There's a little error in the lineart-code.
I did mail the correction to Kevin, the author.
So in his next release of snapscan you will find
the fixes. Kevin has just passed his examinations
and is moving. That leaves him little time for
computing at all.
If you need it right now - without waranty:
around line 2630 yu'll find the code:
case MD_BILEVELCOLOUR:
p->format = SANE_FRAME_RGB;
/* the scanner uses depth 1, but the frontend will use 8 */
p->bytes_per_line = 3 * p->pixels_per_line;
p->depth = 8;
break;
now add the lines:
case MD_LINEART:
{
p->depth = 1;
p->bytes_per_line = ( 3 * p->pixels_per_line ) / 8 ;
}
and continue with the original code:
switch (pss->pdev->model)
{
case SNAPSCAN310:
By the way: I found another bug: When scanning overt the
net with the scanner attached to another computer I'll get
wrong data. Something with sane_net does not work exactly.
I couldn't find the reason yet. Obviously there's a difference
between expected data and received data that mangles the
scan-lines. Even that should be fixed with the next release
of scanimage - and by then even the VUEGO-310-Scanner should
work with the snapscan-interface.
Regards
Wolfgang
-- Source code, list archive, and docs: http://www.mostang.com/sane/ To unsubscribe: echo unsubscribe sane-devel | mail majordomo@mostang.com