Poor choice of words on my part ... but spoken from the viewpoint of someone 
who is a bit intimidated by the idea of subclassing ;-)

Richard

On Nov 12, 2013, at 3:48 PM, Guyer, Jonathan E. Dr. wrote:

> 
> On Nov 12, 2013, at 12:15 PM, Daniel Wheeler <[email protected]> 
> wrote:
> 
>> On Mon, Nov 11, 2013 at 4:30 PM, Richard Gillilan <[email protected]> wrote:
>>> I'm rather new to FiPY too, but my impression is that viewers in FiPy are
>>> very basic ... really intended only for monitoring results during
>>> computation. So they don't have any real ability to customize. I'd love to
>>> learn otherwise.
>> 
>> In a sense that is a true. However you can subclass the viewer. You can
>> also get the matplotlib axis from the viewer and customize that way.
> 
> Yeah. I agree with the first statement: "viewers in FiPy are very basic ... 
> really intended only for monitoring results during computation", but strongly 
> disagree with the second: "they don't have any real ability to customize".
> 
> FiPy viewers are fully object oriented, so can be subclassed and customized 
> at will. Further, because everything is open source, you can use our existing 
> viewers to model a different implementation or to render with a completely 
> different tool than the ones we've set up. 
> 
> See 
> http://www.ctcms.nist.gov/fipy/documentation/FAQ.html#why-doesn-t-the-viewer-look-the-way-i-want
> _______________________________________________
> fipy mailing list
> [email protected]
> http://www.ctcms.nist.gov/fipy
>  [ NIST internal ONLY: https://email.nist.gov/mailman/listinfo/fipy ]
> 



_______________________________________________
fipy mailing list
[email protected]
http://www.ctcms.nist.gov/fipy
  [ NIST internal ONLY: https://email.nist.gov/mailman/listinfo/fipy ]

Reply via email to