[sane-devel] saned issues for remote scanner access

[Herman Kuiper]

Hi,

> - sanetwain should not call sane_cancel() when sane_open()
> hasn't   succeeded. That should be fixed by the sanetwain author.

Thanks for localizing that. I was aware that sometimes a cancel call would 
trigger a segfault, but haven't had tiem to investigate further. BTW, is it 
possible that something in sane itself has changed regarding this during 
the last few versions? I got a couple of reports on this in the last 
months, before that it seemed never to be an issue.

> - the dll backend should be more intelligent when beeing askes for the
> device   name "" (look at the list of devices and take the first one)
> maybe.

OK - maybe it was backend-dependent as well.

> - sanetwain could look for a specific device name. It can use the
>   list gotten from sane_get_devices() and use the first device to do that.

I'm not sure why I try to open the empty device. On first start, it should 
ask for a list of devices and on next calls, use the last selected one. 
I'll check.

> - Tell sanetwain which device to use. I don't know how that works, but
>   it must be possible otherwise sanetwain wouldn't work for anyone.

In the ini file the currently selected device has an entry somewhere. The 
name should be an exact match of what the backend returns as device name.

My list of critical bugs is getting too long for my liking (a handful, but 
still) so I probably will take some time next week to fix the bugs and 
release a new version. Users having problems should'nt hesitate to bug me 
if that takes too long :-)

	Herman

-------------------------------------------------------------------
Herman Kuiper  --  m: herman@ozuzo.net  --  w: http://www.ozuzo.net
Meerhuizenplein 10K - 1078 TC ASD - t: 020-6171885 - f: 020-6142816