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

List:       kde-devel
Subject:    Re: Add LUKS support to media ioslave
From:       Jan =?iso-8859-15?q?Kl=F6tzke?= <jan.kloetzke () freenet ! de>
Date:       2007-10-23 21:44:57
Message-ID: 200710232344.58136.jan.kloetzke () freenet ! de
[Download RAW message or body]

Am Dienstag, 23. Oktober 2007 schrieb Kevin Ottens:
>
> Does it have anything to do with the patch the suse people apply to handle
> encrypted volumes? If it have nothing to do with it, it might be a good
> idea to take a look at it first?

No, it's not related to the patch from suse despite that I took the ui file 
from it. :-) In fact I did have a look at their patch but didn't feel so 
comfortable with it. It only displays a decryption dialog for one time when 
the medium is attached. If you close the dialog by accident you have to 
re-plug the device to see it again. It also doesn't help for encrypted hdd 
partitions and it offers no facility to 'lock' a volume again after using it.

Maybe they improved it in the meantime but I wanted a solution where the 
different volume states (en-/decrypted, (un)mounted) are visible to the user.

>
> Well, I tend to be very picky about adding feature in 3.5.x, but this one
> would be nice to have there indeed.

Nice ;-)

> Well, Solid knows how to handle encrypted volumes, it already works for a
> while in Dolphin for instance, I just have to polish a bit the UI.
> And for the record the "mediamanager" didn't make it for KDE4, it's gone
> (replaced by better facilities). ;-)

The last time I looked I could swear "mediamanager" was still there. ;-o

> For the patch itself (code wise), at a glance it looks fine. Now, I'm not
> sure about the approach since it 2 times more mimetypes... Which is well,
> understandable seeing the approach you chose and the design brokeness of
> kio media. I'm pretty sure the patch applied by the suse people doesn't
> have this problem (it's more a hotplug reaction thing), so it's really
> worth a look before going for a final solution which could make it for a
> potential 3.5.9 (keep in mind we have nothing like this planned at the
> moment).

I see your point. The additional mimetypes are the price to pay for displaying 
(and being able to manipulate) the different states. As there are even no 
plans for 3.5.9 the suse approach may be sufficient for the time being...

/Jan
 
>> Visit http://mail.kde.org/mailman/listinfo/kde-devel#unsub to unsubscribe <<
[prev in list] [next in list] [prev in thread] [next in thread]