Re: May sane_init fail?

Oliver Rauch (oliver.rauch@Wolfsburg.DE)
Sat, 01 May 1999 20:34:59 +0200

Ingo Wilken wrote:

> > Anyway. I'm now writing a patch to sane 1.0.1 with my suggested
> > change. I suggest three valid responses from sane_init():
> > SANE_STATUS_GOOD - Success
> > SANE_STATUS_NO_MEM - Ran out of memory
> > SANE_STATUS_UNSUPPORTED - Not supported on this machines current
> > configuration
> > Would there be need for more return values?
>
> What's the suggested return values for a backend/SANE version mismatch?
> I'm currently using SANE_STATUS_UNSUPPORTED (seems logical), but
> SANE_STATUS_INVAL could be used too.
>
> BTW, it seems that almost no backend checks the version number, and most
> backends distributed with 1.0.x are listed as version 0.something in the
> "supported scanners" webpage - not good.
>

Hi Ingo,

the frontends do have to chacke the version numbers, a backend is not able to do

a right test for it.

xsane does test it and if the backend responds with a sane major version != 1,
xsane aborts with an error message!

Bye
Oliver

--
EMAIL: Oliver.Rauch@Wolfsburg.DE
WWW: http://www.wolfsburg.de/~rauch

--
Source code, list archive, and docs: http://www.mostang.com/sane/
To unsubscribe: echo unsubscribe sane-devel | mail majordomo@mostang.com