[cups] MacOS clients using n-up printing to CUPS server doubles up the n-up option

[Jeff Dyck]

OMG, after month of on and off working on this finally got it working - stupidly my biggest error was pruning back my compiled test environment (thinking it would help me focus on just getting the basics working) only to realize weeks later that NOT having the sssd/realm Active Directory support *AND* having some AD groups set as admin in the cups-files.conf was killing it for me.

Anyway, thanks so much for pointing me in the right direction - not only do I have a fix for this specific issue, I now have an environment where I can easily compile test versions of CUPS.

Probably going to put up a new question asking about the filter wrappers though as I’m also dealing with issues with the PaperCut provider filter (backend?) which doesn’t pass on a bunch of the job attributes properly.

Jeff

> On Feb 23, 2024, at 5:32 PM, Douglas Kosovic <doug at uq.edu.au> wrote:
> 
> Hi Jeff,
> 
>> So just to confirm: this seems to be a side effect of not advertising
>> the queues as _cups (which I definitely am not doing), but changing
>> that to add the _cups PTR would resolve this issue while adding new
>> ones.  The other option is to custom compile CUPS with your patch to
>> skip this (and some other) attribute when the incoming print job is
>> via IPP/Airprint…
> 
> Correct.
> 
>> Now I just need to figure out how to compile CUPS and add the patch ;).
>> That may be complicated somewhat in that I’ve got CUPS within a Docker
>> container… I’ll play around and see what I can figure out.
> 
> 
> Alternatively you could write a wrapper script around the pdftopdf filter and remove the "number-up" option (and others).
> 
> The options I excluded in the patch were based on what the pdftops filter excludes because those options would have already been applied by the pdftopdf filter:
> https://github.com/OpenPrinting/cups-filters/blob/1.28.17/filter/pdftops.c#L67
> 
> I removed "Collate" out of the list of options to exclude in my patch, suspect it might need to be added back.
> 
> 
> Cheers,
> Doug
> 
> _______________________________________________
> cups mailing list
> cups at cups.org
> https://lists.cups.org/mailman/listinfo/cups