Mail Archives: djgpp-workers/2004/10/24/20:40:06
[repost]
On Sat, 20 Mar 2004 13:12:17 +0100 (CET), ams AT ludd DOT ltu DOT se wrote:
>According to Brian Inglis:
>> liblocal.02/src/ansi/locale/setlocal.pat to liblocal.02/src/ansi/locale/setlocal.c
>>
>> liblocal.02/src/ansi/time/strftime.pat to liblocal.02/src/ansi/time/strftime.c
>
>Is this really applicable to djgpp? I've never noticed any
>liblocal. OTOH I haven't been looking.
It's part of the localisation changes by ASA in contrib.
>> CHANGES C99 Time Related Changes
>...
>> liblocal.02/src/ansi/time/strftime.c
>> src/libc/ansi/time/strftime.c
These were and are the same file.
>> #define required constants;
>> add YEAR() conversion macro and ISLEAPYEAR() test macro;
>> add weekday enum definitions;
>> add iso_year() and iso_week() functions;
>> accept and ignore C99 E and O modifiers;
>> add C99 %F, %G, %g, %V formats;
>> add %s (GNU) format;
>> fix %z ISO 8601/RFC 822 time zone results
>>
>>
>> src/libc/ansi/time/strftime.txh
>>
>> add new formats; add all modifiers; make descriptions consistent
>
>These changes should be put into src/docs/kb/wc204.txi.
Single entry for strftime sufficient?
>> tests/libc/ansi/time/makefile
>>
>> add strftiml.c and strftimt.c tests
>>
>>
>> tests/libc/ansi/time/strftime.c
>>
>> change inline macro calls to table driven tests;
>> add new tests and results for all formats, including unsupported letters
>>
>>
>> tests/libc/ansi/time/xstrftm.c
>>
>> add standard and current locale changes; output each locale name
>>
>>
>> tests/libc/ansi/time/strftiml.c
>>
>> list all formats, including unsupported letters
>>
>>
>> tests/libc/ansi/time/strftimt.c
>>
>> exception tester covering all formats;
>> table driven supporting locales and time zones;
>> many corner cases for non/ISO week and weekdays
>
>Possibly these as well, but I don't think we do.
>
>But they are good as a commit messages.
>
>
>> --- setlocal.c 2002-07-17 02:16:36.000000000 -0600
>> +++ setlocal.c 2004-02-24 01:57:18.000000000 -0700
>
>It's not clear where this file resides I (and I suspect the others
>too) muchly prefer if you do a diff from a directory higher up. It's
>much easier to apply the patch. And we see immediately which
>setlocal.c you mean.
Okay, good point, will redo.
>> Index: makefile
>> ===================================================================
>> RCS file: /cvs/djgpp/djgpp/tests/libc/ansi/time/makefile,v
>> retrieving revision 1.3
>> diff -u -t -r1.3 makefile
>> --- makefile 8 Nov 2003 12:19:41 -0000 1.3
>> +++ makefile 17 Mar 2004 06:57:18 -0000
>> @@ -4,6 +4,8 @@
>> SRC += strftime.c
>> SRC += tzinfo.c
>> SRC += xstrftm.c
>> +SRC += strftimt.c
>> +SRC += strftiml.c
>
>Nothing serious, but why not keep them in alphabetical order?
Tend to add new stuff at end of lists, will change.
>I didn't look much at the code as I prefer test programs and it's a
>lot of test cases you have made. Alas it's doubtful if I'll have the
>time to try them.
>
>I looked at the start of the info file patch and it looked
>good. (After a while it became just a list of format parameters which
>I didn't verify.)
I generated some of the non-obvious values e.g. %V %s, as part of the
new test cases, and I cross checked the Standard, docs, tests, and
code. But bugs are always likely.
--
Thanks. Take care, Brian Inglis
Business: +1(403)547-8816 Brian DOT Inglis AT SystematicSW DOT ab DOT ca
Residence: +1(403)239-6520 BWInglis AT Shaw DOT ca
Cellular: +1(403)708-7006 Brian_Inglis AT CompuServe DOT com
Facsimile: +1(403)547-8816 Brian_Inglis AT CSi DOT com
- Raw text -