[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-core-devel
Subject:    Re: KIOSlaves
From:       Waldo Bastian <bastian () suse ! de>
Date:       2000-01-20 18:00:53
[Download RAW message or body]

On Thu, 20 Jan 2000, Stephan Kulow wrote:
> Waldo Bastian wrote:
> > On Thu, 20 Jan 2000, you wrote:
> > > Waldo Bastian wrote:
> > > > Hi Stephan,
> > > >
> > > > What is the status of the IO-slaves in the make-it-cool-branch?
> > > > What is working and what isn't? What needs to be done before it
> > > > can be merged with the head branch?
> > >
> > > Well, what is lacking most is a developer friendly API for
> > > KIOJob.
> >
> > The API beteen KIOJob to the application or the API between KIOJob
> > and the io-slave?
>
> Between application and KIOJob, KIOjob should internally use KIOSlave
> to handle it's own commuication to the slave.

The current API isn't that bad isn't? The major problem of today's 
IO-slaves seems to be the copying stuff which needs to be done in the 
slaves themselves. I thought the idea was to move the recursive 
copying/moving stuff to KIO. That would mean a change in the API 
between KIOJob and the KIOSlave.

What needs to be changed in the API between KIOJob and Application?

> > > KIOJob would be done by David and me, it's just a mater of time.
> > > But also all kioslaves besides kiofile need to be ported to the
> > > new "no URLS, just paths" kind of communication.
> >
> > The way in which kio_file in the make-it-cool-branch communicates
> > is how it is supposed to be? If so I will start on converting the
> > http slave.
>
> Look at kio_protocol and file/file.h.
>
> > > The rest of the job handling shall be done within KIO(Job).
> >
> > What is currently working and what isn't?
>
> Hmm, KIOJob is in the alpha phase of designing :)

What is missing?

Cheers,
Waldo

[prev in list] [next in list] [prev in thread] [next in thread]