Bug#522417: ghostscript: Doesn't work with Brother printers

[Jonathan Nieder <jrnieder@gmail.com>]

Hi,

Antonio Aguilar wrote:

> - Debian unstable (installed a week ago).
[...]
> - Brother HL-2030 - CUPS+Gutenprint v5.2.6 [en] driver
>
> When I try to print with my Brother Hl-2030 it doesn't work.

Even though this may well be the same bug, could you file it
separately so it can be unmerged if appropriate?  I guess the package
to file against would be cups or foomatic-db, since that is the
"front-end" whose maintainers would know what information to get and
what back-end to blame it on.

> I have
> tried with ghostscript from stable (8.71) and unstable (9.01). When
> I try the Mike Simons reproducer it outputs this:
>
> $ /usr/bin/gs -r1200 -g10200x13200 -q -dNOPROMPT \
>	-dNOPAUSE -dSAFER -sDEVICE=bit -sOutputFile=- - -c quit \
>	< /usr/share/cups/data/testprint | wc
> GPL Ghostscript 9.01: Unrecoverable error, exit code 1
>       8      56     630

If you get rid of the "| wc" then the result becomes more obvious:

| $ gs -r1200 -g10200x13200 -q -dNOPROMPT -dNOPAUSE -dSAFER -sDEVICE=bit \
|	-sOutputFile=- - -c quit < /usr/share/cups/data/testprint
| Error: /undefined in #CUPS-BANNER
| Operand stack:
| 
| Execution stack:
|    %interp_exit   .runexec2   --nostringval--   --nostringval--   --nostringval--   2   %stopped_push   --nostringval--   --nostringval--   --nostringval--   false   1   %stopped_push   1910   1   3   %oparray_pop   1909   1   3   %oparray_pop   1893   1   3   %oparray_pop   1787   1   3   %oparray_pop   --nostringval--   %errorexec_pop   .runexec2   --nostringval--   --nostringval--   --nostringval--   2   %stopped_push   --nostringval--
[...]

In other words, the testprint file is not a postscript file.
Debugging would require the full relevant pipeline; see

 http://fedoraproject.org/wiki/Printing/Debugging#Finding_where_the_problem_lies

Looking at /usr/share/foomatic/db/source/printer/Brother-HL-2030.xml,
I see:

  <functionality>B</functionality>
  <driver>hl1250</driver>

> Finally I get working the printer using the foomatic drivers:
> Brother HL - 2030 Foomatic/hl1250

Ah, ok.  I suppose the failures happen with

  <driver>
    <id>gutenprint</id>
  </driver>

My foomatic-db-fu is not so great, so the closest I could find is
/usr/share/foomatic/db/source/driver/gutenprint-ijs.5.2.xml which says

 <execution>
  <filter />
  <prototype>gs -q -dPARANOIDSAFER -dNOPAUSE -dBATCH -sDEVICE=ijs -sIjsServer=ijsgutenprint.5.2 -dIjsUseOutputFD%A -sIjsParams=&quot;STP_VERSION=5.2.6,%B&quot;%Z -sOutputFile=- -</prototype>
 </execution>

i.e., device=ijs, not device=bit.

Changing "debug: 0" to "debug: 1" in /etc/foomatic/filter.conf and
looking at /tmp/foomatic-rip.log might be helpful.

> The CUPS  error's logs contains:
>
> E [09/Apr/2011:11:14:29 +0200] [Job 9] File
> '/System/Library/ColorSync/Profiles/sRGB Profile.icc' not found

This sounds interesting and worth reporting (or finding a similar
bug), too.  I don't know if it's related, either.

Thanks, and hope that helps.
Jonathan