[MOAB-dev] commit/MOAB: janehu: Fixed the bug Andy Davis reported for getting graphics for edges, the first tol is angle tolerance, in degree.

[Jiangtao Hu]

Sorry, I just did a wrong statement, the distance tolerance is important if you have a curved edge which changes curvature, like BezierCurve or spline curve. 
 
Jane Hu


Asst. Researcher
Dept. of Engineering Physics
UW @ Madison


"And we know that for those who love God, that is, for those who are called according to his purpose, all things are working together for good." (Romans 8:28)



On , Jiangtao Hu <jiangtao_ma at yahoo.com> wrote:
 
Yes, the interface before was just one tolerance, which in OCC is interpreted as angle tolerance, cubit didn't give a good documentation. Yet, get_graphics is to get system faceting of geometry, not a user set mesh. So Angle tolerance is more important than distance tolerance for curved edges and less important than distance tolerance for straight large differed edge lengths. 

I don't have a cgm on cubit12.2 build, can Andy double check for me if the change of this tolerance will get different facet data? If not, then the system is not using this option; or can you change the tolerance to 5 for a circular edge to see if you get 73 points; or change the tolerance to 1/36 for the same edge to see if you get 73 points. I wouldn't think it uses the tolerance for a distance, but I could be wrong.

Now in cubit13.1, it added more accuracy to several functions, including reflect and scale, and get_graphics for edge has 2 tolerances and an edge_length option:

angle_tolerance, 

distance_tolerance,

max_edge_length


while in OCC, the edge_length option is not used, since this function is not a mesh function, rather a graphics call. There is no corresponse if this argument get changed. 

If distance tolerance is needed as you have large differences in edge lengths, I can add it separately for cubit13.1 and above. Please let me know.
 
Jane 


Asst. Researcher
Dept. of Engineering Physics
UW @ Madison


"And we know that for those who love God, that is, for those who are called according to his purpose, all things are working together for good." (Romans 8:28)



On Tuesday, December 3, 2013 11:24 AM, Paul Wilson <wilsonp at engr.wisc.edu> wrote:
 
Hi Jane

Did this interface change? In CGM? In ACIS?

Since faceting tolerance is the primary parameter we use, it probably needs to be still passed into the get_graphics method, perhaps as well as a default norm_tol.

Paul

-- --
Professor, Nuclear Engineering
Faculty Director, Advanced Computing Infrastructure
UW-Madison
wilsonp at engr.wisc.edu, 608.263.0807
Sent from my iPhone


On Dec 3, 2013, at 11:17, commits-noreply at bitbucket.org wrote:

> 1 new commit in MOAB:
> 
> https://bitbucket.org/fathomteam/moab/commits/119198dc4d96/
> Changeset:   119198dc4d96
> Branch: 
     master
> User:        janehu
> Date:        2013-12-03 18:15:13
> Summary:     Fixed the bug Andy Davis reported for getting graphics for edges, the first tol is angle tolerance, in degree.
> 
> Affected #:  1 file
> 
> diff --git a/src/io/ReadCGM.cpp b/src/io/ReadCGM.cpp
> index 13b64fb..9b6ba09 100644
> --- a/src/io/ReadCGM.cpp
> +++ b/src/io/ReadCGM.cpp
> @@ -474,7 +474,7 @@ ErrorCode ReadCGM::load_file(const char *cgm_file_name,
>     RefEdge* edge = dynamic_cast<RefEdge*>(ci->first);
>     Curve* curve = edge->get_curve_ptr();
>     data.clean_out();
> - 
   edge->get_graphics( data, faceting_tol);
> +    edge->get_graphics( data, norm_tol);
>     if (CUBIT_SUCCESS != s)
>       return MB_FAILURE;
> 
> Repository URL: https://bitbucket.org/fathomteam/moab/
> 
> --
> 
> This is a commit notification from bitbucket.org. You are receiving
> this because you have the service enabled, addressing the recipient of
> this email.
> 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/moab-dev/attachments/20131203/c6009441/attachment.html>