On Mon, 2009-12-21 at 17:11 +0100, Hans de Goede wrote: > Hi, > > On 12/22/2009 02:26 AM, David Lehman wrote: > > --- > > storage/__init__.py | 2 +- > > 1 files changed, 1 insertions(+), 1 deletions(-) > > > > diff --git a/storage/__init__.py b/storage/__init__.py > > index 2dbbfb8..bc63e43 100644 > > --- a/storage/__init__.py > > +++ b/storage/__init__.py > > @@ -370,7 +370,7 @@ class Storage(object): > > def devices(self): > > """ A list of all the devices in the device tree. """ > > devices = self.devicetree.devices > > - devices.sort(key=lambda d: d.path) > > + devices.sort(key=lambda d: d.path, cmp=isys.compareDrives) > > return devices > > > > @property > > Note that isys.compareDrives only works for > disks, and if something is not recognized it automatically gets > a lower prio then recognized thinks > > So if the bios order is sdb, sda, and you pass in: > > [ "sda", "sda1", "sdb", "sdb1" ] > > You will end up with: > > [ "sdb", "sda", "sda1", "sdb1" ] > > IOW I'm not sure about this one. We already use compareDrives to sort Storage.disks. Why we don't use it to sort Storage.partitioned is beyond me. Comparing by path is worse. It puts the mapper/ devices first, which is just wrong. Granted, I don't have any weird BIOS ordering, but I tried compareDrives and got something like ['sda', 'sdb', 'sda1', 'sda2', 'sdb1', 'sr0', 'VolGroup', 'VolGroup-lv_root', 'VolGroup-lv_swap'], which makes sense to me. So, for disks and for partitioned we should be using compareDrives. For devices, I could live with either alpha-by-name or compareDrives. Opinions? Dave > > Regards, > > Hans > > _______________________________________________ > Anaconda-devel-list mailing list > Anaconda-devel-list@xxxxxxxxxx > https://www.redhat.com/mailman/listinfo/anaconda-devel-list _______________________________________________ Anaconda-devel-list mailing list Anaconda-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/anaconda-devel-list
