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

Diff of /code/trunk/ChangeLog

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

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8        was out-of-date, and there was no check on the pcre_compile() error code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

Legend:
Removed from v.27  
changed lines
  Added in v.433

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12