3D Interoperability

[traceparts]

Hello,

Ok, here's the SAT file.

Thank you very much!

Best regards,

Olivier.

[ybiyani]

Hello traceparts,

You dont need to upgrade your product with R20SP1. You can also verify the behavior using one of the samples included in the product.

Regards,

Yogesh

[ejp]

olivier,

i just verified the problem still exists in r20sp1 and unfortunately using the option to convert analytics to splines does not work so as far as i can tell, this is a bug in the spatial code...

yogesh,

you should be able to reproduce this very easily by taking the acis file olivier attached to his previous message and run it through connect to create a step file.  you will see there are 2 toroidal surfaces created that have negative values for the major radius.  this is totally illegal in step and should be fixed by using the degenerate toroidal surface...

[traceparts]

Hello Yogesh,

I'm sorry but I don't understand your comment...

Which sample? And how can we check our problem using a sample?

Can you give me some pointers?

Best regards,

Olivier.

[ybiyani]

Hello traceparts,

Under the InterOp installation for R20SP1 you will see a samples folder. We have provided some standard samples for the customers to review/test the product. You can find the details of the samples here http://doc.spatial.com/index.php/InterOp:Connect/Using_Samples

We will log this as bug (as we have the all the relevant data - samples and problem description) but as I do not have an incident the best I can do is log as an internal bug for dev review. If you want us to associate this bug number with your organization please can create an incident and refer this post there. As this data was created in ACIS it would be great if you can also journal the acis api (see  http://doc.spatial.com/index.php/Journaling )  used to create this and attach to the incident it would help too.

Meanwhile as your customer is trying to import in V5 is there a way you can ask him to use ACIS->IGES or better ACIS->V5 as an alternative solution!

Regards,

Yogesh

[traceparts]

Hello,

> i just verified the problem still exists in r20sp1 and unfortunately using the option to convert analytics to splines does not work so as far as i can tell, this is a bug in the spatial code

In the meantime we have changed a little bit our code to use api_convert_to_spline( ENTITY* source, ENTITY* newEntity ... ) before calling the STEP translator (we are still using the legacy api_xstep_write() interop interface)

And this works for us! (At least, on this model. For other models, we have to do more tests)

So, if I understand correctly what you did in your test, that would mean that the option you have used with the new Connect interface does not behave the same way as the api_convert_to_spline() API?

Sounds strange...

Also, from our side, we hope that the use of the api_convert_to_spline() API will not break models that don't need this workaround in the first place.... (these models already work fine "as-is")

Best regards,

Olivier.

[ejp]

olivier,

i'll let you submit an incident since this is really your data and your problem...

with respect to the option in connect, i'm not sure why that doesn't do the conversion, maybe yogesh can shed some light on that...