Re: New Control Centre
- From: jamie <jamiemcc blueyonder co uk>
- To: BJörn Lindqvist <bjourne gmail com>
- Cc: Federico Mena Quintero <federico ximian com>, desktop-devel-list gnome org
- Subject: Re: New Control Centre
- Date: Tue, 06 Feb 2007 19:36:11 +0000
On Tue, 2007-02-06 at 20:24 +0100, BJörn Lindqvist wrote:
> On 2/6/07, jamie <jamiemcc blueyonder co uk> wrote:
> > On Mon, 2007-02-05 at 21:19 -0600, Federico Mena Quintero wrote:
> > > El mar, 06-02-2007 a las 12:52 +1100, Russell Shaw escribió:
> > >
> > > > If profiling has to be done to make a menu faster, it is pretty obvious
> > > > the system it is built on is stupidly inefficient and broken, especially
> > > > if said menu is slow on a 10 year old pc.
> > >
> > > Ah, bingo.
> > >
> > > Almost 10 years ago, when GNOME started, we had like three apps we
> > > wanted to put in the menus. So, reading .desktop files from disk didn't
> > > seem like a bad idea.
> > >
> > > Now that we have hundreds of .desktop files, it is not a good idea
> > > anymore to have them scattered all around the disk. You are absolutely
> > > right; the system it is built on is stupidly inefficient and broken!
> >
> > Yes and its fairly easily fixed with tracker once I add .desktop file
> > indexing to it
> >
> > tracker's db is a single file so if it caches all the desktop files so
> > it does not have the disk seek penalty ergo all relevant desktop files
> > could be pulled from it very quickly. Even better is the fact that
> > individual categories can be queried straight from the db so no need to
> > hold everything in memory
> >
> > secondly on systems with spare memory, its db will be in the disk cache
> > and therefore will be as fast as holding it in memory but without the
> > need to waste any (thats why using a db like sqlite is so much smarter -
> > sql is massively more productive to work with and performs tons faster
> > than parsing text files)
>
> I do not understand at all. If Control Center has 40 .desktop files,
> you need to do 80 disk reads. First you read the 40 .desktop files and
> then the 40 icons those files are referring to. How does tracker
> reduce the number of disk reads you need to do? I thought tracker was
> an indexer and a meta data storage.
it is - tracker (or a local sqlite cache) would only eliminate
the .desktop file reads. (The category names would be tags/keywords on
application objects in tracker)
IIRC we already have an multiple icon package format which can be
mmap'ed or has that slipped into disrepair?
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]