[sane-devel] Primefilm 1800i (more ideas)

[Marcio Luis Teixeira]

I forgot to mention, the *absolute* first step is to try to avoid writing
code altogether.
Being lazy is always a good way out, so look for an existing backend, or an
existing
non-sane driver you can adapt. I recommend looking at the Linux Hardware
Database
(lhd.zdnet.com) and the Linux USB hardware pages (www.linux-usb.org).
That's
where I first learned of the command line interface to my scanner, which I
eventually
adapted into a SANE backend. Those pages are also a very good place to
learn  about
other people that would be interested in your backend. Be sure to add your
scanner to
the database if it is not already there. This will be a good way to
advertize your intent
to write a backend. When you get to the "testing" phase you'll appreciate
having people
to help out.

If you can't find any existing code, then one good trick is to open up your
scanner and take
note of the chips in it. Then, ask online to see if anyone has heard of it.
Chances are some
other scanners use the same chipset, and if so, there is an extremely good
chance
that the backend for that scanner will also drive your scanner.

Even if no existing backend supports your scanner, having chipset
information
can be helpful. If you scanner happens to have a "scanner-on-a-chip" IC,
you're very likely to be able to find printed documentation from the
manufacturer.
This may provided you with clues as to the protocol.

In my UMAX wiki, I have a couple links to sites that let you look for chip
information:

    http://umax1220u-sane.sourceforge.net/index.php?UmaxScanners

Anyhow, that's all I can think of for now... (probably a lot more than you
asked for, no?)

Cheers,

Marcio Luis Teixeira