[cups.general] Details of compression=gzip

[Michael Sweet]

On Aug 15, 2011, at 2:13 PM, Steve Bergman wrote:
>> You should get a log entry for every file that gets compressed, and we don't do anything special to omit files that are already compressed, for example, we just copy and compress the file(s) in the job and use the compressed version, even if it would be larger...
> -------
> 
> Thank you for the response. I've read, in another thread here, that raw queues don't get compressed. All of my printers use either the standard postscript ppd, or the okidata/ibm 9 pin ppd. However, some jobs coming, from software that outputs straight ascii to the printer, get queued with 'lpr -l'. Would that prevent them from being compressed?

The IPP backend will only compresses "raw" print files - what we normally recommend is to setup raw queues pointing at the remote CUPS server (these are treated as "hardwired" shared printers) so that all of the filtering happens on the server. This will also ensure that print data is compressed going over the WAN link to your CUPS server.

________________________________________________________________________
Michael Sweet, Senior Printing System Engineer, PWG Chair