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

List:       kfm-devel
Subject:    Re: Should fast mime detection use a stat call?
From:       Mark <markg85 () gmail ! com>
Date:       2013-06-17 13:44:12
Message-ID: CAPd6JnEmxmZEbwSN95uJ6_HVV6=tjzjYVpLqOyB+oDrOneK9aQ () mail ! gmail ! com
[Download RAW message or body]

On Mon, Jun 17, 2013 at 5:38 PM, David Faure <faure@kde.org> wrote:
> Le lundi 17 juin 2013 15:13:36 Mark a =E9crit :
>> @Frank, do you happen to know what dolphin does here? Is it passing
>> this mode property?
>
> I thought you profiled this, don't you have backtraces?

I tested KMimeType directly from a file list without dolphin in
between. A simple sample application that reads out a directory and
puts it in a QList<KUrl>.

>
>> For the rest of this patch. I - sadly - didn't think about looking in
>> the frameworks branch to see if this would still be valid for that.
>> Which it isn't due to QMimeType (awesome work btw, david! Really good
>> to see that stuff in Qt!) so that means this patch is only usable for
>> the KDE 4.x cycle. The patch is a bit invasive and requires quite
>> thorough testing of apps that use mime types. So it won't be accepted
>> for 4.11 which is a long term KDE release. Is it worth it to simplify
>> the patch for some easy speed improvements or should i just close it?
>> Is there even going to be another release after 4.11?
>
> If a patch can speed things up *without* regressions, I see no problems w=
ith
> having it in a 4.11.x bugfix release. But the current patch doesn't reall=
y
> seem to be in that category....
> If there's no way to speed things up without regressions, then yes, it sh=
ould
> be discarded.
>

Oke, i will try to do that asap (this week when time permits it).

> --
> David Faure, faure@kde.org, http://www.davidfaure.fr
> Working on KDE, in particular KDE Frameworks 5
>
[prev in list] [next in list] [prev in thread] [next in thread]