another Service draft

[Aaron J. Seigo]

On Friday 16 May 2008, Kevin Ottens wrote:
> >    void setDestination(const QString &destination);
>
> Why not pass the destination on the call to load? I'm not sure the semantic
> of "setDestination" makes sense here, it's as if I'd be able to move a
> service.

you could. the use pattern is probably going to be:

create a service
tell it what it refers to (e.g. which online account, which local profile, 
etc)
use it over and over.

this will allow one to write generic handlers that use a service that is set 
up for them without them needing to know the destination/address.

> s/destination/address/i ;-)

;)

> >    virtual ServiceJob* createJob(const QString &operation,
> >                                  QMap<QString, QVariant> &parameters) =
> > 0;
>
> Hmmm, or QMap<QString, QVariant> everywhere, or KConfigGroup everywhere.
> I'd be relunctant to mix both. Hmmmm... actually I'm now wondering, what
> about: KConfigGroup operationDescription(const QString &operationName);

i use it here because this is now in the service job where the extra semantics 
that KConfigSkeleton give us are no longer useful.

> >    void setOperationsScheme(QIODevice *xml);
>
> If it's "Operations" here, it should probably be
> "registerOperationsScheme()" above.

good catch. also making the ctors protected.

-- 
Aaron J. Seigo
humru othro a kohnu se
GPG Fingerprint: 8B8B 2209 0C6F 7C47 B1EA  EE75 D6B7 2EB1 A7F1 DB43

KDE core developer sponsored by Trolltech
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 194 bytes
Desc: This is a digitally signed message part.
Url : http://mail.kde.org/pipermail/panel-devel/attachments/20080516/6f99fcd8/attachment.pgp