acdbEcs2Wcs(..) function

[luciano51]

Hello! I never used this function indeed! I normally prefer to roughly set  the Ocs through the command UCS object an then to perform

a acdbUcs2Wcs() function (I now this method is deprecated for object enablers)

Nevertheless I have a curiosity about this function and I wonder if somebody could give me a suggetsion to use it properly.

So, the input parameters for this function are: a point (or a vector) and the normal vector of the planar entity.

I'm rather puzzled indeed... This means, for example,  that two circles, having the same normal, have the same Ocs! In this way the result

of translation will be the same for both circles even if they have different centers....

In other words: how can a coordinate system be determined by the Z axis direction only?

[owenwengerd]

Search for Arbitrary Axis Algorithm. I think that is what you're looking for.

[luciano51]

Thank you Owen! A very challenging matter! But..do you agree with me that this function,  as it is now, is totally useless?

I think is better to implement for my purposes an overloaded function, performing a more complete calculation of the

transforming matrix through the AA algorithm.

On the other hand I found the getEcs(...) method very disappointing....

Do you agree with this solution?

Thanks a lot.

Luciano

[owenwengerd · Accepted solution]

I don't think the function is totally useless. Transforming vectors is independent of position, so your argument only applies when translating points. I think it's fair to say that the function has limited applications.

[luciano51]

You are right, Owen. Programmers are always a bit selfish...

Taking advantage of your kindness I have another question: is there a way in ObjArx to modify the UCS position without using the UCS command?

Thank you very much.

Luciano

[owenwengerd]

I believe Art Cooney has in the past posted code to set the current UCS without using commands. However, I suspect that doing so is a bad idea for your purposes.