Eclipse Community Forums
Forum Search:

Search      Help    Register    Login    Home
Home » Modeling » Graphiti » Problems with respect to printers: Slow performance
Problems with respect to printers: Slow performance [message #716807] Thu, 18 August 2011 10:03 Go to next message
Ivar Refsdal is currently offline Ivar Refsdal
Messages: 24
Registered: May 2011
Junior Member
Hello everybody!

First off: Thanks to the developers for a absolutely very interesting
project. I'm quite thrilled about it.

However, I ran into some small problems.
When selecting between the canvas and nodes/shapes, the performance
was very slow.

Two or three times the following exception occurred:
> java.lang.RuntimeException: WARNING: Prevented recursive attempt to activate part org.eclipse.ui.navigator.ProjectExplorer while still in the middle of activating part org.eclipse.ui.views.ProblemView

The full stack trace is available here: http://pastebin.com/dumpGvE7

Anyhow, the problem turned out to be the following:

> at org.eclipse.swt.widgets.Display.eventProc(Display.java:1238)
> at org.eclipse.swt.internal.gtk.OS._gtk_enumerate_printers(Native Method)
> at org.eclipse.swt.internal.gtk.OS.gtk_enumerate_printers(OS.java:9091)
> at org.eclipse.swt.printing.Printer.getPrinterList(Printer.java:91)
> at org.eclipse.gef.ui.actions.PrintAction.calculateEnabled(PrintAction.java:45)
> at org.eclipse.graphiti.ui.internal.action.PrintGraphicalViewerAction.calculateEnabled(PrintGraphicalViewerAction.java:98)

I suppose the real problem lies within SWT somewhere or this particular
printer setup, but calculateEnabled() gets called very often, for
example every time when you select the canvas or a node, so caching this
would effectively fix the problem.

With making calculateEnabled() return false, the slow performance
disappeared and everything ran very smooth.

Note: There's quite a few printers available (20-30) at this uni, so the
setup is somewhat special. If one chooses "print" from Eclipse,
the dialog where one chooses the printer is also very slow,
but as far as I know this hasn't affected any of the plugins notably,
and also it doesn't crash.

Also an exception occurs if one chooses:
"File" -> "Print" -> "Change printer" -> "Cancel"
See stacktrace here: http://pastebin.com/xjYhm9ep

Should I file a bug report for this or something similar?


Thanks,
Best regards,
Ivar Refsdal
Re: Problems with respect to printers: Slow performance [message #717082 is a reply to message #716807] Fri, 19 August 2011 04:50 Go to previous messageGo to next message
Michael Wenz is currently offline Michael Wenz
Messages: 1527
Registered: July 2009
Location: Walldorf, Germany
Senior Member
Ivar,

thanks for pointing out! In Windows this doesn't seen to be an issue, but
other OS seem to behave differently.

Please open a Bugzilla to track this. What is the OS you are working on?

Thanks,
Michael


"Ivar Refsdal" schrieb im Newsbeitrag news:j2j5f6$8i0$1@news.eclipse.org...

Hello everybody!

First off: Thanks to the developers for a absolutely very interesting
project. I'm quite thrilled about it.

However, I ran into some small problems.
When selecting between the canvas and nodes/shapes, the performance
was very slow.

Two or three times the following exception occurred:
> java.lang.RuntimeException: WARNING: Prevented recursive attempt to
> activate part org.eclipse.ui.navigator.ProjectExplorer while still in the
> middle of activating part org.eclipse.ui.views.ProblemView

The full stack trace is available here: http://pastebin.com/dumpGvE7

Anyhow, the problem turned out to be the following:

> at org.eclipse.swt.widgets.Display.eventProc(Display.java:1238)
> at org.eclipse.swt.internal.gtk.OS._gtk_enumerate_printers(Native Method)
> at org.eclipse.swt.internal.gtk.OS.gtk_enumerate_printers(OS.java:9091)
> at org.eclipse.swt.printing.Printer.getPrinterList(Printer.java:91)
> at
> org.eclipse.gef.ui.actions.PrintAction.calculateEnabled(PrintAction.java:45)
> at
> org.eclipse.graphiti.ui.internal.action.PrintGraphicalViewerAction.calculateEnabled(PrintGraphicalViewerAction.java:98)

I suppose the real problem lies within SWT somewhere or this particular
printer setup, but calculateEnabled() gets called very often, for
example every time when you select the canvas or a node, so caching this
would effectively fix the problem.

With making calculateEnabled() return false, the slow performance
disappeared and everything ran very smooth.

Note: There's quite a few printers available (20-30) at this uni, so the
setup is somewhat special. If one chooses "print" from Eclipse,
the dialog where one chooses the printer is also very slow,
but as far as I know this hasn't affected any of the plugins notably,
and also it doesn't crash.

Also an exception occurs if one chooses:
"File" -> "Print" -> "Change printer" -> "Cancel"
See stacktrace here: http://pastebin.com/xjYhm9ep

Should I file a bug report for this or something similar?


Thanks,
Best regards,
Ivar Refsdal
Re: Problems with respect to printers: Slow performance [message #717870 is a reply to message #717082] Mon, 22 August 2011 11:06 Go to previous message
Ivar Refsdal is currently offline Ivar Refsdal
Messages: 24
Registered: May 2011
Junior Member
Hi Michael,

thanks for you reply!

A bug report has been submitted:
https://bugs.eclipse.org/bugs/show_bug.cgi?id=355401

The OS is:
> $ cat /proc/version
> Linux version 2.6.18-274.el5 (mockbuild@x86-002.build.bos.redhat.com) (gcc
> version 4.1.2 20080704 (Red Hat 4.1.2-51)) #1 SMP Fri Jul 8 17:36:59 EDT 2011

Best,
Ivar

Michael Wenz wrote, on 08/19/2011 10:50 AM:
> Ivar,
>
> thanks for pointing out! In Windows this doesn't seen to be an issue,
> but other OS seem to behave differently.
>
> Please open a Bugzilla to track this. What is the OS you are working on?
>
> Thanks,
> Michael
>
>
> "Ivar Refsdal" schrieb im Newsbeitrag news:j2j5f6$8i0$1@news.eclipse.org...
>
> Hello everybody!
>
> First off: Thanks to the developers for a absolutely very interesting
> project. I'm quite thrilled about it.
>
> However, I ran into some small problems.
> When selecting between the canvas and nodes/shapes, the performance
> was very slow.
>
> Two or three times the following exception occurred:
>> java.lang.RuntimeException: WARNING: Prevented recursive attempt to
>> activate part org.eclipse.ui.navigator.ProjectExplorer while still in
>> the middle of activating part org.eclipse.ui.views.ProblemView
>
> The full stack trace is available here: http://pastebin.com/dumpGvE7
>
> Anyhow, the problem turned out to be the following:
>
>> at org.eclipse.swt.widgets.Display.eventProc(Display.java:1238)
>> at org.eclipse.swt.internal.gtk.OS._gtk_enumerate_printers(Native Method)
>> at org.eclipse.swt.internal.gtk.OS.gtk_enumerate_printers(OS.java:9091)
>> at org.eclipse.swt.printing.Printer.getPrinterList(Printer.java:91)
>> at
>> org.eclipse.gef.ui.actions.PrintAction.calculateEnabled(PrintAction.java:45)
>>
>> at
>> org.eclipse.graphiti.ui.internal.action.PrintGraphicalViewerAction.calculateEnabled(PrintGraphicalViewerAction.java:98)
>>
>
> I suppose the real problem lies within SWT somewhere or this particular
> printer setup, but calculateEnabled() gets called very often, for
> example every time when you select the canvas or a node, so caching this
> would effectively fix the problem.
>
> With making calculateEnabled() return false, the slow performance
> disappeared and everything ran very smooth.
>
> Note: There's quite a few printers available (20-30) at this uni, so the
> setup is somewhat special. If one chooses "print" from Eclipse,
> the dialog where one chooses the printer is also very slow,
> but as far as I know this hasn't affected any of the plugins notably,
> and also it doesn't crash.
>
> Also an exception occurs if one chooses:
> "File" -> "Print" -> "Change printer" -> "Cancel"
> See stacktrace here: http://pastebin.com/xjYhm9ep
>
> Should I file a bug report for this or something similar?
>
>
> Thanks,
> Best regards,
> Ivar Refsdal
Previous Topic:IGaService.creatFont() ?
Next Topic:Error: Graphiti 0.8.0 and Eclipse BPMN2
Goto Forum:
  


Current Time: Mon Apr 21 00:31:41 EDT 2014

Powered by FUDForum. Page generated in 0.13922 seconds