Lev A. Melnikovsky wrote: > ...nice to hear from you > > On Wed, 16 Jul 2008 at 3:08pm, Ren? Rebe wrote: > > RR> The user should not be able to mess the timeouts from the environment. > Actually the user is always able to do _anything_ from the > environment(LD_PRELOAD). > > And we do already have SANE_SCSICMD_TIMEOUT and SANE_NET_TIMEOUT, don't > we? > > RR> The backends should know precisely how long a response takes, > RR> especially to return to the user if the scanner got "stuck". > RR> If one device requires it's own timeouts then go fix the backend to set > it. > Well, I don't feel myself competent enough to suggest specific timeouts > for a few scanner models. My particular unit works much better after I > change libusb_timeout=120*1000. I have no idea if this is going to help > others. I'll be completely satisfied if someone provides a general > solution which just works. > Sure the use can, with an own library implementation, Qemu, whatever. But poking those is like poking your PCI registers with a hex editor.
Go fix the SANE backend to set the timeout to 120* whatever for just your specific model. Joe Ubuntu users do not want to google to death how to get the scanner working reliably by setting a environment variable, anyway. Yours, -- Ren? Rebe - ExactCODE GmbH - Europe, Germany, Berlin http://exactcode.de | http://t2-project.org | http://rene.rebe.name
