CT in DE in CT bug.

CT in DE in CT bug.

Post by webmaste » Wed, 14 Feb 2007 03:26:19


Hello all,

There appears to be some sort of a bug when you nest a CT actor within a
DE actor which is, in turn, within a CT domain. Time does not ever advance
past 0. This problem is new with v6.0, and can be seen with a test-case
I've made at:

http://www.yqcomputer.com/

That example will run normally in v5, time advances and it plots a
straight line as you would expect. In v6 it sits at time 0.

I'm on a Mac, but I assume that's a non-issue for something as fundamental
as this. I have not, however, tested it on any other systems, so it may be
platform-specific.

Thanks,
Xavier Snelgrove


----------------------------------------------------------------------------
Posted to the ptolemy-hackers mailing list. Please send administrative
mail for this list to: XXXX@XXXXX.COM
 
 
 

CT in DE in CT bug.

Post by webmaste » Fri, 02 Mar 2007 14:06:35


----------------------------------------------------------------------------
Posted to the ptolemy-hackers mailing list. Please send administrative
mail for this list to: XXXX@XXXXX.COM
It seems like the problem may be more than that. The test-case I
posted was a stripped-down version of a more complex model in which
the internal CT model did have components which care about the
passage of time.
I've attached a new proof-of-concept showing this.

Xavier Snelgrove

 
 
 

CT in DE in CT bug.

Post by cxh » Sat, 10 Mar 2007 02:23:21

Hi Xavier,

I had a look at this and it looks like your previous example, which is
checked in to the CVS tree as
$PTII/ptolemy/domains/ct/test/CTinDEinCTbase.xml worked in the
2005-10-21 tree, but stopped working by 2005-11-01.

At the time, Haiyang was making a number of changes to the CT domain.
I don't have much confidence in tracking this down further, since the
changes during that time period are complex.

Haiyang, if you have a chance, could you take a look at these two
problems?

They are checked in as:
$PTII/ptolemy/domains/ct/test/CTinDEinCTbase.xml
$PTII/ptolemy/domains/ct/test/CTinDEinCT2.xml

_Christopher
--------


It seems like the problem may be more than that. The test-case I
posted was a stripped-down version of a more complex model in which
the internal CT model did have components which care about the
passage of time.
I've attached a new proof-of-concept showing this.

Xavier Snelgrove

----------------------------------------------------------------------------
Posted to the ptolemy-hackers mailing list. Please send administrative
mail for this list to: XXXX@XXXXX.COM