[9fans] mothra char set

[9fans] mothra char set

[rob]

> I made UTF encoded Japanese Web page for testing this.

> http://basalt.cias.osakafu-u.ac.jp/plan9/p9index.html

Most modern web browsers would be able to read this document
properly if it contained the proper "charset=utf-8" information.

[9fans] mothra char set

[okamoto]

I forgot one.

Is the latin1 charset suitable for those ISO-8859-x language?

Kenji

[9fans] mothra char set

[okamoto]

Now, my mothra can read jis/euc/sjis encoded Japanese Web page.

I want to test this version for other iso-8859-x languages.
If you know some sites showing those languages, please let me
know.

thanks in advance,

Kenji

[9fans] mothra char set

[okamoto]

>Now, I can
>see Japanese document on my screen.

I made UTF encoded Japanese Web page for testing this.

http://basalt.cias.osakafu-u.ac.jp/plan9/p9index.html

Kenji

[9fans] mothra char set

[okamoto]

Mothra again.

I changed a little bit rdhtml.c fot testing to read UTF-encoded
Japanese.  This is the pre-start point, I know.  This is only to
check why we cann't see Japanese text by mothra.   Now, I can
see Japanese document on my screen.

I'd like to know some advise what 'd be better to proceed this line.
Don't we need some mechanism to change font automatically?

1) replaced all the fontlist[4][4]  to 
	"pelm/unicode.8",0,0,0,
	"pelm/unicode.9",0,0,0,
	"pelm/unicode.9",0,0,0,
	"pelm/unicode.9",0,0,0,

for all four sets (this is just for simplicity).

2) edit some lines in latin2utf() in the same file
like
	............
	for(l=latin;*l;) {
		nutf+=chartorune(&r, l);
		l+=runelen(r);
	}
	..............
	for (l=latin;*l;) {
		l+=chartorune(&r, l);
		u+=runetochar(u, &r);
	}

[9fans] mothra char set

[Tom]

On Sep 8,  6:00pm, forsyth@caldo.demon.co.uk wrote:
> Subject: [9fans] mothra char set
> >>latin1 to utf on input.  Obviously if you prefer to
> >>read JIS, you can change latin2utf appropriately.
>
> the content-type should specify charset= to distinguish the
> cases.
Should, but often doesn't.
>  the catch (as always with this multiple-charset clumsiness)
> is that you'd probably need to replicate all the translations of tcs.
> ugh.

Maybe you could pipe text through tcs -- or build a tcs library,
so that this ugh never occurs again.

-- 
Tom Duff.  Hear, ye children, and attend to my lips.

[9fans] mothra char set

[forsyth]

>>latin1 to utf on input.  Obviously if you prefer to
>>read JIS, you can change latin2utf appropriately.

the content-type should specify charset= to distinguish the
cases.  the catch (as always with this multiple-charset clumsiness)
is that you'd probably need to replicate all the translations of tcs.
ugh.