[lsb-discuss] printing-test info

[Jeff Licquia]

First, the good news: after purging and reinstalling the printing tests 
on Debian 4.0, the foomatic-rip tests all passed.  We do have some 
issues in other tests, but I think they're fixable.

Here's the SuSE 10 info.  First:

jeff at xenpc-suse10:~> echo $HOSTNAME
xenpc-suse10
jeff at xenpc-suse10:~> echo $USER
jeff
jeff at xenpc-suse10:~> rpm -qf /usr/bin/foomatic-rip
foomatic-filters-3.0.2-20.5

Journals for the failures:

400|4 9 1 18:59:44|IC Start
200|4 9 18:59:44|TP Start
520|4 9 4478 1 1|Reference FIXME.FIXME-FIXME
520|4 9 4478 1 2|Special entities get substituted by job parameters
520|4 9 4478 1 3|Executing /usr/bin/foomatic-rip --ppd 
/opt/lsb/test/printing/tet_tmp_dir/03772aa/testfoomaticrip/foomatic-test.ppd 
-o 
FilterPath=/opt/lsb/test/printing/tet_tmp_dir/03772aa/testfoomaticrip/ 
/opt/lsb/test/printing/tet_tmp_dir/03772aa/testfoomaticrip/foomatic-test-input-ps.ps 

520|4 9 4478 1 4|Checking: Job parameters
520|4 9 4478 1 5|Output written to stdout did not contain 
"foomatic-test-renderer[^\n\r]*\x20--FoomaticOptionE\x20--user=root\x20--host=xenpc-suse10\x20--title=root\@xenpc-suse10\x20--copies=1\x20--options=\(FilterPath=[^\n\r]*\x20\)\x20--date=\d\d\/\d\d\/\d+\x20--time=\d\d\:\d\d\:\d\d\x20--special=\&\<\>\"" 

220|4 9 1 18:59:44|FAIL
410|4 9 1 18:59:44|IC End

400|4 13 1 18:59:48|IC Start
200|4 13 18:59:48|TP Start
520|4 13 4478 1 1|Reference FIXME.FIXME-FIXME
520|4 13 4478 1 2|foomatic-rip acts as a CUPS filter and recognizes if 
it is
520|4 13 4478 1 3|called as such. Also a test for the full printing 
workflow.
520|4 13 4478 1 4|Creating test print queue "testprinter"
520|4 13 4478 1 5|Default option settings
520|4 13 4478 1 6|Sending print job with option settings ""
520|4 13 4478 1 7|Option setting Option1=Choice1
520|4 13 4478 1 8|Sending print job with option settings "-o 
Option1=Choice1"
520|4 13 4478 1 9|Changing FoomaticOption8
520|4 13 4478 1 10|Sending print job with option settings "-o 
FoomaticOption8=Choice1"
520|4 13 4478 1 11|FoomaticOption4: Valid string (DFhj3545)
520|4 13 4478 1 12|Sending print job with option settings "-o 
FoomaticOption4=DFhj3545"
520|4 13 4478 1 13|FoomaticOption5 set to -7
520|4 13 4478 1 14|Sending print job with option settings "-o 
FoomaticOption5=-7"
520|4 13 4478 1 15|Job parameters
520|4 13 4478 1 16|Sending print job with option settings ""
520|4 13 4478 1 17|Output written to stdout did not contain 
"foomatic-test-renderer[^\n\r]*\x20--FoomaticOptionE\x20--user=root\x20--host=xenpc-suse10\x20--title=TEST\x20--copies=1\x20--options=\([^\n\r]*\)\x20--date=\d\d\/\d\d\/\d+\x20--time=\d\d\:\d\d\:\d\d\x20--special=\&\<\>\"" 

520|4 13 4478 1 18|FoomaticOption2 (PJL) inserted
520|4 13 4478 1 19|Sending print job with option settings ""
520|4 13 4478 1 20|Option6 and FoomaticOption2 (both PJL) merged into 
driver-generated PJL header
520|4 13 4478 1 21|Sending print job with option settings "-o 
FoomaticPJLMergeTest=DriverPJL"
520|4 13 4478 1 22|FoomaticOption1=Choice3 for pages 1-2 and 4
520|4 13 4478 1 23|Sending print job with option settings "-o 
1-2,4:FoomaticOption1=Choice3"
520|4 13 4478 1 24|Removing test print queue "testprinter"
220|4 13 1 19:00:01|FAIL
410|4 13 1 19:00:01|IC End

So it looks like both tests are indeed failing for the same reason.  I 
did run foomatic-rip as indicated on the command line for test 9, and 
the last few lines of that output does seem to indicate what's going on:

%%EOF
% /opt/lsb/test/printing/testfoomaticrip/foomatic-test-renderer 
--option1=choice1 --FoomaticOption7 --option9=choiceC --optionA=choiceF 
--optionC=choiceB --optionD=choiceE --FoomaticOptionE --user=&user; 
--host=&host; --title=&title; --copies=&copies; --options=(&options;) 
--date=&month;/&date;/&year; --time=&hour;:&min;:&sec; --special=&<>"
%-12345X at PJL RESET

All the other tests passed on SuSE.

For Debian 4.0, there are two failures after purging and reinstalling 
the tests:

400|0 22 1 16:58:27|IC Start
200|0 22 16:58:27|TP Start
520|0 22 00001628 1 1|Test 22
520|0 22 00001628 1 2|When cupsLangEncoding is called, the function 
returns the
character encoding for the given language
520|0 22 00001628 1 3|cupsLangEncoding returned 'utf-8', expected 
'iso-8859-1'
220|0 22 1 16:58:27|FAIL
410|0 22 1 16:58:27|IC End

This is probably a test bug; I don't think the spec requires that the 
locale encoding be anything in particular.

400|3 5 1 16:58:31|IC Start
200|3 5 16:58:31|TP Start
520|3 5 1631 1 1|Reference FIXME.FIXME-FIXME
520|3 5 1631 1 2|Checking result of processing a known .ps file using 
the ijs device
520|3 5 1631 1 3|/usr/bin/gs returned 1 using the ijs driver, expected 0
220|3 5 1 16:58:31|FAIL
410|3 5 1 16:58:31|IC End

This is probably an error in my setup.  The gs-esp package in Debian 
reports that it supports ijs, but I may be missing something to get it 
working.

Two additional issues I'll work on:

  - Stew's memory corruption issues, which will require building the 
tests from source.

  - The issue with the test harness freaking itself out when necessary 
components are missing.