Factoring out standard system information

[Adriaan de Groot]

On Friday 30 December 2005 21:10, Richard Moore wrote:
> On 12/30/05, Tomasz Grobelny <grotk at poczta.onet.pl> wrote:
> > What about using external libraries such as libstatgrab
> > (http://www.i-scream.org/libstatgrab/)? It already supports major OSes,
> > is LGPL licensed and provides decent amount of information.
>
> This library seems decent and its platform support is good. I notice that
> the CPU information functions don't seem to allow you to get information
> about each CPU independently.

Does any of the existing applications make use of that? Not that we need to 
limit the API unnecessarily. It does look good, has decent manpages and for 
FreeBSD at least there's a package of it, so as a dependency it's not very 
problematic.

> > by
> > applications or somehow wrapped up by KDE classes? This is more general
> > question, not only about this libstatgrab case.
>
> I would say it should be wrapped in a KDE API then we could replace the
> backend with another one if we need to later. This would also mean that the
> API would be properly integrated with our documentation etc.

Aye. We can wrapper the backend and provide a single consistent API. In 
particular, _even with_ libstatgrab, we need to provide some OS/KDE specific 
functions (getting swap info in libstatgrab requires elevated privileges; you 
can _get_ the same info from /usr/sbin/swapinfo, but that's icky). 

Is there a place for fd.o in this issue if we just _choose_ libstatgrab to use 
as a backend and wrapper it for our own use?

-- 
These are your friends - Adem
    GPG: FEA2 A3FE Adriaan de Groot
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 187 bytes
Desc: not available
URL: <http://mail.kde.org/pipermail/kde-core-devel/attachments/20051231/5d6f87c6/attachment.sig>