[cups.development] [RFE] STR #2280: The lpd backend does not pay attention to -h option of DeviceURI has +banner=yes in it.

[wtautz]

Michael Sweet wrote:
> [STR Closed w/o Resolution]
>
> OK, for starters the "-h" option of the lpr command just turns the CUPS
> banner pages off (job-sheets=none). This simulates the old lpr
> functionality but the backend does not receive any notification that
> "job-sheets=none" was passed with the job - it just gets the final print
> data in one blob or stream.
>
> Now, we *could* have the lpd backend check the job-sheets attribute via an
> IPP request, but that wouldn't produce the intended result since you'd end
> up with double banners when job-sheets != none.
>
> Also, if you have gone to the trouble of setting banner=yes in the device
> URI, chances are (and we're just going off years of user feedback here)
> you don't want users to disable banners, typically because you need them
> for local policy/security reasons.
>
> So, we won't be adding code to set banner=no when job-sheets=none in CUPS.
> You can add code to do this fairly easily (take a look at some of the CUPS
> book examples, or look at the lpstat or lpq code for inspiration), and
> either modify the lpd backend itself or wrap it with your own program that
> passes a modified DEVICE_URI environment variable through to the original
> backend.
>   
 Yes, a wrapper. I just got a copy of your
book. I've been meaning to do some experiments with backends but
other stuff keeps getting in the way. I guess tea4cups could be taught
to do this :-)

Walter