[cups.general] Raw Printing Doesn't work...

[Helge Blischke]

Chris wrote:

> Your suspicions are correct, the r2400 does not understand postscript
> -- however, it does the same thing whether I send it a test page, echo
> text in from the command line or serve the printer up via samba and
> have a windows box with the driver send a job.  The jobs just disappear.
> 
> If I put a ppd on it, and filter locally, it will print a test page.
> But I do not want that...  I need it to be raw, and nothing prints
> when it is set to raw -- keep in mind this the third printer having
> this problem, so it isn't the device.  One of the printers I used to
> have set up this way for more than a year (served from a gentoo system
> though).
> 
> 
> Quoting Paul <paul.conklin at cerner.com>:
> 
>> I do believe I see the problem...
>> I acutally just started testing / playing with 1.4 last week and we
>> use exclusivly raw devices.  the testprint page changed in 1.4 and
>> will no longer work when sent to a raw device.  with 1.3 it would
>> work when sent to a raw device as long as it supported postscript.
>> if you go out and replace /usr/share/cups/data/testprint with
>> /usr/share/cups/data/testprint.ps from 1.3x (or another fully
>> encapsulate print file) and it may work (it also may not after
>> googling your model, it's unclear if the device itself understands
>> Postscript.)  Good luck and let me know
>>> Thanks Paul,
>>>
>>> Please see my previous post for log output.
>>>
>>> Here is the c file:
>>> ^A^A^@^B^@^@^@^A^AG^@^Rattributes-charset^@^Eutf-8H^@^[attributes-
natural-la=
>>> nguage^@^Ben^BE^@^Kprinter-uri^@&ipp://localhost:631/printers/CUPS-
TestB^@^H=
>>> job-name^@      Test
>>> PageI^@^Odocument-format^@^[application/vnd.cups-banner=
>>> !^@^Ljob-priority^@^D^@^@^@2E^@^Hjob-uuid^@-
urn:uuid:90c2de6d-6043-3b43-48b9=
>>> -56d8f9acc485B^@^Yjob-originating-user-name^@
>>> anonymousB^@^Yjob-originatin=
>>> g-host-name^@
>>> localhost!^@^Ptime-at-creation^@^DM=C2=B6=C3=AA<91>^S^@^=
>>> Rtime-at-processing^@^@^S^@^Qtime-at-completed^@^@!^@^Fjob-
id^@^D^@^@^@^G#^@=
>>>     =20
>>> job-state^@^D^@^@^@^C!^@^Zjob-media-sheets-
completed^@^D^@^@^@^@E^@^Ojob-pri=
>>> nter-uri^@)ipp://debian-swan2:631/printers/CUPS-Test!^@^Ljob-k-
octets^@^D^@^=
>>> @^@^AD^@^Njob-hold-until^@^Gno-holdB^@
>>> job-sheets^@^DnoneB^@^@^@^Dnone^C
>>>
>>> Here is the d file:
>>> #CUPS-BANNER
>>> Show printer-name printer-info printer-location printer-make-and-model
>>> =20 printer-driver-name printer-driver-version paper-size imageable-area
>>> Header Printer Test Page
>>> Footer Printer Test Page
>>> Notice This test page was produced by CUPS 1.4.4.
>>> Image images/cups.png
>>> Image images/color-wheel.png
>>>
>>> Like I said, I cannot telnet it, because there is no jetdirect port. 
>>> =20 The printer is an Epson Stylus R2400 (in this instance, but this the
>>> =20 third printer model I have worked with here).
>>>
>>> Thanks!
>>>
>>> Quoting Paul <paul.conklin at cerner.com>:
>>>
>>> > what kind of a printer is it?  if you can post / e-mail the contents
>>> > =20
>>> >  of the d file (i had them backwards when I posted earlier, you want
>>> >  =20 the d, not the c file) i can take a look as well.
>>> >> Thanks Paul,
>>> >>
>>> >> I can try everything except the last part.  These are USB
>>> printers.  =3D2=
>>>
>>> >> That is why we need
>>> >> to share them through cups and/or samba.  (I am assuming that there
>>> >> is =
>>> =3D20
>>> >> no telnet option,
>>> >> since there is no jetdirect port, correct?)
>>> >>
>>>
>>
>> _______________________________________________
>> cups mailing list
>> cups at easysw.com
>> http://lists.easysw.com/mailman/listinfo/cups
>>

>From your error_log snippet I see that the printer accepted the whole bunch 
of data. And, as the printer is supposed to print plain ascii text, I 
suspect that it needs some control code at the end of the text, maibe a form 
feed character or even an eof character (whatever that may be).

Note that, with your raw printing setup, there is no instance to tell the 
printer the raw data are complete and ready to print.

Helge