[sane-devel] [announce] coolscan3 release

[Alessandro Zummo]

On Mon, 17 Dec 2007 08:26:12 -0500
"m. allan noah" <kitno455 at gmail.com> wrote:

> On 12/17/07, Alessandro Zummo <azummo-lists at towertech.it> wrote:
> > On Mon, 17 Dec 2007 10:46:24 +0100
> > René Rebe <rene at exactcode.de> wrote:
> >
> > > > alongside RGB data and this is why you need my program tiffscan
> > > > to read it. that's why the bakend is experimental and not
> > > > suited for mainstream sane. I defined a new frame type,
> > > > SANE_FRAME_RGBX for it.
> 
> there was no consensus at that time, but since then, those of us that
> need these options have plowed ahead and added them to SANE 1. I
> really dont want to see a proliferation of types, but as long as the
> user is required to set some option to get the non-rgb output, and the
> help text for that option explains that the frontend may break, it is
> a workable solution. well, its more workable if we all agree on what
> the new types are :)

 in theory the frontend should not break, since it should check the returned
 frame type. I believe my tiffscan works that way :-D

> maybe you could come back to the table, and we can get this backend into sane.

 I'd love it. 

 from what I've understood we'd need RGBI, GRAYI and RIGIBI. If there's space,
 i' love to have a way to ask the backend for scanner make/model. :)

 if we can make RGBI = 0x10 I will avoid to rewrite the frontend :)


-- 

 Best regards,

 Alessandro Zummo,
  Tower Technologies - Torino, Italy

  http://www.towertech.it