Well, on my two odd macs, I get some pretty strange results.
Observations:
1. Loaded the jmolcalls_difapp.html page of Tim's in a window.
one mac responds only to "direct" button; other only to "applet[]"
2. Loaded the SAME page in a second window, both showing.
both macs respond to both "direct" and "applet[]" calls in THIS window,
still only whatever they did before in the first window.
3. At BOTH Macs, ALL of the following work at the URL line:
javascript:alert(document.debug.script("background red"))
javascript:alert(document.applets.debug.script("background blue"))
javascript:alert(document.applets[0].script("background green"))
4. On a different occasion, starting jmolcalls.html in the
first window, the same mac that first only accepted "direct" commands
now accepts only "applet[]" commands.
Conclusion: Get a PC! Sigh. No, really, something is very messed up deep
in the bowels (AND I MEAN BOWELS) of these systems. Nothing whatsoever to
do with Jmol, of course, but it's OUR problem because it's OUR users who
are going to complain. And if there is a way around, we should implement
it.
What really bothers me is that that page of Tim's runs from a javascript:
command but not a button. Something is really messed up there.
Bob
> On 2005-04-26 (13:05) Bob Hanson wrote:
>
>>Whew! I thought it was just me.... still, what could be going on
>>here? This really is a problem. Tim, we should have one EVEN SIMPLER
>>that does not have messagecallback or pickcallback in the applet
>>code.
>>
>>just the absolute minimum:
>>
>><param name="progressbar" value="true"> <param name="script"
>>value="load 1crn.pdb;color cornflowerblue;">
>>
>
> ok, here are the simple test pages:
>
> uses callbacks, applet in location 1:
> <http://dev.molvisions.com/debugging/jmolcalls.html>
>
> uses callbacks; applet in location 2:
> <http://dev.molvisions.com/debugging/jmolcalls_diffapp.html>
>
>
> no callbacks, applet in location 1:
> <http://dev.molvisions.com/debugging/jmolcalls_nocallback.html>
>
> no callbacks, applet in location 2:
> <http://dev.molvisions.com/debugging/jmolcalls_nocallback_diffapp.html>
>
> I'm interested to hear what people find. I do not see callbacks making
> any difference.
>
> for debugging, please make sure that you clear browser cache and restart
> your browser as soon as you see a failure...
>
>
> regards,
>
> tim
> --
> Timothy Driscoll
> molvisions - see, grasp, learn.
> <http://www.molvisions.com/>
> usa:north carolina:raleigh
>
>
>
>>
>>Bob
>>
>>
>>Philip Bays wrote:
>>
>>>I have tried Tim's new page. Works beautifully, AS LONG AS the
>>applet
>>>has not been previously loaded from another site. If I access
>>>another page and then try Tim's, no action. If I load his first in
>>>a fresh safari, it runs nicely -- with all buttons. But then the
>>>next page
>>will
>>>not -- no matter where it is -- mine, Bob's.... this is true it I
>>>am using tabs for the different pages, or if I try to load
>>>sequentially into the same tab.
>>>
>>>Phil
>>>
>>>On Apr 26, 2005, at 12:14 PM, timothy driscoll wrote:
>>>
>>>>On 2005-04-26 (11:30) Bob Hanson wrote:
>>>>
>>>>>OK, Phil. You are using Mac/Safari, right?
>>>>>
>>>>>Looks like we have the start of a good mystery. Please tell us
>>>>>your page so that others can check. We need a good, reproducible
>>>>>test.
>>>>>
>>>>>Q1: Could this be a jmol version problem? My pages (given below)
>>>>>are both using the applet version: [Jmol Version 10.00
>>>>>2004/12/17 18:55]
>>>>>
>>>>>Q2: Why me? Can this be shown for two sites NOT involving my
>>>>>pages? Who else can point us to sites using messagecallback
>>>>>handling?
>>>>>
>>>>
>>>>hi,
>>>>
>>>>I've posted a very simple test for accessing the JmolApplet. it
>>>>does away with almost all complexity that can cloud this issue.
>>>>you can see it here:
>>>>
>>>><http://dev.molvisions.com/debugging/jmolcalls.html>
>>>>
>>>>it includes three methods:
>>>>
>>>>1. getElementById 2. document.appletname 3. document.applets[]
>>>>
>>>>so far, all three methods work for me in OSX: Safari 1.3, FF 1,
>>>>Moz 1.7.3; also in IE6 (winXP).
>>>>
>>>>
>>>>a few oddities:
>>>>
>>>>1. the messagecallback is somewhat inconsistent in its output.
>>>>that may have to do with the way I implemented it. it does not
>>>>seem to adversely affect the command stream, though.
>>>>
>>>>2. I developed this on my localhost server, and it worked without
>>>>flaw. then I copied it to my LAN www server and, without quitting
>>>>my browser, tried it out. it failed with all sorts of security
>>>>exceptions. I cleared my browser cache, quit and restarted, and
>>>>now it works fine.
>>>>
>>>>
>>>>oddity 2 may point to the issue of accessing different Jmol pages,
>>IOW
>>>>different JmolApplets, and somehow confusing the browser or the
>>>>JVM. but it does not point to a problem with actually sending
>>>>scripts to the JmolApplet. we should spawn a new thread for
>>>>discussing the cross-page problems, IMO.
>>>>
>>>>
>>>>one final note: method 2 above:
>>>>
>>>>document.appletname
>>>>
>>>>must account for the DOM of the page. IOW, if I put my applet
>>>>inside a form named myform, for example, instead of directly under
>>>>the body tag, my call would have to change to
>>>>
>>>>document.myform.appletname
>>>>
>>>>
>>>>the other two methods do not require this change.
>>>>
>>>>
>>>>
>>>>regards,
>>>>
>>>>tim -- Timothy Driscoll molvisions - see, grasp, learn.
>>>><http://www.molvisions.com/> usa:north carolina:raleigh
>>>>
>>>>"As flies to wanton boys are we to the gods. They kill us for
>>>>their sport." - William Shakespeare
>>>>
>>>>
>>>>------------------------------------------------------- SF.Net
>>>>email is sponsored by: Tell us your software development
>>plans!
>>>>Take this survey and enter to win a one-year sub to
>>>>SourceForge.net Plus IDC's 2005 look-ahead and a copy of this
>>>>survey Click here to start!
>>>>http://www.idcswdc.com/cgi-bin/survey?id5hix
>>>>_______________________________________________ Jmol-users mailing
>>>>list [email protected]
>>>>https://lists.sourceforge.net/lists/listinfo/jmol-users
>>>>
>>>>
>>>
>>>
>>>J. Philip Bays Department of Chemistry and Physics Saint Mary's
>>>College Notre Dame IN 46556 [EMAIL PROTECTED]
>>>
>>>
>>>
>>>------------------------------------------------------- SF.Net
>>>email is sponsored by: Tell us your software development plans!
>>>Take this survey and enter to win a one-year sub to SourceForge.net
>>>Plus IDC's 2005 look-ahead and a copy of this survey Click here to
>>>start! http://www.idcswdc.com/cgi-bin/survey?id=105hix
>>>_______________________________________________ Jmol-users mailing
>>>list [email protected]
>>>https://lists.sourceforge.net/lists/listinfo/jmol-users
>>
>
> "Life is pleasant. Death is peaceful. It's the transition that's
> troublesome." - Isaac Asimov
>
>
> -------------------------------------------------------
> SF.Net email is sponsored by: Tell us your software development plans!
> Take this survey and enter to win a one-year sub to SourceForge.net
> Plus IDC's 2005 look-ahead and a copy of this survey
> Click here to start! http://www.idcswdc.com/cgi-bin/survey?id5hix
> _______________________________________________
> Jmol-users mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/jmol-users
>
-------------------------------------------------------
SF.Net email is sponsored by: Tell us your software development plans!
Take this survey and enter to win a one-year sub to SourceForge.net
Plus IDC's 2005 look-ahead and a copy of this survey
Click here to start! http://www.idcswdc.com/cgi-bin/survey?id=105hix
_______________________________________________
Jmol-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/jmol-users
