Path: csiph.com!x330-a1.tempe.blueboxinc.net!feeder1.hal-mli.net!news.glorb.com!news-out.readnews.com!transit3.readnews.com!news-out.news.tds.net!newsreading01.news.tds.net!86597e80!not-for-mail
From: "Andrew Thompson" <andrew.thompson@THRWHITE.remove-dii-this>
Subject: Re: Unpredictable Swing b
Message-ID: <793bc65f2633c@uwe>
X-Comment-To: comp.lang.java.gui
Newsgroups: comp.lang.java.gui
In-Reply-To: <47060CE6.671A@operamail.com>
References: <47060CE6.671A@operamail.com>
Content-Type: text/plain; charset=IBM437
Content-Transfer-Encoding: 8bit
X-Gateway: time.synchro.net [Synchronet 3.15a-Win32 NewsLink 1.92]
Lines: 119
Date: Wed, 27 Apr 2011 15:40:02 GMT
NNTP-Posting-Host: 96.60.20.240
X-Complaints-To: news@tds.net
X-Trace: newsreading01.news.tds.net 1303918802 96.60.20.240 (Wed, 27 Apr 2011 10:40:02 CDT)
NNTP-Posting-Date: Wed, 27 Apr 2011 10:40:02 CDT
Organization: TDS.net
Xref: x330-a1.tempe.blueboxinc.net comp.lang.java.gui:2597

  To: comp.lang.java.gui
H|Nkan Lane wrote:
>Thanks a lot for all your posts. 

Your thanks, for my post at least, can be best expressed
by posting your replies in-line, with trimming, as I am doing,
rather than by 'top-posting'.  I find that most confusing.

>..First of all, the address should be
>http://cafka.econ.au.dk/LMDG_Doc/launcher.jnlp and nothing else. 

That's better, at least in the fact it is a valid URL!

It is, unfortunately, entirely invalid.  It may work, but 
probably not reliably.

Here is what I recommend replacing it with..

<?xml version="1.0" encoding="utf-8"?>
<!-- JNLP file for Cafkadoc app. -->
<jnlp
  spec="1.0+"
  codebase="http://cafka.econ.au.dk/LMDG_Doc/"
  >
<information>
  <title>Cafkadoc Java Application</title>
  <vendor>CLS</vendor>
  <homepage href="/var/sioux/docs/index.html"/>
  <description>Cafka documentation application</description>
  <description kind="short">
    Application for documentation about the DLS projects
  </description>
  <offline-allowed/>
</information>
<security>
  <all-permissions/>
</security>
<resources>
  <j2se version="1.5+"/>
  <jar href="Cafkadoc.jar"/>
</resources>
<application-desc />
</jnlp> 

This one places the application-desc after the resources, where
it should be, and makes a number of other minor tweaks 
(including the comment re SwingSet2!).  The only actual
difference is that I changed it from 1.5 to 1.5+, to account
for my 1.6 based box (why go back to 1.5 now!).

Why is it 'full-permissions'?*

I downloaded the JNLP, and the jar to a local directory,
changed the JNLP to 'sandboxed' and ran it from there.
Got to see the GUI, and even the Info page in a JEditorPane(?)
"Labor Research Group 
School of Economics and Management - 
University of Aarhus"

There is an 'interesting problem' with that window.
I noted that when I dragged the main window left/right,
it obscured a 'proportional degree' of the HTML(?) 
window - and I mean an area entirely beyond the main 
app.'s boundaries!

Are you using null layouts, exact positioning, 
doing clever things with content or glasspanes..?

* Because the (now sandboxed) app. tried to get the 
location of user.home, Java kindly offered to shut it 
down for me, but interestingly, I could just send that 
dialog to the back and keep poking around in the app.!

>The problem was solved by calling pack at the end for my subclass of
>JFrame holding the contents presented to the user. 

Glad it is sorted.

> ... I have
>now reached a problem where a part of the layout (box with some buttons
>in the bottom row) does not show up when I close a "pop up" window. 

What pop-up window?  Besides the problem mentioned 
above, I did not see any rows of buttons at bottom of 
anything.  Is it in one of the 'secure areas' of the program?
(I am not willing to run 'jar's from folks on the net in anything
other than a sandbox - unless I know them well, and trust
both their integrity and competence.)

>..Can
>this thing be mitigated with the same type of commands?

No.  I am not sure about the rows of buttons, it sounds 
suspiciuosly related to the gray window, and I think *that* 
has to do with use of the null layouts and other things 
mentioned earlier.

>I supppose that I need to understand your points about concurrency 

I did not make any, though I expect others in this 
thread, did.  I do not think that is the immediate 
source of the problem I saw.

But to bring you back to the very first suggestion on
the thread, can you supply an SSCCE of this behaviour?
Being a GUI problem, it should be able to display
the effect without reaching outside the sandbox.

-- 
Andrew Thompson
http://www.athompson.info/andrew/

Message posted via JavaKB.com
http://www.javakb.com/Uwe/Forums.aspx/java-gui/200710/1

---
 * Synchronet * The Whitehouse BBS --- whitehouse.hulds.com --- check it out free usenet!
--- Synchronet 3.15a-Win32 NewsLink 1.92
Time Warp of the Future BBS - telnet://time.synchro.net:24