Determining only the inside angles of a closed random curve

Replies to This Discussion

Permalink Reply by Riccardo Majewski on September 28, 2015 at 1:38am

Sorry for late reply, Denis.

The main "problem" here is this: you curve is not just a polyline. The "top" segment is a nurb-curve and you often have small pieces co-linear to others.

My old definition you linked use fragmented patch (mesh) to find the direction of the curve, to determine the inside/outside, fragmented patch doesnt work with nurbs-curve... some fix needed....

Here is a possible solution: convert/clean your curve to a polyline...

1st solution: take start point of each segment and rejoin them into a polyline, strightening any non-linear segment (nurbs-curve segments).

2nd solution: use discontinuity points to rebuild the polyline, this also remove small co-linear segments, like merging them.

Then you can use the output curve/polyline to "feed" my old definition you linked.

This is not properly a correct solution, because the definition "wants" you to adapt the geometry.

I'm working in a definition that accept any shape...

Permalink Reply by Denys Demeshonok on October 1, 2015 at 10:36am

New files

Attachments:

• test_2015_10_01.3dm, 176 KB
• test_2015_10_01.gh, 34 KB

Permalink Reply by Riccardo Majewski on October 1, 2015 at 11:02am

PMing with you, I was talking about my older definition, which you linked here at OP.

It's ugly and dispersive, but at least works.

When I'll have time I will try to do a version with adaptive size for arcs and texts again...

Attachments:

• 3D_polyline_angle_dim_V2.gh, 48 KB

Permalink Reply by Denys Demeshonok on October 1, 2015 at 11:31am

Riccardo thank you!!!!