[sane-devel] [dev] net backend and timing problems? (pretty weird)

[Henning Meier-Geinitz]

Hi,

On Thu, Sep 06, 2001 at 12:34:19PM +0200, mh wrote:
> when I try to scan over saned (localhost) using the net/plustek backend, the
> frontend crashes with a bt like this:

One problem found: pnm doesn't follow the SANE standard in case of
EOF. It just returns SANE_STATUS_EOF without setting *length to 0.

Further more, saned doesn't check the return status of sane_read, it
just increases the pointer to the buffer by length (which is the same
value as in the last sane_read). I will fix pnm in CVS and do some
more tests.

I don't know if it's a similar error in plustek. On first sight, it
looks like plustek sets *length = 0 in any case.

Bye,
  Henning