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

List:       kde-core-devel
Subject:    Re: Reflexion about NetAccess (enter_loop/exit_loop)
From:       Stephan Kulow <coolo () max ! tat ! physik ! uni-tuebingen ! de>
Date:       2000-06-28 10:23:20
[Download RAW message or body]

On Wed, 28 Jun 2000, David Faure wrote:

> On Wed, Jun 28, 2000 at 11:38:38AM +0200, Stephan Kulow wrote:
> > David Faure wrote:
> > > 
> > > Forcing serialisation is fine with me. Just like you can't make
> > > two system stat() at the same time (because they're _synchronous_ !),
> > > you shouldn't be allowed to do two network-transparent stat()
> > > at the same time _IF_ using NetAccess. For async I/O, KIO::Job does the job
> > > just fine.
> > > 
> > I think, caring for it in konqueror is the right thing.
> Not sure what you mean here - after tonight, konqueror won't use
> NetAccess at all anymore (is that what you meant?)
Right. I was saying that konqueror misused it
> 
> > We should only make sure the developer knows that using NetAccess
> > from a slot is causing him problems.
> > This has to be a) documentation and b) an assert in the right place.
> > Then we can only hope people a) care about documentation and b) are
> > testing good enough before releasing :)
> 
> Yes :-)
> 
> Note that NetAccess was written with the opposite in mind: it keeps
> track of a list of temp files, for simultaneous downloads. This will
> go away at the same time, AFAICS.
> 
How was this supposed to work? I always thought the list of temp files
is for deleting at the very end of the application in case noone called
removeTempFile

Greetings, Stephan

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