[sane-devel] frame and batch mode

[m. allan noah]

On Dec 18, 2007 12:37 PM, Jonathan Buzzard <jonathan at buzzard.me.uk> wrote:
>
> On Tue, 2007-12-18 at 08:40 -0500, m. allan noah wrote:
> > On 12/18/07, Giuseppe Sacco <giuseppe at eppesuigoccas.homedns.org> wrote:
> > > Il giorno mar, 18/12/2007 alle 12.00 +0000, Jonathan Buzzard ha scritto:
> > > > On Tue, 2007-12-18 at 12:13 +0100, Giuseppe Sacco wrote:
> > > [...]
> > > > > So, my questions: is there already any standard for those action? Is
> > > > > there any defined rule for how to name a backend parameter like
> > > > > "--frame-number"? Is there any way to know what feeder types are
> > > > > available at a given time?
> > > >
> > > > Good grief, those feelings of déja vu. are pretty strong at the moment.
> > > >
> > > > The basics are that as it stands the SANE standard is heavily geared to
> > > > transmission scanning on flatbed scanners. Understandable as it probably
> > > > accounts for 99% of users requirements.
> > >
> > > Is this problem "solved" with SANE2?
> >
> > would both of you please excuse my ignorance, as i primarily deal with
> > ADF machines, but-
> >
> > why does the front-end need to be involved in the movement at all? can
> > the backend not detect the additional slides and move the feeder
> > automatically? perhaps i am not picturing the mechanism correctly...
> >
>
> Imagine I have just stuck an APS adaptor into my film scanner and loaded
> up a 40 frame APS film. I wish to scan *one* frame which I happen to
> know from the contact print I got when they where developed.
>
> How without the front end telling the scanner which frame to advance to
> and scan do you propose scanning this? From memory a TIFF image from an
> APS frame on my scanner is about 30MB and takes about 1min over 400Mbps
> Firewire. Scanning the lot is utterly impractical.
>
> With 35mm film, I load the strip into a holder and insert the holder. I
> want to scan just two frames from the possible six in the holder, and
> they are frame 2 and 4. Oh and I want to scan 4 first so that it is not
> sticking out the scanner with dust settling on it.
>
> Does that illustrate the point?

yes- though i did have to lookup what APS was :)

the original question was what to name the SANE options that would
control this mess, and i suppose what option type they should be.
sounds almost like a comma-separated list:

4,2 or 4,2-1 if you wanted to skip #3. that sounds a bit like the
gamma vector control that some backends use...

allan
-- 
"The truth is an offense, but not a sin"