_free, are used. They are always called in strict stacking order,
2111        and the size of block requested is always the same.
2112    
2113        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2114        PCRE has been compiled to use the stack or the heap for recursion. The
2115        -C option of pcretest uses this to show which version is compiled.
2116    
2117        A new data escape \S, is added to pcretest; it causes the amounts of store
2118        obtained and freed by both kinds of malloc/free at match time to be added
2119        to the output.
2120    
2121     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2122        what's available on my current Linux desktop machine.
2123    
2124     3. When matching a UTF-8 string, the test for a valid string at the start has
2125        been extended. If start_offset is not zero, PCRE now checks that it points
2126        to a byte that is the start of a UTF-8 character. If not, it returns
2127        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2128        this is necessary because there may be backward assertions in the pattern.
2129        When matching the same subject several times, it may save resources to use
2130        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2131    
2132     4. The code for checking the validity of UTF-8 strings has been tightened so
2133        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2134        containing "overlong sequences".
2135    
2136     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2137        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2138        should have been "&", but it just so happened that all the cases this let
2139        through by mistake were picked up later in the function.
2140    
2141     6. I had used a variable called "isblank" - this is a C99 function, causing
2142        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2143    
2144     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2145        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2146        so the version is shown; (c) stop "make test" echoing "./RunTest".
2147    
2148     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2149    
2150     9. The prototype for memmove() for systems that don't have it was using
2151        size_t, but the inclusion of the header that defines size_t was later. I've
2152        moved the #includes for the C headers earlier to avoid this.
2153    
2154    10. Added some adjustments to the code to make it easier to compiler on certain
2155        special systems:
2156    
2157          (a) Some "const" qualifiers were missing.
2158          (b) Added the macro EXPORT before all exported functions; by default this
2159              is defined to be empty.
2160          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2161              that it reads its output file name as an argument instead of writing
2162              to the standard output and assuming this can be redirected.
2163    
2164    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2165        class containing characters with values greater than 255, PCRE compilation
2166        went into a loop.
2167    
2168    12. A recursive reference to a subpattern that was within another subpattern
2169        that had a minimum quantifier of zero caused PCRE to crash. For example,
2170        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2171        recursion. If the recursively-called subpattern itself had a zero repeat,
2172        that was OK.
2173    
2174    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2175        buffer into which it was copied (for escape processing) was still set at
2176        1024, so long lines caused crashes.
2177    
2178    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2179        "internal error: code overflow...". This applied to any character class
2180        that was followed by a possessive quantifier.
2181    
2182    15. Modified the Makefile to add libpcre.la as a prerequisite for
2183        libpcreposix.la because I was told this is needed for a parallel build to
2184        work.
2185    
2186    16. If a pattern that contained .* following optional items at the start was
2187        studied, the wrong optimizing data was generated, leading to matching
2188        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2189        matching string must start with a or b or c. The correct conclusion for
2190        this pattern is that a match can start with any character.
2191    
2192    
2193    Version 4.4 13-Aug-03
2194    ---------------------
2195    
2196     1. In UTF-8 mode, a character class containing characters with values between
2197        127 and 255 was not handled correctly if the compiled pattern was studied.
2198        In fixing this, I have also improved the studying algorithm for such
2199        classes (slightly).
2200    
2201     2. Three internal functions had redundant arguments passed to them. Removal
2202        might give a very teeny performance improvement.
2203    
2204     3. Documentation bug: the value of the capture_top field in a callout is *one
2205        more than* the number of the hightest numbered captured substring.
2206    
2207     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2208        in incorrectly linking with a previously installed version. They now link
2209        explicitly with libpcre.la.
2210    
2211     5. configure.in no longer needs to recognize Cygwin specially.
2212    
2213     6. A problem in pcre.in for Windows platforms is fixed.
2214    
2215     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2216        pcretest, it used to include the size of the study block as part of its
2217        output. Unfortunately, the structure contains a field that has a different
2218        size on different hardware architectures. This meant that the tests that
2219        showed this size failed. As the block is currently always of a fixed size,
2220        this information isn't actually particularly useful in pcretest output, so
2221        I have just removed it.
2222    
2223     8. Three pre-processor statements accidentally did not start in column 1.
2224        Sadly, there are *still* compilers around that complain, even though
2225        standard C has not required this for well over a decade. Sigh.
2226    
2227     9. In pcretest, the code for checking callouts passed small integers in the
2228        callout_data field, which is a void * field. However, some picky compilers
2229        complained about the casts involved for this on 64-bit systems. Now
2230        pcretest passes the address of the small integer instead, which should get
2231        rid of the warnings.
2232    
2233    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2234        both compile and run time, and gives an error if an invalid UTF-8 sequence
2235        is found. There is a option for disabling this check in cases where the
2236        string is known to be correct and/or the maximum performance is wanted.
2237    
2238    11. In response to a bug report, I changed one line in Makefile.in from
2239    
2240            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2241        to
2242            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2243    
2244        to look similar to other lines, but I have no way of telling whether this
2245        is the right thing to do, as I do not use Windows. No doubt I'll get told
2246        if it's wrong...
2247    
2248    
2249    Version 4.3 21-May-03
2250    ---------------------
2251    
2252    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2253       Makefile.
2254    
2255    2. Some refactoring to improve the quality of the code:
2256    
2257       (i)   The utf8_table... variables are now declared "const".
2258    
2259       (ii)  The code for \cx, which used the "case flipping" table to upper case
2260             lower case letters, now just substracts 32. This is ASCII-specific,
2261             but the whole concept of \cx is ASCII-specific, so it seems
2262             reasonable.
2263    
2264       (iii) PCRE was using its character types table to recognize decimal and
2265             hexadecimal digits in the pattern. This is silly, because it handles
2266             only 0-9, a-f, and A-F, but the character types table is locale-
2267             specific, which means strange things might happen. A private
2268             table is now used for this - though it costs 256 bytes, a table is
2269             much faster than multiple explicit tests. Of course, the standard
2270             character types table is still used for matching digits in subject
2271             strings against \d.
2272    
2273       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2274             ending in _t are). So I've renamed it as ESC_tee.
2275    
2276    3. The first argument for regexec() in the POSIX wrapper should have been
2277       defined as "const".
2278    
2279    4. Changed pcretest to use malloc() for its buffers so that they can be
2280       Electric Fenced for debugging.
2281    
2282    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2283       to read one or more bytes before the start of the subject string. Often this
2284       had no effect on PCRE's behaviour, but in some circumstances it could
2285       provoke a segmentation fault.
2286    
2287    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2288       to try to read one or more bytes before the start of the subject string.
2289    
2290    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2291       UTF-8 support could misbehave in various ways if the subject string
2292       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2293       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2294       back over UTF-8 characters.)
2295    
2296    
2297    Version 4.2 14-Apr-03
2298    ---------------------
2299    
2300    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2301    
2302    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2303         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2304         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2305         [WIN_PREFIX]: new variable, "cyg" for Cygwin
2306         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2307           and BUILD_EXEEXT
2308         Note: automatic setting of the BUILD variables is not yet working
2309         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2310           compile-time but not at link-time
2311         [LINK]: use for linking executables only
2312         make different versions for Windows and non-Windows
2313         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2314           libraries
2315         [LINK_FOR_BUILD]: new variable
2316         [OBJEXT]: use throughout
2317         [EXEEXT]: use throughout
2318         <winshared>: new target
2319         <wininstall>: new target
2320         <dftables.o>: use native compiler
2321         <dftables>: use native linker
2322         <install>: handle Windows platform correctly
2323         <clean>: ditto
2324         <check>: ditto
2325         copy DLL to top builddir before testing
2326    
2327       As part of these changes, -no-undefined was removed again. This was reported
2328       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2329       in any case.
2330    
2331    3. Some tidies to get rid of compiler warnings:
2332    
2333       . In the match_data structure, match_limit was an unsigned long int, whereas
2334         match_call_count was an int. I've made them both unsigned long ints.
2335    
2336       . In pcretest the fact that a const uschar * doesn't automatically cast to
2337         a void * provoked a warning.
2338    
2339       . Turning on some more compiler warnings threw up some "shadow" variables
2340         and a few more missing casts.
2341    
2342    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2343       option, a class that contained a single character with a value between 128
2344       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2345    
2346    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2347       option, a class that contained several characters, but with at least one
2348       whose value was between 128 and 255 caused PCRE to crash.
2349    
2350    
2351    Version 4.1 12-Mar-03
2352    ---------------------
2353    
2354    1. Compiling with gcc -pedantic found a couple of places where casts were
2355    needed, and a string in dftables.c that was longer than standard compilers are
2356    required to support.
2357    
2358    2. Compiling with Sun's compiler found a few more places where the code could
2359    be tidied up in order to avoid warnings.
2360    
2361    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2362    first of these names is deprecated in the latest Autoconf in favour of the name
2363    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2364    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2365    analogy I have changed it to CFLAGS_FOR_BUILD.
2366    
2367    4. Added -no-undefined to the linking command in the Makefile, because this is
2368    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2369    linking step for the pcreposix library.
2370    
2371    5. PCRE was failing to diagnose the case of two named groups with the same
2372    name.
2373    
2374    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2375    literal character that is needed in the subject for a match, and scans along to
2376    ensure that it is present before embarking on the full matching process. This
2377    saves time in cases of nested unlimited repeats that are never going to match.
2378    Problem: the scan can take a lot of time if the subject is very long (e.g.
2379    megabytes), thus penalizing straightforward matches. It is now done only if the
2380    amount of subject to be scanned is less than 1000 bytes.
2381    
2382    7. A lesser problem with the same optimization is that it was recording the
2383    first character of an anchored pattern as "needed", thus provoking a search
2384    right along the subject, even when the first match of the pattern was going to
2385    fail. The "needed" character is now not set for anchored patterns, unless it
2386    follows something in the pattern that is of non-fixed length. Thus, it still
2387    fulfils its original purpose of finding quick non-matches in cases of nested
2388    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2389    
2390    
2391    Version 4.0 17-Feb-03
2392    ---------------------
2393    
2394    1. If a comment in an extended regex that started immediately after a meta-item
2395    extended to the end of string, PCRE compiled incorrect data. This could lead to
2396    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2397    
2398    2. Moved to autoconf 2.53 and libtool 1.4.2.
2399    
2400    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2401    the special perltest8 script is no longer needed - all the tests can be run
2402    from a single perltest script.
2403    
2404    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2405    by \s. It has now been removed in PCRE. This means it isn't recognized as
2406    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2407    class [:space:] *does* include VT, thereby creating a mess.
2408    
2409    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2410    space and tab.
2411    
2412    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2413    its new features into the main test script, reducing the number of scripts.
2414    
2415    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2416    were backward compatible, and made the (?i) apply to the whole pattern, as if
2417    /i were given. Now it behaves more logically, and applies the option setting
2418    only to what follows. PCRE has been changed to follow suit. However, if it
2419    finds options settings right at the start of the pattern, it extracts them into
2420    the global options, as before. Thus, they show up in the info data.
2421    
2422    8. Added support for the \Q...\E escape sequence. Characters in between are
2423    treated as literals. This is slightly different from Perl in that $ and @ are
2424    also handled as literals inside the quotes. In Perl, they will cause variable
2425    interpolation. Note the following examples:
2426    
2427        Pattern            PCRE matches      Perl matches
2428    
2429        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
2430        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2431        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2432    
2433    For compatibility with Perl, \Q...\E sequences are recognized inside character
2434    classes as well as outside them.
2435    
2436    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2437    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2438    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2439    signed/unsigned warnings.
2440    
2441    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2442    option for pcretest, so I've replaced it by a simple function that does just
2443    that job.
2444    
2445    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2446    "pcregrep -".
2447    
2448    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2449    Java package. This provides some syntactic sugar for simple cases of what my
2450    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2451    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2452    item, you can use this simplified notation. Note that only makes sense with
2453    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2454    greediness, whatever the setting of the PCRE_UNGREEDY option.
2455    
2456    13. A change of greediness default within a pattern was not taking effect at
2457    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2458    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2459    was abstracted outside.
2460    
2461    14. PCRE now supports the \G assertion. It is true when the current matching
2462    position is at the start point of the match. This differs from \A when the
2463    starting offset is non-zero. Used with the /g option of pcretest (or similar
2464    code), it works in the same way as it does for Perl's /g option. If all
2465    alternatives of a regex begin with \G, the expression is anchored to the start
2466    match position, and the "anchored" flag is set in the compiled expression.
2467    
2468    15. Some bugs concerning the handling of certain option changes within patterns
2469    have been fixed. These applied to options other than (?ims). For example,
2470    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2471    been the other way round. Some of this was related to change 7 above.
2472    
2473    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2474    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2475    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2476    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2477    
2478    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2479    mode. Unlike ".", it always matches newline, whatever the setting of
2480    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2481    assertions. Perl allows it, but it doesn't (in general) work because it can't
2482    calculate the length of the lookbehind. At least, that's the case for Perl
2483    5.8.0 - I've been told they are going to document that it doesn't work in
2484    future.
2485    
2486    18. Added an error diagnosis for escapes that PCRE does not support: these are
2487    \L, \l, \N, \P, \p, \U, \u, and \X.
2488    
2489    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2490    reading past the end of the pattern in cases such as /[abcd/.
2491    
2492    20. PCRE was getting more memory than necessary for patterns with classes that
2493    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2494    
2495    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2496    compiling PCRE for use with Virtual Pascal.
2497    
2498    22. Small fix to the Makefile to make it work properly if the build is done
2499    outside the source tree.
2500    
2501    23. Added a new extension: a condition to go with recursion. If a conditional
2502    subpattern starts with (?(R) the "true" branch is used if recursion has
2503    happened, whereas the "false" branch is used only at the top level.
2504    
2505    24. When there was a very long string of literal characters (over 255 bytes
2506    without UTF support, over 250 bytes with UTF support), the computation of how
2507    much memory was required could be incorrect, leading to segfaults or other
2508    strange effects.
2509    
2510    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2511    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2512    there was a subsequent back reference to those brackets. This meant that, for
2513    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2514    possible to check for precisely this case. All we can do is abandon the
2515    optimization if .* occurs inside capturing brackets when there are any back
2516    references whatsoever. (See below for a better fix that came later.)
2517    
2518    26. The handling of the optimization for finding the first character of a
2519    non-anchored pattern, and for finding a character that is required later in the
2520    match were failing in some cases. This didn't break the matching; it just
2521    failed to optimize when it could. The way this is done has been re-implemented.
2522    
2523    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2524    
2525    28. Added a new feature that provides some of the functionality that Perl
2526    provides with (?{...}). The facility is termed a "callout". The way it is done
2527    in PCRE is for the caller to provide an optional function, by setting
2528    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2529    global variable. By default it is unset, which disables all calling out. To get
2530    the function called, the regex must include (?C) at appropriate points. This
2531    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2532    This provides a means of identifying different callout points. When PCRE
2533    reaches such a point in the regex, if pcre_callout has been set, the external
2534    function is called. It is provided with data in a structure called
2535    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2536    matching continues; if it returns a non-zero value, the match at the current
2537    point fails. However, backtracking will occur if possible. [This was changed
2538    later and other features added - see item 49 below.]
2539    
2540    29. pcretest is upgraded to test the callout functionality. It provides a
2541    callout function that displays information. By default, it shows the start of
2542    the match and the current position in the text. There are some new data escapes
2543    to vary what happens:
2544    
2545        \C+         in addition, show current contents of captured substrings
2546        \C-         do not supply a callout function
2547        \C!n        return 1 when callout number n is reached
2548        \C!n!m      return 1 when callout number n is reached for the mth time
2549    
2550    30. If pcregrep was called with the -l option and just a single file name, it
2551    output "<stdin>" if a match was found, instead of the file name.
2552    
2553    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2554    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2555    pcre_exec(). This saves a malloc/free per call. The default value of
2556    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2557    when configuring.
2558    
2559    32. The default maximum size of a compiled pattern is 64K. There have been a
2560    few cases of people hitting this limit. The code now uses macros to handle the
2561    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2562    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2563    configuring. Tests 2 and 5 work only with 2-byte links because they output
2564    debugging information about compiled patterns.
2565    
2566    33. Internal code re-arrangements:
2567    
2568    (a) Moved the debugging function for printing out a compiled regex into
2569        its own source file (printint.c) and used #include to pull it into
2570        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2571        separate copies.
2572    
2573    (b) Defined the list of op-code names for debugging as a macro in
2574        internal.h so that it is next to the definition of the opcodes.
2575    
2576    (c) Defined a table of op-code lengths for simpler skipping along compiled
2577        code. This is again a macro in internal.h so that it is next to the
2578        definition of the opcodes.
2579    
2580    34. Added support for recursive calls to individual subpatterns, along the
2581    lines of Robin Houston's patch (but implemented somewhat differently).
2582    
2583    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2584    allow it to read and process whole directories in Win32. This code was
2585    contributed by Lionel Fourquaux; it has not been tested by me.
2586    
2587    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2588    used to name a group. Names consist of alphanumerics and underscores, and must
2589    be unique. Back references use the syntax (?P=name) and recursive calls use
2590    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2591    numbers. The function pcre_fullinfo() can be used after compilation to extract
2592    a name/number map. There are three relevant calls:
2593    
2594      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2595      PCRE_INFO_NAMECOUNT            yields the number of entries
2596      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2597    
2598    The map is a vector of fixed-size entries. The size of each entry depends on
2599    the length of the longest name used. The first two bytes of each entry are the
2600    group number, most significant byte first. There follows the corresponding
2601    name, zero terminated. The names are in alphabetical order.
2602    
2603    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2604    case instead of 255. Making it the same both with and without UTF-8 support
2605    means that the same test output works with both.
2606    
2607    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2608    calling malloc() with a zero argument.
2609    
2610    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2611    optimization. I've improved things by keeping a bitmap of backreferences with
2612    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2613    fact referenced, the optimization can be applied. It is unlikely that a
2614    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2615    the match to follow \n) will appear inside brackets with a number greater than
2616    31, but if it does, any back reference > 31 suppresses the optimization.
2617    
2618    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2619    of disabling numbered capturing parentheses. Any opening parenthesis that is
2620    not followed by ? behaves as if it were followed by ?: but named parentheses
2621    can still be used for capturing (and they will acquire numbers in the usual
2622    way).
2623    
2624    41. Redesigned the return codes from the match() function into yes/no/error so
2625    that errors can be passed back from deep inside the nested calls. A malloc
2626    failure while inside a recursive subpattern call now causes the
2627    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2628    
2629    42. It is now possible to set a limit on the number of times the match()
2630    function is called in a call to pcre_exec(). This facility makes it possible to
2631    limit the amount of recursion and backtracking, though not in a directly
2632    obvious way, because the match() function is used in a number of different
2633    circumstances. The count starts from zero for each position in the subject
2634    string (for non-anchored patterns). The default limit is, for compatibility, a
2635    large number, namely 10 000 000. You can change this in two ways:
2636    
2637    (a) When configuring PCRE before making, you can use --with-match-limit=n
2638        to set a default value for the compiled library.
2639    
2640    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2641        a different value is set. See 45 below.
2642    
2643    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2644    
2645    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2646    of things that can be changed at compile time. The first argument specifies
2647    what is wanted and the second points to where the information is to be placed.
2648    The current list of available information is:
2649    
2650      PCRE_CONFIG_UTF8
2651    
2652    The output is an integer that is set to one if UTF-8 support is available;
2653    otherwise it is set to zero.
2654    
2655      PCRE_CONFIG_NEWLINE
2656    
2657    The output is an integer that it set to the value of the code that is used for
2658    newline. It is either LF (10) or CR (13).
2659    
2660      PCRE_CONFIG_LINK_SIZE
2661    
2662    The output is an integer that contains the number of bytes used for internal
2663    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2664    
2665      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2666    
2667    The output is an integer that contains the threshold above which the POSIX
2668    interface uses malloc() for output vectors. See item 31 above.
2669    
2670      PCRE_CONFIG_MATCH_LIMIT
2671    
2672    The output is an unsigned integer that contains the default limit of the number
2673    of match() calls in a pcre_exec() execution. See 42 above.
2674    
2675    44. pcretest has been upgraded by the addition of the -C option. This causes it
2676    to extract all the available output from the new pcre_config() function, and to
2677    output it. The program then exits immediately.
2678    
2679    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2680    order to support additional features. One way would have been to define
2681    pcre_exec2() (for example) with extra arguments, but this would not have been
2682    extensible, and would also have required all calls to the original function to
2683    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2684    is used for passing in "extra" data from pcre_study().
2685    
2686    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2687    contains the following fields:
2688    
2689      flags         a bitmap indicating which of the following fields are set
2690      study_data    opaque data from pcre_study()
2691      match_limit   a way of specifying a limit on match() calls for a specific
2692                      call to pcre_exec()
2693      callout_data  data for callouts (see 49 below)
2694    
2695    The flag bits are also defined in pcre.h, and are
2696    
2697      PCRE_EXTRA_STUDY_DATA
2698      PCRE_EXTRA_MATCH_LIMIT
2699      PCRE_EXTRA_CALLOUT_DATA
2700    
2701    The pcre_study() function now returns one of these new pcre_extra blocks, with
2702    the actual study data pointed to by the study_data field, and the
2703    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2704    before. That is, this change is entirely upwards-compatible and requires no
2705    change to existing code.
2706    
2707    If you want to pass in additional data to pcre_exec(), you can either place it
2708    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2709    block.
2710    
2711    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2712    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2713    times with different match limits, until it finds the minimum value needed for
2714    pcre_exec() to complete. The value is then output. This can be instructive; for
2715    most simple matches the number is quite small, but for pathological cases it
2716    gets very large very quickly.
2717    
2718    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2719    returns the size of the data block pointed to by the study_data field in a
2720    pcre_extra block, that is, the value that was passed as the argument to
2721    pcre_malloc() when PCRE was getting memory in which to place the information
2722    created by pcre_study(). The fourth argument should point to a size_t variable.
2723    pcretest has been extended so that this information is shown after a successful
2724    pcre_study() call when information about the compiled regex is being displayed.
2725    
2726    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2727    because what follows is always an absolute path. (Later: it turns out that this
2728    is more than cosmetic for MinGW, because it doesn't like empty path
2729    components.)
2730    
2731    49. Some changes have been made to the callout feature (see 28 above):
2732    
2733    (i)  A callout function now has three choices for what it returns:
2734    
2735           0  =>  success, carry on matching
2736         > 0  =>  failure at this point, but backtrack if possible
2737         < 0  =>  serious error, return this value from pcre_exec()
2738    
2739         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2740         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2741         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2742         use by callout functions. It will never be used by PCRE itself.
2743    
2744    (ii) The pcre_extra structure (see 45 above) has a void * field called
2745         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2746         pcre_callout_block structure has a field of the same name. The contents of
2747         the field passed in the pcre_extra structure are passed to the callout
2748         function in the corresponding field in the callout block. This makes it
2749         easier to use the same callout-containing regex from multiple threads. For
2750         testing, the pcretest program has a new data escape
2751    
2752           \C*n        pass the number n (may be negative) as callout_data
2753    
2754         If the callout function in pcretest receives a non-zero value as
2755         callout_data, it returns that value.
2756    
2757    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2758    there were some redundant $(CFLAGS) in commands that are now specified as
2759    $(LINK), which already includes $(CFLAGS).
2760    
2761    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2762    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2763    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2764    one-byte characters throughout. Note that case-insensitive matching applies
2765    only to characters whose values are less than 256. PCRE doesn't support the
2766    notion of cases for higher-valued characters.
2767    
2768    (i)   A character class whose characters are all within 0-255 is handled as
2769          a bit map, and the map is inverted for negative classes. Previously, a
2770          character > 255 always failed to match such a class; however it should
2771          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2772    
2773    (ii)  A negated character class with a single character < 255 is coded as
2774          "not this character" (OP_NOT). This wasn't working properly when the test
2775          character was multibyte, either singly or repeated.
2776    
2777    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2778          mode, for example: \x{100}{2,3}.
2779    
2780    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2781          singly or repeated) now correctly test multibyte characters. However,
2782          PCRE doesn't recognize any characters with values greater than 255 as
2783          digits, spaces, or word characters. Such characters always match \D, \S,
2784          and \W, and never match \d, \s, or \w.
2785    
2786    (v)   Classes may now contain characters and character ranges with values
2787          greater than 255. For example: [ab\x{100}-\x{400}].
2788    
2789    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2790          PCRE in UTF-8 mode.
2791    
2792    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2793    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2794    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2795    value.)
2796    
2797    53. The single man page has become too large. I have therefore split it up into
2798    a number of separate man pages. These also give rise to individual HTML pages;
2799    these are now put in a separate directory, and there is an index.html page that
2800    lists them all. Some hyperlinking between the pages has been installed.
2801    
2802    54. Added convenience functions for handling named capturing parentheses.
2803    
2804    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2805    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2806    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2807    are faulted.
2808    
2809    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2810    calling configure. These values are used when compiling the dftables.c program
2811    which is run to generate the source of the default character tables. They
2812    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2813    you will need to set these values.
2814    
2815    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2816    
2817    
2818    Version 3.9 02-Jan-02
2819    ---------------------
2820    
2821    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2822    
2823    2. If --disable-static was given, the building process failed when trying to
2824    build pcretest and pcregrep. (For some reason it was using libtool to compile
2825    them, which is not right, as they aren't part of the library.)
2826    
2827    
2828    Version 3.8 18-Dec-01
2829    ---------------------
2830    
2831    1. The experimental UTF-8 code was completely screwed up. It was packing the
2832    bytes in the wrong order. How dumb can you get?
2833    
2834    
2835    Version 3.7 29-Oct-01
2836    ---------------------
2837    
2838    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2839    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2840    this didn't happen under Solaris 8, where I normally test things.
2841    
2842    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2843    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2844    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2845    this fix an hour or so after the initial 3.7 release.)
2846    
2847    
2848    Version 3.6 23-Oct-01
2849    ---------------------
2850    
2851    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2852    offsets passed as NULL with zero offset count.
2853    
2854    2. The config.guess and config.sub files had not been updated when I moved to
2855    the latest autoconf.
2856    
2857    
2858    Version 3.5 15-Aug-01
2859    ---------------------
2860    
2861    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2862    had been forgotten.
2863    
2864    2. By using declared but undefined structures, we can avoid using "void"
2865    definitions in pcre.h while keeping the internal definitions of the structures
2866    private.
2867    
2868    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2869    user point of view, this means that both static and shared libraries are built
2870    by default, but this can be individually controlled. More of the work of
2871    handling this static/shared cases is now inside libtool instead of PCRE's make
2872    file.
2873    
2874    4. The pcretest utility is now installed along with pcregrep because it is
2875    useful for users (to test regexs) and by doing this, it automatically gets
2876    relinked by libtool. The documentation has been turned into a man page, so
2877    there are now .1, .txt, and .html versions in /doc.
2878    
2879    5. Upgrades to pcregrep:
2880       (i)   Added long-form option names like gnu grep.
2881       (ii)  Added --help to list all options with an explanatory phrase.
2882       (iii) Added -r, --recursive to recurse into sub-directories.
2883       (iv)  Added -f, --file to read patterns from a file.
2884    
2885    6. pcre_exec() was referring to its "code" argument before testing that
2886    argument for NULL (and giving an error if it was NULL).
2887    
2888    7. Upgraded Makefile.in to allow for compiling in a different directory from
2889    the source directory.
2890    
2891    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2892    options bits, the pointer it was passed was to an int instead of to an unsigned
2893    long int. This mattered only on 64-bit systems.
2894    
2895    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2896    generated) instead of pcre.in, which it its source. Also made the same change
2897    in several of the .c files.
2898    
2899    10. A new release of gcc defines printf() as a macro, which broke pcretest
2900    because it had an ifdef in the middle of a string argument for printf(). Fixed
2901    by using separate calls to printf().
2902    
2903    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2904    script, to force use of CR or LF instead of \n in the source. On non-Unix
2905    systems, the value can be set in config.h.
2906    
2907    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2908    absolute limit. Changed the text of the error message to make this clear, and
2909    likewise updated the man page.
2910    
2911    13. The limit of 99 on the number of capturing subpatterns has been removed.
2912    The new limit is 65535, which I hope will not be a "real" limit.
2913    
2914    
2915    Version 3.4 22-Aug-00
2916    ---------------------
2917    
2918    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2919    
2920    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2921    
2922    
2923    Version 3.3 01-Aug-00
2924    ---------------------
2925    
2926    1. If an octal character was given, but the value was greater than \377, it
2927    was not getting masked to the least significant bits, as documented. This could
2928    lead to crashes in some systems.
2929    
2930    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2931    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2932    
2933    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2934    These just pass their arguments on to (pcre_free)(), but they are provided
2935    because some uses of PCRE bind it to non-C systems that can call its functions,
2936    but cannot call free() or pcre_free() directly.
2937    
2938    4. Add "make test" as a synonym for "make check". Corrected some comments in
2939    the Makefile.
2940    
2941    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2942    Makefile.
2943    
2944    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2945    command called pgrep for grepping around the active processes.
2946    
2947    7. Added the beginnings of support for UTF-8 character strings.
2948    
2949    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2950    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2951    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2952    out for the ar command.)
2953    
2954    
2955    Version 3.2 12-May-00
2956    ---------------------
2957    
2958    This is purely a bug fixing release.
2959    
2960    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2961    of ZA. This was just one example of several cases that could provoke this bug,
2962    which was introduced by change 9 of version 2.00. The code for breaking
2963    infinite loops after an iteration that matches an empty string was't working
2964    correctly.
2965    
2966    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2967    when matched against abbab (for example). After matching an empty string, it
2968    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2969    caused it to match further down the string than it should.
2970    
2971    3. The code contained an inclusion of sys/types.h. It isn't clear why this
2972    was there because it doesn't seem to be needed, and it causes trouble on some
2973    systems, as it is not a Standard C header. It has been removed.
2974    
2975    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2976    were reported on the Macintosh. The changes were from
2977    
2978      while ((c = *(++ptr)) != 0 && c != '\n');
2979    to
2980      while ((c = *(++ptr)) != 0 && c != '\n') ;
2981    
2982    Totally extraordinary, but if that's what it takes...
2983    
2984    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2985    available. Added HAVE_BCOPY and an autoconf test for it; if neither
2986    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2987    assumes the way PCRE uses memmove() (always moving upwards).
2988    
2989    6. PCRE is being used in one environment where strchr() is not available. There
2990    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2991    faster code anyway.
2992    
2993    
2994    Version 3.1 09-Feb-00
2995    ---------------------
2996    
2997    The only change in this release is the fixing of some bugs in Makefile.in for
2998    the "install" target:
2999    
3000    (1) It was failing to install pcreposix.h.
3001    
3002    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3003    
3004    
3005    Version 3.0 01-Feb-00
3006    ---------------------
3007    
3008    1. Add support for the /+ modifier to perltest (to output $` like it does in
3009    pcretest).
3010    
3011    2. Add support for the /g modifier to perltest.
3012    
3013    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3014    matches null strings.
3015    
3016    4. Fix perltest so that it doesn't do unwanted things when fed an empty
3017    pattern. Perl treats empty patterns specially - it reuses the most recent
3018    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3019    effect.
3020    
3021    5. The POSIX interface was broken in that it was just handing over the POSIX
3022    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3023    required a bigger vector, with some working space on the end. This means that
3024    the POSIX wrapper now has to get and free some memory, and copy the results.
3025    
3026    6. Added some simple autoconf support, placing the test data and the
3027    documentation in separate directories, re-organizing some of the
3028    information files, and making it build pcre-config (a GNU standard). Also added
3029    libtool support for building PCRE as a shared library, which is now the
3030    default.
3031    
3032    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3033    09 are not valid octal constants. Single digits will be used for minor values
3034    less than 10.
3035    
3036    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3037    existing programs that set these in the POSIX interface can use PCRE without
3038    modification.
3039    
3040    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3041    return all that pcre_info() returns, plus additional data. The pcre_info()
3042    function is retained for compatibility, but is considered to be obsolete.
3043    
3044    10. Added experimental recursion feature (?R) to handle one common case that
3045    Perl 5.6 will be able to do with (?p{...}).
3046    
3047    11. Added support for POSIX character classes like [:alpha:], which Perl is
3048    adopting.
3049    
3050    
3051    Version 2.08 31-Aug-99
3052    ----------------------
3053    
3054    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3055    trying to match at the startoffset position, but instead was moving forward to
3056    the next newline as if a previous match had failed.
3057    
3058    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3059    and could get into a loop if a null string was matched other than at the start
3060    of the subject.
3061    
3062    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3063    be distinguished at compile time, and for completeness also added PCRE_DATE.
3064    
3065    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3066    in GnuWin32 environments.
3067    
3068    
3069    Version 2.07 29-Jul-99
3070    ----------------------
3071    
3072    1. The documentation is now supplied in plain text form and HTML as well as in
3073    the form of man page sources.
3074    
3075    2. C++ compilers don't like assigning (void *) values to other pointer types.
3076    In particular this affects malloc(). Although there is no problem in Standard
3077    C, I've put in casts to keep C++ compilers happy.
3078    
3079    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3080    should be (const char *).
3081    
3082    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3083    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3084    However, I haven't made this a standard facility. The documentation doesn't
3085    mention it, and the Makefile doesn't support it.
3086    
3087    5. The Makefile now contains an "install" target, with editable destinations at
3088    the top of the file. The pcretest program is not installed.
3089    
3090    6. pgrep -V now gives the PCRE version number and date.
3091    
3092    7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3093    causing the entire string to be ignored, instead of just the last character.
3094    
3095    8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3096    non-matching string, it can take a very, very long time, even for strings of
3097    quite modest length, because of the nested recursion. PCRE now does better in
3098    some of these cases. It does this by remembering the last required literal
3099    character in the pattern, and pre-searching the subject to ensure it is present
3100    before running the real match. In other words, it applies a heuristic to detect
3101    some types of certain failure quickly, and in the above example, if presented
3102    with a string that has no trailing " it gives "no match" very quickly.
3103    
3104    9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3105    other alternatives are tried instead.
3106    
3107    
3108    Version 2.06 09-Jun-99
3109    ----------------------
3110    
3111    1. Change pcretest's output for amount of store used to show just the code
3112    space, because the remainder (the data block) varies in size between 32-bit and
3113    64-bit systems.
3114    
3115    2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3116    start matching at. This allows lookbehinds to work when searching for multiple
3117    occurrences in a string.
3118    
3119    3. Added additional options to pcretest for testing multiple occurrences:
3120    
3121       /+   outputs the rest of the string that follows a match
3122       /g   loops for multiple occurrences, using the new startoffset argument
3123       /G   loops for multiple occurrences by passing an incremented pointer
3124    
3125    4. PCRE wasn't doing the "first character" optimization for patterns starting
3126    with \b or \B, though it was doing it for other lookbehind assertions. That is,
3127    it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3128    the letter 'x'. On long subject strings, this gives a significant speed-up.
3129    
3130    
3131    Version 2.05 21-Apr-99
3132    ----------------------
3133    
3134    1. Changed the type of magic_number from int to long int so that it works
3135    properly on 16-bit systems.
3136    
3137    2. Fixed a bug which caused patterns starting with .* not to work correctly
3138    when the subject string contained newline characters. PCRE was assuming
3139    anchoring for such patterns in all cases, which is not correct because .* will
3140    not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3141    DOTALL is set at top level; otherwise it knows that patterns starting with .*
3142    must be retried after every newline in the subject.
3143    
3144    
3145    Version 2.04 18-Feb-99
3146    ----------------------
3147    
3148    1. For parenthesized subpatterns with repeats whose minimum was zero, the
3149    computation of the store needed to hold the pattern was incorrect (too large).
3150    If such patterns were nested a few deep, this could multiply and become a real
3151    problem.
3152    
3153    2. Added /M option to pcretest to show the memory requirement of a specific
3154    pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3155    
3156    3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3157    compiled in such a way that the backtracking after subsequent failure was
3158    pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3159    ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3160    
3161    
3162    Version 2.03 02-Feb-99
3163    ----------------------
3164    
3165    1. Fixed typo and small mistake in man page.
3166    
3167    2. Added 4th condition (GPL supersedes if conflict) and created separate
3168    LICENCE file containing the conditions.
3169    
3170    3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3171    Perl, that is the internal \ allows the delimiter to be included in the
3172    pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3173    the final delimiter, add \ to the end of the pattern (to test the error).
3174    
3175    4. Added the convenience functions for extracting substrings after a successful
3176    match. Updated pcretest to make it able to test these functions.
3177    
3178    
3179    Version 2.02 14-Jan-99
3180    ----------------------
3181    
3182    1. Initialized the working variables associated with each extraction so that
3183    their saving and restoring doesn't refer to uninitialized store.
3184    
3185    2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3186    compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3187    fix the problem.
3188    
3189    3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3190    calls, and wasn't printing the correct value for compiling calls. Increased the
3191    default value of LOOPREPEAT, and the number of significant figures in the
3192    times.
3193    
3194    4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3195    
3196    5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3197    a building problem on Windows NT with a FAT file system.
3198    
3199    
3200    Version 2.01 21-Oct-98
3201    ----------------------
3202    
3203    1. Changed the API for pcre_compile() to allow for the provision of a pointer
3204    to character tables built by pcre_maketables() in the current locale. If NULL
3205    is passed, the default tables are used.
3206    
3207    
3208    Version 2.00 24-Sep-98
3209    ----------------------
3210    
3211    1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3212    it any more.
3213    
3214    2. Allow quantification of (?>) groups, and make it work correctly.
3215    
3216    3. The first character computation wasn't working for (?>) groups.
3217    
3218    4. Correct the implementation of \Z (it is permitted to match on the \n at the
3219    end of the subject) and add 5.005's \z, which really does match only at the
3220    very end of the subject.
3221    
3222    5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3223    
3224    6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3225    DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3226    localized options. All options to pcre_study() were also removed.
3227    
3228    7. Add other new features from 5.005:
3229    
3230       $(?<=           positive lookbehind
3231       $(?<!           negative lookbehind
3232       (?imsx-imsx)    added the unsetting capability
3233                       such a setting is global if at outer level; local otherwise
3234       (?imsx-imsx:)   non-capturing groups with option setting
3235       (?(cond)re|re)  conditional pattern matching
3236    
3237       A backreference to itself in a repeated group matches the previous
3238       captured string.
3239    
3240    8. General tidying up of studying (both automatic and via "study")
3241    consequential on the addition of new assertions.
3242    
3243    9. As in 5.005, unlimited repeated groups that could match an empty substring
3244    are no longer faulted at compile time. Instead, the loop is forcibly broken at
3245    runtime if any iteration does actually match an empty substring.
3246    
3247    10. Include the RunTest script in the distribution.
3248    
3249    11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3250    discrepancies, some of which were old and were also with respect to 5.004. They
3251    have now been fixed.
3252    
3253    
3254  Version 1.09 28 Apr-98  Version 1.09 28-Apr-98
3255  ----------------------  ----------------------
3256    
3257  1. A negated single character class followed by a quantifier with a minimum  1. A negated single character class followed by a quantifier with a minimum

Legend:
Removed from v.21  
changed lines
  Added in v.427

  ViewVC Help
Powered by ViewVC 1.1.5