On 1/7/06, Ted Bardusch <[EMAIL PROTECTED]> wrote: > This is all on the exact same machine, yes. The libusb is from > darwinports, looks like 0.1.10a_1 > > On 2.0.2 of NUT the driver connected, but had no awareness of an APC > (as the docs state, newhidups on 2.0.2 only knew about MGE). 2.0.3 > however fails to "steal" the USB connection from the default USB UPS > handler in OS X as the 2.0.2 version did. Exact same machine, only > difference is the newer version of NUT (latest download of 2.0.3).
Hmm, then the question in my mind is whether 2.0.2 really did claim the UPS, or if you just found a corner case where it got some generic USB information, but didn't actually do any HID transactions. The thing is, although the libhid docs say to claim the interface first, you can sometimes still do other operations on the device. If the OS power management routines see your UPS, though, chances are that those other operations are going to interfere with what the OS is doing. On OS X, with supported UPSes (APC, MGE, possibly others), the solution might be to connect to the power management routines, and avoid the whole "steal the USB interface" dilemma. I'll look into that later. For now, a quicker solution might be to create a stub driver that prevents the HID toolbox from claiming the device. I'll try and dig up that code later today. -- - Charles Lepple _______________________________________________ Nut-upsuser mailing list [email protected] http://lists.alioth.debian.org/mailman/listinfo/nut-upsuser
