Bug 518756

Thanks for finding that Ubuntu bug report.  Can you see if downgrading to ghostscript-8.64-12.fc11 fixes it? (I have a printer that exhibits this problem here so I'll get to it myself eventually if not...)

The problem remained after downgrading to ghostscript-8.64-12.fc11 (from koji) and restarting cups.

I seem to have made some sort of progress today:

I edited hpcups.drv so that the stuff in the generated ppd for PageSize, for my printer and just normal A4 paper, matched the stuff for PageRegion.

This gets cupsInteger0 set to 26.

This appeared fix it!

I have no idea what the proper fix should be (or even what component).

Created attachment 358614 [details]
The .ppd file an unmodified hpcups generates for my printer

Created attachment 358615 [details]
My "fixed (just for A4)" .ppd file

Created attachment 358616 [details]
The change to hpcups.drv

And now I've noticed that compared to the hpcups ppd shipped in the hplip tarball, and the generated .ppd, this line is missing from the generated one:

*RequiresPageRegion All: True

Undoing my previous fix and adding that to the ppd in /etc/cups/ppd/ should also solve this problem.

Brilliant, that's the fix.  Thanks!

In detail: /cupsInteger0 is used by hpcups for holding the PCL Page Size ID.  In this case, 26 corresponds to A4.  The PageRegion codes all contain the cupsInteger0 setting, but the PageSize codes do not.

The CUPS pstops filter decides whether to emit PageSize or PageRegion codes based on a variety of factors:

/*
  * This function determines what page size code to use, if any, for the
  * current media size, InputSlot, and ManualFeed selections.
  *
  * We use the PageSize code if:
  *
  * 1. A custom media size is selected.
  * 2. ManualFeed and InputSlot are not selected (or do not exist).
  * 3. ManualFeed is selected but is False and InputSlot is not selected or
  *    the selection has no code - the latter check done to support "auto" or
  *    "printer default" InputSlot options.
  *
  * We use the PageRegion code if:
  *
  * 4. RequiresPageRegion does not exist and the PPD contains cupsFilter
  *    keywords, indicating this is a CUPS-based driver.
  * 5. RequiresPageRegion exists for the selected InputSlot (or "All" for any
  *    InputSlot or ManualFeed selection) and is True.
  *
  * If none of the 5 conditions are true, no page size code is used and we
  * unmark any existing PageSize or PageRegion choices.
  */

So when the PPD contains '*RequiresPageRegion All: True' it forces the PageRegion code (including the cupsInteger0 setting for the PCL Page Size ID) to be emitted.  This allows the hpcups driver to know to send the appropriate PCL command to the device.

This can be achieved for all PPDs by adding this to hpcups.drv:

Attribute "RequiresPageRegion" "All" "True"

*** Bug 515470 has been marked as a duplicate of this bug. ***

hplip-3.9.8-10.fc11 has been pushed to the Fedora 11 testing repository.  If problems still persist, please make note of it in this bug report.
 If you want to test the update, you can install it with 
 su -c 'yum --enablerepo=updates-testing update hplip'.  You can provide feedback for this update here: http://admin.fedoraproject.org/updates/F11/FEDORA-2009-6959

The patch in hplip-3.9.8-10.fc11 fixes all the "HP" PPDs, but not the "Apollo" ones, there is another Manufacturer section hidden way down at the bottom of hpcups.drv.

Oops, I put it a little further down in the file than I intended to.  Thanks for catching that!

hplip-3.9.8-12.fc11 has been pushed to the Fedora 11 testing repository.  If problems still persist, please make note of it in this bug report.
 If you want to test the update, you can install it with 
 su -c 'yum --enablerepo=updates-testing update hplip'.  You can provide feedback for this update here: http://admin.fedoraproject.org/updates/F11/FEDORA-2009-6959

hplip-3.9.8-12.fc11 has been pushed to the Fedora 11 stable repository.  If problems still persist, please make note of it in this bug report.