Attached are two files.
mail_sample_shift_jis.txt is the one I got from Japan. I pasted the content
in a html form (the browser is IE 5.5 and the encoding set is default which
is Western European (Windows)).
My java servlet takes it as any other form name, value pair and dumps in a
unicode support SQL server. Then I view it back in the browser, the data
looks like in shift_jis_corrupted.txt
I also tried to put the data directly into the database and sent the email
out. The email is full of question marks but looks a little better than the
previous one.
btw what are CR and LF associated with charset and where can I include it in
the header Content-type?
Thanks.
> -----Original Message-----
> From: Marco.Cimarosti@icl.com [mailto:Marco.Cimarosti@icl.com]
> Sent: Thursday, October 12, 2000 5:16 AM
> To: Raghu Kolluru; unicode@unicode.org
> Subject: RE: [OT] problem with shift_jis
>
>
> Raghu Kolluru wrote:
> > My email delivery programs works with most of the charsets
> > but not with
> > shift_jis.
> > Here are the steps that I do,
> > 1) I get a text file from Japan which as the content in the
> > encoded charset.
> > 2) I paste this content in web based UI and store it in SQL server
> > 3) Then I send it out with appropriate content header.
>
> Could you attach one of the shift-JIS files that you received
> (step 1), and
> the same text as you send it out (step 3).
>
> > This works fine with big5,euc-jp, etc but not with shift_jis.
>
> Are you sure that it really works fine?
>
> Notice that random sequences of bytes (over 0x7F) often
> *look* like genuine
> Asian text to someone who cannot read it. Can you (or someone
> else in your
> team) read Chinese/Japanese to see that the text actually
> makes sense? If
> not, you may wish to also send some sample of these.
>
> _ Marco
>
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:21:14 EDT