print bold in openoffice gives bad output

Bug #314018 reported by Marcus
14
This bug affects 1 person
Affects Status Importance Assigned to Milestone
GS-GPL
Invalid
Critical
ghostscript (Ubuntu)
Fix Released
High
Till Kamppeter

Bug Description

Binary package hint: openoffice.org-writer

I'm not sure if this is related to openoffice or cups.

Any time I print a document in OpenOffice writer that has one of the following fonts:
Swenson (http://img.dafont.com/dl/?f=swenson), Domestic manners or Swift (probably more fonts could be listed here) and the text is in BOLD, I get very strange output. I can print the same document in MSWord without any problem.

I have a "Canon mp610" printer, but I can repeat this by using the "print to file" option in the printer dialog. If a look at the produced postscript-file in okular it looks ok, but if I run ps2pdf and then look at the result. It looks as strange as the result from the printer.

This bug can also be repeated by adding other printers in cups, but if I stop cups and choose the "Generic printer" in openoffice and then "print to file". The output is ok!

"Export to pdf" in openoffice gives me nice output. This what I have to do to be able to print.

I attach a ps-file created with "print to file" in openoffice. If this file is converted to pdf with ps2pdf it is very easy to see the terrible output.

# lsb_release -rd
Description: Ubuntu 8.10
Release: 8.10

# apt-cache policy openoffice.org-writer
openoffice.org-writer:
  Installed: 1:2.4.1-11ubuntu2.1
  Candidate: 1:2.4.1-11ubuntu2.1
  Version table:
 *** 1:2.4.1-11ubuntu2.1 0
        500 http://se.archive.ubuntu.com intrepid-updates/main Packages
        500 http://security.ubuntu.com intrepid-security/main Packages
        100 /var/lib/dpkg/status
     1:2.4.1-11ubuntu2 0
        500 http://se.archive.ubuntu.com intrepid/main Packages

Related branches

Revision history for this message
Marcus (marcus-liljedahl) wrote :
Revision history for this message
Kazupon (kazupon-up) wrote :

Ubuntu 8.10, Canon PIXMA MP610

When I print out a bold Japanese letter,the letter is too bold that I can't read them.
It happens in Landscape mode too.

When I installed Canon's MP610 driver(cnijfilter--common_2.80-1_i386.deb, cnijfilter-mp610series-pd-2.80-1_i386.deb), it was fixed in portrait mode.
But in Landscape mode it still too bold.

Sorry about my poor english.

Revision history for this message
Kazupon (kazupon-up) wrote :
Revision history for this message
Kazupon (kazupon-up) wrote :
Revision history for this message
Chris Cheney (ccheney) wrote :

Installing different driver fixed most of the users issue so it sounds like it is a cups issue.

Revision history for this message
Till Kamppeter (till-kamppeter) wrote :

This is not a problem of CUPS but of Ghostscript. Both ps2pdf and CUPS convert the incoming PostScript data stream to PDF using Ghostscript. I will report it to Ghostscript upstream.

Changed in cups:
importance: Undecided → High
Revision history for this message
Till Kamppeter (till-kamppeter) wrote :

Reported upstream. See link above.

Changed in gs-gpl:
status: Unknown → New
Revision history for this message
Till Kamppeter (till-kamppeter) wrote :

This is fixed in Ghostscript 8.64, to be released 1st of February.

Changed in ghostscript:
status: New → In Progress
assignee: nobody → till-kamppeter
Revision history for this message
Marcus (marcus-liljedahl) wrote : Re: [Bug 314018] Re: print bold in openoffice gives bad output

Is it possible to receive a patch for 8.63?

Revision history for this message
Kazupon (kazupon-up) wrote :

Thanks for information.
I will wait for GS 8.64. zzz

Revision history for this message
Till Kamppeter (till-kamppeter) wrote :

Marcus, Kazupon, please try the following workaround: Replace your file /usr/lib/cups/filter/pstopdf by the attached one. Make sure that the new file is also world-readable and world-executable. Do your documents print correctly now?

Revision history for this message
Till Kamppeter (till-kamppeter) wrote :
Revision history for this message
Till Kamppeter (till-kamppeter) wrote :

An additional remark: The problem does not occur with standard fonts which are defined both in a standard and in a bold version. It only occurs if a bold version does not exist and OpenOffice.org has to boldify the font by itself. So for reproducing the bug you have to use some more special fonts.

Changed in gs-gpl:
status: New → Invalid
Revision history for this message
Martin Pitt (pitti) wrote :

Can this extra pstops calling be done in /etc/cups/oopstops.convs, or in the oopstops filter itself?

Revision history for this message
Till Kamppeter (till-kamppeter) wrote :

pitti, this is no problem, then there is no extra implementation needed of the checking whether the input file is from OOo. We only need to check whether ps2ps conserves the paper dimensions without needing to repeat them on the command line. Otherwise we would need to repeat a lot of code of pstopdf in oopstops, so that we can supply the paper dimensions on the command line of oopstops.

The overall effect and the shortcomings do not change. We can still get bitten by http://bugs.ghostscript.com/show_bug.cgi?id=690025.

Revision history for this message
Martin Pitt (pitti) wrote :

Till Kamppeter [2009-01-06 19:59 -0000]:
> Otherwise we would need to repeat a lot of code of pstopdf in
> oopstops, so that we can supply the paper dimensions on the command
> line of oopstops.

That would be bad, let's not do that.

> The overall effect and the shortcomings do not change. We can still get
> bitten by http://bugs.ghostscript.com/show_bug.cgi?id=690025.

Okay, if it causes any regression, it's not appropriate for an SRU,
and we should rather live with the existing bug in intrepid (which is
far from grave) and fix it properly in Jaunty.

Revision history for this message
Till Kamppeter (till-kamppeter) wrote :

Perhaps we should make a backport of GS 8.64 then and not an SRU. This way it is not automatically installed on everyones box but people who have problems can install it.

Revision history for this message
Martin Pitt (pitti) wrote :

Till Kamppeter [2009-01-06 20:59 -0000]:
> Perhaps we should make a backport of GS 8.64 then and not an SRU. This
> way it is not automatically installed on everyones box but people who
> have problems can install it.

Works for me.

Revision history for this message
Kazupon (kazupon-up) wrote :

Pstopdf give me very good out for me. I love it.

Thanks Till.

Revision history for this message
Kazupon (kazupon-up) wrote :

Here's my landscape output before I replace pstopdf.

Revision history for this message
Marcus (marcus-liljedahl) wrote :

The new pstopdf works fine for me too as long as I print directly to the printer. Great work! If I use "print to ps" and then run ps2pdf (not pstopdf) the problem is still there, but I suppose this is as it should be?

Revision history for this message
Till Kamppeter (till-kamppeter) wrote :

Yes, that is absolutely correct. The pstopdf replacement only covers printing. It does not fix ps2pdf. If you want to have a PDF file, use "Export to PDF" in OpenOffice.org.

Revision history for this message
Launchpad Janitor (janitor) wrote :
Download full text (3.2 KiB)

This bug was fixed in the package ghostscript - 8.64.dfsg.1~svn9377-0ubuntu1

---------------
ghostscript (8.64.dfsg.1~svn9377-0ubuntu1) jaunty; urgency=low

  * New upstream release (SVN rev 9377)
     o Fixes many bugs concerning PDF rendering, to make the PDF printing
       workflow correctly working.
     o Fixes long-standing bugs in many drivers, like input paper tray and
       duplex options not working for the built-in PCL 4, 5, 5c, 5e, and
       6/XL drivers, PDF input not working for bjc600, bjc800, and cups
       output devices, several options not working and uninitialized
       memory with cups output device.
     o Merged nearly all patches of the Ubuntu and Debian packages upstream.
     o Fixes LP: #317810, LP: #314439, LP: #314018.
  * debian/patches/03_libpaper_support.dpatch,
    debian/patches/11_gs-cjk_font_glyph_handling_fix.dpatch,
    debian/patches/12_gs-cjk_vertical_writing_metrics_fix.dpatch,
    debian/patches/13_gs-cjk_cjkps_examples.dpatch,
    debian/patches/20_bbox_segv_fix.dpatch,
    debian/patches/21_brother_7x0_gdi_fix.dpatch,
    debian/patches/22_epsn_margin_workaround.dpatch,
    debian/patches/24_gs_man_fix.dpatch,
    debian/patches/25_toolbin_insecure_tmp_usage_fix.dpatch,
    debian/patches/26_assorted_script_fixes.dpatch,
    debian/patches/29_gs_css_fix.dpatch,
    debian/patches/30_ps2pdf_man_improvement.dpatch,
    debian/patches/31_fix-gc-sigbus.dpatch,
    debian/patches/34_ftbfs-on-hurd-fix.dpatch,
    debian/patches/35_disable_libcairo.dpatch,
    debian/patches/38_pxl-duplex.dpatch,
    debian/patches/39_pxl-resolution.dpatch,
    debian/patches/42_gs-init-ps-delaybind-fix.dpatch,
    debian/patches/45_bjc600-bjc800-pdf-input.dpatch,
    debian/patches/48_cups-output-device-pdf-duplex-uninitialized-memory-fix.dpatch,
    debian/patches/50_lips4-floating-point-exception.dpatch,
    debian/patches/52_cups-device-logging.dpatch,
    debian/patches/55_pcl-input-slot-fix.dpatch,
    debian/patches/57_pxl-input-slot-fix.dpatch,
    debian/patches/60_pxl-cups-driver-pdf.dpatch,
    debian/patches/62_onebitcmyk-pdf.dpatch,
    debian/patches/65_too-big-temp-files-1.dpatch,
    debian/patches/67_too-big-temp-files-2.dpatch,
    debian/patches/70_take-into-account-data-in-stream-buffer-before-refill.dpatch:
    Removed, applied upstream.
  * debian/patches/01_docdir_fix_for_debian.dpatch,
    debian/patches/02_gs_man_fix_debian.dpatch,
    debian/patches/01_docdir-fix-for-debian.dpatch,
    debian/patches/02_docdir-fix-for-debian.dpatch: Renamed patches to
    make merging with Debian easier.
  * debian/patches/32_improve-handling-of-media-size-changes-from-gv.dpatch,
    debian/patches/33_bad-params-to-xinitimage-on-large-bitmaps.dpatch:
    regenerated for new source directory structure.
  * debian/rules: Corrected paths to remove cidfmap (it is in Resource/Init/
    in GS 8.64) and to install headers (source paths are psi/ and base/ now).
  * debian/rules: Remove all fontmaps, as DeFoMa replaces them.
  * debian/local/pdftoraster/pdftoraster.c,
    debian/local/pdftoraster/pdftoraster.convs, debian/rules: Removed
    added pdftoraster filter and use the one which comes with Ghostscript.
  * de...

Read more...

Changed in ghostscript:
status: In Progress → Fix Released
Revision history for this message
Ralph Corderoy (ralph-inputplus) wrote :

Ubuntu 8.04, ghostscript 8.61.dfsg.1-1ubuntu3.

ps2pdf turns Courier into Courier Bold Italic.

    echo foo |
    a2ps --borders=no -RB --columns 1 -o foo.ps
    ps2pdf foo.ps

evince displays foo.ps correctly, but foo.pdf shows the fault.

    bash /usr/share/doc/libcupsys2/examples/pstopdf job-id user title 1 '' foo.ps >foo.pdf

also produces faulty output in the same way. Using
http://launchpadlibrarian.net/20954728/pstopdf instead produces a
PDF with the same font as the PostScript.

So 8.04 LTS apparently also has the same fault.

Changed in gs-gpl:
status: Invalid → Unknown
Changed in gs-gpl:
importance: Unknown → Critical
Changed in gs-gpl:
status: Unknown → Invalid
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.