[cups] USB connections and the printer's device ID.

[Brian Potkin]

On Wed 05 Dec 2018 at 12:20:35 +0100, Johannes Meixner wrote:

> 
> Hello,
> 
> On Dec 4 19:02 Brian Potkin wrote (excerpt):
> > I have never experienced it, but believe there are printers that do not
> > supply, for whatever reason, a device ID to CUPS. In that situation I am
> > led to believe that CUPS cannot create a USB URI. Correct?
> > 
> > I think that one way to print to such a printer would be to set up a
> > queue with
> > 
> > lpadmin -p testq -v file:/dev/usb/lp0 -E -m <PPD>
> > 
> > I know this works, but is there anything to militate against using it?
> 
> you may have a look at
> https://en.opensuse.org/SDB:Using_Your_Own_Backends_to_Print_with_CUPS
> therein in particular the section about
> "A careless backend for a single USB printer"
> and
> "A backend that sends its input into a file for debugging"
> that also talks a bit about the 'file' DeviceURI.

"A careless backend for a single USB printer" is very helpful and puts
the issue into perspective for me. A nicely written section. Thank you
for writing it.

As for my second query: I do not think this will work as the URI is not
associated with /dev/usb/lp0. Making one up is not sufficient. It isn't
important for me but did come into my mind while looking at an HPLIP bug
report which seemed to indicate that the printer was discovered by the
hp backend but not the usb one.

Regards,

Brian.