Note by Toltec

Tue Jul 13 20:05:42 CEST 2004

On Tuesday 13 July 2004 19:49, Joon Radley wrote:
> Hi David,
> 
> > > Please find attached a kolab.note object encapsulated in a 
> > mail messages.
> > > 
> > > (I am so afraid to ask this) Is this what we discussed?
> > 
> > Looks ok to me.
> > 
> > I'm surprised to see that the XML is base64 encoded - I guess 
> > that's fine, it just doesn't help debugging :)
> 
> The XML library already does some quote encoding and I truly do not want
> have issues around this. Secondly the MIME standard recommends that
> application MIME types should be base64 encoded.

OK. No problem. I opened the mail with KMail, and noticed something though:
The main Content-Type header had a newline in the middle of the boundary,
which broke decoding the message. I had to remove that newline for kmail
to parse the mail correctly. I hope the newline was added by some tool you
used for saving, but wasn't in the original email?

> This avoids encoding issues. Headers must be UTF-7. Since I can not
> guarantee that this wall always be the case I quote encode it. When it is
> parsed by the MIME library it should strip the quotes.

Our MIME library doesn't strip such quotes, I see a subject of
"1D53F6A800E34252B8BA11398FDE6C73" in kmail. Which makes sense,
otherwise actual quotes in the subject (added by users) would be removed !?
Since it's only letters and numbers, is there really a need to quote-encode it?

Now that I can check the XML: it doesn't use the latest version of the tags at
http://www.sonofthor.dk/~bo/kolab/kolabformat.ps
All the tags are mixed-case (like CreationDate) instead of lower-case (like creation-date).

<Body>Kolab Test Note
</Body>
This looks like \r\n, where the \r got escaped... I don't remember off-hand but 
I suppose the XML standard says a newline is enough, i.e. the &#13 shouldn't be there?

-- 
David Faure -- faure at kde.org, dfaure at klaralvdalens-datakonsult.se
Qt/KDE/KOffice developer
Klarälvdalens Datakonsult AB, Platform-independent software solutions