Divide Curve

Shouldn't it be 0 to 1?

Replies to This Discussion

Permalink Reply by David Rutten on January 28, 2016 at 10:56am

Is the curve closed? Can you post it?

Permalink Reply by Fred Becquelin on January 28, 2016 at 2:37pm

Yes it is a foil profile.

Attachments:

Exemple FoilData.3dm, 20 KB

Permalink Reply by Andrew Heumann on January 30, 2016 at 10:46am

It being closed makes sense then. If divide produced points at param 0 and param 1 on a closed curve, you'd always wind up with a duplicate point at the curve seam. With an open curve, you would expect to see the params go from 0 to 1. There are other differences in behavior - dividing an open curve by N segments, you get N+1 points whereas with a closed curve you get only N points.

Permalink Reply by Fred Becquelin on January 30, 2016 at 11:42am

Sorry Andrew, but if I read right, N is the number of segments, so I should expect N+1 points in any situation. That's not the case. I asked for ten segments and I got nine.

If you then plug a polyline component, the result will always be open. Isn't that inconsistent?

If there is a good reason for this, please explain again.

I mean, there IS a point at t=1... What's wrong with a duplicate point?

Permalink Reply by David Rutten on January 30, 2016 at 12:15pm

I asked for ten segments and I got nine.

You didn't get any segments actually, they are implied to exist in between the points (and parameters) you do get. When the curve is closed, the component simply omits the last point (as it would be coincident with the first one). The Rhino _Divide command works in the same way.

Permalink Reply by Fred Becquelin on January 30, 2016 at 12:49pm

Thanks David.

You're saying that it's not a bug, but a feature?

I know how the divide command works in rhino, and the Curve.DivideByCount method.

(including the MarkEnds / includeEnds option). I understand why duplicate points are an issue in Rhino, because the order in the list is irrelevant.

My point is : is it the behaviour we expect in grasshopper?

Now I changed my mind, I don't want points anymore, but perp frames.

On my foil profile (which has a kink at the start/end), the first and the last planes are not the same! Yet it behaves in the same way.

Permalink Reply by David Rutten on January 30, 2016 at 1:56pm

Yes it is designed this way. I assumed that one wouldn't want the same point twice when dividing a curve up. I can see though that sometimes you do want this and then you have to jump through a bunch of extra hoops.

Permalink Reply by Fred Becquelin on January 30, 2016 at 2:57pm

Ok sry for being a nuisance

Permalink Reply by David Rutten on January 30, 2016 at 3:27pm

Sorry there's no easy way to do what you need to do...

Permalink Reply by Pieter Segeren on January 30, 2016 at 4:06pm

Hi Fred, I think I might have something related lying around. It's an oldie so I'm not entirely sure it'll fit or contribute to your goal, but it reminded me of the trailing edge issue when using airfoils.

Attachments: