2003-june.txt

acis说明文档

I would also consider reviewing the Rhino data.  How was it constructed? (as
a solid?  or as just a bunch of faces?)  How did you export it?  (what
options were chosen during export?)  If you can export a correct solid to
begin with, you avoid all the trouble.  If not, well, you have your work cut
out for you.

Arlo

-----Original Message-----
From: B.J. Fregly [mailto:[e-mail]]
Sent: Tuesday, June 24, 2003 8:46 PM
To: Ames, Arlo
Cc: Yanhong Bei; [e-mail]
Subject: Re: [Acis-alliance] connectivity of several bodies


Arlo,

Thanks for your response to Yanhong.

This actually is a topology rather than geometry problem.

We are reading in SAT files created in Rhino 2.0, and the files contain an
unknown number of faces. Some of these faces share common edge curves and
form shells (i.e., groups of contiguous faces), but ACIS does not seem to
recognize this connectivity when the SAT file is read in (though maybe we
are missing something here).

Our goal is to automatically determine which faces should be grouped
together into shells (and eventually bodies for use with various api
functions like api_raytest_body).

I just checked the on-line help, and it appears that a function like
"api_stitch" may do the job for us. We'll poke around there some more and
see if that will work.

Thanks again for your input on this problem.

B.J.

B.J. Fregly, Ph.D.
Assistant Professor
Department of Mechanical & Aerospace Engineering,
Department of Biomedical Engineering, and
Department of Orthopaedics & Rehabilitation (adjunct)
231 MAE-A Building
P.O. Box 116250
University of Florida
Gainesville, FL  32611-6250
Tel: [phone]
Fax: [phone]
E-mail: [e-mail]
Web: http://www.mae.ufl.edu/~fregly

----- Original Message ----- 
From: "Ames, Arlo" <[e-mail]>
To: <mailto:[e-mail]>
Cc: <[e-mail]>
Sent: Tuesday, June 24, 2003 7:01 PM
Subject: RE: [Acis-alliance] connectivity of several bodies


> Can you be more specific about your definition of "connected together"?
> Functions like
> api_entity_entity_distance should be sufficient unless you're looking for
> overlap, in which case api_slice/api_imprint/api_boolean_stage_one would
> perhaps be better.  The boolean-related calls are rather slow compared to
> distance-type calls, but give more complete information (including
locations
> of the overlaps) in cases where the objects overlap.  If no overlap
occurs,
> the distance functions will tell you about near misses.  Of course,
> "connected" can also mean that they share topology, instead of thinking in
> purely geometric terms.  This interpretation is less likely, but I've seen
> it.  Detection of topology sharing isn't likely to be found in an api
> function.
>
> Arlo Ames
>
>
> -----Original Message-----
> From: Yanhong Bei [mailto:[e-mail]]
> Sent: Tuesday, June 24, 2003 2:08 PM
> To: [e-mail]
> Subject: [Acis-alliance] connectivity of several bodies
>
> Dear All,
>
> I have several entities ( actually they are bodies) in an entity list. I
> am
> trying to find out among them the entities that are connected together.
> I
> tried api_entity_entity_touch() and api_entity_entity_distance(),
> however,
> neither was able to give me the correct results.
>
> Do you think there are other ways to test the connectivity between
> entities?
>
> Thanks for your help,
>
> Yanhong Bei
> Ph.D Student
> The Department of  Mechanical
>  and Aerospace Engineering
> University of Florida
> Gainesville, FL, 32611
> (Tel: [phone])
>
> _______________________________________________
> Acis-alliance mailing list
> [e-mail]
> http://lists.cs.columbia.edu/mailman/listinfo/acis-alliance
>
> _______________________________________________
> Acis-alliance mailing list
> [e-mail]
> http://lists.cs.columbia.edu/mailman/listinfo/acis-alliance
>
> _______________________________________________
> Acis-alliance mailing list
> [e-mail]
> http://lists.cs.columbia.edu/mailman/listinfo/acis-alliance
>

_______________________________________________
Acis-alliance mailing list
[e-mail]
http://lists.cs.columbia.edu/mailman/listinfo/acis-alliance

From yhbei at mae.ufl.edu  Wed Jun 25 15:14:33 2003
From: yhbei at mae.ufl.edu (Yanhong Bei)
Date: Wed Jun 25 14:16:49 2003
Subject: [Acis-alliance] connectivity of several bodies
References: <[e-mail]>
Message-ID: <00c201c33b45$a33e09c0$cd2ae30a@Calcaneus>

Hi,

First, I'd like to clarify the problem a little bit.

I import a sat file that I know has some big patches, but not know the
number of the patches.  Each of these big patches consists of dozens of
little surfaces which are connected with its neighbors.  Just like a chest
board except that the patches are 3D and curved.  When I read in the sat
file, all I get is an ENTITY_LIST (the ENTITIES are either BODYs or FACEs).
My goals are to (1) find the entities that belong to a certain big patch and
(2) build a body out of them.

For (1), I posted the question and asked if there is a way in ACIS to test
the connectivity of the entities in the ENTITY_LIST.  Actually,
api_entity_entity_touch() can do the job.  And I used the algorithm of
breadth-first search to group them.  This has been done.

For (2), api_stitch() seems to be a good choice, but I have not got a chance
to test it.

Thank you for your help with the problem.  We really appreciate it.

Yanhong
----- Original Message -----
From: "Ames, Arlo" <[e-mail]>
To: "'B.J. Fregly'" <[e-mail]>; "Ames, Arlo" <[e-mail]>
Cc: "Yanhong Bei" <[e-mail]>; <[e-mail]>
Sent: Wednesday, June 25, 2003 12:30 PM
Subject: RE: [Acis-alliance] connectivity of several bodies


> If you have imported a single solid body, you should have a BODY object
with
> likely 1 LUMP, which has 1 SHELL, which has a FACE which likely has a next
> FACE, and so on.  Each face should have at least 1 LOOP, each of which
> should have 1 or more COEDGEs.  Each COEDGE should have an EDGE, and for a
> closed object, each COEDGE should have a partner, which refers to the same
> edge it did, with opposite sense.
>
> api_check_entity (sp?) can check to see whether you have a valid, closed
> solid, and will complain about deviation from what's stated above (along
> with complaining about accuracy issues and so on as well.
>
> If you have a bunch of disconnected faces, api_stitch will do the job, IF
> they have identical (or near identical) EDGEs to stitch together.  It's
> likely your best bet.
>
> I would also consider reviewing the Rhino data.  How was it constructed?
(as
> a solid?  or as just a bunch of faces?)  How did you export it?  (what
> options were chosen during export?)  If you can export a correct solid to
> begin with, you avoid all the trouble.  If not, well, you have your work
cut
> out for you.
>
> Arlo
>
> -----Original Message-----
> From: B.J. Fregly [mailto:[e-mail]]
> Sent: Tuesday, June 24, 2003 8:46 PM
> To: Ames, Arlo
> Cc: Yanhong Bei; [e-mail]
> Subject: Re: [Acis-alliance] connectivity of several bodies
>
>
> Arlo,
>
> Thanks for your response to Yanhong.
>
> This actually is a topology rather than geometry problem.
>
> We are reading in SAT files created in Rhino 2.0, and the files contain an
> unknown number of faces. Some of these faces share common edge curves and
> form shells (i.e., groups of contiguous faces), but ACIS does not seem to
> recognize this connectivity when the SAT file is read in (though maybe we
> are missing something here).
>
> Our goal is to automatically determine which faces should be grouped
> together into shells (and eventually bodies for use with various api
> functions like api_raytest_body).
>
> I just checked the on-line help, and it appears that a function like
> "api_stitch" may do the job for us. We'll poke around there some more and
> see if that will work.
>
> Thanks again for your input on this problem.
>
> B.J.
>
> B.J. Fregly, Ph.D.
> Assistant Professor
> Department of Mechanical & Aerospace Engineering,
> Department of Biomedical Engineering, and
> Department of Orthopaedics & Rehabilitation (adjunct)
> 231 MAE-A Building
> P.O. Box 116250
> University of Florida
> Gainesville, FL  32611-6250
> Tel: [phone]
> Fax: [phone]
> E-mail: [e-mail]
> Web: http://www.mae.ufl.edu/~fregly
>
> ----- Original Message -----
> From: "Ames, Arlo" <[e-mail]>
> To: <mailto:[e-mail]>
> Cc: <[e-mail]>
> Sent: Tuesday, June 24, 2003 7:01 PM
> Subject: RE: [Acis-alliance] connectivity of several bodies
>
>
> > Can you be more specific about your definition of "connected together"?
> > Functions like
> > api_entity_entity_distance should be sufficient unless you're looking
for
> > overlap, in which case api_slice/api_imprint/api_boolean_stage_one would
> > perhaps be better.  The boolean-related calls are rather slow compared
to
> > distance-type calls, but give more complete information (including
> locations
> > of the overlaps) in cases where the objects overlap.  If no overlap
> occurs,
> > the distance functions will tell you about near misses.  Of course,
> > "connected" can also mean that they share topology, instead of thinking
in
> > purely geometric terms.  This interpretation is less likely, but I've
seen
> > it.  Detection of topology sharing isn't likely to be found in an api
> > function.
> >
> > Arlo Ames
> >
> >
> > -----Original Message-----
> > From: Yanhong Bei [mailto:[e-mail]]
> > Sent: Tuesday, June 24, 2003 2:08 PM
> > To: [e-mail]
> > Subject: [Acis-alliance] connectivity of several bodies
> >
> > Dear All,
> >
> > I have several entities ( actually they are bodies) in an entity list. I
> > am
> > trying to find out among them the entities that are connected together.
> > I
> > tried api_entity_entity_touch() and api_entity_entity_distance(),
> > however,
> > neither was able to give me the correct results.
> >
> > Do you think there are other ways to test the connectivity between
> > entities?
> >
> > Thanks for your help,
> >
> > Yanhong Bei
> > Ph.D Student
> > The Department of  Mechanical
> >  and Aerospace Engineering
> > University of Florida
> > Gainesville, FL, 32611
> > (Tel: [phone])
> >
> > _______________________________________________
> > Acis-alliance mailing list
> > [e-mail]
> > http://lists.cs.columbia.edu/mailman/listinfo/acis-alliance
> >
> > _______________________________________________
> > Acis-alliance mailing list
> > [e-mail]
> > http://lists.cs.columbia.edu/mailman/listinfo/acis-alliance
> >
> > _______________________________________________
> > Acis-alliance mailing list
> > [e-mail]
> > http://lists.cs.columbia.edu/mailman/listinfo/acis-alliance
> >
>
> _______________________________________________
> Acis-alliance mailing list
> [e-mail]
> http://lists.cs.columbia.edu/mailman/listinfo/acis-alliance
>
> _______________________________________________
> Acis-alliance mailing list
> [e-mail]
> http://lists.cs.columbia.edu/mailman/listinfo/acis-alliance
>