Robert Bradshaw wrote:

> On Dec 30, 2007, at 10:38 AM, Bob Hanson wrote:
>
>> Robert Bradshaw wrote:
>>
>
> Yep, I think this is the way to go. You mentioned curves, arrows, and  
> ellipses. What are the commands for curves/arrows? (If the  
> documentation is clear, just point me there...) And I couldn't figure  
> out a way to get a non-axis aligned ellipsoid (or one with three  
> separate radii).
>
curves, arrows, lines, points (spheres), planes -- all simple DRAW 
objects. Curves and arrows may be any number of segments, which are 
connected using a Hermite. Lines are two points. Planes are three 
(triangle) or four (quadrilateral) points.

>
>>>
>> this sounds very much like a simple set of Jmol script + associated  
>> files.
>
>
> Yep.
>
>> In fact, just recently I expanded the zip file reading to ANY file of
>> any sort read by Jmol. For example, right now in Jmol 11.3.65 you can
>> give the command:
>>
>>   script "spt.zip|ta.spt"
>>
>> which runs the script "ta.spt" found in the zip file "spt.zip".
>
>
> Cool. How does ta.spt refer to other files in that zip file?


Right now there's no way to specify "in my zip file" -- but I should add 
that --  so you would just specify the file name again, as for the 
script file.

  pmesh "spt.zip|meshes|mesh1.pmesh"

something like that.


> Just going to binary is going to do that (on both ends). The ascii  
> pmesh format is very simple and efficient to implement, with the  
> exception of the decimal conversions and having to close up the polygon.
>
> I would just as soon only allow one mesh per file--I'm thinking of  
> the TIFF file format that allows multiple images in a single file and  
> this greatly complicates the interface (not worried about the format,  
> the interface)--when reading a TIFF one always has to worry about  
> there being multiple images (though 99% of the time you want one) and  
> if there is more than one you have to worry about which one you want,  
> or how to show/pass two distinct images on to the user.
>
yes - well, zip files are fine, too, and a good binary header would 
allow jumping, but one problem is that web browsers must stream -- no 
random access!

> E.g. If I did "load binary_pmesh.blah" and there were two meshes, how  
> would the script distinguish them. And if it couldn't, why not call  
> it all the same pmesh (as it doesn't have to be connected)? Or is it  
> an issue of being able to reset the vertex counters (e.g. specify  
> some vertices, then some faces, then some more vertices, then some  
> more faces, etc.)
>
keeping them separate allows full control over color of each, 
translucency, mesh vs. fill, etc.
Quite possibly one-per-file and within a zip file would be the best. You 
get the compression, plus binary, plus readability. So then the files 
them selves could be extraordinarily simple:

PM[0x00][0x01]
nVertices
nPolyhedra
a few other items
float[] vertices.....
int[] polyhedra....

That would be about it.



> If you think this kind of thing would be useful, than we could still  
> do it though.
>
>> 2) we use as much already-present Jmol functionality as possible, not
>> duplicating that
>> 3) it provides what you really need, not an approximation
>
>
> I think Jmol + binary pmesh + zip archives will do an excellent job.
>
Let's go for that. I have exactly 24 hours to implement something like 
this just so you can try it out, then I'm off to Jamaica for a month 
without a whole lot of time for anything.


> - Robert
>


-- 
Robert M. Hanson
Professor of Chemistry
St. Olaf College
Northfield, MN
http://www.stolaf.edu/people/hansonr


If nature does not answer first what we want,
it is better to take what answer we get. 

-- Josiah Willard Gibbs, Lecture XXX, Monday, February 5, 1900



-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Jmol-developers mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/jmol-developers

Reply via email to