While I discovered this using the srpms from fisher on linuxppc, I suspect the same is true on the other platforms as well. It appears that there is no equivalent to the 'send EOF to eject page at end of print job' option in printconf as there was in printtool. The net result is that the page is not ejected at the end of a printjob. This occurs with the Epson 740 print drivers. I noticed a comment in the magic filters for the 760 drivers suggesting that the author had been able to figure out how to make this sending if the EOF work. This problem definitely has to be resolved if Redhat intends to move over to these magicfilters.
We (Red Hat) should really try to fix this before the next release.
Okay, the 'correct' way to send Postscript to a printer is to perform the LF -> CRLF translation, AND to send EOF at the end of the job. Many Postscript printers are smart enough to do this themselves. magicfilter does this, but the ghostscript wrapper we are using, mfomatic, does not. I will make it do this.
Hmm, this is tricker than I thought. Because of the nature of the foomatic system, it is difficult to tell if I am sending Postscript output after it passes through mfomatic. I could unconditionally stick an EOF on everything I send, but I don't know if this breaks PCL, PJL, PPA, etc. printers. Add to this the nature of the piping system in mfomatic, and this gets rather complex. Hmm.
*** Bug 26471 has been marked as a duplicate of this bug. ***
Yes, it needs to be an option, like in printtool. One setting doesn't work for everyone. :-(
Yes, one setting /does/ work for everyone. It is safe to always send that EOF at the end of Postscript jobs. I'm working with Grant Taylor to figgure out how.
Crutcher, we need an option for this. Sending EOF at the end of a PS job isn't going to help my text print job, is it?
*** Bug 27755 has been marked as a duplicate of this bug. ***
*** Bug 27813 has been marked as a duplicate of this bug. ***
In Wolverine, printing PostScript files to my HP Deskjet 890C using the deskjet, cdj550, or cdj890 drivers works fine. Printing text is another story. I'll create a new bug.
Hmmm, Okay. PostScript jobs want a terminating EOT (octal 004) Text jobs want a terminating FF (octal 014) I'm going to try hard to shoehorn these into the general case, because otherwise I have to extend the config front end to allow for non-generated options, which would be useful, but isn't currently supported. Will finnish after some sleep.
Text jobs now get their terminating '\014' form feed. Working on '\004' end of transmission for postscript.
fixed. ick