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

List:       kde-devel
Subject:    Re: Status on indexing system
From:       John Tapsell <john () geola ! co ! uk>
Date:       2005-02-24 13:54:07
Message-ID: 200502241354.07894.john () geola ! co ! uk
[Download RAW message or body]

I fully agree it should be in a seperate process, but apps should be able to 
use it by just linking in a library.  The library can use rpc or soap or dcop 
or shared memory or anything else, but that's transparent to the app.



On Thursday 24 February 2005 13:10, Matteo Merli wrote:
> To acheve better timing (especially on startup) is better to leave the
> search system in a separate process.. so that you don't have to open
> files for the db or other things... (more over the disk reads are
> cached at the DB level so it would very slow to have all in a libray)
>
> Then you can have whatever interface to the search system ( xmlrpc,
> soap, dcop, ....)
>
> On Thu, 24 Feb 2005 11:47:57 +0000, John Tapsell <john@geola.co.uk> wrote:
> > Apps that use an indexing system will probably want to just include a
> > header file, and call c++ functions, rather than talk rpc or soap
> > directly.
 
>> Visit http://mail.kde.org/mailman/listinfo/kde-devel#unsub to unsubscribe <<
[prev in list] [next in list] [prev in thread] [next in thread]