[Kst] [Bug 85594] kst should allow datasource via fish (and perhaps others)

[George Staikos]

------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.
      
http://bugs.kde.org/show_bug.cgi?id=85594      




------- Additional Comments From staikos kde org  2004-07-21 19:19 -------
On Wednesday 21 July 2004 11:16, Matthew Truch wrote:
> I agree, for live data, it would 'suck' (technical term).  But for static
> data, it would definatly be useful; and polling shouldn't hurt much (other
> than network bandwidth and longer latencies in checkinig) in this case, as
> the data isn't changinig.

   The bigger problem is this:

- Data sources do their own I/O so each would have to support KIO
- Most data sources use external libraries to read the data, so they have to 
copy the data file down to a temporary file and hopefully can clean it up 
properly
- Therefore we might as well just copy the file ourselves before we call out 
to the data source
- Completion and the wizard "Next" disable functionality basically breaks here 
in the KURLRequester because we can't try to open a new remote file each time 
the user types a character

So it's messy, but still conceivable.