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

Diff of /code/trunk/ChangeLog

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

revision 41 by nigel, Sat Feb 24 21:39:17 2007 UTC revision 426 by ph10, Wed Aug 26 15:38:32 2009 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 8.00 ??-???-??
 Version 2.09 14-Sep-99  
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8        was out-of-date, and there was no check on the pcre_compile() error code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    
55    Version 7.9 11-Apr-09
56    ---------------------
57    
58    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
59        (pcretest), all targets were linked against these libraries. This included
60        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
61        libraries. This caused unwanted dependencies to be created. This problem
62        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
63        pcretest is linked with readline.
64    
65    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
66        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
67        moved outside it again, because FALSE and TRUE are already defined in AIX,
68        but BOOL is not.
69    
70    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
71        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
72    
73    4.  The pcregrep documentation said spaces were inserted as well as colons (or
74        hyphens) following file names and line numbers when outputting matching
75        lines. This is not true; no spaces are inserted. I have also clarified the
76        wording for the --colour (or --color) option.
77    
78    5.  In pcregrep, when --colour was used with -o, the list of matching strings
79        was not coloured; this is different to GNU grep, so I have changed it to be
80        the same.
81    
82    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
83        each matching line was coloured. Now it goes on to look for further matches
84        of any of the test patterns, which is the same behaviour as GNU grep.
85    
86    7.  A pattern that could match an empty string could cause pcregrep to loop; it
87        doesn't make sense to accept an empty string match in pcregrep, so I have
88        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
89        seems to be how GNU grep behaves.
90    
91    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
92        start or after a newline", because the conditional assertion was not being
93        correctly handled. The rule now is that both the assertion and what follows
94        in the first alternative must satisfy the test.
95    
96    9.  If auto-callout was enabled in a pattern with a conditional group whose
97        condition was an assertion, PCRE could crash during matching, both with
98        pcre_exec() and pcre_dfa_exec().
99    
100    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
101        used for matching.
102    
103    11. Unicode property support in character classes was not working for
104        characters (bytes) greater than 127 when not in UTF-8 mode.
105    
106    12. Added the -M command line option to pcretest.
107    
108    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
109    
110    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
111    
112    16. Added comments and documentation about mis-use of no_arg in the C++
113        wrapper.
114    
115    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
116        from Martin Jerabek that uses macro names for all relevant character and
117        string constants.
118    
119    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
120        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
121        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
122        these, but not everybody uses configure.
123    
124    19. A conditional group that had only one branch was not being correctly
125        recognized as an item that could match an empty string. This meant that an
126        enclosing group might also not be so recognized, causing infinite looping
127        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
128        with the subject "ab", where knowledge that the repeated group can match
129        nothing is needed in order to break the loop.
130    
131    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
132        exec(), but without supplying a callout function, matching went wrong.
133    
134    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
135        leak if the size of the offset vector was greater than 30. When the vector
136        is smaller, the saved offsets during recursion go onto a local stack
137        vector, but for larger vectors malloc() is used. It was failing to free
138        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
139        error, in fact).
140    
141    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
142        heapframe that is used only when UTF-8 support is enabled. This caused no
143        problem, but was untidy.
144    
145    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
146        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
147        included within another project.
148    
149    24. Steven Van Ingelgem's patches to add more options to the CMake support,
150        slightly modified by me:
151    
152          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
153              not building pcregrep.
154    
155          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
156              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
157    
158    25. Forward references, both numeric and by name, in patterns that made use of
159        duplicate group numbers, could behave incorrectly or give incorrect errors,
160        because when scanning forward to find the reference group, PCRE was not
161        taking into account the duplicate group numbers. A pattern such as
162        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
163    
164    26. Changed a few more instances of "const unsigned char *" to USPTR, making
165        the feature of a custom pointer more persuasive (as requested by a user).
166    
167    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
168        pcretest.c, inside #ifndefs, because it seems they are sometimes already
169        pre-defined.
170    
171    28. Added support for (*UTF8) at the start of a pattern.
172    
173    29. Arrange for flags added by the "release type" setting in CMake to be shown
174        in the configuration summary.
175    
176    
177    Version 7.8 05-Sep-08
178    ---------------------
179    
180    1.  Replaced UCP searching code with optimized version as implemented for Ad
181        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
182        stage table and inline lookup instead of a function, giving speed ups of 2
183        to 5 times on some simple patterns that I tested. Permission was given to
184        distribute the MultiStage2.py script that generates the tables (it's not in
185        the tarball, but is in the Subversion repository).
186    
187    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
188        scripts.
189    
190    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
191        a group with a zero qualifier. The result of the study could be incorrect,
192        or the function might crash, depending on the pattern.
193    
194    4.  Caseless matching was not working for non-ASCII characters in back
195        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
196        It now works when Unicode Property Support is available.
197    
198    5.  In pcretest, an escape such as \x{de} in the data was always generating
199        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
200        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
201        truncation.
202    
203    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
204    
205    7.  Added two (int) casts to pcregrep when printing the difference of two
206        pointers, in case they are 64-bit values.
207    
208    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
209        test 2 if it fails.
210    
211    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
212        and a #define of that name to empty if it is not externally set. This is to
213        allow users of MSVC to set it if necessary.
214    
215    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
216        the convenience functions in the pcre_get.c source file.
217    
218    11. An option change at the start of a pattern that had top-level alternatives
219        could cause overwriting and/or a crash. This command provoked a crash in
220        some environments:
221    
222          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
223    
224        This potential security problem was recorded as CVE-2008-2371.
225    
226    12. For a pattern where the match had to start at the beginning or immediately
227        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
228        pcre_dfa_exec() could read past the end of the passed subject if there was
229        no match. To help with detecting such bugs (e.g. with valgrind), I modified
230        pcretest so that it places the subject at the end of its malloc-ed buffer.
231    
232    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
233        exec() might read past the end of the data buffer in UTF-8 mode.
234    
235    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
236        the data contained the byte 0x85 as part of a UTF-8 character within its
237        first line. This applied both to normal and DFA matching.
238    
239    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
240        /^[^d]*?$/8 failed to match "abc".
241    
242    16. Added a missing copyright notice to pcrecpp_internal.h.
243    
244    17. Make it more clear in the documentation that values returned from
245        pcre_exec() in ovector are byte offsets, not character counts.
246    
247    18. Tidied a few places to stop certain compilers from issuing warnings.
248    
249    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
250        supplied by Stefan Weber. I made a further small update for 7.8 because
251        there is a change of source arrangements: the pcre_searchfuncs.c module is
252        replaced by pcre_ucd.c.
253    
254    
255    Version 7.7 07-May-08
256    ---------------------
257    
258    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
259        a string to a long long, pretend we don't even have a long long." This is
260        done by checking for the strtoq, strtoll, and _strtoi64 functions.
261    
262    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
263        pre-7.6 versions, which defined a global no_arg variable instead of putting
264        it in the RE class. (See also #8 below.)
265    
266    3.  Remove a line of dead code, identified by coverity and reported by Nuno
267        Lopes.
268    
269    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
270    
271        (1) The include/exclude patterns were being applied to the whole pathnames
272            of files, instead of just to the final components.
273    
274        (2) If there was more than one level of directory, the subdirectories were
275            skipped unless they satisfied the include/exclude conditions. This is
276            inconsistent with GNU grep (and could even be seen as contrary to the
277            pcregrep specification - which I improved to make it absolutely clear).
278            The action now is always to scan all levels of directory, and just
279            apply the include/exclude patterns to regular files.
280    
281    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
282        --exclude_dir in the tests to avoid scanning .svn directories.
283    
284    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
285        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
286        doesn't support NULs in patterns.
287    
288    7.  Added some missing "const"s to declarations of static tables in
289        pcre_compile.c and pcre_dfa_exec.c.
290    
291    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
292        caused by fix #2  above. (Subsequently also a second patch to fix the
293        first patch. And a third patch - this was a messy problem.)
294    
295    9.  Applied Craig's patch to remove the use of push_back().
296    
297    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
298        matching function regexec().
299    
300    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
301        which, however, unlike Perl's \g{...}, are subroutine calls, not back
302        references. PCRE supports relative numbers with this syntax (I don't think
303        Oniguruma does).
304    
305    12. Previously, a group with a zero repeat such as (...){0} was completely
306        omitted from the compiled regex. However, this means that if the group
307        was called as a subroutine from elsewhere in the pattern, things went wrong
308        (an internal error was given). Such groups are now left in the compiled
309        pattern, with a new opcode that causes them to be skipped at execution
310        time.
311    
312    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
313        to the way PCRE behaves:
314    
315        (a) A lone ] character is dis-allowed (Perl treats it as data).
316    
317        (b) A back reference to an unmatched subpattern matches an empty string
318            (Perl fails the current match path).
319    
320        (c) A data ] in a character class must be notated as \] because if the
321            first data character in a class is ], it defines an empty class. (In
322            Perl it is not possible to have an empty class.) The empty class []
323            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
324            The negative empty class [^] matches any one character, independently
325            of the DOTALL setting.
326    
327    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
328        non-existent subpattern following a character class starting with ']' and
329        containing () gave an internal compiling error instead of "reference to
330        non-existent subpattern". Fortunately, when the pattern did exist, the
331        compiled code was correct. (When scanning forwards to check for the
332        existencd of the subpattern, it was treating the data ']' as terminating
333        the class, so got the count wrong. When actually compiling, the reference
334        was subsequently set up correctly.)
335    
336    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
337        it was being rejected as not supported by pcre_dfa_exec(), even though
338        other assertions are supported. I have made pcre_dfa_exec() support
339        (*FAIL).
340    
341    16. The implementation of 13c above involved the invention of a new opcode,
342        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
343        cannot be changed at match time, I realized I could make a small
344        improvement to matching performance by compiling OP_ALLANY instead of
345        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
346        on the OP_ANY path.
347    
348    17. Compiling pcretest on Windows with readline support failed without the
349        following two fixes: (1) Make the unistd.h include conditional on
350        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
351    
352    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
353        ncurses library to be included for pcretest when ReadLine support is
354        requested, but also to allow for it to be overridden. This patch came from
355        Daniel Bergström.
356    
357    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
358        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
359        any errors with the current Unicode tables. Thanks to Peter Kankowski for
360        spotting this.
361    
362    
363    Version 7.6 28-Jan-08
364    ---------------------
365    
366    1.  A character class containing a very large number of characters with
367        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
368        overflow.
369    
370    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
371        HAVE_LONG_LONG is not defined.
372    
373    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
374        bring them up to date and include new features. This patch includes:
375    
376        - Fixed PH's badly added libz and libbz2 support.
377        - Fixed a problem with static linking.
378        - Added pcredemo. [But later removed - see 7 below.]
379        - Fixed dftables problem and added an option.
380        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
381            HAVE_LONG_LONG.
382        - Added readline support for pcretest.
383        - Added an listing of the option settings after cmake has run.
384    
385    4.  A user submitted a patch to Makefile that makes it easy to create
386        "pcre.dll" under mingw when using Configure/Make. I added stuff to
387        Makefile.am that cause it to include this special target, without
388        affecting anything else. Note that the same mingw target plus all
389        the other distribution libraries and programs are now supported
390        when configuring with CMake (see 6 below) instead of with
391        Configure/Make.
392    
393    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
394        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
395        exported in the Windows port". It has not yet been confirmed that the patch
396        solves the problem, but it does no harm.
397    
398    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
399        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
400        with CMake, and also correct the comment about stack recursion.
401    
402    7.  Remove the automatic building of pcredemo from the ./configure system and
403        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
404        of a program that users should build themselves after PCRE is installed, so
405        building it automatically is not really right. What is more, it gave
406        trouble in some build environments.
407    
408    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
409    
410    
411    Version 7.5 10-Jan-08
412    ---------------------
413    
414    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
415        values in parens when parsing an RE using the C++ wrapper."
416    
417    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
418        Characters greater than 255 were excluded from the class instead of being
419        included.
420    
421    3.  The same bug as (2) above applied to negated POSIX classes such as
422        [:^space:].
423    
424    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
425        defined or documented. It seems to have been a typo for PCRE_STATIC, so
426        I have changed it.
427    
428    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
429        first named subpattern) and a construct such as (?&a) would reference the
430        first named subpattern whose name started with "a" (in other words, the
431        length check was missing). Both these problems are fixed. "Subpattern name
432        expected" is now given for (?&) (a zero-length name), and this patch also
433        makes it give the same error for \k'' (previously it complained that that
434        was a reference to a non-existent subpattern).
435    
436    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
437        this is right because (?- can be followed by option settings as well as by
438        digits. I have, however, made the messages clearer.
439    
440    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
441        than the number used in the conditional) now cause a compile-time error.
442        This is actually not compatible with Perl, which accepts such patterns, but
443        treats the conditional as always being FALSE (as PCRE used to), but it
444        seems to me that giving a diagnostic is better.
445    
446    8.  Change "alphameric" to the more common word "alphanumeric" in comments
447        and messages.
448    
449    9.  Fix two occurrences of "backslash" in comments that should have been
450        "backspace".
451    
452    10. Remove two redundant lines of code that can never be obeyed (their function
453        was moved elsewhere).
454    
455    11. The program that makes PCRE's Unicode character property table had a bug
456        which caused it to generate incorrect table entries for sequences of
457        characters that have the same character type, but are in different scripts.
458        It amalgamated them into a single range, with the script of the first of
459        them. In other words, some characters were in the wrong script. There were
460        thirteen such cases, affecting characters in the following ranges:
461    
462          U+002b0 - U+002c1
463          U+0060c - U+0060d
464          U+0061e - U+00612
465          U+0064b - U+0065e
466          U+0074d - U+0076d
467          U+01800 - U+01805
468          U+01d00 - U+01d77
469          U+01d9b - U+01dbf
470          U+0200b - U+0200f
471          U+030fc - U+030fe
472          U+03260 - U+0327f
473          U+0fb46 - U+0fbb1
474          U+10450 - U+1049d
475    
476    12. The -o option (show only the matching part of a line) for pcregrep was not
477        compatible with GNU grep in that, if there was more than one match in a
478        line, it showed only the first of them. It now behaves in the same way as
479        GNU grep.
480    
481    13. If the -o and -v options were combined for pcregrep, it printed a blank
482        line for every non-matching line. GNU grep prints nothing, and pcregrep now
483        does the same. The return code can be used to tell if there were any
484        non-matching lines.
485    
486    14. Added --file-offsets and --line-offsets to pcregrep.
487    
488    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
489        infinitely looping recursion. The bug was that positive lookaheads were not
490        being skipped when checking for a possible empty match (negative lookaheads
491        and both kinds of lookbehind were skipped).
492    
493    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
494        inclusion of <windows.h> to before rather than after the definition of
495        INVALID_FILE_ATTRIBUTES (patch from David Byron).
496    
497    17. Specifying a possessive quantifier with a specific limit for a Unicode
498        character property caused pcre_compile() to compile bad code, which led at
499        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
500        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
501        caused the error; without that there was no problem.
502    
503    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
504    
505    19. Added --enable-pcretest-libreadline.
506    
507    20. In pcrecpp.cc, the variable 'count' was incremented twice in
508        RE::GlobalReplace(). As a result, the number of replacements returned was
509        double what it should be. I removed one of the increments, but Craig sent a
510        later patch that removed the other one (the right fix) and added unit tests
511        that check the return values (which was not done before).
512    
513    21. Several CMake things:
514    
515        (1) Arranged that, when cmake is used on Unix, the libraries end up with
516            the names libpcre and libpcreposix, not just pcre and pcreposix.
517    
518        (2) The above change means that pcretest and pcregrep are now correctly
519            linked with the newly-built libraries, not previously installed ones.
520    
521        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
522    
523    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
524        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
525        UTF-8 newline character). The key issue is that the pattern starts .*;
526        this means that the match must be either at the beginning, or after a
527        newline. The bug was in the code for advancing after a failed match and
528        checking that the new position followed a newline. It was not taking
529        account of UTF-8 characters correctly.
530    
531    23. PCRE was behaving differently from Perl in the way it recognized POSIX
532        character classes. PCRE was not treating the sequence [:...:] as a
533        character class unless the ... were all letters. Perl, however, seems to
534        allow any characters between [: and :], though of course it rejects as
535        unknown any "names" that contain non-letters, because all the known class
536        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
537        for example, whereas PCRE did not - it did not recognize a POSIX character
538        class. This seemed a bit dangerous, so the code has been changed to be
539        closer to Perl. The behaviour is not identical to Perl, because PCRE will
540        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
541        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
542        Perl does, and where it didn't before.
543    
544    24. Rewrite so as to remove the single use of %n from pcregrep because in some
545        Windows environments %n is disabled by default.
546    
547    
548    Version 7.4 21-Sep-07
549    ---------------------
550    
551    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
552        means that a class such as [\s] counted as "explicit reference to CR or
553        LF". That isn't really right - the whole point of the change was to try to
554        help when there was an actual mention of one of the two characters. So now
555        the change happens only if \r or \n (or a literal CR or LF) character is
556        encountered.
557    
558    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
559        of both had grown to the point where there were only 3 bits left.
560        Fortunately, there was spare space in the data structure, and so I have
561        moved the internal flags into a new 16-bit field to free up more option
562        bits.
563    
564    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
565        but did not set the internal JCHANGED flag - either of these is enough to
566        control the way the "get" function works - but the PCRE_INFO_JCHANGED
567        facility is supposed to tell if (?J) was ever used, so now (?J) at the
568        start sets both bits.
569    
570    4.  Added options (at build time, compile time, exec time) to change \R from
571        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
572    
573    5.  doc/pcresyntax.html was missing from the distribution.
574    
575    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
576        compatibility, even though it is no longer used.
577    
578    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
579        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
580        windows.h file is present (where different names are used). [This was
581        reversed later after testing - see 16 below.]
582    
583    8.  Changed all #include <config.h> to #include "config.h". There were also
584        some further <pcre.h> cases that I changed to "pcre.h".
585    
586    9.  When pcregrep was used with the --colour option, it missed the line ending
587        sequence off the lines that it output.
588    
589    10. It was pointed out to me that arrays of string pointers cause lots of
590        relocations when a shared library is dynamically loaded. A technique of
591        using a single long string with a table of offsets can drastically reduce
592        these. I have refactored PCRE in four places to do this. The result is
593        dramatic:
594    
595          Originally:                          290
596          After changing UCP table:            187
597          After changing error message table:   43
598          After changing table of "verbs"       36
599          After changing table of Posix names   22
600    
601        Thanks to the folks working on Gregex for glib for this insight.
602    
603    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
604        unicode-properties was also set.
605    
606    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
607    
608    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
609        checked only for CRLF.
610    
611    14. Added casts to pcretest.c to avoid compiler warnings.
612    
613    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
614    
615    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
616        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
617        entirely. This removes changes made in 7 above.
618    
619    17. The CMake files have been updated, and there is now more information about
620        building with CMake in the NON-UNIX-USE document.
621    
622    
623    Version 7.3 28-Aug-07
624    ---------------------
625    
626     1. In the rejigging of the build system that eventually resulted in 7.1, the
627        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
628        brackets there is not right, since it causes compilers to look for an
629        installed pcre.h, not the version that is in the source that is being
630        compiled (which of course may be different). I have changed it back to:
631    
632          #include "pcre.h"
633    
634        I have a vague recollection that the change was concerned with compiling in
635        different directories, but in the new build system, that is taken care of
636        by the VPATH setting the Makefile.
637    
638     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
639        when the subject happened to end in the byte 0x85 (e.g. if the last
640        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
641        characters but of course it shouldn't be taken as a newline when it is part
642        of another character. The bug was that, for an unlimited repeat of . in
643        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
644        characters when looking for a newline.
645    
646     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
647    
648     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
649        in debug output.
650    
651     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
652        long printing in the pcrecpp unittest when running under MinGW.
653    
654     6. ESC_K was left out of the EBCDIC table.
655    
656     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
657        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
658        limit also applies to "virtual nesting" when a pattern is recursive, and in
659        this case 1000 isn't so big. I have been able to remove this limit at the
660        expense of backing off one optimization in certain circumstances. Normally,
661        when pcre_exec() would call its internal match() function recursively and
662        immediately return the result unconditionally, it uses a "tail recursion"
663        feature to save stack. However, when a subpattern that can match an empty
664        string has an unlimited repetition quantifier, it no longer makes this
665        optimization. That gives it a stack frame in which to save the data for
666        checking that an empty string has been matched. Previously this was taken
667        from the 1000-entry workspace that had been reserved. So now there is no
668        explicit limit, but more stack is used.
669    
670     8. Applied Daniel's patches to solve problems with the import/export magic
671        syntax that is required for Windows, and which was going wrong for the
672        pcreposix and pcrecpp parts of the library. These were overlooked when this
673        problem was solved for the main library.
674    
675     9. There were some crude static tests to avoid integer overflow when computing
676        the size of patterns that contain repeated groups with explicit upper
677        limits. As the maximum quantifier is 65535, the maximum group length was
678        set at 30,000 so that the product of these two numbers did not overflow a
679        32-bit integer. However, it turns out that people want to use groups that
680        are longer than 30,000 bytes (though not repeat them that many times).
681        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
682        made it possible to implement the integer overflow checks in a much more
683        dynamic way, which I have now done. The artificial limitation on group
684        length has been removed - we now have only the limit on the total length of
685        the compiled pattern, which depends on the LINK_SIZE setting.
686    
687    10. Fixed a bug in the documentation for get/copy named substring when
688        duplicate names are permitted. If none of the named substrings are set, the
689        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
690        empty string.
691    
692    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
693        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
694        because the ] is interpreted as the first data character and the
695        terminating ] is not found. PCRE has been made compatible with Perl in this
696        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
697        cause memory overwriting.
698    
699    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
700        string has been matched (to stop an infinite loop). It was not recognizing
701        a conditional subpattern that could match an empty string if that
702        subpattern was within another subpattern. For example, it looped when
703        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
704        condition was not nested. This bug has been fixed.
705    
706    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
707        past the start of the subject in the presence of bytes with the top bit
708        set, for example "\x8aBCD".
709    
710    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
711        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
712    
713    14. Optimized (?!) to (*FAIL).
714    
715    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
716        This restricts code points to be within the range 0 to 0x10FFFF, excluding
717        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
718        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
719        does: it's just the validity check that is more restrictive.
720    
721    16. Inserted checks for integer overflows during escape sequence (backslash)
722        processing, and also fixed erroneous offset values for syntax errors during
723        backslash processing.
724    
725    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
726        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
727    
728    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
729        caused an overrun.
730    
731    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
732        something other than just ASCII characters) inside a group that had an
733        unlimited repeat caused a loop at compile time (while checking to see
734        whether the group could match an empty string).
735    
736    20. Debugging a pattern containing \p or \P could cause a crash. For example,
737        [\P{Any}] did so. (Error in the code for printing property names.)
738    
739    21. An orphan \E inside a character class could cause a crash.
740    
741    22. A repeated capturing bracket such as (A)? could cause a wild memory
742        reference during compilation.
743    
744    23. There are several functions in pcre_compile() that scan along a compiled
745        expression for various reasons (e.g. to see if it's fixed length for look
746        behind). There were bugs in these functions when a repeated \p or \P was
747        present in the pattern. These operators have additional parameters compared
748        with \d, etc, and these were not being taken into account when moving along
749        the compiled data. Specifically:
750    
751        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
752            length.
753    
754        (b) An item such as \pL+ within a repeated group could cause crashes or
755            loops.
756    
757        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
758            "reference to non-existent subpattern" error.
759    
760        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
761    
762    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
763        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
764    
765    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
766    
767    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
768        character were causing crashes (broken optimization).
769    
770    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
771        \p or \P) caused a compile-time loop.
772    
773    28. More problems have arisen in unanchored patterns when CRLF is a valid line
774        break. For example, the unstudied pattern [\r\n]A does not match the string
775        "\r\nA" because change 7.0/46 below moves the current point on by two
776        characters after failing to match at the start. However, the pattern \nA
777        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
778        the same is true. There doesn't seem any very clean way out of this, but
779        what I have chosen to do makes the common cases work: PCRE now takes note
780        of whether there can be an explicit match for \r or \n anywhere in the
781        pattern, and if so, 7.0/46 no longer applies. As part of this change,
782        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
783        pattern has explicit CR or LF references.
784    
785    29. Added (*CR) etc for changing newline setting at start of pattern.
786    
787    
788    Version 7.2 19-Jun-07
789    ---------------------
790    
791     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
792        which is apparently normally available under Windows.
793    
794     2. Re-jig the pcregrep tests with different newline settings in an attempt
795        to make them independent of the local environment's newline setting.
796    
797     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
798    
799     4. Some of the "internals" tests were previously cut out when the link size
800        was not 2, because the output contained actual offsets. The recent new
801        "Z" feature of pcretest means that these can be cut out, making the tests
802        usable with all link sizes.
803    
804     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
805        stack recursion. This gives a massive performance boost under BSD, but just
806        a small improvement under Linux. However, it saves one field in the frame
807        in all cases.
808    
809     6. Added more features from the forthcoming Perl 5.10:
810    
811        (a) (?-n) (where n is a string of digits) is a relative subroutine or
812            recursion call. It refers to the nth most recently opened parentheses.
813    
814        (b) (?+n) is also a relative subroutine call; it refers to the nth next
815            to be opened parentheses.
816    
817        (c) Conditions that refer to capturing parentheses can be specified
818            relatively, for example, (?(-2)... or (?(+3)...
819    
820        (d) \K resets the start of the current match so that everything before
821            is not part of it.
822    
823        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
824    
825        (f) \g{name} is another synonym - part of Perl 5.10's unification of
826            reference syntax.
827    
828        (g) (?| introduces a group in which the numbering of parentheses in each
829            alternative starts with the same number.
830    
831        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
832    
833     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
834        PCRE_INFO_JCHANGED.
835    
836     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
837        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
838        for detecting groups that can match an empty string.
839    
840     9. A pattern with a very large number of alternatives (more than several
841        hundred) was running out of internal workspace during the pre-compile
842        phase, where pcre_compile() figures out how much memory will be needed. A
843        bit of new cunning has reduced the workspace needed for groups with
844        alternatives. The 1000-alternative test pattern now uses 12 bytes of
845        workspace instead of running out of the 4096 that are available.
846    
847    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
848    
849    11. Applied patch from Google to remove an optimization that didn't quite work.
850        The report of the bug said:
851    
852          pcrecpp::RE("a*").FullMatch("aaa") matches, while
853          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
854          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
855    
856    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
857        it matched the wrong number of bytes.
858    
859    
860    Version 7.1 24-Apr-07
861    ---------------------
862    
863     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
864        that is more "standard", making use of automake and other Autotools. There
865        is some re-arrangement of the files and adjustment of comments consequent
866        on this.
867    
868     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
869        for recursive directory scanning broke on some systems because the files
870        are not scanned in any specific order and on different systems the order
871        was different. A call to "sort" has been inserted into RunGrepTest for the
872        approprate test as a short-term fix. In the longer term there may be an
873        alternative.
874    
875     3. I had an email from Eric Raymond about problems translating some of PCRE's
876        man pages to HTML (despite the fact that I distribute HTML pages, some
877        people do their own conversions for various reasons). The problems
878        concerned the use of low-level troff macros .br and .in. I have therefore
879        removed all such uses from the man pages (some were redundant, some could
880        be replaced by .nf/.fi pairs). The 132html script that I use to generate
881        HTML has been updated to handle .nf/.fi and to complain if it encounters
882        .br or .in.
883    
884     4. Updated comments in configure.ac that get placed in config.h.in and also
885        arranged for config.h to be included in the distribution, with the name
886        config.h.generic, for the benefit of those who have to compile without
887        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
888    
889     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
890        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
891        makevp.bat for the current PCRE, using the additional files
892        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
893    
894     6. A Windows user reported a minor discrepancy with test 2, which turned out
895        to be caused by a trailing space on an input line that had got lost in his
896        copy. The trailing space was an accident, so I've just removed it.
897    
898     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
899        that is needed.
900    
901     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
902        as "const" (a) because they are and (b) because it helps the PHP
903        maintainers who have recently made a script to detect big data structures
904        in the php code that should be moved to the .rodata section. I remembered
905        to update Builducptable as well, so it won't revert if ucptable.h is ever
906        re-created.
907    
908     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
909        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
910        order to be able to cut out the UTF-8 tables in the latter when UTF-8
911        support is not required. This saves 1.5-2K of code, which is important in
912        some applications.
913    
914        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
915        so as not to refer to the tables, even though these functions will never be
916        called when UTF-8 support is disabled. Otherwise there are problems with a
917        shared library.
918    
919    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
920    
921        (a) It was defining its arguments as char * instead of void *.
922    
923        (b) It was assuming that all moves were upwards in memory; this was true
924            a long time ago when I wrote it, but is no longer the case.
925    
926        The emulated memove() is provided for those environments that have neither
927        memmove() nor bcopy(). I didn't think anyone used it these days, but that
928        is clearly not the case, as these two bugs were recently reported.
929    
930    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
931        and Detrail to create the HTML documentation, the .txt form of the man
932        pages, and it removes trailing spaces from listed files. It also creates
933        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
934        case, it wraps all the #defines with #ifndefs. This script should be run
935        before "make dist".
936    
937    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
938        with Unicode property support.
939    
940        (a) For a maximizing quantifier, if the two different cases of the
941            character were of different lengths in their UTF-8 codings (there are
942            some cases like this - I found 11), and the matching function had to
943            back up over a mixture of the two cases, it incorrectly assumed they
944            were both the same length.
945    
946        (b) When PCRE was configured to use the heap rather than the stack for
947            recursion during matching, it was not correctly preserving the data for
948            the other case of a UTF-8 character when checking ahead for a match
949            while processing a minimizing repeat. If the check also involved
950            matching a wide character, but failed, corruption could cause an
951            erroneous result when trying to check for a repeat of the original
952            character.
953    
954    13. Some tidying changes to the testing mechanism:
955    
956        (a) The RunTest script now detects the internal link size and whether there
957            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
958            values substituted by "configure". (The RunGrepTest script already did
959            this for UTF-8.) The configure.ac script no longer substitutes the
960            relevant variables.
961    
962        (b) The debugging options /B and /D in pcretest show the compiled bytecode
963            with length and offset values. This means that the output is different
964            for different internal link sizes. Test 2 is skipped for link sizes
965            other than 2 because of this, bypassing the problem. Unfortunately,
966            there was also a test in test 3 (the locale tests) that used /B and
967            failed for link sizes other than 2. Rather than cut the whole test out,
968            I have added a new /Z option to pcretest that replaces the length and
969            offset values with spaces. This is now used to make test 3 independent
970            of link size. (Test 2 will be tidied up later.)
971    
972    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
973        segmentation fault instead of returning the appropriate error message.
974    
975    15. In multiline mode when the newline sequence was set to "any", the pattern
976        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
977        This doesn't seem right; it now treats the CRLF combination as the line
978        ending, and so does not match in that case. It's only a pattern such as ^$
979        that would hit this one: something like ^ABC$ would have failed after \r
980        and then tried again after \r\n.
981    
982    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
983        in an attempt to make files that differ only in their line terminators
984        compare equal. This works on Linux.
985    
986    17. Under certain error circumstances pcregrep might try to free random memory
987        as it exited. This is now fixed, thanks to valgrind.
988    
989    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
990        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
991        was because its rules for how to advance for /g after matching an empty
992        string at the end of a line did not allow for this case. They now check for
993        it specially.
994    
995    20. pcretest is supposed to handle patterns and data of any length, by
996        extending its buffers when necessary. It was getting this wrong when the
997        buffer for a data line had to be extended.
998    
999    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1000        CRLF as a newline sequence.
1001    
1002    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1003        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1004        I have nevertheless tidied it up.
1005    
1006    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1007    
1008    24. Added a man page for pcre-config.
1009    
1010    
1011    Version 7.0 19-Dec-06
1012    ---------------------
1013    
1014     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1015        moving to gcc 4.1.1.
1016    
1017     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1018        sys/time.h, which is documented as needed for this function. It doesn't
1019        seem to matter on Linux, but it showed up on some releases of OS X.
1020    
1021     3. It seems that there are systems where bytes whose values are greater than
1022        127 match isprint() in the "C" locale. The "C" locale should be the
1023        default when a C program starts up. In most systems, only ASCII printing
1024        characters match isprint(). This difference caused the output from pcretest
1025        to vary, making some of the tests fail. I have changed pcretest so that:
1026    
1027        (a) When it is outputting text in the compiled version of a pattern, bytes
1028            other than 32-126 are always shown as hex escapes.
1029    
1030        (b) When it is outputting text that is a matched part of a subject string,
1031            it does the same, unless a different locale has been set for the match
1032            (using the /L modifier). In this case, it uses isprint() to decide.
1033    
1034     4. Fixed a major bug that caused incorrect computation of the amount of memory
1035        required for a compiled pattern when options that changed within the
1036        pattern affected the logic of the preliminary scan that determines the
1037        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1038        that the computed length was too small. The symptoms of this bug were
1039        either the PCRE error "internal error: code overflow" from pcre_compile(),
1040        or a glibc crash with a message such as "pcretest: free(): invalid next
1041        size (fast)". Examples of patterns that provoked this bug (shown in
1042        pcretest format) are:
1043    
1044          /(?-x: )/x
1045          /(?x)(?-x: \s*#\s*)/
1046          /((?i)[\x{c0}])/8
1047          /(?i:[\x{c0}])/8
1048    
1049        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1050        is now done differently.
1051    
1052     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1053        wrapper classes; (b) implement a new function in the C++ scanner that is
1054        more efficient than the old way of doing things because it avoids levels of
1055        recursion in the regex matching; (c) add a paragraph to the documentation
1056        for the FullMatch() function.
1057    
1058     6. The escape sequence \n was being treated as whatever was defined as
1059        "newline". Not only was this contrary to the documentation, which states
1060        that \n is character 10 (hex 0A), but it also went horribly wrong when
1061        "newline" was defined as CRLF. This has been fixed.
1062    
1063     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1064        was being set to -1 for the "end of line" case (supposedly a value that no
1065        character can have). Though this value is never used (the check for end of
1066        line is "zero bytes in current character"), it caused compiler complaints.
1067        I've changed it to 0xffffffff.
1068    
1069     8. In pcre_version.c, the version string was being built by a sequence of
1070        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1071        string (as it is for production releases) called a macro with an empty
1072        argument. The C standard says the result of this is undefined. The gcc
1073        compiler treats it as an empty string (which was what was wanted) but it is
1074        reported that Visual C gives an error. The source has been hacked around to
1075        avoid this problem.
1076    
1077     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1078        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1079        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1080        of them did).
1081    
1082    10. Originally, pcretest opened its input and output without "b"; then I was
1083        told that "b" was needed in some environments, so it was added for release
1084        5.0 to both the input and output. (It makes no difference on Unix-like
1085        systems.) Later I was told that it is wrong for the input on Windows. I've
1086        now abstracted the modes into two macros, to make it easier to fiddle with
1087        them, and removed "b" from the input mode under Windows.
1088    
1089    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1090    
1091    12. Added -help and --help to pcretest as an official way of being reminded
1092        of the options.
1093    
1094    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1095        and pcrecpp.cc because they annoy compilers at high warning levels.
1096    
1097    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1098    
1099    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1100        scripts are not C programs :-) and which was not noticed because it works
1101        on Linux.
1102    
1103    16. pcretest is supposed to handle any length of pattern and data line (as one
1104        line or as a continued sequence of lines) by extending its input buffer if
1105        necessary. This feature was broken for very long pattern lines, leading to
1106        a string of junk being passed to pcre_compile() if the pattern was longer
1107        than about 50K.
1108    
1109    17. I have done a major re-factoring of the way pcre_compile() computes the
1110        amount of memory needed for a compiled pattern. Previously, there was code
1111        that made a preliminary scan of the pattern in order to do this. That was
1112        OK when PCRE was new, but as the facilities have expanded, it has become
1113        harder and harder to keep it in step with the real compile phase, and there
1114        have been a number of bugs (see for example, 4 above). I have now found a
1115        cunning way of running the real compile function in a "fake" mode that
1116        enables it to compute how much memory it would need, while actually only
1117        ever using a few hundred bytes of working memory and without too many
1118        tests of the mode. This should make future maintenance and development
1119        easier. A side effect of this work is that the limit of 200 on the nesting
1120        depth of parentheses has been removed (though this was never a serious
1121        limitation, I suspect). However, there is a downside: pcre_compile() now
1122        runs more slowly than before (30% or more, depending on the pattern). I
1123        hope this isn't a big issue. There is no effect on runtime performance.
1124    
1125    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1126        newline (only possible for the last line of a file) and it was a
1127        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1128    
1129    19. Added additional timing features to pcretest. (1) The -tm option now times
1130        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1131        separate command line item, by a number that specifies the number of
1132        repeats to use when timing. The default is 50000; this gives better
1133        precision, but takes uncomfortably long for very large patterns.
1134    
1135    20. Extended pcre_study() to be more clever in cases where a branch of a
1136        subpattern has no definite first character. For example, (a*|b*)[cd] would
1137        previously give no result from pcre_study(). Now it recognizes that the
1138        first character must be a, b, c, or d.
1139    
1140    21. There was an incorrect error "recursive call could loop indefinitely" if
1141        a subpattern (or the entire pattern) that was being tested for matching an
1142        empty string contained only one non-empty item after a nested subpattern.
1143        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1144        incorrectly, because the \d was being skipped in the check.
1145    
1146    22. The pcretest program now has a new pattern option /B and a command line
1147        option -b, which is equivalent to adding /B to every pattern. This causes
1148        it to show the compiled bytecode, without the additional information that
1149        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1150        is the same as /B/I).
1151    
1152    23. A new optimization is now able automatically to treat some sequences such
1153        as a*b as a*+b. More specifically, if something simple (such as a character
1154        or a simple class like \d) has an unlimited quantifier, and is followed by
1155        something that cannot possibly match the quantified thing, the quantifier
1156        is automatically "possessified".
1157    
1158    24. A recursive reference to a subpattern whose number was greater than 39
1159        went wrong under certain circumstances in UTF-8 mode. This bug could also
1160        have affected the operation of pcre_study().
1161    
1162    25. Realized that a little bit of performance could be had by replacing
1163        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1164    
1165    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1166    
1167    27. Possessive quantifiers such as a++ were previously implemented by turning
1168        them into atomic groups such as ($>a+). Now they have their own opcodes,
1169        which improves performance. This includes the automatically created ones
1170        from 23 above.
1171    
1172    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1173        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1174        the first matched character to be a colon. This applied both to named and
1175        numbered groups.
1176    
1177    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1178    
1179    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1180        building PCRE on Windows easier, so I have included it in the distribution.
1181    
1182    31. There is now a check in pcretest against a ridiculously large number being
1183        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1184        loop, the loop is abandoned.
1185    
1186    32. Forward references to subpatterns in conditions such as (?(2)...) where
1187        subpattern 2 is defined later cause pcre_compile() to search forwards in
1188        the pattern for the relevant set of parentheses. This search went wrong
1189        when there were unescaped parentheses in a character class, parentheses
1190        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1191    
1192    33. "Subroutine" calls and backreferences were previously restricted to
1193        referencing subpatterns earlier in the regex. This restriction has now
1194        been removed.
1195    
1196    34. Added a number of extra features that are going to be in Perl 5.10. On the
1197        whole, these are just syntactic alternatives for features that PCRE had
1198        previously implemented using the Python syntax or my own invention. The
1199        other formats are all retained for compatibility.
1200    
1201        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1202            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1203            also .NET compatible.
1204    
1205        (b) A recursion or subroutine call to a named group can now be defined as
1206            (?&name) as well as (?P>name).
1207    
1208        (c) A backreference to a named group can now be defined as \k<name> or
1209            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1210            5.10, are also .NET compatible.
1211    
1212        (d) A conditional reference to a named group can now use the syntax
1213            (?(<name>) or (?('name') as well as (?(name).
1214    
1215        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1216            groups (named and numbered) that are never evaluated inline, but can be
1217            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1218            is always false. There may be only one alternative in such a group.
1219    
1220        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1221            as the simple (?(R). The condition is true only if the most recent
1222            recursion is that of the given number or name. It does not search out
1223            through the entire recursion stack.
1224    
1225        (g) The escape \gN or \g{N} has been added, where N is a positive or
1226            negative number, specifying an absolute or relative reference.
1227    
1228    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1229        some "unreachable code" warnings.
1230    
1231    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1232        things, this adds five new scripts.
1233    
1234    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1235        There were also incompatibilities regarding the handling of \Q..\E inside
1236        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1237        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1238    
1239    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1240        matches an empty string, and forcibly breaks the loop. There were bugs in
1241        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1242        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1243        separate and independent bugs (that affected different cases) have been
1244        fixed.
1245    
1246    39. Refactored the code to abolish the use of different opcodes for small
1247        capturing bracket numbers. This is a tidy that I avoided doing when I
1248        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1249        The new approach is not only tidier, it makes it possible to reduce the
1250        memory needed to fix the previous bug (38).
1251    
1252    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1253        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1254        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1255        mode.
1256    
1257    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1258        report.
1259    
1260    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1261        copy construction and assignment in the C++ wrapper.
1262    
1263    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1264        couple of bugs that could have given wrong results in the "--newline=crlf"
1265        case.
1266    
1267    44. Added a number of casts and did some reorganization of signed/unsigned int
1268        variables following suggestions from Dair Grant. Also renamed the variable
1269        "this" as "item" because it is a C++ keyword.
1270    
1271    45. Arranged for dftables to add
1272    
1273          #include "pcre_internal.h"
1274    
1275        to pcre_chartables.c because without it, gcc 4.x may remove the array
1276        definition from the final binary if PCRE is built into a static library and
1277        dead code stripping is activated.
1278    
1279    46. For an unanchored pattern, if a match attempt fails at the start of a
1280        newline sequence, and the newline setting is CRLF or ANY, and the next two
1281        characters are CRLF, advance by two characters instead of one.
1282    
1283    
1284    Version 6.7 04-Jul-06
1285    ---------------------
1286    
1287     1. In order to handle tests when input lines are enormously long, pcretest has
1288        been re-factored so that it automatically extends its buffers when
1289        necessary. The code is crude, but this _is_ just a test program. The
1290        default size has been increased from 32K to 50K.
1291    
1292     2. The code in pcre_study() was using the value of the re argument before
1293        testing it for NULL. (Of course, in any sensible call of the function, it
1294        won't be NULL.)
1295    
1296     3. The memmove() emulation function in pcre_internal.h, which is used on
1297        systems that lack both memmove() and bcopy() - that is, hardly ever -
1298        was missing a "static" storage class specifier.
1299    
1300     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1301        containing an extended class (one that cannot be represented by a bitmap
1302        because it contains high-valued characters or Unicode property items, e.g.
1303        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1304        pattern, but PCRE should not loop if you do not (it no longer does).
1305        [Detail: two cases were found: (a) a repeated subpattern containing an
1306        extended class; (b) a recursive reference to a subpattern that followed a
1307        previous extended class. It wasn't skipping over the extended class
1308        correctly when UTF-8 mode was not set.]
1309    
1310     5. A negated single-character class was not being recognized as fixed-length
1311        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1312        compile error "lookbehind assertion is not fixed length".
1313    
1314     6. The RunPerlTest auxiliary script was showing an unexpected difference
1315        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1316        write a Perl script that can interpret lines of an input file either as
1317        byte characters or as UTF-8, which is what "perltest" was being required to
1318        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1319        can't do is switch easily at run time between having the "use utf8;" pragma
1320        or not. In the end, I fudged it by using the RunPerlTest script to insert
1321        "use utf8;" explicitly for the UTF-8 tests.
1322    
1323     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1324        the end of the subject string, contrary to the documentation and to what
1325        Perl does. This was true of both matching functions. Now it matches only at
1326        the start of the subject and immediately after *internal* newlines.
1327    
1328     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1329        a pointer to an int instead of a pointer to an unsigned long int. This
1330        caused problems on 64-bit systems.
1331    
1332     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1333        instance of the 'standard' template library not being so standard".
1334    
1335    10. There was no check on the number of named subpatterns nor the maximum
1336        length of a subpattern name. The product of these values is used to compute
1337        the size of the memory block for a compiled pattern. By supplying a very
1338        long subpattern name and a large number of named subpatterns, the size
1339        computation could be caused to overflow. This is now prevented by limiting
1340        the length of names to 32 characters, and the number of named subpatterns
1341        to 10,000.
1342    
1343    11. Subpatterns that are repeated with specific counts have to be replicated in
1344        the compiled pattern. The size of memory for this was computed from the
1345        length of the subpattern and the repeat count. The latter is limited to
1346        65535, but there was no limit on the former, meaning that integer overflow
1347        could in principle occur. The compiled length of a repeated subpattern is
1348        now limited to 30,000 bytes in order to prevent this.
1349    
1350    12. Added the optional facility to have named substrings with the same name.
1351    
1352    13. Added the ability to use a named substring as a condition, using the
1353        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1354        are numbers (not recommended). Forward references are permitted.
1355    
1356    14. Added forward references in named backreferences (if you see what I mean).
1357    
1358    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1359        pattern could run off the end of the subject. For example, the pattern
1360        "(?s)(.{1,5})"8 did this with the subject "ab".
1361    
1362    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1363        PCRE_CASELESS was set when matching characters that were quantified with ?
1364        or *.
1365    
1366    17. A character class other than a single negated character that had a minimum
1367        but no maximum quantifier - for example [ab]{6,} - was not handled
1368        correctly by pce_dfa_exec(). It would match only one character.
1369    
1370    18. A valid (though odd) pattern that looked like a POSIX character
1371        class but used an invalid character after [ (for example [[,abc,]]) caused
1372        pcre_compile() to give the error "Failed: internal error: code overflow" or
1373        in some cases to crash with a glibc free() error. This could even happen if
1374        the pattern terminated after [[ but there just happened to be a sequence of
1375        letters, a binary zero, and a closing ] in the memory that followed.
1376    
1377    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1378        over the years. Originally (before any Unicode support), just the bottom 8
1379        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1380        output from "man perlunicode" includes this:
1381    
1382          The regular expression compiler produces polymorphic opcodes.  That
1383          is, the pattern adapts to the data and automatically switches to
1384          the Unicode character scheme when presented with Unicode data--or
1385          instead uses a traditional byte scheme when presented with byte
1386          data.
1387    
1388        Sadly, a wide octal escape does not cause a switch, and in a string with
1389        no other multibyte characters, these octal escapes are treated as before.
1390        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1391        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1392        Unicode string.
1393    
1394        I have not perpetrated such confusion in PCRE. Up till now, it took just
1395        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1396        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1397        translate to the appropriate multibyte character.
1398    
1399    29. Applied some refactoring to reduce the number of warnings from Microsoft
1400        and Borland compilers. This has included removing the fudge introduced
1401        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1402        a warning about an unused variable.
1403    
1404    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1405        characters since release 4.0, because Perl (from release 5.004) does not.
1406        [Or at least, is documented not to: some releases seem to be in conflict
1407        with the documentation.] However, when a pattern was studied with
1408        pcre_study() and all its branches started with \s, PCRE still included VT
1409        as a possible starting character. Of course, this did no harm; it just
1410        caused an unnecessary match attempt.
1411    
1412    22. Removed a now-redundant internal flag bit that recorded the fact that case
1413        dependency changed within the pattern. This was once needed for "required
1414        byte" processing, but is no longer used. This recovers a now-scarce options
1415        bit. Also moved the least significant internal flag bit to the most-
1416        significant bit of the word, which was not previously used (hangover from
1417        the days when it was an int rather than a uint) to free up another bit for
1418        the future.
1419    
1420    23. Added support for CRLF line endings as well as CR and LF. As well as the
1421        default being selectable at build time, it can now be changed at runtime
1422        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1423        specify that it is scanning data with non-default line endings.
1424    
1425    24. Changed the definition of CXXLINK to make it agree with the definition of
1426        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1427    
1428    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1429        recursions. This makes a big different to stack usage for some patterns.
1430    
1431    26. If a subpattern containing a named recursion or subroutine reference such
1432        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1433        the space required for the compiled pattern went wrong and gave too small a
1434        value. Depending on the environment, this could lead to "Failed: internal
1435        error: code overflow at offset 49" or "glibc detected double free or
1436        corruption" errors.
1437    
1438    27. Applied patches from Google (a) to support the new newline modes and (b) to
1439        advance over multibyte UTF-8 characters in GlobalReplace.
1440    
1441    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1442        difference for some implementation of PCRE in some Windows version.
1443    
1444    29. Added some extra testing facilities to pcretest:
1445    
1446        \q<number>   in a data line sets the "match limit" value
1447        \Q<number>   in a data line sets the "match recursion limt" value
1448        -S <number>  sets the stack size, where <number> is in megabytes
1449    
1450        The -S option isn't available for Windows.
1451    
1452    
1453    Version 6.6 06-Feb-06
1454    ---------------------
1455    
1456     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1457        in pcreposix.h. I have copied the definition from pcre.h.
1458    
1459     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1460        because pcre.h is no longer a built file.
1461    
1462     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1463        not normally included in the compiled code.
1464    
1465    
1466    Version 6.5 01-Feb-06
1467    ---------------------
1468    
1469     1. When using the partial match feature with pcre_dfa_exec(), it was not
1470        anchoring the second and subsequent partial matches at the new starting
1471        point. This could lead to incorrect results. For example, with the pattern
1472        /1234/, partially matching against "123" and then "a4" gave a match.
1473    
1474     2. Changes to pcregrep:
1475    
1476        (a) All non-match returns from pcre_exec() were being treated as failures
1477            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1478            error message is output. Some extra information is given for the
1479            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1480            probably the only errors that are likely to be caused by users (by
1481            specifying a regex that has nested indefinite repeats, for instance).
1482            If there are more than 20 of these errors, pcregrep is abandoned.
1483    
1484        (b) A binary zero was treated as data while matching, but terminated the
1485            output line if it was written out. This has been fixed: binary zeroes
1486            are now no different to any other data bytes.
1487    
1488        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1489            used to set a locale for matching. The --locale=xxxx long option has
1490            been added (no short equivalent) to specify a locale explicitly on the
1491            pcregrep command, overriding the environment variables.
1492    
1493        (d) When -B was used with -n, some line numbers in the output were one less
1494            than they should have been.
1495    
1496        (e) Added the -o (--only-matching) option.
1497    
1498        (f) If -A or -C was used with -c (count only), some lines of context were
1499            accidentally printed for the final match.
1500    
1501        (g) Added the -H (--with-filename) option.
1502    
1503        (h) The combination of options -rh failed to suppress file names for files
1504            that were found from directory arguments.
1505    
1506        (i) Added the -D (--devices) and -d (--directories) options.
1507    
1508        (j) Added the -F (--fixed-strings) option.
1509    
1510        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1511    
1512        (l) Added the --colo(u)r option.
1513    
1514        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1515            is not present by default.
1516    
1517     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1518        items such as (?R) or (?1), when the recursion could match a number of
1519        alternatives. If it matched one of the alternatives, but subsequently,
1520        outside the recursion, there was a failure, the code tried to back up into
1521        the recursion. However, because of the way PCRE is implemented, this is not
1522        possible, and the result was an incorrect result from the match.
1523    
1524        In order to prevent this happening, the specification of recursion has
1525        been changed so that all such subpatterns are automatically treated as
1526        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1527    
1528     4. I had overlooked the fact that, in some locales, there are characters for
1529        which isalpha() is true but neither isupper() nor islower() are true. In
1530        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1531        and ordfeminine) are like this. This affected the treatment of \w and \W
1532        when they appeared in character classes, but not when they appeared outside
1533        a character class. The bit map for "word" characters is now created
1534        separately from the results of isalnum() instead of just taking it from the
1535        upper, lower, and digit maps. (Plus the underscore character, of course.)
1536    
1537     5. The above bug also affected the handling of POSIX character classes such as
1538        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1539        permanent tables. Instead, the bit maps for such a class were previously
1540        created as the appropriate unions of the upper, lower, and digit bitmaps.
1541        Now they are created by subtraction from the [[:word:]] class, which has
1542        its own bitmap.
1543    
1544     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1545        It is created by subtracting the vertical space characters (\x09, \x0a,
1546        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1547        subtraction was done in the overall bitmap for a character class, meaning
1548        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1549        be recognized. This bug has been fixed.
1550    
1551     7. Patches from the folks at Google:
1552    
1553          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1554          real life, but is still worth protecting against".
1555    
1556          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1557          regular expressions".
1558    
1559          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1560          have it.
1561    
1562          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1563          "configure" and the latter not, in order to fix a problem somebody had
1564          with compiling the Arg class on HP-UX.
1565    
1566          (e) Improve the error-handling of the C++ wrapper a little bit.
1567    
1568          (f) New tests for checking recursion limiting.
1569    
1570     8. The pcre_memmove() function, which is used only if the environment does not
1571        have a standard memmove() function (and is therefore rarely compiled),
1572        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1573        returning a result (though PCRE never actually uses the result).
1574    
1575     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1576        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1577        returned instead of calling malloc() with an overflowing number that would
1578        most likely cause subsequent chaos.
1579    
1580    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1581    
1582    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1583        with this option is matched, the nmatch and pmatch options of regexec() are
1584        ignored.
1585    
1586    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1587        provided in case anyone wants to the the POSIX interface with UTF-8
1588        strings.
1589    
1590    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1591        C++ linking (needed for some HP-UX environments).
1592    
1593    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1594        (unused parameter) and in the pcre_printint() function (omitted "default"
1595        switch label when the default is to do nothing).
1596    
1597    15. Added some code to make it possible, when PCRE is compiled as a C++
1598        library, to replace subject pointers for pcre_exec() with a smart pointer
1599        class, thus making it possible to process discontinuous strings.
1600    
1601    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1602        much the same function. They were added by different people who were trying
1603        to make PCRE easy to compile on non-Unix systems. It has been suggested
1604        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1605        for compiling on Windows systems. I have therefore replaced it with
1606        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1607        defaults to "extern" for C or "extern C" for C++, which works fine on
1608        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1609        SCOPE with something explicit in config.h. In addition:
1610    
1611        (a) pcreposix.h still had just "extern" instead of either of these macros;
1612            I have replaced it with PCRE_DATA_SCOPE.
1613    
1614        (b) Functions such as _pcre_xclass(), which are internal to the library,
1615            but external in the C sense, all had PCRE_EXPORT in their definitions.
1616            This is apparently wrong for the Windows case, so I have removed it.
1617            (It makes no difference on Unix-like systems.)
1618    
1619    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1620        of recursive calls to match(). This is different to MATCH_LIMIT because
1621        that limits the total number of calls to match(), not all of which increase
1622        the depth of recursion. Limiting the recursion depth limits the amount of
1623        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1624        when PCRE is compiled, and changed at run time. A patch from Google adds
1625        this functionality to the C++ interface.
1626    
1627    18. Changes to the handling of Unicode character properties:
1628    
1629        (a) Updated the table to Unicode 4.1.0.
1630    
1631        (b) Recognize characters that are not in the table as "Cn" (undefined).
1632    
1633        (c) I revised the way the table is implemented to a much improved format
1634            which includes recognition of ranges. It now supports the ranges that
1635            are defined in UnicodeData.txt, and it also amalgamates other
1636            characters into ranges. This has reduced the number of entries in the
1637            table from around 16,000 to around 3,000, thus reducing its size
1638            considerably. I realized I did not need to use a tree structure after
1639            all - a binary chop search is just as efficient. Having reduced the
1640            number of entries, I extended their size from 6 bytes to 8 bytes to
1641            allow for more data.
1642    
1643        (d) Added support for Unicode script names via properties such as \p{Han}.
1644    
1645    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1646        matching that character.
1647    
1648    20. When matching a repeated Unicode property with a minimum greater than zero,
1649        (for example \pL{2,}), PCRE could look past the end of the subject if it
1650        reached it while seeking the minimum number of characters. This could
1651        happen only if some of the characters were more than one byte long, because
1652        there is a check for at least the minimum number of bytes.
1653    
1654    21. Refactored the implementation of \p and \P so as to be more general, to
1655        allow for more different types of property in future. This has changed the
1656        compiled form incompatibly. Anybody with saved compiled patterns that use
1657        \p or \P will have to recompile them.
1658    
1659    22. Added "Any" and "L&" to the supported property types.
1660    
1661    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1662        but give a compile time error if the value is greater than 0xff.
1663    
1664    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1665        accidentally not being installed or uninstalled.
1666    
1667    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1668        made were to insert the current release number. This seemed silly, because
1669        it made things harder for people building PCRE on systems that don't run
1670        "configure". I have turned pcre.h into a distributed file, no longer built
1671        by "configure", with the version identification directly included. There is
1672        no longer a pcre.h.in file.
1673    
1674        However, this change necessitated a change to the pcre-config script as
1675        well. It is built from pcre-config.in, and one of the substitutions was the
1676        release number. I have updated configure.ac so that ./configure now finds
1677        the release number by grepping pcre.h.
1678    
1679    26. Added the ability to run the tests under valgrind.
1680    
1681    
1682    Version 6.4 05-Sep-05
1683    ---------------------
1684    
1685     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1686        "--" to be printed when multiple files were scanned, even when none of the
1687        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1688        consider it to be a bug, and have restored the previous behaviour.
1689    
1690     2. A couple of code tidies to get rid of compiler warnings.
1691    
1692     3. The pcretest program used to cheat by referring to symbols in the library
1693        whose names begin with _pcre_. These are internal symbols that are not
1694        really supposed to be visible externally, and in some environments it is
1695        possible to suppress them. The cheating is now confined to including
1696        certain files from the library's source, which is a bit cleaner.
1697    
1698     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1699        file's purpose clearer.
1700    
1701     5. Reorganized pcre_ucp_findchar().
1702    
1703    
1704    Version 6.3 15-Aug-05
1705    ---------------------
1706    
1707     1. The file libpcre.pc.in did not have general read permission in the tarball.
1708    
1709     2. There were some problems when building without C++ support:
1710    
1711        (a) If C++ support was not built, "make install" and "make test" still
1712            tried to test it.
1713    
1714        (b) There were problems when the value of CXX was explicitly set. Some
1715            changes have been made to try to fix these, and ...
1716    
1717        (c) --disable-cpp can now be used to explicitly disable C++ support.
1718    
1719        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1720            backslash in a target when C++ was disabled. This confuses some
1721            versions of "make", apparently. Using an intermediate variable solves
1722            this. (Same for CPP_LOBJ.)
1723    
1724     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1725        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1726        necessary on certain architectures.
1727    
1728     4. Added a setting of -export-symbols-regex to the link command to remove
1729        those symbols that are exported in the C sense, but actually are local
1730        within the library, and not documented. Their names all begin with
1731        "_pcre_". This is not a perfect job, because (a) we have to except some
1732        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1733        available (and never for static libraries). I have made a note to try to
1734        find a way round (a) in the future.
1735    
1736    
1737    Version 6.2 01-Aug-05
1738    ---------------------
1739    
1740     1. There was no test for integer overflow of quantifier values. A construction
1741        such as {1111111111111111} would give undefined results. What is worse, if
1742        a minimum quantifier for a parenthesized subpattern overflowed and became
1743        negative, the calculation of the memory size went wrong. This could have
1744        led to memory overwriting.
1745    
1746     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1747    
1748     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1749        operating environments where this matters.
1750    
1751     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1752        PCRE options from within the C++ wrapper.
1753    
1754     5. Named capturing subpatterns were not being correctly counted when a pattern
1755        was compiled. This caused two problems: (a) If there were more than 100
1756        such subpatterns, the calculation of the memory needed for the whole
1757        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1758        back references of the form \12, where the number was greater than 9, were
1759        not recognized as back references, even though there were sufficient
1760        previous subpatterns.
1761    
1762     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1763        versions of gcc, e.g. 2.95.4.
1764    
1765    
1766    Version 6.1 21-Jun-05
1767    ---------------------
1768    
1769     1. There was one reference to the variable "posix" in pcretest.c that was not
1770        surrounded by "#if !defined NOPOSIX".
1771    
1772     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1773        the cross-check on the old pcre_info() function, for the benefit of the
1774        cut-down version of PCRE that is currently imported into Exim.
1775    
1776     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1777        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1778        patterns that start (?i) but I don't think that matters. The use of (?i) is
1779        just an example; this all applies to the other options as well.
1780    
1781     4. Since libtool seems to echo the compile commands it is issuing, the output
1782        from "make" can be reduced a bit by putting "@" in front of each libtool
1783        compile command.
1784    
1785     5. Patch from the folks at Google for configure.in to be a bit more thorough
1786        in checking for a suitable C++ installation before trying to compile the
1787        C++ stuff. This should fix a reported problem when a compiler was present,
1788        but no suitable headers.
1789    
1790     6. The man pages all had just "PCRE" as their title. I have changed them to
1791        be the relevant file name. I have also arranged that these names are
1792        retained in the file doc/pcre.txt, which is a concatenation in text format
1793        of all the man pages except the little individual ones for each function.
1794    
1795     7. The NON-UNIX-USE file had not been updated for the different set of source
1796        files that come with release 6. I also added a few comments about the C++
1797        wrapper.
1798    
1799    
1800    Version 6.0 07-Jun-05
1801    ---------------------
1802    
1803     1. Some minor internal re-organization to help with my DFA experiments.
1804    
1805     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1806        didn't matter for the library itself when fully configured, but did matter
1807        when compiling without UCP support, or within Exim, where the ucp files are
1808        not imported.
1809    
1810     3. Refactoring of the library code to split up the various functions into
1811        different source modules. The addition of the new DFA matching code (see
1812        below) to a single monolithic source would have made it really too
1813        unwieldy, quite apart from causing all the code to be include in a
1814        statically linked application, when only some functions are used. This is
1815        relevant even without the DFA addition now that patterns can be compiled in
1816        one application and matched in another.
1817    
1818        The downside of splitting up is that there have to be some external
1819        functions and data tables that are used internally in different modules of
1820        the library but which are not part of the API. These have all had their
1821        names changed to start with "_pcre_" so that they are unlikely to clash
1822        with other external names.
1823    
1824     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1825        a different (DFA) algorithm. Although it is slower than the original
1826        function, it does have some advantages for certain types of matching
1827        problem.
1828    
1829     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1830        including restarting after a partial match.
1831    
1832     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1833        defined when compiling for Windows was sent to me. I have put it into the
1834        code, though I have no means of testing or verifying it.
1835    
1836     7. Added the pcre_refcount() auxiliary function.
1837    
1838     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1839        match before or at the first newline in the subject string. In pcretest,
1840        the /f option on a pattern can be used to set this.
1841    
1842     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1843        would behave wrongly. This has been present in PCRE since release 4.0.
1844    
1845    10. A number of changes to the pcregrep command:
1846    
1847        (a) Refactored how -x works; insert ^(...)$ instead of setting
1848            PCRE_ANCHORED and checking the length, in preparation for adding
1849            something similar for -w.
1850    
1851        (b) Added the -w (match as a word) option.
1852    
1853        (c) Refactored the way lines are read and buffered so as to have more
1854            than one at a time available.
1855    
1856        (d) Implemented a pcregrep test script.
1857    
1858        (e) Added the -M (multiline match) option. This allows patterns to match
1859            over several lines of the subject. The buffering ensures that at least
1860            8K, or the rest of the document (whichever is the shorter) is available
1861            for matching (and similarly the previous 8K for lookbehind assertions).
1862    
1863        (f) Changed the --help output so that it now says
1864    
1865              -w, --word-regex(p)
1866    
1867            instead of two lines, one with "regex" and the other with "regexp"
1868            because that confused at least one person since the short forms are the
1869            same. (This required a bit of code, as the output is generated
1870            automatically from a table. It wasn't just a text change.)
1871    
1872        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1873            option but starts with a hyphen. Could be a pattern or a path name
1874            starting with a hyphen, for instance.
1875    
1876        (h) "-" can be given as a file name to represent stdin.
1877    
1878        (i) When file names are being printed, "(standard input)" is used for
1879            the standard input, for compatibility with GNU grep. Previously
1880            "<stdin>" was used.
1881    
1882        (j) The option --label=xxx can be used to supply a name to be used for
1883            stdin when file names are being printed. There is no short form.
1884    
1885        (k) Re-factored the options decoding logic because we are going to add
1886            two more options that take data. Such options can now be given in four
1887            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1888    
1889        (l) Added the -A, -B, and -C options for requesting that lines of context
1890            around matches be printed.
1891    
1892        (m) Added the -L option to print the names of files that do not contain
1893            any matching lines, that is, the complement of -l.
1894    
1895        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1896            continue to scan other files.
1897    
1898        (o) The -s option was incorrectly implemented. For compatibility with other
1899            greps, it now suppresses the error message for a non-existent or non-
1900            accessible file (but not the return code). There is a new option called
1901            -q that suppresses the output of matching lines, which was what -s was
1902            previously doing.
1903    
1904        (p) Added --include and --exclude options to specify files for inclusion
1905            and exclusion when recursing.
1906    
1907    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1908        Hopefully, it now does.
1909    
1910    12. Missing cast in pcre_study().
1911    
1912    13. Added an "uninstall" target to the makefile.
1913    
1914    14. Replaced "extern" in the function prototypes in Makefile.in with
1915        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1916        world, but is set differently for Windows.
1917    
1918    15. Added a second compiling function called pcre_compile2(). The only
1919        difference is that it has an extra argument, which is a pointer to an
1920        integer error code. When there is a compile-time failure, this is set
1921        non-zero, in addition to the error test pointer being set to point to an
1922        error message. The new argument may be NULL if no error number is required
1923        (but then you may as well call pcre_compile(), which is now just a
1924        wrapper). This facility is provided because some applications need a
1925        numeric error indication, but it has also enabled me to tidy up the way
1926        compile-time errors are handled in the POSIX wrapper.
1927    
1928    16. Added VPATH=.libs to the makefile; this should help when building with one
1929        prefix path and installing with another. (Or so I'm told by someone who
1930        knows more about this stuff than I do.)
1931    
1932    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1933        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1934        match everything, including newlines. This is not POSIX-compatible, but
1935        somebody wanted the feature. From pcretest it can be activated by using
1936        both the P and the s flags.
1937    
1938    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1939    
1940    19. libpcre.pc was being incorrectly installed as executable.
1941    
1942    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1943        it now also looks for '\r' so that it will work unmodified on Windows.
1944    
1945    21. Added Google's contributed C++ wrapper to the distribution.
1946    
1947    22. Added some untidy missing memory free() calls in pcretest, to keep
1948        Electric Fence happy when testing.
1949    
1950    
1951    
1952    Version 5.0 13-Sep-04
1953    ---------------------
1954    
1955     1. Internal change: literal characters are no longer packed up into items
1956        containing multiple characters in a single byte-string. Each character
1957        is now matched using a separate opcode. However, there may be more than one
1958        byte in the character in UTF-8 mode.
1959    
1960     2. The pcre_callout_block structure has two new fields: pattern_position and
1961        next_item_length. These contain the offset in the pattern to the next match
1962        item, and its length, respectively.
1963    
1964     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1965        insertion of callouts before each pattern item. Added the /C option to
1966        pcretest to make use of this.
1967    
1968     4. On the advice of a Windows user, the lines
1969    
1970          #if defined(_WIN32) || defined(WIN32)
1971          _setmode( _fileno( stdout ), 0x8000 );
1972          #endif  /* defined(_WIN32) || defined(WIN32) */
1973    
1974        have been added to the source of pcretest. This apparently does useful
1975        magic in relation to line terminators.
1976    
1977     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1978        for the benefit of those environments where the "b" makes a difference.
1979    
1980     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1981        to know about it. I have put a hack into configure.in that adds in code
1982        to set GCC=yes if CC=icc. This seems to end up at a point in the
1983        generated configure script that is early enough to affect the setting of
1984        compiler options, which is what is needed, but I have no means of testing
1985        whether it really works. (The user who reported this had patched the
1986        generated configure script, which of course I cannot do.)
1987    
1988        LATER: After change 22 below (new libtool files), the configure script
1989        seems to know about icc (and also ecc). Therefore, I have commented out
1990        this hack in configure.in.
1991    
1992     7. Added support for pkg-config (2 patches were sent in).
1993    
1994     8. Negated POSIX character classes that used a combination of internal tables
1995        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1996        [[:^ascii]]. Typically, they would match almost any characters. The other
1997        POSIX classes were not broken in this way.
1998    
1999     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2000        to find the match, as PCRE was deluded into thinking that the match had to
2001        start at the start point or following a newline. The same bug applied to
2002        patterns with negative forward assertions or any backward assertions
2003        preceding ".*" at the start, unless the pattern required a fixed first
2004        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2005    
2006    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2007        starting at the last subject character, bytes beyond the end of the subject
2008        string were read.
2009    
2010    11. Renamed the variable "class" as "classbits" to make life easier for C++
2011        users. (Previously there was a macro definition, but it apparently wasn't
2012        enough.)
2013    
2014    12. Added the new field "tables" to the extra data so that tables can be passed
2015        in at exec time, or the internal tables can be re-selected. This allows
2016        a compiled regex to be saved and re-used at a later time by a different
2017        program that might have everything at different addresses.
2018    
2019    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2020        -R library as well as a -L library.
2021    
2022    14. The debugging options of pcretest (-d on the command line or D on a
2023        pattern) showed incorrect output for anything following an extended class
2024        that contained multibyte characters and which was followed by a quantifier.
2025    
2026    15. Added optional support for general category Unicode character properties
2027        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2028        support. It adds about 90K to the size of the library. The meanings of the
2029        inbuilt class escapes such as \d and \s have NOT been changed.
2030    
2031    16. Updated pcredemo.c to include calls to free() to release the memory for the
2032        compiled pattern.
2033    
2034    17. The generated file chartables.c was being created in the source directory
2035        instead of in the building directory. This caused the build to fail if the
2036        source directory was different from the building directory, and was
2037        read-only.
2038    
2039    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2040        file. No doubt somebody will tell me if they don't make sense... Also added
2041        Dan Mooney's comments about building on OpenVMS.
2042    
2043    19. Added support for partial matching via the PCRE_PARTIAL option for
2044        pcre_exec() and the \P data escape in pcretest.
2045    
2046    20. Extended pcretest with 3 new pattern features:
2047    
2048        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2049              write the compiled pattern to the file whose name is "rest-of-line".
2050              This is a straight binary dump of the data, with the saved pointer to
2051              the character tables forced to be NULL. The study data, if any, is
2052              written too. After writing, pcretest reads a new pattern.
2053    
2054        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2055              compiled pattern from the given file. There must not be any
2056              occurrences of "<" in the file name (pretty unlikely); if there are,
2057              pcretest will instead treat the initial "<" as a pattern delimiter.
2058              After reading in the pattern, pcretest goes on to read data lines as
2059              usual.
2060    
2061        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2062              and 16-bit fields in a compiled pattern, to simulate a pattern that
2063              was compiled on a host of opposite endianness.
2064    
2065    21. The pcre-exec() function can now cope with patterns that were compiled on
2066        hosts of opposite endianness, with this restriction:
2067    
2068          As for any compiled expression that is saved and used later, the tables
2069          pointer field cannot be preserved; the extra_data field in the arguments
2070          to pcre_exec() should be used to pass in a tables address if a value
2071          other than the default internal tables were used at compile time.
2072    
2073    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2074        now diagnosed as an error. Previously, most of the time, a negative number
2075        would have been treated as zero, but if in addition "ovector" was passed as
2076        NULL, a crash could occur.
2077    
2078    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2079        new versions from the libtool 1.5 distribution (the last one is a copy of
2080        a file called libtool.m4). This seems to have fixed the need to patch
2081        "configure" to support Darwin 1.3 (which I used to do). However, I still
2082        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2083        workstation).
2084    
2085    24. Changed the PCRE licence to be the more standard "BSD" licence.
2086    
2087    
2088    Version 4.5 01-Dec-03
2089    ---------------------
2090    
2091     1. There has been some re-arrangement of the code for the match() function so
2092        that it can be compiled in a version that does not call itself recursively.
2093        Instead, it keeps those local variables that need separate instances for
2094        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2095        needs to "recurse". Keeping track of where control must go is done by means
2096        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2097        hide most of the details from the main code, and operates only if
2098        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2099        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2100        operating.
2101    
2102        To make it easier for callers to provide specially tailored get/free
2103        functions for this usage, two new functions, pcre_stack_malloc, and
2104        pcre_stack_free, are used. They are always called in strict stacking order,
2105        and the size of block requested is always the same.
2106    
2107        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2108        PCRE has been compiled to use the stack or the heap for recursion. The
2109        -C option of pcretest uses this to show which version is compiled.
2110    
2111        A new data escape \S, is added to pcretest; it causes the amounts of store
2112        obtained and freed by both kinds of malloc/free at match time to be added
2113        to the output.
2114    
2115     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2116        what's available on my current Linux desktop machine.
2117    
2118     3. When matching a UTF-8 string, the test for a valid string at the start has
2119        been extended. If start_offset is not zero, PCRE now checks that it points
2120        to a byte that is the start of a UTF-8 character. If not, it returns
2121        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2122        this is necessary because there may be backward assertions in the pattern.
2123        When matching the same subject several times, it may save resources to use
2124        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2125    
2126     4. The code for checking the validity of UTF-8 strings has been tightened so
2127        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2128        containing "overlong sequences".
2129    
2130     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2131        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2132        should have been "&", but it just so happened that all the cases this let
2133        through by mistake were picked up later in the function.
2134    
2135     6. I had used a variable called "isblank" - this is a C99 function, causing
2136        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2137    
2138     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2139        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2140        so the version is shown; (c) stop "make test" echoing "./RunTest".
2141    
2142     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2143    
2144     9. The prototype for memmove() for systems that don't have it was using
2145        size_t, but the inclusion of the header that defines size_t was later. I've
2146        moved the #includes for the C headers earlier to avoid this.
2147    
2148    10. Added some adjustments to the code to make it easier to compiler on certain
2149        special systems:
2150    
2151          (a) Some "const" qualifiers were missing.
2152          (b) Added the macro EXPORT before all exported functions; by default this
2153              is defined to be empty.
2154          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2155              that it reads its output file name as an argument instead of writing
2156              to the standard output and assuming this can be redirected.
2157    
2158    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2159        class containing characters with values greater than 255, PCRE compilation
2160        went into a loop.
2161    
2162    12. A recursive reference to a subpattern that was within another subpattern
2163        that had a minimum quantifier of zero caused PCRE to crash. For example,
2164        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2165        recursion. If the recursively-called subpattern itself had a zero repeat,
2166        that was OK.
2167    
2168    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2169        buffer into which it was copied (for escape processing) was still set at
2170        1024, so long lines caused crashes.
2171    
2172    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2173        "internal error: code overflow...". This applied to any character class
2174        that was followed by a possessive quantifier.
2175    
2176    15. Modified the Makefile to add libpcre.la as a prerequisite for
2177        libpcreposix.la because I was told this is needed for a parallel build to
2178        work.
2179    
2180    16. If a pattern that contained .* following optional items at the start was
2181        studied, the wrong optimizing data was generated, leading to matching
2182        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2183        matching string must start with a or b or c. The correct conclusion for
2184        this pattern is that a match can start with any character.
2185    
2186    
2187    Version 4.4 13-Aug-03
2188    ---------------------
2189    
2190     1. In UTF-8 mode, a character class containing characters with values between
2191        127 and 255 was not handled correctly if the compiled pattern was studied.
2192        In fixing this, I have also improved the studying algorithm for such
2193        classes (slightly).
2194    
2195     2. Three internal functions had redundant arguments passed to them. Removal
2196        might give a very teeny performance improvement.
2197    
2198     3. Documentation bug: the value of the capture_top field in a callout is *one
2199        more than* the number of the hightest numbered captured substring.
2200    
2201     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2202        in incorrectly linking with a previously installed version. They now link
2203        explicitly with libpcre.la.
2204    
2205     5. configure.in no longer needs to recognize Cygwin specially.
2206    
2207     6. A problem in pcre.in for Windows platforms is fixed.
2208    
2209     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2210        pcretest, it used to include the size of the study block as part of its
2211        output. Unfortunately, the structure contains a field that has a different
2212        size on different hardware architectures. This meant that the tests that
2213        showed this size failed. As the block is currently always of a fixed size,
2214        this information isn't actually particularly useful in pcretest output, so
2215        I have just removed it.
2216    
2217     8. Three pre-processor statements accidentally did not start in column 1.
2218        Sadly, there are *still* compilers around that complain, even though
2219        standard C has not required this for well over a decade. Sigh.
2220    
2221     9. In pcretest, the code for checking callouts passed small integers in the
2222        callout_data field, which is a void * field. However, some picky compilers
2223        complained about the casts involved for this on 64-bit systems. Now
2224        pcretest passes the address of the small integer instead, which should get
2225        rid of the warnings.
2226    
2227    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2228        both compile and run time, and gives an error if an invalid UTF-8 sequence
2229        is found. There is a option for disabling this check in cases where the
2230        string is known to be correct and/or the maximum performance is wanted.
2231    
2232    11. In response to a bug report, I changed one line in Makefile.in from
2233    
2234            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2235        to
2236            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2237    
2238        to look similar to other lines, but I have no way of telling whether this
2239        is the right thing to do, as I do not use Windows. No doubt I'll get told
2240        if it's wrong...
2241    
2242    
2243    Version 4.3 21-May-03
2244    ---------------------
2245    
2246    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2247       Makefile.
2248    
2249    2. Some refactoring to improve the quality of the code:
2250    
2251       (i)   The utf8_table... variables are now declared "const".
2252    
2253       (ii)  The code for \cx, which used the "case flipping" table to upper case
2254             lower case letters, now just substracts 32. This is ASCII-specific,
2255             but the whole concept of \cx is ASCII-specific, so it seems
2256             reasonable.
2257    
2258       (iii) PCRE was using its character types table to recognize decimal and
2259             hexadecimal digits in the pattern. This is silly, because it handles
2260             only 0-9, a-f, and A-F, but the character types table is locale-
2261             specific, which means strange things might happen. A private
2262             table is now used for this - though it costs 256 bytes, a table is
2263             much faster than multiple explicit tests. Of course, the standard
2264             character types table is still used for matching digits in subject
2265             strings against \d.
2266    
2267       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2268             ending in _t are). So I've renamed it as ESC_tee.
2269    
2270    3. The first argument for regexec() in the POSIX wrapper should have been
2271       defined as "const".
2272    
2273    4. Changed pcretest to use malloc() for its buffers so that they can be
2274       Electric Fenced for debugging.
2275    
2276    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2277       to read one or more bytes before the start of the subject string. Often this
2278       had no effect on PCRE's behaviour, but in some circumstances it could
2279       provoke a segmentation fault.
2280    
2281    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2282       to try to read one or more bytes before the start of the subject string.
2283    
2284    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2285       UTF-8 support could misbehave in various ways if the subject string
2286       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2287       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2288       back over UTF-8 characters.)
2289    
2290    
2291    Version 4.2 14-Apr-03
2292    ---------------------
2293    
2294    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2295    
2296    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2297         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2298         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2299         [WIN_PREFIX]: new variable, "cyg" for Cygwin
2300         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2301           and BUILD_EXEEXT
2302         Note: automatic setting of the BUILD variables is not yet working
2303         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2304           compile-time but not at link-time
2305         [LINK]: use for linking executables only
2306         make different versions for Windows and non-Windows
2307         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2308           libraries
2309         [LINK_FOR_BUILD]: new variable
2310         [OBJEXT]: use throughout
2311         [EXEEXT]: use throughout
2312         <winshared>: new target
2313         <wininstall>: new target
2314         <dftables.o>: use native compiler
2315         <dftables>: use native linker
2316         <install>: handle Windows platform correctly
2317         <clean>: ditto
2318         <check>: ditto
2319         copy DLL to top builddir before testing
2320    
2321       As part of these changes, -no-undefined was removed again. This was reported
2322       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2323       in any case.
2324    
2325    3. Some tidies to get rid of compiler warnings:
2326    
2327       . In the match_data structure, match_limit was an unsigned long int, whereas
2328         match_call_count was an int. I've made them both unsigned long ints.
2329    
2330       . In pcretest the fact that a const uschar * doesn't automatically cast to
2331         a void * provoked a warning.
2332    
2333       . Turning on some more compiler warnings threw up some "shadow" variables
2334         and a few more missing casts.
2335    
2336    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2337       option, a class that contained a single character with a value between 128
2338       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2339    
2340    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2341       option, a class that contained several characters, but with at least one
2342       whose value was between 128 and 255 caused PCRE to crash.
2343    
2344    
2345    Version 4.1 12-Mar-03
2346    ---------------------
2347    
2348    1. Compiling with gcc -pedantic found a couple of places where casts were
2349    needed, and a string in dftables.c that was longer than standard compilers are
2350    required to support.
2351    
2352    2. Compiling with Sun's compiler found a few more places where the code could
2353    be tidied up in order to avoid warnings.
2354    
2355    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2356    first of these names is deprecated in the latest Autoconf in favour of the name
2357    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2358    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2359    analogy I have changed it to CFLAGS_FOR_BUILD.
2360    
2361    4. Added -no-undefined to the linking command in the Makefile, because this is
2362    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2363    linking step for the pcreposix library.
2364    
2365    5. PCRE was failing to diagnose the case of two named groups with the same
2366    name.
2367    
2368    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2369    literal character that is needed in the subject for a match, and scans along to
2370    ensure that it is present before embarking on the full matching process. This
2371    saves time in cases of nested unlimited repeats that are never going to match.
2372    Problem: the scan can take a lot of time if the subject is very long (e.g.
2373    megabytes), thus penalizing straightforward matches. It is now done only if the
2374    amount of subject to be scanned is less than 1000 bytes.
2375    
2376    7. A lesser problem with the same optimization is that it was recording the
2377    first character of an anchored pattern as "needed", thus provoking a search
2378    right along the subject, even when the first match of the pattern was going to
2379    fail. The "needed" character is now not set for anchored patterns, unless it
2380    follows something in the pattern that is of non-fixed length. Thus, it still
2381    fulfils its original purpose of finding quick non-matches in cases of nested
2382    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2383    
2384    
2385    Version 4.0 17-Feb-03
2386    ---------------------
2387    
2388    1. If a comment in an extended regex that started immediately after a meta-item
2389    extended to the end of string, PCRE compiled incorrect data. This could lead to
2390    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2391    
2392    2. Moved to autoconf 2.53 and libtool 1.4.2.
2393    
2394    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2395    the special perltest8 script is no longer needed - all the tests can be run
2396    from a single perltest script.
2397    
2398    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2399    by \s. It has now been removed in PCRE. This means it isn't recognized as
2400    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2401    class [:space:] *does* include VT, thereby creating a mess.
2402    
2403    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2404    space and tab.
2405    
2406    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2407    its new features into the main test script, reducing the number of scripts.
2408    
2409    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2410    were backward compatible, and made the (?i) apply to the whole pattern, as if
2411    /i were given. Now it behaves more logically, and applies the option setting
2412    only to what follows. PCRE has been changed to follow suit. However, if it
2413    finds options settings right at the start of the pattern, it extracts them into
2414    the global options, as before. Thus, they show up in the info data.
2415    
2416    8. Added support for the \Q...\E escape sequence. Characters in between are
2417    treated as literals. This is slightly different from Perl in that $ and @ are
2418    also handled as literals inside the quotes. In Perl, they will cause variable
2419    interpolation. Note the following examples:
2420    
2421        Pattern            PCRE matches      Perl matches
2422    
2423        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
2424        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2425        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2426    
2427    For compatibility with Perl, \Q...\E sequences are recognized inside character
2428    classes as well as outside them.
2429    
2430    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2431    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2432    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2433    signed/unsigned warnings.
2434    
2435    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2436    option for pcretest, so I've replaced it by a simple function that does just
2437    that job.
2438    
2439    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2440    "pcregrep -".
2441    
2442    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2443    Java package. This provides some syntactic sugar for simple cases of what my
2444    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2445    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2446    item, you can use this simplified notation. Note that only makes sense with
2447    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2448    greediness, whatever the setting of the PCRE_UNGREEDY option.
2449    
2450    13. A change of greediness default within a pattern was not taking effect at
2451    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2452    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2453    was abstracted outside.
2454    
2455    14. PCRE now supports the \G assertion. It is true when the current matching
2456    position is at the start point of the match. This differs from \A when the
2457    starting offset is non-zero. Used with the /g option of pcretest (or similar
2458    code), it works in the same way as it does for Perl's /g option. If all
2459    alternatives of a regex begin with \G, the expression is anchored to the start
2460    match position, and the "anchored" flag is set in the compiled expression.
2461    
2462    15. Some bugs concerning the handling of certain option changes within patterns
2463    have been fixed. These applied to options other than (?ims). For example,
2464    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2465    been the other way round. Some of this was related to change 7 above.
2466    
2467    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2468    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2469    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2470    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2471    
2472    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2473    mode. Unlike ".", it always matches newline, whatever the setting of
2474    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2475    assertions. Perl allows it, but it doesn't (in general) work because it can't
2476    calculate the length of the lookbehind. At least, that's the case for Perl
2477    5.8.0 - I've been told they are going to document that it doesn't work in
2478    future.
2479    
2480    18. Added an error diagnosis for escapes that PCRE does not support: these are
2481    \L, \l, \N, \P, \p, \U, \u, and \X.
2482    
2483    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2484    reading past the end of the pattern in cases such as /[abcd/.
2485    
2486    20. PCRE was getting more memory than necessary for patterns with classes that
2487    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2488    
2489    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2490    compiling PCRE for use with Virtual Pascal.
2491    
2492    22. Small fix to the Makefile to make it work properly if the build is done
2493    outside the source tree.
2494    
2495    23. Added a new extension: a condition to go with recursion. If a conditional
2496    subpattern starts with (?(R) the "true" branch is used if recursion has
2497    happened, whereas the "false" branch is used only at the top level.
2498    
2499    24. When there was a very long string of literal characters (over 255 bytes
2500    without UTF support, over 250 bytes with UTF support), the computation of how
2501    much memory was required could be incorrect, leading to segfaults or other
2502    strange effects.
2503    
2504    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2505    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2506    there was a subsequent back reference to those brackets. This meant that, for
2507    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2508    possible to check for precisely this case. All we can do is abandon the
2509    optimization if .* occurs inside capturing brackets when there are any back
2510    references whatsoever. (See below for a better fix that came later.)
2511    
2512    26. The handling of the optimization for finding the first character of a
2513    non-anchored pattern, and for finding a character that is required later in the
2514    match were failing in some cases. This didn't break the matching; it just
2515    failed to optimize when it could. The way this is done has been re-implemented.
2516    
2517    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2518    
2519    28. Added a new feature that provides some of the functionality that Perl
2520    provides with (?{...}). The facility is termed a "callout". The way it is done
2521    in PCRE is for the caller to provide an optional function, by setting
2522    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2523    global variable. By default it is unset, which disables all calling out. To get
2524    the function called, the regex must include (?C) at appropriate points. This
2525    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2526    This provides a means of identifying different callout points. When PCRE
2527    reaches such a point in the regex, if pcre_callout has been set, the external
2528    function is called. It is provided with data in a structure called
2529    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2530    matching continues; if it returns a non-zero value, the match at the current
2531    point fails. However, backtracking will occur if possible. [This was changed
2532    later and other features added - see item 49 below.]
2533    
2534    29. pcretest is upgraded to test the callout functionality. It provides a
2535    callout function that displays information. By default, it shows the start of
2536    the match and the current position in the text. There are some new data escapes
2537    to vary what happens:
2538    
2539        \C+         in addition, show current contents of captured substrings
2540        \C-         do not supply a callout function
2541        \C!n        return 1 when callout number n is reached
2542        \C!n!m      return 1 when callout number n is reached for the mth time
2543    
2544    30. If pcregrep was called with the -l option and just a single file name, it
2545    output "<stdin>" if a match was found, instead of the file name.
2546    
2547    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2548    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2549    pcre_exec(). This saves a malloc/free per call. The default value of
2550    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2551    when configuring.
2552    
2553    32. The default maximum size of a compiled pattern is 64K. There have been a
2554    few cases of people hitting this limit. The code now uses macros to handle the
2555    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2556    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2557    configuring. Tests 2 and 5 work only with 2-byte links because they output
2558    debugging information about compiled patterns.
2559    
2560    33. Internal code re-arrangements:
2561    
2562    (a) Moved the debugging function for printing out a compiled regex into
2563        its own source file (printint.c) and used #include to pull it into
2564        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2565        separate copies.
2566    
2567    (b) Defined the list of op-code names for debugging as a macro in
2568        internal.h so that it is next to the definition of the opcodes.
2569    
2570    (c) Defined a table of op-code lengths for simpler skipping along compiled
2571        code. This is again a macro in internal.h so that it is next to the
2572        definition of the opcodes.
2573    
2574    34. Added support for recursive calls to individual subpatterns, along the
2575    lines of Robin Houston's patch (but implemented somewhat differently).
2576    
2577    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2578    allow it to read and process whole directories in Win32. This code was
2579    contributed by Lionel Fourquaux; it has not been tested by me.
2580    
2581    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2582    used to name a group. Names consist of alphanumerics and underscores, and must
2583    be unique. Back references use the syntax (?P=name) and recursive calls use
2584    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2585    numbers. The function pcre_fullinfo() can be used after compilation to extract
2586    a name/number map. There are three relevant calls:
2587    
2588      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2589      PCRE_INFO_NAMECOUNT            yields the number of entries
2590      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2591    
2592    The map is a vector of fixed-size entries. The size of each entry depends on
2593    the length of the longest name used. The first two bytes of each entry are the
2594    group number, most significant byte first. There follows the corresponding
2595    name, zero terminated. The names are in alphabetical order.
2596    
2597    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2598    case instead of 255. Making it the same both with and without UTF-8 support
2599    means that the same test output works with both.
2600    
2601    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2602    calling malloc() with a zero argument.
2603    
2604    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2605    optimization. I've improved things by keeping a bitmap of backreferences with
2606    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2607    fact referenced, the optimization can be applied. It is unlikely that a
2608    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2609    the match to follow \n) will appear inside brackets with a number greater than
2610    31, but if it does, any back reference > 31 suppresses the optimization.
2611    
2612    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2613    of disabling numbered capturing parentheses. Any opening parenthesis that is
2614    not followed by ? behaves as if it were followed by ?: but named parentheses
2615    can still be used for capturing (and they will acquire numbers in the usual
2616    way).
2617    
2618    41. Redesigned the return codes from the match() function into yes/no/error so
2619    that errors can be passed back from deep inside the nested calls. A malloc
2620    failure while inside a recursive subpattern call now causes the
2621    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2622    
2623    42. It is now possible to set a limit on the number of times the match()
2624    function is called in a call to pcre_exec(). This facility makes it possible to
2625    limit the amount of recursion and backtracking, though not in a directly
2626    obvious way, because the match() function is used in a number of different
2627    circumstances. The count starts from zero for each position in the subject
2628    string (for non-anchored patterns). The default limit is, for compatibility, a
2629    large number, namely 10 000 000. You can change this in two ways:
2630    
2631    (a) When configuring PCRE before making, you can use --with-match-limit=n
2632        to set a default value for the compiled library.
2633    
2634    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2635        a different value is set. See 45 below.
2636    
2637    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2638    
2639    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2640    of things that can be changed at compile time. The first argument specifies
2641    what is wanted and the second points to where the information is to be placed.
2642    The current list of available information is:
2643    
2644      PCRE_CONFIG_UTF8
2645    
2646    The output is an integer that is set to one if UTF-8 support is available;
2647    otherwise it is set to zero.
2648    
2649      PCRE_CONFIG_NEWLINE
2650    
2651    The output is an integer that it set to the value of the code that is used for
2652    newline. It is either LF (10) or CR (13).
2653    
2654      PCRE_CONFIG_LINK_SIZE
2655    
2656    The output is an integer that contains the number of bytes used for internal
2657    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2658    
2659      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2660    
2661    The output is an integer that contains the threshold above which the POSIX
2662    interface uses malloc() for output vectors. See item 31 above.
2663    
2664      PCRE_CONFIG_MATCH_LIMIT
2665    
2666    The output is an unsigned integer that contains the default limit of the number
2667    of match() calls in a pcre_exec() execution. See 42 above.
2668    
2669    44. pcretest has been upgraded by the addition of the -C option. This causes it
2670    to extract all the available output from the new pcre_config() function, and to
2671    output it. The program then exits immediately.
2672    
2673    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2674    order to support additional features. One way would have been to define
2675    pcre_exec2() (for example) with extra arguments, but this would not have been
2676    extensible, and would also have required all calls to the original function to
2677    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2678    is used for passing in "extra" data from pcre_study().
2679    
2680    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2681    contains the following fields:
2682    
2683      flags         a bitmap indicating which of the following fields are set
2684      study_data    opaque data from pcre_study()
2685      match_limit   a way of specifying a limit on match() calls for a specific
2686                      call to pcre_exec()
2687      callout_data  data for callouts (see 49 below)
2688    
2689    The flag bits are also defined in pcre.h, and are
2690    
2691      PCRE_EXTRA_STUDY_DATA
2692      PCRE_EXTRA_MATCH_LIMIT
2693      PCRE_EXTRA_CALLOUT_DATA
2694    
2695    The pcre_study() function now returns one of these new pcre_extra blocks, with
2696    the actual study data pointed to by the study_data field, and the
2697    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2698    before. That is, this change is entirely upwards-compatible and requires no
2699    change to existing code.
2700    
2701    If you want to pass in additional data to pcre_exec(), you can either place it
2702    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2703    block.
2704    
2705    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2706    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2707    times with different match limits, until it finds the minimum value needed for
2708    pcre_exec() to complete. The value is then output. This can be instructive; for
2709    most simple matches the number is quite small, but for pathological cases it
2710    gets very large very quickly.
2711    
2712    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2713    returns the size of the data block pointed to by the study_data field in a
2714    pcre_extra block, that is, the value that was passed as the argument to
2715    pcre_malloc() when PCRE was getting memory in which to place the information
2716    created by pcre_study(). The fourth argument should point to a size_t variable.
2717    pcretest has been extended so that this information is shown after a successful
2718    pcre_study() call when information about the compiled regex is being displayed.
2719    
2720    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2721    because what follows is always an absolute path. (Later: it turns out that this
2722    is more than cosmetic for MinGW, because it doesn't like empty path
2723    components.)
2724    
2725    49. Some changes have been made to the callout feature (see 28 above):
2726    
2727    (i)  A callout function now has three choices for what it returns:
2728    
2729           0  =>  success, carry on matching
2730         > 0  =>  failure at this point, but backtrack if possible
2731         < 0  =>  serious error, return this value from pcre_exec()
2732    
2733         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2734         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2735         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2736         use by callout functions. It will never be used by PCRE itself.
2737    
2738    (ii) The pcre_extra structure (see 45 above) has a void * field called
2739         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2740         pcre_callout_block structure has a field of the same name. The contents of
2741         the field passed in the pcre_extra structure are passed to the callout
2742         function in the corresponding field in the callout block. This makes it
2743         easier to use the same callout-containing regex from multiple threads. For
2744         testing, the pcretest program has a new data escape
2745    
2746           \C*n        pass the number n (may be negative) as callout_data
2747    
2748         If the callout function in pcretest receives a non-zero value as
2749         callout_data, it returns that value.
2750    
2751    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2752    there were some redundant $(CFLAGS) in commands that are now specified as
2753    $(LINK), which already includes $(CFLAGS).
2754    
2755    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2756    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2757    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2758    one-byte characters throughout. Note that case-insensitive matching applies
2759    only to characters whose values are less than 256. PCRE doesn't support the
2760    notion of cases for higher-valued characters.
2761    
2762    (i)   A character class whose characters are all within 0-255 is handled as
2763          a bit map, and the map is inverted for negative classes. Previously, a
2764          character > 255 always failed to match such a class; however it should
2765          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2766    
2767    (ii)  A negated character class with a single character < 255 is coded as
2768          "not this character" (OP_NOT). This wasn't working properly when the test
2769          character was multibyte, either singly or repeated.
2770    
2771    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2772          mode, for example: \x{100}{2,3}.
2773    
2774    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2775          singly or repeated) now correctly test multibyte characters. However,
2776          PCRE doesn't recognize any characters with values greater than 255 as
2777          digits, spaces, or word characters. Such characters always match \D, \S,
2778          and \W, and never match \d, \s, or \w.
2779    
2780    (v)   Classes may now contain characters and character ranges with values
2781          greater than 255. For example: [ab\x{100}-\x{400}].
2782    
2783    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2784          PCRE in UTF-8 mode.
2785    
2786    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2787    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2788    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2789    value.)
2790    
2791    53. The single man page has become too large. I have therefore split it up into
2792    a number of separate man pages. These also give rise to individual HTML pages;
2793    these are now put in a separate directory, and there is an index.html page that
2794    lists them all. Some hyperlinking between the pages has been installed.
2795    
2796    54. Added convenience functions for handling named capturing parentheses.
2797    
2798    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2799    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2800    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2801    are faulted.
2802    
2803    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2804    calling configure. These values are used when compiling the dftables.c program
2805    which is run to generate the source of the default character tables. They
2806    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2807    you will need to set these values.
2808    
2809    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2810    
2811    
2812    Version 3.9 02-Jan-02
2813    ---------------------
2814    
2815    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2816    
2817    2. If --disable-static was given, the building process failed when trying to
2818    build pcretest and pcregrep. (For some reason it was using libtool to compile
2819    them, which is not right, as they aren't part of the library.)
2820    
2821    
2822    Version 3.8 18-Dec-01
2823    ---------------------
2824    
2825    1. The experimental UTF-8 code was completely screwed up. It was packing the
2826    bytes in the wrong order. How dumb can you get?
2827    
2828    
2829    Version 3.7 29-Oct-01
2830    ---------------------
2831    
2832    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2833    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2834    this didn't happen under Solaris 8, where I normally test things.
2835    
2836    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2837    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2838    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2839    this fix an hour or so after the initial 3.7 release.)
2840    
2841    
2842    Version 3.6 23-Oct-01
2843    ---------------------
2844    
2845    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2846    offsets passed as NULL with zero offset count.
2847    
2848    2. The config.guess and config.sub files had not been updated when I moved to
2849    the latest autoconf.
2850    
2851    
2852    Version 3.5 15-Aug-01
2853    ---------------------
2854    
2855    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2856    had been forgotten.
2857    
2858    2. By using declared but undefined structures, we can avoid using "void"
2859    definitions in pcre.h while keeping the internal definitions of the structures
2860    private.
2861    
2862    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2863    user point of view, this means that both static and shared libraries are built
2864    by default, but this can be individually controlled. More of the work of
2865    handling this static/shared cases is now inside libtool instead of PCRE's make
2866    file.
2867    
2868    4. The pcretest utility is now installed along with pcregrep because it is
2869    useful for users (to test regexs) and by doing this, it automatically gets
2870    relinked by libtool. The documentation has been turned into a man page, so
2871    there are now .1, .txt, and .html versions in /doc.
2872    
2873    5. Upgrades to pcregrep:
2874       (i)   Added long-form option names like gnu grep.
2875       (ii)  Added --help to list all options with an explanatory phrase.
2876       (iii) Added -r, --recursive to recurse into sub-directories.
2877       (iv)  Added -f, --file to read patterns from a file.
2878    
2879    6. pcre_exec() was referring to its "code" argument before testing that
2880    argument for NULL (and giving an error if it was NULL).
2881    
2882    7. Upgraded Makefile.in to allow for compiling in a different directory from
2883    the source directory.
2884    
2885    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2886    options bits, the pointer it was passed was to an int instead of to an unsigned
2887    long int. This mattered only on 64-bit systems.
2888    
2889    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2890    generated) instead of pcre.in, which it its source. Also made the same change
2891    in several of the .c files.
2892    
2893    10. A new release of gcc defines printf() as a macro, which broke pcretest
2894    because it had an ifdef in the middle of a string argument for printf(). Fixed
2895    by using separate calls to printf().
2896    
2897    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2898    script, to force use of CR or LF instead of \n in the source. On non-Unix
2899    systems, the value can be set in config.h.
2900    
2901    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2902    absolute limit. Changed the text of the error message to make this clear, and
2903    likewise updated the man page.
2904    
2905    13. The limit of 99 on the number of capturing subpatterns has been removed.
2906    The new limit is 65535, which I hope will not be a "real" limit.
2907    
2908    
2909    Version 3.4 22-Aug-00
2910    ---------------------
2911    
2912    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2913    
2914    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2915    
2916    
2917    Version 3.3 01-Aug-00
2918    ---------------------
2919    
2920    1. If an octal character was given, but the value was greater than \377, it
2921    was not getting masked to the least significant bits, as documented. This could
2922    lead to crashes in some systems.
2923    
2924    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2925    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2926    
2927    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2928    These just pass their arguments on to (pcre_free)(), but they are provided
2929    because some uses of PCRE bind it to non-C systems that can call its functions,
2930    but cannot call free() or pcre_free() directly.
2931    
2932    4. Add "make test" as a synonym for "make check". Corrected some comments in
2933    the Makefile.
2934    
2935    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2936    Makefile.
2937    
2938    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2939    command called pgrep for grepping around the active processes.
2940    
2941    7. Added the beginnings of support for UTF-8 character strings.
2942    
2943    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2944    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2945    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2946    out for the ar command.)
2947    
2948    
2949    Version 3.2 12-May-00
2950    ---------------------
2951    
2952    This is purely a bug fixing release.
2953    
2954    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2955    of ZA. This was just one example of several cases that could provoke this bug,
2956    which was introduced by change 9 of version 2.00. The code for breaking
2957    infinite loops after an iteration that matches an empty string was't working
2958    correctly.
2959    
2960    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2961    when matched against abbab (for example). After matching an empty string, it
2962    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2963    caused it to match further down the string than it should.
2964    
2965    3. The code contained an inclusion of sys/types.h. It isn't clear why this
2966    was there because it doesn't seem to be needed, and it causes trouble on some
2967    systems, as it is not a Standard C header. It has been removed.
2968    
2969    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2970    were reported on the Macintosh. The changes were from
2971    
2972      while ((c = *(++ptr)) != 0 && c != '\n');
2973    to
2974      while ((c = *(++ptr)) != 0 && c != '\n') ;
2975    
2976    Totally extraordinary, but if that's what it takes...
2977    
2978    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2979    available. Added HAVE_BCOPY and an autoconf test for it; if neither
2980    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2981    assumes the way PCRE uses memmove() (always moving upwards).
2982    
2983    6. PCRE is being used in one environment where strchr() is not available. There
2984    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2985    faster code anyway.
2986    
2987    
2988    Version 3.1 09-Feb-00
2989    ---------------------
2990    
2991    The only change in this release is the fixing of some bugs in Makefile.in for
2992    the "install" target:
2993    
2994    (1) It was failing to install pcreposix.h.
2995    
2996    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2997    
2998    
2999    Version 3.0 01-Feb-00
3000    ---------------------
3001    
3002  1. Add support for the /+ modifier to perltest (to output $` like it does in  1. Add support for the /+ modifier to perltest (to output $` like it does in
3003  pcretest).  pcretest).
3004    
# Line 23  captured string vector to pcre_exec(), b Line 3017  captured string vector to pcre_exec(), b
3017  required a bigger vector, with some working space on the end. This means that  required a bigger vector, with some working space on the end. This means that
3018  the POSIX wrapper now has to get and free some memory, and copy the results.  the POSIX wrapper now has to get and free some memory, and copy the results.
3019    
3020    6. Added some simple autoconf support, placing the test data and the
3021    documentation in separate directories, re-organizing some of the
3022    information files, and making it build pcre-config (a GNU standard). Also added
3023    libtool support for building PCRE as a shared library, which is now the
3024    default.
3025    
3026    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3027    09 are not valid octal constants. Single digits will be used for minor values
3028    less than 10.
3029    
3030    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3031    existing programs that set these in the POSIX interface can use PCRE without
3032    modification.
3033    
3034    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3035    return all that pcre_info() returns, plus additional data. The pcre_info()
3036    function is retained for compatibility, but is considered to be obsolete.
3037    
3038    10. Added experimental recursion feature (?R) to handle one common case that
3039    Perl 5.6 will be able to do with (?p{...}).
3040    
3041    11. Added support for POSIX character classes like [:alpha:], which Perl is
3042    adopting.
3043    
3044    
3045  Version 2.08 31-Aug-99  Version 2.08 31-Aug-99
3046  ----------------------  ----------------------

Legend:
Removed from v.41  
changed lines
  Added in v.426

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12