Comment # 34
on bug 954884
from Rodney Peters
There has been no additional comment on this thread since March 2016, during
which period many progressive updates have occurred. I retested the current
configuration and it appears that the bug has been resolved. I have not been
testing intermediate releases and can't pinpoint where the fix might have
occurred
I was still using Leap 42.1 on the test PC, but with LibreOffice & CUPS on
progressive updates from http://download.opensuse.org/repositories Printing &
LibreOffice, taking these up to CUPS 2.1.3 & Build ID:
20m0(Build:3)respectively.
I have changed only one setting in LibreOffice
Tools-> Options -> General -> Print to make "PDF as standard print job format".
I verified that it applied in the per-job print dialog from Properties button
and Device tab, where is shows as "Printer Language type: Automatic PDF"
I've concentrated on the spreadsheet output, that being my main use of
landscape prints. Prints to 3 real printers all produce the desired landscape
print, on paper:
Brother MFC-9120CN (BR-script3)
Brother MFC-J6910DW (page description language unknown)
Dataproducts Typhoon 8 (Postscript 2 device, using CUPS generic PS Level 2 PPD)
In an effort to establish what might have changed, I disabled the CUPS queue
for the Brother MFC-J6910DW. Test prints from Writer, Calc & Draw all produce
a PDF file.
rod@jw870:/tmp> file /tmp/print-job-odtdata-20170103.save
/tmp/print-job-odtdata-20170103.save: PDF document, version 1.4
rod@jw870:/tmp> file
/tmp/print-job-odsmfcj6910dwlpr-3.0.0-1.i386.rpmdata-20170103.save
/tmp/print-job-odsdata-20170103.save: PDF document, version 1.4
rod@jw870:/tmp> file /tmp/print-job-odgdata-20170103.save
/tmp/print-job-odgdata-20170103.save: PDF document, version 1.4
With each file, Ghostscript displays multiple copies of the image in a single
page file, each copy being truncated at approximately 2/3 of true width of
landscape page with the text displayed in portrait mode - ie the same faulty
output as previously
rod@jw870:/tmp> gs -r50 -sPAPERSIZE=a4 -dFIXEDMEDIA
/tmp/print-job-odtdata-20170103.save
GPL Ghostscript 9.20 (2016-09-26)
Copyright (C) 2016 Artifex Software, Inc. All rights reserved.
This software comes with NO WARRANTY: see the file PUBLIC for details.
Processing pages 1 through 1.
Page 1
>>showpage, press <return> to continue<<
GS>quit
XIO: fatal IO error 11 (Resource temporarily unavailable) on X server ":0"
after 68 requests (68 known processed) with 1 events remaining.
Although it appears to be a valid PDF file:
rod@jw870:/tmp> gs -r50 -dNOPAUSE -dBATCH -sDEVICE=nullpage
/tmp/print-job-odtdata-20170103.save
GPL Ghostscript 9.20 (2016-09-26)
Copyright (C) 2016 Artifex Software, Inc. All rights reserved.
This software comes with NO WARRANTY: see the file PUBLIC for details.
Processing pages 1 through 1.
Page 1
The test to generic Postscript printer:
rod@jw870:/tmp> gs -r50 -sPAPERSIZE=a4 -dFIXEDMEDIA
/tmp/testq.odsout-20170103.ps
GPL Ghostscript 9.20 (2016-09-26)
Copyright (C) 2016 Artifex Software, Inc. All rights reserved.
This software comes with NO WARRANTY: see the file PUBLIC for details.
>>showpage, press <return> to continue<<
GS>
produces an X11 window displaying the complete text of the landscape page,
oriented in portrait mode. My interpretation of
https://en.opensuse.org/SDB:Landscape_Printing
is that this is correct - the application is doing the rotation for landscape
mode ?
rod@jw870:/tmp> gs -r50 -dNOPAUSE -dBATCH -sDEVICE=nullpage
/tmp/testq.odsout-20170103.ps
GPL Ghostscript 9.20 (2016-09-26)
Copyright (C) 2016 Artifex Software, In
c. All rights reserved.
This software comes with NO WARRANTY: see the file PUBLIC for details.
For the first test to a real printer, I used the inkjet, Brother MFC J6910DW,
as in March 2016.
The PPD & filter come in an RPM from Brother web site -
http://support.brother.com.
mfcj6910dwcupswrapper-3.0.0-1.i386.rpm
mfcj6910dwlpr-3.0.0-1.i386.rpm
The RPM have not changed since 201211 and are the same as used for tests in
201603
I then repeated the tests to another real printer - the Brother product, MFC
9120CN, that has a Postscript interpreter. "print-job-data.save" files,
produced as above, also report as file type PDF and give similar, truncated
output with
gs -r50 -sPAPERSIZE=a4 -dFIXEDMEDIA
As this printer has three readily available PPD files I tried all. All gave
similar results with Ghostscript tests. Results to real paper were:
Brother's PPD from http://support.brother.com - correct hard copy
(what appears to be a copy of the latter in
/usr/share/cups/model/openPrinting/Brother) - locks up the printer
foomatic PPD from /usr/share/cups/model/openPrinting/Brother- correct hard
copy
Which raises two questions
1 can this bug be closed ?
2 is "gs -r50 -sPAPERSIZE=a4 -dFIXEDMEDIA" still a reliable tool for
troubleshooting ?