On 12 May 2010 15:39, Damyan Ivanov <[email protected]> wrote: > No reason to Conflict/Replace/Break $self, I guess? Is there a problem > with a normal upgrade?
I think so, since the new version depends on acpi-support, the latter needs to be installed, but it can't, since the old version conflicts with it. In this way, the old eeepc-acpi-scripts is first removed, than the new eeepc-acpi-scripts preinst removes old configuration files that would conflict with acpi-support, than the new packages are installed. It could be that removing obsolete conffiles in preinst is enough though, I'll give it a try. > True. Ideally, toggling of wireless (and everything else) should be > done by the kernel. I think things are moving in that direction, at > least the kernel does its job on model 901 and our scripts are used > only for notification. I submitted #581312 I'm working on a simple script that should work with decently recent kernels - with or without rfkill-input. I'm afraid of the quirks of rt2860 though, and since I've changed my card with an atheros (best 20$ investment you can do on a 901; ath9k has reached an amazing level of functionality on pci-e cards) I can't test them. If this change is to be made, I'd suggest making it land in unstable as soon as possible: since testing on a wide range of systems will most probably show some regressions, the farther we are from the release the safer we play. Cheers, Luca _______________________________________________ Debian-eeepc-devel mailing list [email protected] http://lists.alioth.debian.org/mailman/listinfo/debian-eeepc-devel
