Good evening, I need some help, I should split an open mesh with curve as in the pictures. How do I do it with grasshopper?
Thank you
Definitely going to need a gh file with internalized geometry.
prova split.gh (2.8 MB)
thank you
Looks like something that @PeterFotiadis would know how to do.
He is a pro at scripting this kind of stuff.
You might try extruding your curve along the z axis and then splitting with the resulting extrusion… Considered trying that but the computing was taking to long.
thank you, i tried to do as you suggested but it requires a lot of time.
No need for code (well … in theory [and assuming lot’s of right Karma]). Just turn crv into a Mesh and split the ugly thing (or things):
Of course doing this via code would allow lot’s of checks and progress reports etc etc.
I’ve tried many things, often through ignorance about meshes, sometimes way too slow. Here is something that works quickly and reveals some things about the curve; it’s not always on the mesh surface (faces) and in some cases doubles back on itself.
prova split_2023Oct2b.gh (2.8 MB)
Improve the mesh or improve the curve? LoL - as it’s been pointed out, there are parts where the curve does not touch the mesh - then if this curve is projected to the mesh, the result is quite bad.
If you project the curve then clean it up a bit and extrude as a solid mesh the result is still not very useful.
Not necessarily too slow but not good either…but hey, it’s split. Split region is invalid, though. Use Combine&Clean afterwards I guess.
I didn’t even want to try mesh difference.
prova split.gh (2.8 MB)
Good luck
Thank You very much. So you have a .gh file to understand how to do?
If you can read C# … maybe. If you can’t consider a big switch (i.e. start walking the C# walk, time flies).
Mesh_Split_JustAStupidTest_V1.gh (389.8 KB)
TIP: Put your stuff as near the global origin as possible (Meshes work with floats … meaning possible failures due to accuracy etc etc).
NOTE: Avoid at any cost to overdue your splitter Polyline. This means: avoid small segments, too much “detail” etc etc. After all … all that are in a way like using multi Clip planes (so to speak). In the atteched no elaborated tests for the splitter are included (for instance: max N of vertices, min segment length, self Ccx Events, check splitterMesh for disjoined pieces [life sucks] … etc etc).
NOTE: If inspection is your goal then a rather better option (via OpenGL etc) is using collections of Clip planes. Rhino supports that kind of stuff (but in a very primitive way).
Like:
NOTE: Mesh Split works when it works (Karma a must).
Update: added some stuff more (and a splitter List support). But still you can get rather easily bananas since Mesh.Split is a bit temperamental (at best). That said the standard RC Mesh.Split Method using a List of splitters … well … it’s not the thing to use (thus an “explicit” Method is used).
Mesh_Split_JustAStupidTest_V1B.gh (389.9 KB)