32 lines
1.6 KiB
Plaintext
32 lines
1.6 KiB
Plaintext
|
Random plans and ideas for 1.1 and beyond:
|
||
|
|
||
|
* We need set label support for fat16/32. Ideally, we take the code from mtools
|
||
|
that deals with labels and try to get it out of the package. If
|
||
|
that's entirely impossible, maybe a kludge like GParted uses should be
|
||
|
considered *shudder*
|
||
|
|
||
|
* To say it bluntly: libparted is a complete dead end. What we have to do is
|
||
|
separate the backend logic from the core and write backend interfaces so
|
||
|
partitionmanager can make use of other backends apart from libparted. KDE's
|
||
|
own Solid would of course be candidate #1 here; unfortunately, it doesn't
|
||
|
implement much of what HAL does regarding disks, partitions and file systems.
|
||
|
Also, DeviceKit-disks is around the corner and implementations using it for
|
||
|
solid begin to materialize.
|
||
|
|
||
|
* People want mount management (i.e. fstab management) in partition manager.
|
||
|
I've written a small kcm, Mount Manager, as a proof of concept how this could
|
||
|
look like, but I'm not yet happy with it. Eventually, this will / should /
|
||
|
can me merged into partition manager.
|
||
|
|
||
|
* The whole lvm/dm debacle. Nothing much useful can be done about it without
|
||
|
the backend stuff mentioned above, though.
|
||
|
|
||
|
* The interface between core and the file systems needs some more thought.
|
||
|
Ideally we'd be free to use either external helper apps (like we do now) or
|
||
|
libraries to achieve the same things. Libraries in this case being libntfs-3g
|
||
|
and libext2fs. There's one for reiserfs too, iirc. Using libraries we'd
|
||
|
hopefully be able to get meaningful progress status reporting while resizing,
|
||
|
checking and so on.
|
||
|
|
||
|
|