[sane-devel] Epson Perfection 1250/Photo 64-bit

[Ralph Little]

Hi,

On 2020-09-14 6:49 a.m., Charles Lindsey wrote:
>
>
> On 13/09/2020 19:42, Ralph Little wrote:
>
>> I did drag out my 1260 and I cannot reproduce this problem on my
>> machine. According to the backend setup parameters, they do appear to be
>> practically identical machines internally, but unfortunately I cannot
>> see this problem.
>> What I *do* see is a narrow yellow band on the right side of the scan
>> both in preview and not. It is a hard band which is likely to be caused
>> by a bug rather than something on my test print or scan aperture.
>> An example is attached.
>> I will spend a bit of time (probably not today) trying to track that
>> down, and who knows? It might be related to your issue.
>
> Before we get too involved, we should establish whether we are using
> the same version of 'xsane'. I have upgraded my libsane1 to 1.0.31 as
> already discussed, including various dependencies. But xsane was not a
> dependency, so I am still using what was there before. 'About' merely
> says it is version 0.999, which is not very helpful, since I gather no
> serious upgrades have happened since 2013.
>
> I am using Ubuntu Bionic LTS, and Ubuntu says its version is
> 0.999-5ubuntu2.
> Ubuntu Focal LTS (which I believe uses 1.0.31) says its version is
> 0.999-8ubuntu2, and that may be closer to what you are using. I could
> easily download the source and compile either of those. Looking at the
> output of "nm -nD `which xsane`", it would seem that it calls
> sane_read (or maybe sane_start) to actually set a scan in progress.
>

Well we can establish what versions of the backends that xsane is using
by looking at the diag output.
xsane should be using exactly the same backends as scanimage.

If you want to be absolutely sure, then you can run xsane thusly in a
terminal:

SANE_DEBUG_DLL=10 xsane >diag.out 2>&1

...and for double certainty make sure that the preview still gives you
the sane effect.

Then attach the diag.out file here.

Cheers,
Ralph