From:  Postmaster <>
Date:  21 Dec 2005 03:54:48 Hong Kong Time

Re: Saving accents and umlauts using midas


Hi Rolf,

I recently spoke to a translation service about multilingual data-driven
sites and they recommended that the database you are submitting your content
to should be set to UTF-8 encoding.

I beleive you also use contentType="text/html;charset=UTF-8" on the page
that contains the editing component.

If you are submitting to a MS SQL server for example the data fields to use
should be nvarcar (for simple text) and ntext (for memo).

I did some tests with our midas editing component and Japanese Kangi
characters stored in a SQL database and it seems to work fine.

I also have my browser encoding set to "auto-select" language display.

Search Google for more details on UTF-8 encoding and multilingual content.

I am not sure how your java-servlet plays into the mix. The key is to make
sure all parts of the systems are set to the same encoding.

I hope this helps. I am not an expert on the multilingual CMS systems, It
would be great to hear from other on this subject.

Please post a follow up on this if you resolve your issue.

Best regards,

On 12/19/05 5:09 AM, in article, "Rolf Wouters"

> Hi
> I'm incorporating the midas editing component of mozilla in our CMS tool.
> The problem I'm having is that when I save a page containing letters
> with accents or umlauts (I work for a German company) using JavaScript,
> it seems as if the page is being saved with the wrong encoding.
> Anyone an idea on how I can specify that these characters should be
> saved "as is" ?
> After the content is grabbed with .innerHTML it is passed to a
> java-servlet, which saves the actual page.
> greetz
> Rolf