newsgroups-index (beta)

Current group: opera.linux

Opera 8.0 Beta 1 printing problem

Opera 8.0 Beta 1 printing problem  
James Finnall
 Re: Opera 8.0 Beta 1 printing problem  
Eirik Byrkjeflot Anonsen
From:James Finnall
Subject:Opera 8.0 Beta 1 printing problem
Date:Mon, 24 Jan 2005 00:33:00 -0500
Hello all,

Attempted to use 8.0 Beta 1 and all is working well except printing with
CUPS of course. The printer I am using is an old LaserJet 4+ with
postsript simm and JetDirect card installed. The CUPS print queue is setup
for raw mode. So all output is sent straight to the printer. The printer
will handle postscript, PCL, or just ASCII. My other system particulars, QT
3.3.2, Slackware 10 Linux, with kernel 2.6.10 and udev to handle device
management.

When I use print preview the output presented looks acceptable on the
screen.

When I attempt to print; the desired printer is listed and indicates a CUPS
printer. When I select it and print, the printer prints an offending
message page. The printed material is just my personal home page and the
printer is easily capable of printing it. The error page indicates that
postscript output was actually sent, but bad output.

Next I tried to use the custom program and direct the output directly into
kprinter. Kprinter dialog appears with a filename in the window. Kprinter
doesn't understand what to do with it and provides an error message. Like
something was passed to it on the command line interface. I delete the
unusual file in the list and then print but nothing prints. As though the
stdin input was ignored because a filename was passed on the CLI.

Next, I tried to use the shell script that I used under version 6 of Opera.
The shell script reads stdin device and captures output to a file. Then
passes the file to kprinter on the CLI. This method prints the same
offending messsage page as earlier. Here I also tried to view the captured
file with KGhostView and received the following error message:

Error: /syntaxerror
in --%ztokenexec_continue--

Operand stack:
HL --nostringval--
Execution stack:
%interp_exit .runexec2 --nostringval-- --nostringval-- --nos
tringval-- 2 %stopped_push --nostringval-- --nostringval--
--nostringval-- false 1 %stopped_push 1 3 %opar
ray_pop 1 3 %oparray_pop 1 3 %oparray_pop .runexec2
--nostringval-- --nostringval-- --nostringval-- 2 %s
topped_push --nostringval-- --nostringval-- --nostringval--
Dictionary stack:
--dict:1052/1417(ro)(G)-- --dict:0/20(G
)-- --dict:126/200(L)--
Current allocation mode is local
Last OS error: 2
ESP Ghostscript
7.07.1
:
Unrecoverable error, exit code 1

Next, I try to print directly to a file. KGhostView opens the output file
great. But is missing several graphic images. These images did appear in
the print preview. And I can send the output file to the printer and it is
printed great with the exception of the missing graphic images. When
comparing the two postscript files they are not even the same size. The
broken file is 41,781 bytes and the printable file is 44,189 bytes. The
printable file is larger and is missing the graphic images.

The file size even indicates that the output is not the same when printing
to a file and what is sent to a printer directly. But at least printing
direct to a file will produce a printable file.

So I have had to create an icon on my desktop that calls kprinter and passes
the output file as an argument. Or I can have a shell script that cats the
output file and pipe it into kprinter by using the icon. Either icon
method works well, but requires two operations to complete a print job.
But it is the only acceptable print output I have been able to achieve.

So I set the margins to .2 cm and scale output by 95% and print again to a
file. The output file now jumps up to 115,615 bytes. Original settings
were 100% and 0.0 cm margins. KGhostView now displays the missing images
and the printer prints the missing images as well. Except the printer now
waits for about 3 minutes after receiving the data to print the last page.
I had this same wait period on version 6 of Opera as well when sending
output directly to the printer. So I directed the output to another print
queue for the same printer but on a Linux print server queue that is using
lpd. The pages printed as desired with all graphics and without any
undesirable wait time at the end.

After making the latest changes to the output settings to get all the
graphics included I went back to the first print option and retested
printing through CUPS directly to the laser printer. Same results as
before with the offending error message page. This error page produces
nothing that is really usable that I can tell but I will provide below to
complete the report.

ERROR: syntaxerror
OFFENDING COMMAND:>

STACK:

/HL

Perhaps it will mean something to somebody else.

Please advise if I can provide any additional information or perform other
tests / methods desired.

Thank you,
James@Finnall.net
From:Eirik Byrkjeflot Anonsen
Subject:Re: Opera 8.0 Beta 1 printing problem
Date:Mon, 24 Jan 2005 10:27:46 +0100
James Finnall writes:

> Hello all,
>
[ ... ]
> When I attempt to print; the desired printer is listed and indicates a CUPS
> printer. When I select it and print, the printer prints an offending
> message page. The printed material is just my personal home page and the
> printer is easily capable of printing it. The error page indicates that
> postscript output was actually sent, but bad output.
[ ... ]
> Next, I try to print directly to a file. KGhostView opens the output file
> great. But is missing several graphic images. These images did appear in
> the print preview. And I can send the output file to the printer and it is
> printed great with the exception of the missing graphic images. When
> comparing the two postscript files they are not even the same size. The
> broken file is 41,781 bytes and the printable file is 44,189 bytes. The
> printable file is larger and is missing the graphic images.
[ ... ]


There is a bug in Qt that can cause printing to a program to drop a
piece of the postscript data. This also happens when printing to a
printer, since Qt will do this by starting a program (e.g. lpr).

I have reported this bug to Trolltech.


I don't know why you have the other problems.


eirik
   

Copyright © 2006 newsgroups-index   -   All rights reserved   -   Impressum