Note by Toltec
dfaure at klaralvdalens-datakonsult.se
Tue Jul 13 21:09:57 CEST 2004
On Tuesday 13 July 2004 20:58, Joon Radley wrote:
> Changed. We need to define the legal format of a UID to avoid encoding
Yes. But I think it's simple to define: us-ascii only.
> > All the tags are mixed-case (like CreationDate) instead of
> > lower-case (like creation-date).
> Fixed. Me and old specifications. :-)
The validation script (see doc/kolab-formats/validation/kolablint) says
that it's still <Note> instead of <note>, and Version instead of version.
After fixing those two things, the XML validates.
> > <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
shouldn't be there?
> The XML library encodes the carriage return. This is present in the body of
> the note.
I have no doubt about that. I was just wondering if this was going to lead
to windows<->unix compatibility problems (e.g. a strange square might appear
on unix for this \r character). There are two ways to deal with that:
1) making the reader code on both sides robust (ignoring \r on unix and adding them on windows)
2) defining how the XML should be, e.g. no \r in the XML.
Hmm, given that I can't hack base64-encoded data to fix the title-case tags,
I can't test how the KDE notes program reacts to this note, so I don't know
if this is a real problem yet.
David Faure -- faure at kde.org, dfaure at klaralvdalens-datakonsult.se
Klarälvdalens Datakonsult AB, Platform-independent software solutions
More information about the Kolab-format