Hi,
On Fri, May 11, 2001 at 01:11:57PM +0000, Paul Nixon wrote:
> Maybe there's something messed up in my libraries. What should libsane.so.1
> be linked to?
Nothing. There should be no /usr/local/lib/sane/libsane.so*.
"ls /usr/local/lib/libsane*" should output something like that:
-rw-r--r-- 1 root staff 162912 Mai 12 12:49 /usr/local/lib/libsane.a
-rwxr-xr-x 1 root staff 664 Mai 12 12:49 /usr/local/lib/libsane.la
lrwxrwxrwx 1 root staff 16 Mai 12 12:49 /usr/local/lib/libsane.so -> libsane.so.1.0.4
lrwxrwxrwx 1 root staff 16 Mai 12 12:49 /usr/local/lib/libsane.so.1 -> libsane.so.1.0.4
-rwxr-xr-x 1 root staff 81965 Mai 12 12:49 /usr/local/lib/libsane.so.1.0.4
> Please clarify. I would have thought that since scanimage is able to produce
> a perfect full-color scan (albeit 26MB) that this would indicate a frontend
> problem. No?
Maybe. But there are other possibilities: Timing problems, the way the
backend is accessed may be a bit different between frondends,
different backends are used (/usr/lib and /usr/local/lib).
> > No, scanimage and xscanimage are separate front ends, the only similarity
> > is the name and common contributors AFAIK.
>
> You're right. It's just confusing that scanimage is part of the backends but
> is in the frontend subdir.
I think the idea was to add at least one working frontend to the
package and put the more complicated and/or advanced ones into
seperate packages. Scanimage doesn't need X, gtk or gimp so
sane-backends doesn't depend on them.
> Is there something about kernel 2.4.x that has broken Sane?
Yes, the timeout problems concerning SCSI generic. But I don' know if
there is any relation to USB.
Bye,
Henning
-- 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 : Sat May 12 2001 - 03:42:27 PDT