[Jmol-users] Jmol.js prototype: Jmol-new.js and jmolScriptWait
It's probably time I told people about some of the prototype Jmol.js
niceties in
http://www.stolaf.edu/people/hansonr/jmol/test/proto/Jmol-new.js
Here are a few, at least:
*JAR File Option*
An enhanced jmolInitialize() function that allows for specification of a
particular Jar file as the second parameter.
jmolInitialize(., JmolAppletProto.jar)
This then allows testing of different versions. Leaving that parameter
off defaults to JmolApplet0.jar, which then also requires 1-6. But
setting it to a copy of the JmolApplet.jar file from any release pr
prerelease of Jmol or from your own compiling bypasses the 1-6 jar files
and directs Jmol to use the Jar file you specifically want it to use.
Note that this may only require a single Jar file in the specified
directory. But if you specify the release's JmolApplet0.jar file, then
all the othere, 1-6, must be there as well.
*Synchronous Scripting*
OK, the main subject of this message deals with scripting. It answers
Eric's question about how one can get a pop-up message if a script has
an error along with several other possibilities.
One of the problems with scripting the applet has always been that the
script runs in a separate, parallel thread. This means that you can't
expect the applet to have done anything at all right after executing
jmolScript(.)
For example, if you use:
jmolScript(load myfile)
alert(testing
the alert will appear long before the file is actually loaded.
The new option
result = jmolScriptWait()
1) waits until the script has completed running to return, and
2) returns valuable status messages in the assigned variable (result
in this case).
jmolScriptWait() turns out to be VERY powerful. Consider, for example:
var fileContents = jmolScriptWait(show file)
or
var returnStatus = jmolScriptWait(selct not hydrogen)
if(status.indexOf(ERROR)=0)alert(returnStatus)
I'm starting to use jmolScriptWait() a LOT in my work because I can get
information from the applet and do something with it. For many
applications, jmolScriptWait() removes the need for callbacks. Here are
a few examples:
*Save/Restore orientation*
One of the nicest script commands is show orientation, which returns
the moveTo command required to set an orientation as the first line,
ending with a semicolon.
So consider this sequence:
var thisOrientation = jmolScriptWait(show orientation).split(;)[0]
..
later
..
jmolScriptWait(thisOrientation)
That give you any ideas? You can check this out at
http://www.stolaf.edu/people/hansonr/jmol/xtalx
I built this into Jmol-new.js as two new functions:
jmolSaveOrientation(id, targetSuffix)
and
jmolRestoreOrientation(id, targetSuffix)
The idea of the id is that you can then name different save/restore
points under different identifiers. Call one initial and one final,
for example.
*getProperty*
I'm not going to detail this at this time, but something new you can do
in the prototype is get all sorts of information just using
jmolScriptWait(). Here are a few examples only:
var info = jmolScriptWait(getProperty appletinfo)
This returns what at first may look rather peculiar:
{appletinfo: {version: 10.3.BH,date: 2006-05-01
20:08,javaVersion: 1.5.0_06,htmlName: jmolApplet0,javaVendor:
Sun Microsystems Inc.,operatingSystem: Windows XP}}
This is called JavaScript Object Notation, or JSON,
http://www.json.org/ and it is just a compact form of a relatively new
way to work with JavaScript. Interestingly, JSON is a lot like XML, but
is it more compact and more easily handled in common browsers. For these
reasons JSON is starting to replace XML in certain applications.
If you unpack this returned string, you will see that it is what is
called a JavaScript object literal:
{appletinfo: {
version: 10.3.BH,
date: 2006-05-01 20:08,
javaVersion: 1.5.0_06,
htmlName: jmolApplet0,
javaVendor: Sun Microsystems Inc.,
operatingSystem: Windows XP
}}
That's actual JavaScript. It is returned as a string, but you can
evaluate it by placing it in parentheses:
eval((+info+))
Now you can check
appletinfo.version
appletinfo.date
appletinfo.javaVersion
etc.
I've built several different properties that you can get. I'll elaborate
on these in a future message, but if you want to play with this, some of
the more useful ones are given below:
appletinfo
animationinfo
modelinfo
shapeinfo
measurementinfo
boundboxinfo
centerinfo
orientationinfo
transforminfo
zoominfo
atominfo
bondinfo
polymerinfo
chaininfo
stateinfo
The following three are special in that they return straight text, not JSON:
fileName
fileHeader
fileContents
extractModel
(That last one will be discussed below.)
Some of these can take an additional argument that specifies WHICH
information. You must specify this information in IN PARENTHESES. For
example:
getproperty atominfo atomno=32
might return:
{atominfo: [{
visible: true,
model: 1,
radius: 0.875,
Re: [Jmol-users] Jmol.js prototype: Jmol-new.js and jmolScriptWait
binQfg41Vl0od.bin Description: undefined
Re: [Jmol-users] Jmol.js prototype: Jmol-new.js and jmolScriptWait
At 5/5/06, you wrote: It's probably time I told people about some of the prototype Jmol.js niceties in http://www.stolaf.edu/people/hansonr/jmol/test/proto/Jmol-new.js Dear Bob, This all looks extremely useful. At last there is a mechanism to keep track of when the applet has completed a task! And better yet, whether it completed the task successfully. 1. Is there any way to tell when the applet is initially loaded and ready for the first script? My inability to do this caused lots of head scratching when I began designing FirstGlance. 2. Is there any way to tell when the Protein Data Bank does not supply the PDB file requested in a load command? FirstGlance presently just leaves a blank applet (white, Jmol frank disappears), a very ungraceful failure. Unfortunately, I cannot immediately take advantage of the new jmol.js ... 3. LACK OF JMOL.JS DOCUMENTATION. When I started coding FirstGlance, I decided not to use jmol.js. This was partly because it didn't support some functions I needed, but mainly because it is undocumented and I was afraid to use it. From inspection of the contents of jmol.js, it appears to do some sort of browser/java version testing, but I cannot tell what this is supposed to accomplish. Until someone documents the situations this is supposed to handle and what it is attempting to do, I will continue to be reluctant to use jmol.js for FirstGlance. I want to avoid a situation where a (common) client situation that I haven't tested causes an unanticipated failure mode. For example, no java in Windows, or a very old version of java. Right now, FirstGlance would fail to work at all, or to work correctly, without any explicit error messages. That is certainly not optimal. But I can't tell whether jmol.js would do better. One bit of code in jmol.js that particularly concerns me (and is inscrutable to me) is this: windowsClassId: clsid:8AD9C840-044E-11D1-B3E9-00805F499D93, windowsCabUrl: http://java.sun.com/update/1.5.0/jinstall-1_5_0_07-windows-i586.cab;, I wish someone would explain to me what this is designed to do, and how I could test it. More generally, I think we need thorough on-line documentation, in one place, on what jmol.js is designed to do, how to use the functions in it, etc. -Eric --- 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=lnkkid=120709bid=263057dat=121642 ___ Jmol-users mailing list Jmol-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jmol-users
Re: [Jmol-users] Jmol.js prototype: Jmol-new.js and jmolScriptWait
Eric Martz wrote: At 5/5/06, you wrote: It's probably time I told people about some of the prototype Jmol.js niceties in http://www.stolaf.edu/people/hansonr/jmol/test/proto/Jmol-new.js Dear Bob, This all looks extremely useful. At last there is a mechanism to keep track of when the applet has completed a task! And better yet, whether it completed the task successfully. 1. Is there any way to tell when the applet is initially loaded and ready for the first script? My inability to do this caused lots of head scratching when I began designing FirstGlance. This needs experimentation on a variety of platforms. 2. Is there any way to tell when the Protein Data Bank does not supply the PDB file requested in a load command? FirstGlance presently just leaves a blank applet (white, Jmol frank disappears), a very ungraceful failure. myVar = jmolScriptWait(load jmol.js) The file exist, but the return message is ERROR: ScriptException:unrecognized file format | Script line:load jmol.js 3. LACK OF JMOL.JS DOCUMENTATION. When I started coding FirstGlance, I decided not to use jmol.js. This was partly because it didn't support some functions I needed, but mainly because it is undocumented and I was afraid to use it. From inspection of the contents of jmol.js, it appears to do some sort of browser/java version testing, but I cannot tell what this is supposed to accomplish. Until someone documents the situations this is supposed to handle and what it is attempting to do, I will continue to be reluctant to use jmol.js for FirstGlance. I want to avoid a situation where a (common) client situation that I haven't tested causes an unanticipated failure mode. For example, no java in Windows, or a very old version of java. Right now, FirstGlance would fail to work at all, or to work correctly, without any explicit error messages. That is certainly not optimal. But I can't tell whether jmol.js would do better. You have to give Miguel credit for writing a VERY GOOD browser testing sequence into Jmol.js. Initially I felt the same as you -- too much of a black box. After using Jmol.js with touches of modification here and there, I can tell you that it is SUPERB. I don't think I'll ever write any page using Jmol without it now. I think you should sit down, take the time to understand it, and use it. No documentation will ever provide what you are asking for, I think -- a description of HOW it works. I'm not sure what more you could want than http://jmol.sourceforge.net/jslibrary/ One bit of code in jmol.js that particularly concerns me (and is inscrutable to me) is this: windowsClassId: clsid:8AD9C840-044E-11D1-B3E9-00805F499D93, windowsCabUrl: http://java.sun.com/update/1.5.0/jinstall-1_5_0_07-windows-i586.cab;, I wish someone would explain to me what this is designed to do, and how I could test it. Miguel can do that. More generally, I think we need thorough on-line documentation, in one place, on what jmol.js is designed to do, how to use the functions in it, etc. Is it possible you haven't seen http://jmol.sourceforge.net/jslibrary/ ? Bob -- Robert M. Hanson, [EMAIL PROTECTED], 507-646-3107 Professor of Chemistry, St. Olaf College 1520 St. Olaf Ave., Northfield, MN 55057 mailto:[EMAIL PROTECTED] http://www.stolaf.edu/people/hansonr Imagination is more important than knowledge. - Albert Einstein --- 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=lnkkid=120709bid=263057dat=121642 ___ Jmol-users mailing list Jmol-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jmol-users
