On Tue, Jun 23, 2009 at 6:29 AM, Ozan ?a?layan<ozan at pardus.org.tr> wrote: > Ozan ?a?layan wrote On 23-06-2009 12:48: >> backend/avision.c: Mark Fujitsu ScanPartner 620C with >> AV_LIGHT_CHECK_BOGUS as it currently refuses to work through avision >> after 90 tries: >> > > > BTW, the scanner works well with sp15c backend and it's marked as *good* > in avision.c. So it's possible that the previous patch is not the > optimal solution to the problem if it was working OK some time ago. > > Tested with 1.0.19 and 1.0.20.
I will commit your patch, since it is easy to revert. The avision backend author may be able to comment further on the status. > > And an additional question: Both avision and sp15c seem to support this > device. What is the policy about those situations in sane-backends? Is > there a some sort of *recommended* flag to mark a backend more stable > than other or is the backend selection completely left to the user? > Well, in this particular case I think the sp15c backend probably supported these Avision-built Fujitsu machines better at one time, but that driver is unmaintained now. On the other hand, the avision backend supports a huge number of machines, so unintended regressions can sometimes be a problem. allan -- "The truth is an offense, but not a sin"
