occ6.5.3 BRepAlgoAPI_Cut gives wrong result - Forum Open Cascade Technology

Jane Hu
Tue, 08/14/2012 - 23:53

Forums:

I also found that BRepAlgoAPI_Cut code used to work for occ6.5.2, now using it with occ6.5.3 we have several testcases fail. It sends the shapes into the cutter, and did return a resultant shape, however, the BRepGProp::VolumeProperties(final_shape, myProps) returns myProps.Mass() = 0, which means the subject is gone after cut, which is wrong in many of our testcases.

I am not sure if the return shape is correct, just the volume property didn't get updated correctly or the return shape is wrong already, as in Timo's thread
http://www.opencascade.org/org/forum/thread_23599/?forum=3

Jane

Forum supervisor
Wed, 08/15/2012 - 12:55

Dear Jane Hu,
You also are welcome to register the issue in Mantis Bugtracker which is available via the Collaborative portal - http://dev.opencascade.org/index.php?q=home/get_involved.
Don't forget to put a data allowing to reproduce your case.
Regards

Jane Hu
Wed, 09/05/2012 - 23:43

Dear Forum Supervisor:

I've registered in the new website, however, it didn't grant me the right to report bugs.

Here, I attach 2 brep files, one contains a block, the other contains a sphere. They intersect each other. I want to use the box to cut the sphere, leaving part of the sphere as a result.

I used occ6.5.3, BRepAlgoAPI_Cut function, giving the from_shape as the sphere, and the tool_shape as the block. It returns a circular surface, instead of a solid.

This didn't happen in occ6.5.2 version.

Thanks for investigating it.

Jane

Attachments:

cutting_box.brep

Jane Hu
Wed, 09/05/2012 - 23:44

Another file is attached here. Thanks.

Attachments:

sphere.brep

Pawel
Thu, 09/06/2012 - 14:19

There's been the following fix integrated into the master branch this Monday:

* 0023374: BOP Common between a sphere and a box gives wrong result

I haven't compiled it yet but it might address the issue.

Pawel

Forum supervisor
Thu, 09/06/2012 - 16:10

Dear Jan Hu,
I would like to inform you that:
- the bug is reproduced in OCCT 6.5.3;
- the problem does not reproduced in current Git master
(most likely Pawel is right that it was fixed by #23374);
- anyway registering an issue in Mantis would be useful as it would gave one more test case
- I suggest you to contact Bugmaster via "Contact Us" form on dev portal in case of any problems using Mantis. He will help you with pleasure.
Regards

Jane Hu
Fri, 09/07/2012 - 00:02

Thank you for both Pawel and Forum supervisor. I've downloaded the fix and verified it works.

I have another bug related to Unite operation, in which a wheel is cut into two parts, and now I want to unite them into a wheel. The boolean operation gives a part of the wheel's surface as the result of the boolean, which is obviously wrong. But seems a lot like this Cut problem where a surface is returned for a solid.

I am wondering if there's a fix for it too.

Here I attach both the parts and 'unite' fails.

Jane Hu
Fri, 09/07/2012 - 00:03

please see the following thread for the models that reproduce the problem.

Thanks.

Jane

http://www.opencascade.org/org/forum/thread_22294/?forum=3