At Tue, 30 May 2000 01:04:56 +0200 (CEST),
Urban Widmark <urban@svenskatest.se> wrote:
> > hmm, 2.4.0test-ac4 has missing map tables of cp932, cp936, cp949, c950.
> > so i made additional patch for Urban's + Andrzej's.
> > The patch is found at:
> > http://www2.fe.dis.titech.ac.jp/~gotom/patch/linux/nls16maps.patch.gz
>
> Hello, a few comments:
Thanks your comments.
> I changed 'void uni2char' to 'int uni2char' in my "forward port" since
> there are those who want to know if the result was ok (ncpfs). IMHO
> returning -1 on error is a lot nicer than just writing '?' to the output
> buffer. It's a simple change to make for these as well.
I fully agreed. Writing '?' is not good way.
> 2.3 doesn't have 'dec/inc_use_count' in the struct nls_table. So those
> should be removed (and the functions).
Removed.
> There are 2 arrays for converting characters from lower to upper and upper
> to lower in the 2.3/2.4 struct nls_table. These probably need to be
> changed from 2 arrays into 2 functions, so that the interface allows
> variable size characters for upper/lower (as it does for the uni<->char).
cp932 have wide-width ASCII characters, so adding
lower/upper functions needs modifying tables.
> There is a little more work to this since a few fs access the arrays (and
> expect an 8-bit value). vfat, ncpfs & I think Andrzej's patch makes msdos
> use these arrays.
Hmm, OK.
> Feel free to take my patch + the missing codepages and work on this if you
> like. I'm certain you know more about how 16-bit charsets work in real
> life than I do.
Thanks. I shall hack it, since current patch cannot handle
wide charsets well yet...
-- gotom
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed May 31 2000 - 21:00:24 EST