Boundary Surface fails to create the surface within the closed curve?

[Intuos]

I closing a bunch of curve pairs by connecting their open ends with line segments. However, there is this one case which fails; the joined curve consists of 5 vertices (see explode curve component) and as a result, the Boundary Surface fails outside the closed curve.

Boundary surface fails.gh (19.3 KB)

[leopoldomonzani]

After a rebuild it works, but the bug remains.

[inno]

wow, nice spot!

meanwhile the bug gets solved I’d go for network surface:

Boundary surface fails_Re.gh (12.5 KB)

[oana.angh]

edge surface command also works; i think it has to do with the input curves - if you use the close curve command, it will create two closed planar curves, then apparently trims the smaller out of the large one

[oana.angh]

the original code works with two randomly drawn curves in rhino; i’m not sure why it doesn’t work for the curves that are internalized.

[Intuos]

This method has always worked for me for years and years, but has become buggy for some reason. I’ve never had this issue before.

Version 8 SR23 (8.23.25251.13001, 2025-09-08)

[martinsiegrist]

Fit Curve helps

Boundary surface fit.gh (6.3 KB)

[oana.angh]

are the internalized curves two trimmed curves?

[Intuos]

They are offsets.

[oana.angh]

my feeling is there is something to do with the degrees of the input curves? since one is an offset of the first.

[Intuos]

Looks like the offset flipped one of the curve domains:

E: even Fitcurve leaves behind a 5th vertex. With 4 curve segments, there should be 4 vertices (in this case). The problem is that there is an overlapping vertex.

E2: A rectangle has 5 vertices too when exploded, so nvm.

[oana.angh]

but a deconstruct brep gives you 4 vertices i think when you explode a closed curve, it gives you the last vertex identical to the first one: you have n curves, and n+1 vertices as each curve has two end points. i think the issue is with the input curves, specifically with the second being an offset of the first, but i can’t yet say what.

[Volker_Rakow]

or Pull Curve or Sweep

@Intuos Yeah. The seam of a closed curve is repeated in the vertices.

[martinsiegrist]

Interesting. How did you do the offset?

Offsetting the outer curve in Grasshopper results in a correct solution.

Boundary surface offset.gh (4.7 KB)

[Volker_Rakow]

Yes, it clearly is the curves that are the problem, that much I’ve figured as well.

[martinsiegrist]

@Volker_Rakow

the curves

It would be nice to know how the offset was created.

[Intuos]

Sorry had to leave earlier.
Here’s the file with the offset with an internalised curve. Notice that you can bake the curve to replicate the screencast I posted earlier.
Boundary surface fails.gh (14.8 KB)

[martinsiegrist]

@Intuos

the file with the offset with an internalise

Here’s what I see when I open your file.

PS: I baked your previous curves so I could look at them in Rhino but could not see anything special.

[Intuos]

What’s the error?
My file is in meters FYI.

E: I should increase the tolerance a little.

[martinsiegrist]

Disconnecting the plane input solves the problem on my end.

What Rhino file do you use with the definition?