Dave Hill wrote:
>
> Karl Heinz Kremer wrote:
> >
> > On Mon, Jan 24, 2000 at 09:43:39AM +0000, John Vickers wrote:
> > > If I then do:
> > > /usr/local/bin/scanimage -v --device epson:/dev/scanner -T
> > > that also PASSes (scanner whirrs, etc), but only the first time I try it.
> >
> > ... that's a known problem. You can only do this once and then you have to
> > reset the scanner. I don't have an idea yet why, but this is on my list
> > of things to work out. It's no real problem, because it only happens
> > when you use the -T switch, the "normal" scanning works just fine (at
> > least for me).
> >
>
> There are two problems here.
>
> One is that scanimage -T somehow leaves the scanner in a state
> that it will not respond to the "inquiry" command. I noticed
> this with my GT7000 but everything else worked OK.
I found that the "initialise" command, issued just before the "inquiry"
command also fails, but the status of the "initialise" command isn't
tested
in the snapshot or in 0.1.7
Are we leaving the scanner in a strange state, or the sg SCSI driver ?
(I don't know).
> The second problem is the "segmentation fault". You will find
> that this will also happen if your scanner fails to identify
> itself for any reason (including being powered off 8-)).
>
> The fix for this is as follows. Sorry I haven't downloaded the
> latest snapshot yet, but the attached diff is how I fixed
> Christian's 0.1.3 version.
>
> The reason for moving the lines is that if the dev_name is
> filled in, the sane main code assumes that the driver init'ed
> OK and all fields are valid. By deferring filling in the name
> until the end, that is the case!!
I don't have the source in front of me, but it looks like the right idea.
A SEGV is always a bug. Anyone going to patch the snapshot ?
Maybe there should be an error message saying "Is your scanner powered on
?"
John.
-- Source code, list archive, and docs: http://www.mostang.com/sane/ To unsubscribe: echo unsubscribe sane-devel | mail majordomo@mostang.com
This archive was generated by hypermail 2b29 : Wed Jan 26 2000 - 01:44:56 PST