Re: Shift_JIS conflict with CSS?

From: Shez (UseReplyAddress_at_nospam.uk.invalid)
Date: 09/01/04 

Date: Wed, 1 Sep 2004 20:38:18 +0100

In the faraway land of sci.lang.japan, dc <d3ntaku@gmail.com> said:
>> Anyhow, if anyone has tips on using Open Office (western version) for
>> mixed language documents and webpages that include Japanese, I'd be
>> interested to hear them. My needs are basically for English stuff with
>
>I find OpenOff/write basically creates awful HTML, the preview is
>nothing like a browser, even for basic word-processing, and it rips
>out style *** references and reformats everything.

That's odd, I found the preview looked almost exactly the same as in
Firefox and IE. However the fact that it replaces external stylesheets
with internal ones is indeed a pain, as I have to reinstate the "link"
metatag each time.

My previous experience with WYSIWYG programmes was that they wrote
dreadfully convoluted HTML, however with OO the only cluttering it did
was because it was trying to make my HTML compliant by sticking
<p>...</p> round any loose bits of text and repeating my font tags
within every paragraph. I usually stick in one global font tag that
spans from <body> to </body>, but despite its elegance this is
unfortunately considered bad HTML. (One of my niggles with the evolution
of HTML is that elegant and intuitive constructions are being replaced
by obscure long winded ones, e.g.

        background=pic.jpg
by
        STYLE="background: url(pic.jpg) repeat scroll"

>Nvu does a good job for S-Jis pages:
>http://www.nvu.com/

I don't use Shift_JIS any more as I found it caused my other special
characters such as opening & closing quotes, em-dashes and so on being
rendered as mojibake. UTF-8 fixes all this, but until OO the only editor
I had which understood UTF-8 was JWPce.

>but for UTF8 my text editor comes up with gibberish (editplus).

Ordinary ASCII type editors show UTF-8 as triplets or pairs of bytes,
however this isn't a problem as long as you didn't want to insert any
new unicode or proof-read what's there: I find I can cut and paste the
stuff and resave the file without any corruption of the unicode. So for
general tweaking of webpages I often just use Notepad or somesuch
without any problems even though they show the Japanese etc as
indecipherable strings of hieroglyphics. Much worse are programs that
are aware of UTF-8 but can't edit it, e.g. NoteTab, which bizarrely
tells you "this is a Unicode file" and then proceeds to strip out all
the Unicode from it.

The bottom line is that if you want a plain text editor that can input
and display Japanese, do dictionary look ups, and save your work as
UTF-8 or Shift_JIS, then JWPce fits the bill very nicely. The only
drawbacks are that it's a bit of a pain if you're doing much English
writing due to lack of smart quotes etc, and the fact that it displays
unicode punctuation in a weird monospacing as if it were Japanese. Also
you have to know your charset metatag syntax as it won't stick them in
for you since it wasn't intended as a HTML editor.

-Shez. 

-- 
______________________________________________________
Revenge is a dish best served cold -- Klingon saying
______________________________________________________
Take a break at the Last Stop Cafe: http://www.xerez.demon.co.uk/
Reply-to address for email: mailreply AT xerez.demon.co.uk