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

List:       kde-core-devel
Subject:    Re: PATCH: slavebase.cpp [Correct patch]
From:       Dawit Alemayehu <adawit () kde ! org>
Date:       2001-07-23 9:39:24
[Download RAW message or body]

On Monday 23 July 2001 00:32, Waldo Bastian wrote:
> On Sunday 22 July 2001 04:34 pm, Dawit Alemayehu wrote:
> > On Sunday 22 July 2001 18:14, Waldo Bastian wrote:
> > > On Saturday 21 July 2001 11:38 pm, Dawit Alemayehu wrote:
> > > > Hi,
> > > >
> > > > Ooops, sorry that was the worng patch :(  Here is
> > > > the correct one.
> > >
> > > I don't think this is correct. Can you explain to me why it is
> > > necassery to send medata to the slave _after_ the mimetype has been
> > > send?
> >
> > Well this is because some clients cannot or do not connect to the
> > mimetype signal and thus will never get an essential metadata.  khtml is
> > an example. And connecting to that signal did not work in khtml for me
> > for whatever reason even after Dirk corrected my typo :)  Hmm... perhaps
> > I should try it again since my setup was sort of messed up the past few
> > days.  anyways, I will try to re-do this at the khtml level and if that
> > works I'll revert this patch.
>
> This doesn't answer the question, why does the application need to send
> metadata when the request has already been processed? When the mimetype
> signal is sent to the application, the request has already been send to the
> server, why do you want to send metaData at this late stage?
>
> > > The correct way would IMO, be a simair approach as followed for
> > > incoming meta data where the metadata of mltiple meta-data cmd's is
> > > accumulated. Your patch results in unnnecassery communication if I
> > > understand it correctly.
> >
> > Yes it does result in an extra call.  However, we cannot fix this using
> > the accumulation approach since it would break the ability to get a
> > correct filename to the save as dialog when kio_http receives a
> > Content-Disposition header.  IIRC this is the reason why the eimttion of
> > the meta-data was added to the mime-type in the first palce ?
>
> I have no idea. Your patch changes the way how metaData is sent to the
> io-slave. What you describe is about sending metaData FROM the io-slave to
> the application.

Never mind.  While trying to reply to this response I actually figured out what you
meant about accumulating the metadata.  Doh!!  :) Reverting patch.  Sorry, I am
a bit slow at times...

Regards,
Dawit A.

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