[EMAIL PROTECTED] wrote:
>when you use "show draw" you don't add any qualifiers
>
>draw border1  #sort of a "select" for DRAW -- sets "current" object
>show draw     #shows current draw object
>
>should give something like:
>
>draw border1 {23.890585 32.710503 10.786414} {21.773388 34.16974 10.07222}
>{20.224413 32.710503 11.682588} {22.34161 31.251266 12.396782}
>
>
>I suspect the reason it seems unpredictable is that sometimes you had the
>object selected, and sometimes not. But if I'm wrong, reply with the exact
>wording of the script that is causing the problem.
>
>Also, something I keep doing is I sometimes accidentally put the $ in:
>
>  draw $border1
>
>DON'T DO THAT.
>
>  
This didn't make a difference. But I found out that the coordinates are
only provided if a specific frame is displayed, not with 'frame all' set
before.
Since we add amino acid variants as new models in our Jmol viewer,
'frame all' is always set initially.

In this context I noticed that Jmol freezes if the frame number is not
specified and not all atoms specified in the 'draw' command are present
in all frames. The java console shows the following error message:

java.lang.NullPointerException
    at javax.vecmath.Tuple3f.sub(Tuple3f.java:225)
    at org.jmol.viewer.Draw.scaleDrawing(Draw.java:462)
    at org.jmol.viewer.Draw.setProperty(Draw.java:197)
    at org.jmol.viewer.Frame.setShapeProperty(Frame.java:1092)
    at org.jmol.viewer.ModelManager.setShapeProperty(ModelManager.java:386)
    at org.jmol.viewer.Viewer.setShapeProperty(Viewer.java:2463)
    at org.jmol.viewer.Eval.draw(Eval.java:5039)
    at org.jmol.viewer.Eval.instructionDispatchLoop(Eval.java:634)
    at org.jmol.viewer.Eval.runEval(Eval.java:104)
    at org.jmol.viewer.Viewer.evalStringWaitStatus(Viewer.java:2289)
    at org.jmol.viewer.ScriptManager.runScript(ScriptManager.java:130)
    at org.jmol.viewer.ScriptManager.runNextScript(ScriptManager.java:119)
    at
org.jmol.viewer.ScriptManager$ScriptQueueRunnable.run(ScriptManager.java:147)
    at java.lang.Thread.run(Thread.java:534)


>>The next issue (if  'show draw' works correctly) is to read the
>>information from Jmol and to generate the corresponding 'echo' command.
>>To avoid parsing the message callback stream, I tried the
>>'jmolScriptWait' command for the first time. But it totally freezes
>>Firefox 1.5.0.7 with Java 1.4.2_11 on SuSE linux 9.3. The browser
>>freezes independant of the script command I used.
>>    
>
>Interesting -- this means that those browsers are using the event queue.
>David Evans at Eli Lilly had this problem with the application (as opposed
>to the applet). The solution for applications was to make sure you call
>scriptWait() from a new thread rather than the one associated with a user
>action, like clicking a button.
>
>Something to try: If the jmolScriptWait function is being run because the
>user has clicked on a link or button, try making sure that the button
>method looks like this:
>
>
>  onClick="setTimeout('clickAction()', 100)"
>
>
>the 100-ms delay won't be noticed by the user, but a new thread is started
>that isn't the event queue thread. So it shouldn't hang the program. Then
>
>function clickAction() {
>  var info = jmolScriptWait("draw $border1;show draw")
>}
>
>If anything will work, that stands the best chance.
>
>I seem to remember having this sort of problem with Opera browsers some
>eons ago, even with regular scripting, and the only solution was to use
>setTimeout() this way with every user-derived action.
>
>
>  
Unfortunately it doesn't work. Without 'jmolScriptWait' in 'clickAction'
the method works, but with it the browser freezes.

Regards,
Rolf

-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Jmol-users mailing list
Jmol-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jmol-users

Reply via email to