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

Diff of /code/trunk/ChangeLog

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

revision 89 by nigel, Sat Feb 24 21:41:27 2007 UTC revision 420 by ph10, Wed Aug 12 17:32:27 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8        was out-of-date, and there was no check on the pcre_compile() error code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    
25    Version 7.9 11-Apr-09
26    ---------------------
27    
28    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
29        (pcretest), all targets were linked against these libraries. This included
30        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
31        libraries. This caused unwanted dependencies to be created. This problem
32        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
33        pcretest is linked with readline.
34    
35    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
36        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
37        moved outside it again, because FALSE and TRUE are already defined in AIX,
38        but BOOL is not.
39    
40    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
41        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
42    
43    4.  The pcregrep documentation said spaces were inserted as well as colons (or
44        hyphens) following file names and line numbers when outputting matching
45        lines. This is not true; no spaces are inserted. I have also clarified the
46        wording for the --colour (or --color) option.
47    
48    5.  In pcregrep, when --colour was used with -o, the list of matching strings
49        was not coloured; this is different to GNU grep, so I have changed it to be
50        the same.
51    
52    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
53        each matching line was coloured. Now it goes on to look for further matches
54        of any of the test patterns, which is the same behaviour as GNU grep.
55    
56    7.  A pattern that could match an empty string could cause pcregrep to loop; it
57        doesn't make sense to accept an empty string match in pcregrep, so I have
58        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
59        seems to be how GNU grep behaves.
60    
61    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
62        start or after a newline", because the conditional assertion was not being
63        correctly handled. The rule now is that both the assertion and what follows
64        in the first alternative must satisfy the test.
65    
66    9.  If auto-callout was enabled in a pattern with a conditional group whose
67        condition was an assertion, PCRE could crash during matching, both with
68        pcre_exec() and pcre_dfa_exec().
69    
70    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
71        used for matching.
72    
73    11. Unicode property support in character classes was not working for
74        characters (bytes) greater than 127 when not in UTF-8 mode.
75    
76    12. Added the -M command line option to pcretest.
77    
78    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
79    
80    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
81    
82    16. Added comments and documentation about mis-use of no_arg in the C++
83        wrapper.
84    
85    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
86        from Martin Jerabek that uses macro names for all relevant character and
87        string constants.
88    
89    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
90        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
91        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
92        these, but not everybody uses configure.
93    
94    19. A conditional group that had only one branch was not being correctly
95        recognized as an item that could match an empty string. This meant that an
96        enclosing group might also not be so recognized, causing infinite looping
97        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
98        with the subject "ab", where knowledge that the repeated group can match
99        nothing is needed in order to break the loop.
100    
101    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
102        exec(), but without supplying a callout function, matching went wrong.
103    
104    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
105        leak if the size of the offset vector was greater than 30. When the vector
106        is smaller, the saved offsets during recursion go onto a local stack
107        vector, but for larger vectors malloc() is used. It was failing to free
108        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
109        error, in fact).
110    
111    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
112        heapframe that is used only when UTF-8 support is enabled. This caused no
113        problem, but was untidy.
114    
115    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
116        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
117        included within another project.
118    
119    24. Steven Van Ingelgem's patches to add more options to the CMake support,
120        slightly modified by me:
121    
122          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
123              not building pcregrep.
124    
125          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
126              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
127    
128    25. Forward references, both numeric and by name, in patterns that made use of
129        duplicate group numbers, could behave incorrectly or give incorrect errors,
130        because when scanning forward to find the reference group, PCRE was not
131        taking into account the duplicate group numbers. A pattern such as
132        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
133    
134    26. Changed a few more instances of "const unsigned char *" to USPTR, making
135        the feature of a custom pointer more persuasive (as requested by a user).
136    
137    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
138        pcretest.c, inside #ifndefs, because it seems they are sometimes already
139        pre-defined.
140    
141    28. Added support for (*UTF8) at the start of a pattern.
142    
143    29. Arrange for flags added by the "release type" setting in CMake to be shown
144        in the configuration summary.
145    
146    
147    Version 7.8 05-Sep-08
148    ---------------------
149    
150    1.  Replaced UCP searching code with optimized version as implemented for Ad
151        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
152        stage table and inline lookup instead of a function, giving speed ups of 2
153        to 5 times on some simple patterns that I tested. Permission was given to
154        distribute the MultiStage2.py script that generates the tables (it's not in
155        the tarball, but is in the Subversion repository).
156    
157    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
158        scripts.
159    
160    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
161        a group with a zero qualifier. The result of the study could be incorrect,
162        or the function might crash, depending on the pattern.
163    
164    4.  Caseless matching was not working for non-ASCII characters in back
165        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
166        It now works when Unicode Property Support is available.
167    
168    5.  In pcretest, an escape such as \x{de} in the data was always generating
169        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
170        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
171        truncation.
172    
173    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
174    
175    7.  Added two (int) casts to pcregrep when printing the difference of two
176        pointers, in case they are 64-bit values.
177    
178    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
179        test 2 if it fails.
180    
181    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
182        and a #define of that name to empty if it is not externally set. This is to
183        allow users of MSVC to set it if necessary.
184    
185    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
186        the convenience functions in the pcre_get.c source file.
187    
188    11. An option change at the start of a pattern that had top-level alternatives
189        could cause overwriting and/or a crash. This command provoked a crash in
190        some environments:
191    
192          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
193    
194        This potential security problem was recorded as CVE-2008-2371.
195    
196    12. For a pattern where the match had to start at the beginning or immediately
197        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
198        pcre_dfa_exec() could read past the end of the passed subject if there was
199        no match. To help with detecting such bugs (e.g. with valgrind), I modified
200        pcretest so that it places the subject at the end of its malloc-ed buffer.
201    
202    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
203        exec() might read past the end of the data buffer in UTF-8 mode.
204    
205    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
206        the data contained the byte 0x85 as part of a UTF-8 character within its
207        first line. This applied both to normal and DFA matching.
208    
209    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
210        /^[^d]*?$/8 failed to match "abc".
211    
212    16. Added a missing copyright notice to pcrecpp_internal.h.
213    
214    17. Make it more clear in the documentation that values returned from
215        pcre_exec() in ovector are byte offsets, not character counts.
216    
217    18. Tidied a few places to stop certain compilers from issuing warnings.
218    
219    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
220        supplied by Stefan Weber. I made a further small update for 7.8 because
221        there is a change of source arrangements: the pcre_searchfuncs.c module is
222        replaced by pcre_ucd.c.
223    
224    
225    Version 7.7 07-May-08
226    ---------------------
227    
228    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
229        a string to a long long, pretend we don't even have a long long." This is
230        done by checking for the strtoq, strtoll, and _strtoi64 functions.
231    
232    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
233        pre-7.6 versions, which defined a global no_arg variable instead of putting
234        it in the RE class. (See also #8 below.)
235    
236    3.  Remove a line of dead code, identified by coverity and reported by Nuno
237        Lopes.
238    
239    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
240    
241        (1) The include/exclude patterns were being applied to the whole pathnames
242            of files, instead of just to the final components.
243    
244        (2) If there was more than one level of directory, the subdirectories were
245            skipped unless they satisfied the include/exclude conditions. This is
246            inconsistent with GNU grep (and could even be seen as contrary to the
247            pcregrep specification - which I improved to make it absolutely clear).
248            The action now is always to scan all levels of directory, and just
249            apply the include/exclude patterns to regular files.
250    
251    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
252        --exclude_dir in the tests to avoid scanning .svn directories.
253    
254    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
255        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
256        doesn't support NULs in patterns.
257    
258    7.  Added some missing "const"s to declarations of static tables in
259        pcre_compile.c and pcre_dfa_exec.c.
260    
261    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
262        caused by fix #2  above. (Subsequently also a second patch to fix the
263        first patch. And a third patch - this was a messy problem.)
264    
265    9.  Applied Craig's patch to remove the use of push_back().
266    
267    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
268        matching function regexec().
269    
270    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
271        which, however, unlike Perl's \g{...}, are subroutine calls, not back
272        references. PCRE supports relative numbers with this syntax (I don't think
273        Oniguruma does).
274    
275    12. Previously, a group with a zero repeat such as (...){0} was completely
276        omitted from the compiled regex. However, this means that if the group
277        was called as a subroutine from elsewhere in the pattern, things went wrong
278        (an internal error was given). Such groups are now left in the compiled
279        pattern, with a new opcode that causes them to be skipped at execution
280        time.
281    
282    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
283        to the way PCRE behaves:
284    
285        (a) A lone ] character is dis-allowed (Perl treats it as data).
286    
287        (b) A back reference to an unmatched subpattern matches an empty string
288            (Perl fails the current match path).
289    
290        (c) A data ] in a character class must be notated as \] because if the
291            first data character in a class is ], it defines an empty class. (In
292            Perl it is not possible to have an empty class.) The empty class []
293            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
294            The negative empty class [^] matches any one character, independently
295            of the DOTALL setting.
296    
297    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
298        non-existent subpattern following a character class starting with ']' and
299        containing () gave an internal compiling error instead of "reference to
300        non-existent subpattern". Fortunately, when the pattern did exist, the
301        compiled code was correct. (When scanning forwards to check for the
302        existencd of the subpattern, it was treating the data ']' as terminating
303        the class, so got the count wrong. When actually compiling, the reference
304        was subsequently set up correctly.)
305    
306    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
307        it was being rejected as not supported by pcre_dfa_exec(), even though
308        other assertions are supported. I have made pcre_dfa_exec() support
309        (*FAIL).
310    
311    16. The implementation of 13c above involved the invention of a new opcode,
312        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
313        cannot be changed at match time, I realized I could make a small
314        improvement to matching performance by compiling OP_ALLANY instead of
315        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
316        on the OP_ANY path.
317    
318    17. Compiling pcretest on Windows with readline support failed without the
319        following two fixes: (1) Make the unistd.h include conditional on
320        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
321    
322    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
323        ncurses library to be included for pcretest when ReadLine support is
324        requested, but also to allow for it to be overridden. This patch came from
325        Daniel Bergström.
326    
327    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
328        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
329        any errors with the current Unicode tables. Thanks to Peter Kankowski for
330        spotting this.
331    
332    
333    Version 7.6 28-Jan-08
334    ---------------------
335    
336    1.  A character class containing a very large number of characters with
337        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
338        overflow.
339    
340    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
341        HAVE_LONG_LONG is not defined.
342    
343    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
344        bring them up to date and include new features. This patch includes:
345    
346        - Fixed PH's badly added libz and libbz2 support.
347        - Fixed a problem with static linking.
348        - Added pcredemo. [But later removed - see 7 below.]
349        - Fixed dftables problem and added an option.
350        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
351            HAVE_LONG_LONG.
352        - Added readline support for pcretest.
353        - Added an listing of the option settings after cmake has run.
354    
355    4.  A user submitted a patch to Makefile that makes it easy to create
356        "pcre.dll" under mingw when using Configure/Make. I added stuff to
357        Makefile.am that cause it to include this special target, without
358        affecting anything else. Note that the same mingw target plus all
359        the other distribution libraries and programs are now supported
360        when configuring with CMake (see 6 below) instead of with
361        Configure/Make.
362    
363    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
364        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
365        exported in the Windows port". It has not yet been confirmed that the patch
366        solves the problem, but it does no harm.
367    
368    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
369        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
370        with CMake, and also correct the comment about stack recursion.
371    
372    7.  Remove the automatic building of pcredemo from the ./configure system and
373        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
374        of a program that users should build themselves after PCRE is installed, so
375        building it automatically is not really right. What is more, it gave
376        trouble in some build environments.
377    
378    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
379    
380    
381    Version 7.5 10-Jan-08
382    ---------------------
383    
384    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
385        values in parens when parsing an RE using the C++ wrapper."
386    
387    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
388        Characters greater than 255 were excluded from the class instead of being
389        included.
390    
391    3.  The same bug as (2) above applied to negated POSIX classes such as
392        [:^space:].
393    
394    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
395        defined or documented. It seems to have been a typo for PCRE_STATIC, so
396        I have changed it.
397    
398    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
399        first named subpattern) and a construct such as (?&a) would reference the
400        first named subpattern whose name started with "a" (in other words, the
401        length check was missing). Both these problems are fixed. "Subpattern name
402        expected" is now given for (?&) (a zero-length name), and this patch also
403        makes it give the same error for \k'' (previously it complained that that
404        was a reference to a non-existent subpattern).
405    
406    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
407        this is right because (?- can be followed by option settings as well as by
408        digits. I have, however, made the messages clearer.
409    
410    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
411        than the number used in the conditional) now cause a compile-time error.
412        This is actually not compatible with Perl, which accepts such patterns, but
413        treats the conditional as always being FALSE (as PCRE used to), but it
414        seems to me that giving a diagnostic is better.
415    
416    8.  Change "alphameric" to the more common word "alphanumeric" in comments
417        and messages.
418    
419    9.  Fix two occurrences of "backslash" in comments that should have been
420        "backspace".
421    
422    10. Remove two redundant lines of code that can never be obeyed (their function
423        was moved elsewhere).
424    
425    11. The program that makes PCRE's Unicode character property table had a bug
426        which caused it to generate incorrect table entries for sequences of
427        characters that have the same character type, but are in different scripts.
428        It amalgamated them into a single range, with the script of the first of
429        them. In other words, some characters were in the wrong script. There were
430        thirteen such cases, affecting characters in the following ranges:
431    
432          U+002b0 - U+002c1
433          U+0060c - U+0060d
434          U+0061e - U+00612
435          U+0064b - U+0065e
436          U+0074d - U+0076d
437          U+01800 - U+01805
438          U+01d00 - U+01d77
439          U+01d9b - U+01dbf
440          U+0200b - U+0200f
441          U+030fc - U+030fe
442          U+03260 - U+0327f
443          U+0fb46 - U+0fbb1
444          U+10450 - U+1049d
445    
446    12. The -o option (show only the matching part of a line) for pcregrep was not
447        compatible with GNU grep in that, if there was more than one match in a
448        line, it showed only the first of them. It now behaves in the same way as
449        GNU grep.
450    
451    13. If the -o and -v options were combined for pcregrep, it printed a blank
452        line for every non-matching line. GNU grep prints nothing, and pcregrep now
453        does the same. The return code can be used to tell if there were any
454        non-matching lines.
455    
456    14. Added --file-offsets and --line-offsets to pcregrep.
457    
458    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
459        infinitely looping recursion. The bug was that positive lookaheads were not
460        being skipped when checking for a possible empty match (negative lookaheads
461        and both kinds of lookbehind were skipped).
462    
463    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
464        inclusion of <windows.h> to before rather than after the definition of
465        INVALID_FILE_ATTRIBUTES (patch from David Byron).
466    
467    17. Specifying a possessive quantifier with a specific limit for a Unicode
468        character property caused pcre_compile() to compile bad code, which led at
469        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
470        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
471        caused the error; without that there was no problem.
472    
473    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
474    
475    19. Added --enable-pcretest-libreadline.
476    
477    20. In pcrecpp.cc, the variable 'count' was incremented twice in
478        RE::GlobalReplace(). As a result, the number of replacements returned was
479        double what it should be. I removed one of the increments, but Craig sent a
480        later patch that removed the other one (the right fix) and added unit tests
481        that check the return values (which was not done before).
482    
483    21. Several CMake things:
484    
485        (1) Arranged that, when cmake is used on Unix, the libraries end up with
486            the names libpcre and libpcreposix, not just pcre and pcreposix.
487    
488        (2) The above change means that pcretest and pcregrep are now correctly
489            linked with the newly-built libraries, not previously installed ones.
490    
491        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
492    
493    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
494        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
495        UTF-8 newline character). The key issue is that the pattern starts .*;
496        this means that the match must be either at the beginning, or after a
497        newline. The bug was in the code for advancing after a failed match and
498        checking that the new position followed a newline. It was not taking
499        account of UTF-8 characters correctly.
500    
501    23. PCRE was behaving differently from Perl in the way it recognized POSIX
502        character classes. PCRE was not treating the sequence [:...:] as a
503        character class unless the ... were all letters. Perl, however, seems to
504        allow any characters between [: and :], though of course it rejects as
505        unknown any "names" that contain non-letters, because all the known class
506        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
507        for example, whereas PCRE did not - it did not recognize a POSIX character
508        class. This seemed a bit dangerous, so the code has been changed to be
509        closer to Perl. The behaviour is not identical to Perl, because PCRE will
510        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
511        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
512        Perl does, and where it didn't before.
513    
514    24. Rewrite so as to remove the single use of %n from pcregrep because in some
515        Windows environments %n is disabled by default.
516    
517    
518    Version 7.4 21-Sep-07
519    ---------------------
520    
521    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
522        means that a class such as [\s] counted as "explicit reference to CR or
523        LF". That isn't really right - the whole point of the change was to try to
524        help when there was an actual mention of one of the two characters. So now
525        the change happens only if \r or \n (or a literal CR or LF) character is
526        encountered.
527    
528    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
529        of both had grown to the point where there were only 3 bits left.
530        Fortunately, there was spare space in the data structure, and so I have
531        moved the internal flags into a new 16-bit field to free up more option
532        bits.
533    
534    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
535        but did not set the internal JCHANGED flag - either of these is enough to
536        control the way the "get" function works - but the PCRE_INFO_JCHANGED
537        facility is supposed to tell if (?J) was ever used, so now (?J) at the
538        start sets both bits.
539    
540    4.  Added options (at build time, compile time, exec time) to change \R from
541        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
542    
543    5.  doc/pcresyntax.html was missing from the distribution.
544    
545    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
546        compatibility, even though it is no longer used.
547    
548    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
549        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
550        windows.h file is present (where different names are used). [This was
551        reversed later after testing - see 16 below.]
552    
553    8.  Changed all #include <config.h> to #include "config.h". There were also
554        some further <pcre.h> cases that I changed to "pcre.h".
555    
556    9.  When pcregrep was used with the --colour option, it missed the line ending
557        sequence off the lines that it output.
558    
559    10. It was pointed out to me that arrays of string pointers cause lots of
560        relocations when a shared library is dynamically loaded. A technique of
561        using a single long string with a table of offsets can drastically reduce
562        these. I have refactored PCRE in four places to do this. The result is
563        dramatic:
564    
565          Originally:                          290
566          After changing UCP table:            187
567          After changing error message table:   43
568          After changing table of "verbs"       36
569          After changing table of Posix names   22
570    
571        Thanks to the folks working on Gregex for glib for this insight.
572    
573    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
574        unicode-properties was also set.
575    
576    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
577    
578    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
579        checked only for CRLF.
580    
581    14. Added casts to pcretest.c to avoid compiler warnings.
582    
583    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
584    
585    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
586        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
587        entirely. This removes changes made in 7 above.
588    
589    17. The CMake files have been updated, and there is now more information about
590        building with CMake in the NON-UNIX-USE document.
591    
592    
593    Version 7.3 28-Aug-07
594    ---------------------
595    
596     1. In the rejigging of the build system that eventually resulted in 7.1, the
597        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
598        brackets there is not right, since it causes compilers to look for an
599        installed pcre.h, not the version that is in the source that is being
600        compiled (which of course may be different). I have changed it back to:
601    
602          #include "pcre.h"
603    
604        I have a vague recollection that the change was concerned with compiling in
605        different directories, but in the new build system, that is taken care of
606        by the VPATH setting the Makefile.
607    
608     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
609        when the subject happened to end in the byte 0x85 (e.g. if the last
610        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
611        characters but of course it shouldn't be taken as a newline when it is part
612        of another character. The bug was that, for an unlimited repeat of . in
613        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
614        characters when looking for a newline.
615    
616     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
617    
618     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
619        in debug output.
620    
621     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
622        long printing in the pcrecpp unittest when running under MinGW.
623    
624     6. ESC_K was left out of the EBCDIC table.
625    
626     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
627        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
628        limit also applies to "virtual nesting" when a pattern is recursive, and in
629        this case 1000 isn't so big. I have been able to remove this limit at the
630        expense of backing off one optimization in certain circumstances. Normally,
631        when pcre_exec() would call its internal match() function recursively and
632        immediately return the result unconditionally, it uses a "tail recursion"
633        feature to save stack. However, when a subpattern that can match an empty
634        string has an unlimited repetition quantifier, it no longer makes this
635        optimization. That gives it a stack frame in which to save the data for
636        checking that an empty string has been matched. Previously this was taken
637        from the 1000-entry workspace that had been reserved. So now there is no
638        explicit limit, but more stack is used.
639    
640     8. Applied Daniel's patches to solve problems with the import/export magic
641        syntax that is required for Windows, and which was going wrong for the
642        pcreposix and pcrecpp parts of the library. These were overlooked when this
643        problem was solved for the main library.
644    
645     9. There were some crude static tests to avoid integer overflow when computing
646        the size of patterns that contain repeated groups with explicit upper
647        limits. As the maximum quantifier is 65535, the maximum group length was
648        set at 30,000 so that the product of these two numbers did not overflow a
649        32-bit integer. However, it turns out that people want to use groups that
650        are longer than 30,000 bytes (though not repeat them that many times).
651        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
652        made it possible to implement the integer overflow checks in a much more
653        dynamic way, which I have now done. The artificial limitation on group
654        length has been removed - we now have only the limit on the total length of
655        the compiled pattern, which depends on the LINK_SIZE setting.
656    
657    10. Fixed a bug in the documentation for get/copy named substring when
658        duplicate names are permitted. If none of the named substrings are set, the
659        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
660        empty string.
661    
662    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
663        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
664        because the ] is interpreted as the first data character and the
665        terminating ] is not found. PCRE has been made compatible with Perl in this
666        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
667        cause memory overwriting.
668    
669    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
670        string has been matched (to stop an infinite loop). It was not recognizing
671        a conditional subpattern that could match an empty string if that
672        subpattern was within another subpattern. For example, it looped when
673        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
674        condition was not nested. This bug has been fixed.
675    
676    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
677        past the start of the subject in the presence of bytes with the top bit
678        set, for example "\x8aBCD".
679    
680    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
681        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
682    
683    14. Optimized (?!) to (*FAIL).
684    
685    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
686        This restricts code points to be within the range 0 to 0x10FFFF, excluding
687        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
688        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
689        does: it's just the validity check that is more restrictive.
690    
691    16. Inserted checks for integer overflows during escape sequence (backslash)
692        processing, and also fixed erroneous offset values for syntax errors during
693        backslash processing.
694    
695    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
696        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
697    
698    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
699        caused an overrun.
700    
701    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
702        something other than just ASCII characters) inside a group that had an
703        unlimited repeat caused a loop at compile time (while checking to see
704        whether the group could match an empty string).
705    
706    20. Debugging a pattern containing \p or \P could cause a crash. For example,
707        [\P{Any}] did so. (Error in the code for printing property names.)
708    
709    21. An orphan \E inside a character class could cause a crash.
710    
711    22. A repeated capturing bracket such as (A)? could cause a wild memory
712        reference during compilation.
713    
714    23. There are several functions in pcre_compile() that scan along a compiled
715        expression for various reasons (e.g. to see if it's fixed length for look
716        behind). There were bugs in these functions when a repeated \p or \P was
717        present in the pattern. These operators have additional parameters compared
718        with \d, etc, and these were not being taken into account when moving along
719        the compiled data. Specifically:
720    
721        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
722            length.
723    
724        (b) An item such as \pL+ within a repeated group could cause crashes or
725            loops.
726    
727        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
728            "reference to non-existent subpattern" error.
729    
730        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
731    
732    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
733        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
734    
735    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
736    
737    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
738        character were causing crashes (broken optimization).
739    
740    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
741        \p or \P) caused a compile-time loop.
742    
743    28. More problems have arisen in unanchored patterns when CRLF is a valid line
744        break. For example, the unstudied pattern [\r\n]A does not match the string
745        "\r\nA" because change 7.0/46 below moves the current point on by two
746        characters after failing to match at the start. However, the pattern \nA
747        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
748        the same is true. There doesn't seem any very clean way out of this, but
749        what I have chosen to do makes the common cases work: PCRE now takes note
750        of whether there can be an explicit match for \r or \n anywhere in the
751        pattern, and if so, 7.0/46 no longer applies. As part of this change,
752        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
753        pattern has explicit CR or LF references.
754    
755    29. Added (*CR) etc for changing newline setting at start of pattern.
756    
757    
758    Version 7.2 19-Jun-07
759    ---------------------
760    
761     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
762        which is apparently normally available under Windows.
763    
764     2. Re-jig the pcregrep tests with different newline settings in an attempt
765        to make them independent of the local environment's newline setting.
766    
767     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
768    
769     4. Some of the "internals" tests were previously cut out when the link size
770        was not 2, because the output contained actual offsets. The recent new
771        "Z" feature of pcretest means that these can be cut out, making the tests
772        usable with all link sizes.
773    
774     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
775        stack recursion. This gives a massive performance boost under BSD, but just
776        a small improvement under Linux. However, it saves one field in the frame
777        in all cases.
778    
779     6. Added more features from the forthcoming Perl 5.10:
780    
781        (a) (?-n) (where n is a string of digits) is a relative subroutine or
782            recursion call. It refers to the nth most recently opened parentheses.
783    
784        (b) (?+n) is also a relative subroutine call; it refers to the nth next
785            to be opened parentheses.
786    
787        (c) Conditions that refer to capturing parentheses can be specified
788            relatively, for example, (?(-2)... or (?(+3)...
789    
790        (d) \K resets the start of the current match so that everything before
791            is not part of it.
792    
793        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
794    
795        (f) \g{name} is another synonym - part of Perl 5.10's unification of
796            reference syntax.
797    
798        (g) (?| introduces a group in which the numbering of parentheses in each
799            alternative starts with the same number.
800    
801        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
802    
803     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
804        PCRE_INFO_JCHANGED.
805    
806     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
807        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
808        for detecting groups that can match an empty string.
809    
810     9. A pattern with a very large number of alternatives (more than several
811        hundred) was running out of internal workspace during the pre-compile
812        phase, where pcre_compile() figures out how much memory will be needed. A
813        bit of new cunning has reduced the workspace needed for groups with
814        alternatives. The 1000-alternative test pattern now uses 12 bytes of
815        workspace instead of running out of the 4096 that are available.
816    
817    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
818    
819    11. Applied patch from Google to remove an optimization that didn't quite work.
820        The report of the bug said:
821    
822          pcrecpp::RE("a*").FullMatch("aaa") matches, while
823          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
824          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
825    
826    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
827        it matched the wrong number of bytes.
828    
829    
830    Version 7.1 24-Apr-07
831    ---------------------
832    
833     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
834        that is more "standard", making use of automake and other Autotools. There
835        is some re-arrangement of the files and adjustment of comments consequent
836        on this.
837    
838     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
839        for recursive directory scanning broke on some systems because the files
840        are not scanned in any specific order and on different systems the order
841        was different. A call to "sort" has been inserted into RunGrepTest for the
842        approprate test as a short-term fix. In the longer term there may be an
843        alternative.
844    
845     3. I had an email from Eric Raymond about problems translating some of PCRE's
846        man pages to HTML (despite the fact that I distribute HTML pages, some
847        people do their own conversions for various reasons). The problems
848        concerned the use of low-level troff macros .br and .in. I have therefore
849        removed all such uses from the man pages (some were redundant, some could
850        be replaced by .nf/.fi pairs). The 132html script that I use to generate
851        HTML has been updated to handle .nf/.fi and to complain if it encounters
852        .br or .in.
853    
854     4. Updated comments in configure.ac that get placed in config.h.in and also
855        arranged for config.h to be included in the distribution, with the name
856        config.h.generic, for the benefit of those who have to compile without
857        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
858    
859     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
860        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
861        makevp.bat for the current PCRE, using the additional files
862        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
863    
864     6. A Windows user reported a minor discrepancy with test 2, which turned out
865        to be caused by a trailing space on an input line that had got lost in his
866        copy. The trailing space was an accident, so I've just removed it.
867    
868     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
869        that is needed.
870    
871     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
872        as "const" (a) because they are and (b) because it helps the PHP
873        maintainers who have recently made a script to detect big data structures
874        in the php code that should be moved to the .rodata section. I remembered
875        to update Builducptable as well, so it won't revert if ucptable.h is ever
876        re-created.
877    
878     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
879        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
880        order to be able to cut out the UTF-8 tables in the latter when UTF-8
881        support is not required. This saves 1.5-2K of code, which is important in
882        some applications.
883    
884        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
885        so as not to refer to the tables, even though these functions will never be
886        called when UTF-8 support is disabled. Otherwise there are problems with a
887        shared library.
888    
889    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
890    
891        (a) It was defining its arguments as char * instead of void *.
892    
893        (b) It was assuming that all moves were upwards in memory; this was true
894            a long time ago when I wrote it, but is no longer the case.
895    
896        The emulated memove() is provided for those environments that have neither
897        memmove() nor bcopy(). I didn't think anyone used it these days, but that
898        is clearly not the case, as these two bugs were recently reported.
899    
900    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
901        and Detrail to create the HTML documentation, the .txt form of the man
902        pages, and it removes trailing spaces from listed files. It also creates
903        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
904        case, it wraps all the #defines with #ifndefs. This script should be run
905        before "make dist".
906    
907    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
908        with Unicode property support.
909    
910        (a) For a maximizing quantifier, if the two different cases of the
911            character were of different lengths in their UTF-8 codings (there are
912            some cases like this - I found 11), and the matching function had to
913            back up over a mixture of the two cases, it incorrectly assumed they
914            were both the same length.
915    
916        (b) When PCRE was configured to use the heap rather than the stack for
917            recursion during matching, it was not correctly preserving the data for
918            the other case of a UTF-8 character when checking ahead for a match
919            while processing a minimizing repeat. If the check also involved
920            matching a wide character, but failed, corruption could cause an
921            erroneous result when trying to check for a repeat of the original
922            character.
923    
924    13. Some tidying changes to the testing mechanism:
925    
926        (a) The RunTest script now detects the internal link size and whether there
927            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
928            values substituted by "configure". (The RunGrepTest script already did
929            this for UTF-8.) The configure.ac script no longer substitutes the
930            relevant variables.
931    
932        (b) The debugging options /B and /D in pcretest show the compiled bytecode
933            with length and offset values. This means that the output is different
934            for different internal link sizes. Test 2 is skipped for link sizes
935            other than 2 because of this, bypassing the problem. Unfortunately,
936            there was also a test in test 3 (the locale tests) that used /B and
937            failed for link sizes other than 2. Rather than cut the whole test out,
938            I have added a new /Z option to pcretest that replaces the length and
939            offset values with spaces. This is now used to make test 3 independent
940            of link size. (Test 2 will be tidied up later.)
941    
942    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
943        segmentation fault instead of returning the appropriate error message.
944    
945    15. In multiline mode when the newline sequence was set to "any", the pattern
946        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
947        This doesn't seem right; it now treats the CRLF combination as the line
948        ending, and so does not match in that case. It's only a pattern such as ^$
949        that would hit this one: something like ^ABC$ would have failed after \r
950        and then tried again after \r\n.
951    
952    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
953        in an attempt to make files that differ only in their line terminators
954        compare equal. This works on Linux.
955    
956    17. Under certain error circumstances pcregrep might try to free random memory
957        as it exited. This is now fixed, thanks to valgrind.
958    
959    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
960        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
961        was because its rules for how to advance for /g after matching an empty
962        string at the end of a line did not allow for this case. They now check for
963        it specially.
964    
965    20. pcretest is supposed to handle patterns and data of any length, by
966        extending its buffers when necessary. It was getting this wrong when the
967        buffer for a data line had to be extended.
968    
969    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
970        CRLF as a newline sequence.
971    
972    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
973        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
974        I have nevertheless tidied it up.
975    
976    23. Added some casts to kill warnings from HP-UX ia64 compiler.
977    
978    24. Added a man page for pcre-config.
979    
980    
981    Version 7.0 19-Dec-06
982    ---------------------
983    
984     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
985        moving to gcc 4.1.1.
986    
987     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
988        sys/time.h, which is documented as needed for this function. It doesn't
989        seem to matter on Linux, but it showed up on some releases of OS X.
990    
991     3. It seems that there are systems where bytes whose values are greater than
992        127 match isprint() in the "C" locale. The "C" locale should be the
993        default when a C program starts up. In most systems, only ASCII printing
994        characters match isprint(). This difference caused the output from pcretest
995        to vary, making some of the tests fail. I have changed pcretest so that:
996    
997        (a) When it is outputting text in the compiled version of a pattern, bytes
998            other than 32-126 are always shown as hex escapes.
999    
1000        (b) When it is outputting text that is a matched part of a subject string,
1001            it does the same, unless a different locale has been set for the match
1002            (using the /L modifier). In this case, it uses isprint() to decide.
1003    
1004     4. Fixed a major bug that caused incorrect computation of the amount of memory
1005        required for a compiled pattern when options that changed within the
1006        pattern affected the logic of the preliminary scan that determines the
1007        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1008        that the computed length was too small. The symptoms of this bug were
1009        either the PCRE error "internal error: code overflow" from pcre_compile(),
1010        or a glibc crash with a message such as "pcretest: free(): invalid next
1011        size (fast)". Examples of patterns that provoked this bug (shown in
1012        pcretest format) are:
1013    
1014          /(?-x: )/x
1015          /(?x)(?-x: \s*#\s*)/
1016          /((?i)[\x{c0}])/8
1017          /(?i:[\x{c0}])/8
1018    
1019        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1020        is now done differently.
1021    
1022     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1023        wrapper classes; (b) implement a new function in the C++ scanner that is
1024        more efficient than the old way of doing things because it avoids levels of
1025        recursion in the regex matching; (c) add a paragraph to the documentation
1026        for the FullMatch() function.
1027    
1028     6. The escape sequence \n was being treated as whatever was defined as
1029        "newline". Not only was this contrary to the documentation, which states
1030        that \n is character 10 (hex 0A), but it also went horribly wrong when
1031        "newline" was defined as CRLF. This has been fixed.
1032    
1033     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1034        was being set to -1 for the "end of line" case (supposedly a value that no
1035        character can have). Though this value is never used (the check for end of
1036        line is "zero bytes in current character"), it caused compiler complaints.
1037        I've changed it to 0xffffffff.
1038    
1039     8. In pcre_version.c, the version string was being built by a sequence of
1040        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1041        string (as it is for production releases) called a macro with an empty
1042        argument. The C standard says the result of this is undefined. The gcc
1043        compiler treats it as an empty string (which was what was wanted) but it is
1044        reported that Visual C gives an error. The source has been hacked around to
1045        avoid this problem.
1046    
1047     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1048        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1049        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1050        of them did).
1051    
1052    10. Originally, pcretest opened its input and output without "b"; then I was
1053        told that "b" was needed in some environments, so it was added for release
1054        5.0 to both the input and output. (It makes no difference on Unix-like
1055        systems.) Later I was told that it is wrong for the input on Windows. I've
1056        now abstracted the modes into two macros, to make it easier to fiddle with
1057        them, and removed "b" from the input mode under Windows.
1058    
1059    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1060    
1061    12. Added -help and --help to pcretest as an official way of being reminded
1062        of the options.
1063    
1064    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1065        and pcrecpp.cc because they annoy compilers at high warning levels.
1066    
1067    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1068    
1069    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1070        scripts are not C programs :-) and which was not noticed because it works
1071        on Linux.
1072    
1073    16. pcretest is supposed to handle any length of pattern and data line (as one
1074        line or as a continued sequence of lines) by extending its input buffer if
1075        necessary. This feature was broken for very long pattern lines, leading to
1076        a string of junk being passed to pcre_compile() if the pattern was longer
1077        than about 50K.
1078    
1079    17. I have done a major re-factoring of the way pcre_compile() computes the
1080        amount of memory needed for a compiled pattern. Previously, there was code
1081        that made a preliminary scan of the pattern in order to do this. That was
1082        OK when PCRE was new, but as the facilities have expanded, it has become
1083        harder and harder to keep it in step with the real compile phase, and there
1084        have been a number of bugs (see for example, 4 above). I have now found a
1085        cunning way of running the real compile function in a "fake" mode that
1086        enables it to compute how much memory it would need, while actually only
1087        ever using a few hundred bytes of working memory and without too many
1088        tests of the mode. This should make future maintenance and development
1089        easier. A side effect of this work is that the limit of 200 on the nesting
1090        depth of parentheses has been removed (though this was never a serious
1091        limitation, I suspect). However, there is a downside: pcre_compile() now
1092        runs more slowly than before (30% or more, depending on the pattern). I
1093        hope this isn't a big issue. There is no effect on runtime performance.
1094    
1095    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1096        newline (only possible for the last line of a file) and it was a
1097        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1098    
1099    19. Added additional timing features to pcretest. (1) The -tm option now times
1100        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1101        separate command line item, by a number that specifies the number of
1102        repeats to use when timing. The default is 50000; this gives better
1103        precision, but takes uncomfortably long for very large patterns.
1104    
1105    20. Extended pcre_study() to be more clever in cases where a branch of a
1106        subpattern has no definite first character. For example, (a*|b*)[cd] would
1107        previously give no result from pcre_study(). Now it recognizes that the
1108        first character must be a, b, c, or d.
1109    
1110    21. There was an incorrect error "recursive call could loop indefinitely" if
1111        a subpattern (or the entire pattern) that was being tested for matching an
1112        empty string contained only one non-empty item after a nested subpattern.
1113        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1114        incorrectly, because the \d was being skipped in the check.
1115    
1116    22. The pcretest program now has a new pattern option /B and a command line
1117        option -b, which is equivalent to adding /B to every pattern. This causes
1118        it to show the compiled bytecode, without the additional information that
1119        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1120        is the same as /B/I).
1121    
1122    23. A new optimization is now able automatically to treat some sequences such
1123        as a*b as a*+b. More specifically, if something simple (such as a character
1124        or a simple class like \d) has an unlimited quantifier, and is followed by
1125        something that cannot possibly match the quantified thing, the quantifier
1126        is automatically "possessified".
1127    
1128    24. A recursive reference to a subpattern whose number was greater than 39
1129        went wrong under certain circumstances in UTF-8 mode. This bug could also
1130        have affected the operation of pcre_study().
1131    
1132    25. Realized that a little bit of performance could be had by replacing
1133        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1134    
1135    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1136    
1137    27. Possessive quantifiers such as a++ were previously implemented by turning
1138        them into atomic groups such as ($>a+). Now they have their own opcodes,
1139        which improves performance. This includes the automatically created ones
1140        from 23 above.
1141    
1142    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1143        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1144        the first matched character to be a colon. This applied both to named and
1145        numbered groups.
1146    
1147    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1148    
1149    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1150        building PCRE on Windows easier, so I have included it in the distribution.
1151    
1152    31. There is now a check in pcretest against a ridiculously large number being
1153        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1154        loop, the loop is abandoned.
1155    
1156    32. Forward references to subpatterns in conditions such as (?(2)...) where
1157        subpattern 2 is defined later cause pcre_compile() to search forwards in
1158        the pattern for the relevant set of parentheses. This search went wrong
1159        when there were unescaped parentheses in a character class, parentheses
1160        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1161    
1162    33. "Subroutine" calls and backreferences were previously restricted to
1163        referencing subpatterns earlier in the regex. This restriction has now
1164        been removed.
1165    
1166    34. Added a number of extra features that are going to be in Perl 5.10. On the
1167        whole, these are just syntactic alternatives for features that PCRE had
1168        previously implemented using the Python syntax or my own invention. The
1169        other formats are all retained for compatibility.
1170    
1171        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1172            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1173            also .NET compatible.
1174    
1175        (b) A recursion or subroutine call to a named group can now be defined as
1176            (?&name) as well as (?P>name).
1177    
1178        (c) A backreference to a named group can now be defined as \k<name> or
1179            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1180            5.10, are also .NET compatible.
1181    
1182        (d) A conditional reference to a named group can now use the syntax
1183            (?(<name>) or (?('name') as well as (?(name).
1184    
1185        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1186            groups (named and numbered) that are never evaluated inline, but can be
1187            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1188            is always false. There may be only one alternative in such a group.
1189    
1190        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1191            as the simple (?(R). The condition is true only if the most recent
1192            recursion is that of the given number or name. It does not search out
1193            through the entire recursion stack.
1194    
1195        (g) The escape \gN or \g{N} has been added, where N is a positive or
1196            negative number, specifying an absolute or relative reference.
1197    
1198    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1199        some "unreachable code" warnings.
1200    
1201    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1202        things, this adds five new scripts.
1203    
1204    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1205        There were also incompatibilities regarding the handling of \Q..\E inside
1206        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1207        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1208    
1209    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1210        matches an empty string, and forcibly breaks the loop. There were bugs in
1211        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1212        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1213        separate and independent bugs (that affected different cases) have been
1214        fixed.
1215    
1216    39. Refactored the code to abolish the use of different opcodes for small
1217        capturing bracket numbers. This is a tidy that I avoided doing when I
1218        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1219        The new approach is not only tidier, it makes it possible to reduce the
1220        memory needed to fix the previous bug (38).
1221    
1222    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1223        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1224        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1225        mode.
1226    
1227    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1228        report.
1229    
1230    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1231        copy construction and assignment in the C++ wrapper.
1232    
1233    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1234        couple of bugs that could have given wrong results in the "--newline=crlf"
1235        case.
1236    
1237    44. Added a number of casts and did some reorganization of signed/unsigned int
1238        variables following suggestions from Dair Grant. Also renamed the variable
1239        "this" as "item" because it is a C++ keyword.
1240    
1241    45. Arranged for dftables to add
1242    
1243          #include "pcre_internal.h"
1244    
1245        to pcre_chartables.c because without it, gcc 4.x may remove the array
1246        definition from the final binary if PCRE is built into a static library and
1247        dead code stripping is activated.
1248    
1249    46. For an unanchored pattern, if a match attempt fails at the start of a
1250        newline sequence, and the newline setting is CRLF or ANY, and the next two
1251        characters are CRLF, advance by two characters instead of one.
1252    
1253    
1254    Version 6.7 04-Jul-06
1255    ---------------------
1256    
1257     1. In order to handle tests when input lines are enormously long, pcretest has
1258        been re-factored so that it automatically extends its buffers when
1259        necessary. The code is crude, but this _is_ just a test program. The
1260        default size has been increased from 32K to 50K.
1261    
1262     2. The code in pcre_study() was using the value of the re argument before
1263        testing it for NULL. (Of course, in any sensible call of the function, it
1264        won't be NULL.)
1265    
1266     3. The memmove() emulation function in pcre_internal.h, which is used on
1267        systems that lack both memmove() and bcopy() - that is, hardly ever -
1268        was missing a "static" storage class specifier.
1269    
1270     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1271        containing an extended class (one that cannot be represented by a bitmap
1272        because it contains high-valued characters or Unicode property items, e.g.
1273        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1274        pattern, but PCRE should not loop if you do not (it no longer does).
1275        [Detail: two cases were found: (a) a repeated subpattern containing an
1276        extended class; (b) a recursive reference to a subpattern that followed a
1277        previous extended class. It wasn't skipping over the extended class
1278        correctly when UTF-8 mode was not set.]
1279    
1280     5. A negated single-character class was not being recognized as fixed-length
1281        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1282        compile error "lookbehind assertion is not fixed length".
1283    
1284     6. The RunPerlTest auxiliary script was showing an unexpected difference
1285        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1286        write a Perl script that can interpret lines of an input file either as
1287        byte characters or as UTF-8, which is what "perltest" was being required to
1288        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1289        can't do is switch easily at run time between having the "use utf8;" pragma
1290        or not. In the end, I fudged it by using the RunPerlTest script to insert
1291        "use utf8;" explicitly for the UTF-8 tests.
1292    
1293     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1294        the end of the subject string, contrary to the documentation and to what
1295        Perl does. This was true of both matching functions. Now it matches only at
1296        the start of the subject and immediately after *internal* newlines.
1297    
1298     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1299        a pointer to an int instead of a pointer to an unsigned long int. This
1300        caused problems on 64-bit systems.
1301    
1302     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1303        instance of the 'standard' template library not being so standard".
1304    
1305    10. There was no check on the number of named subpatterns nor the maximum
1306        length of a subpattern name. The product of these values is used to compute
1307        the size of the memory block for a compiled pattern. By supplying a very
1308        long subpattern name and a large number of named subpatterns, the size
1309        computation could be caused to overflow. This is now prevented by limiting
1310        the length of names to 32 characters, and the number of named subpatterns
1311        to 10,000.
1312    
1313    11. Subpatterns that are repeated with specific counts have to be replicated in
1314        the compiled pattern. The size of memory for this was computed from the
1315        length of the subpattern and the repeat count. The latter is limited to
1316        65535, but there was no limit on the former, meaning that integer overflow
1317        could in principle occur. The compiled length of a repeated subpattern is
1318        now limited to 30,000 bytes in order to prevent this.
1319    
1320    12. Added the optional facility to have named substrings with the same name.
1321    
1322    13. Added the ability to use a named substring as a condition, using the
1323        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1324        are numbers (not recommended). Forward references are permitted.
1325    
1326    14. Added forward references in named backreferences (if you see what I mean).
1327    
1328    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1329        pattern could run off the end of the subject. For example, the pattern
1330        "(?s)(.{1,5})"8 did this with the subject "ab".
1331    
1332    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1333        PCRE_CASELESS was set when matching characters that were quantified with ?
1334        or *.
1335    
1336    17. A character class other than a single negated character that had a minimum
1337        but no maximum quantifier - for example [ab]{6,} - was not handled
1338        correctly by pce_dfa_exec(). It would match only one character.
1339    
1340    18. A valid (though odd) pattern that looked like a POSIX character
1341        class but used an invalid character after [ (for example [[,abc,]]) caused
1342        pcre_compile() to give the error "Failed: internal error: code overflow" or
1343        in some cases to crash with a glibc free() error. This could even happen if
1344        the pattern terminated after [[ but there just happened to be a sequence of
1345        letters, a binary zero, and a closing ] in the memory that followed.
1346    
1347    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1348        over the years. Originally (before any Unicode support), just the bottom 8
1349        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1350        output from "man perlunicode" includes this:
1351    
1352          The regular expression compiler produces polymorphic opcodes.  That
1353          is, the pattern adapts to the data and automatically switches to
1354          the Unicode character scheme when presented with Unicode data--or
1355          instead uses a traditional byte scheme when presented with byte
1356          data.
1357    
1358        Sadly, a wide octal escape does not cause a switch, and in a string with
1359        no other multibyte characters, these octal escapes are treated as before.
1360        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1361        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1362        Unicode string.
1363    
1364        I have not perpetrated such confusion in PCRE. Up till now, it took just
1365        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1366        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1367        translate to the appropriate multibyte character.
1368    
1369    29. Applied some refactoring to reduce the number of warnings from Microsoft
1370        and Borland compilers. This has included removing the fudge introduced
1371        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1372        a warning about an unused variable.
1373    
1374    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1375        characters since release 4.0, because Perl (from release 5.004) does not.
1376        [Or at least, is documented not to: some releases seem to be in conflict
1377        with the documentation.] However, when a pattern was studied with
1378        pcre_study() and all its branches started with \s, PCRE still included VT
1379        as a possible starting character. Of course, this did no harm; it just
1380        caused an unnecessary match attempt.
1381    
1382    22. Removed a now-redundant internal flag bit that recorded the fact that case
1383        dependency changed within the pattern. This was once needed for "required
1384        byte" processing, but is no longer used. This recovers a now-scarce options
1385        bit. Also moved the least significant internal flag bit to the most-
1386        significant bit of the word, which was not previously used (hangover from
1387        the days when it was an int rather than a uint) to free up another bit for
1388        the future.
1389    
1390    23. Added support for CRLF line endings as well as CR and LF. As well as the
1391        default being selectable at build time, it can now be changed at runtime
1392        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1393        specify that it is scanning data with non-default line endings.
1394    
1395    24. Changed the definition of CXXLINK to make it agree with the definition of
1396        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1397    
1398    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1399        recursions. This makes a big different to stack usage for some patterns.
1400    
1401    26. If a subpattern containing a named recursion or subroutine reference such
1402        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1403        the space required for the compiled pattern went wrong and gave too small a
1404        value. Depending on the environment, this could lead to "Failed: internal
1405        error: code overflow at offset 49" or "glibc detected double free or
1406        corruption" errors.
1407    
1408    27. Applied patches from Google (a) to support the new newline modes and (b) to
1409        advance over multibyte UTF-8 characters in GlobalReplace.
1410    
1411    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1412        difference for some implementation of PCRE in some Windows version.
1413    
1414    29. Added some extra testing facilities to pcretest:
1415    
1416        \q<number>   in a data line sets the "match limit" value
1417        \Q<number>   in a data line sets the "match recursion limt" value
1418        -S <number>  sets the stack size, where <number> is in megabytes
1419    
1420        The -S option isn't available for Windows.
1421    
1422    
1423  Version 6.6 06-Feb-06  Version 6.6 06-Feb-06
1424  ---------------------  ---------------------
1425    

Legend:
Removed from v.89  
changed lines
  Added in v.420

  ViewVC Help
Powered by ViewVC 1.1.5