split surface - domain issue

Hi,

I am trying to split a sphere with different curves lying on the surface.

The area that lies at the start or end of the surface domain will be automatically split in 2 parts. I guess the reason is that the split surface component reads the surface from the domain 0 to the domain 1 and therefore splits it although geometrically should be one piece.

Any suggestions how to solve this?

Many thanks,

kiara

Attachments:

split surface.jpg, 162 KB

Replies to This Discussion

Permalink Reply by Danny Boyes on March 22, 2011 at 11:23am

rotate either your sphere or your splitting curves so that they coincide with the seam.

Permalink Reply by kiara on March 22, 2011 at 11:28am

thanks Danny,

unfortunately these curves are the result of another process and will vary always location. I also need to consider the option of having just horizontal or vertical curves.

Probably I need to set up the sphere in a way that it adapts to this variable curve alignment.... I'll think it through...

many thanks,

Permalink Reply by Danny Boyes on March 22, 2011 at 11:35am

Another way might be to rotate one of the other subsrfs to replace the split one.

Permalink Reply by kiara on March 22, 2011 at 11:49am

they could alla be different (see image attached)...

Attachments:

irregular curves.jpg, 98 KB

Permalink Reply by Yoann Mescam (Systemiq) on March 22, 2011 at 12:26pm

I may have a fix, but it needs testing.

I solve the intersection for the splitting fragments and the domain start/end curve(isocurve V on u,v = 0,0). The fragments with only one intersection are joined.

Edit : too much joining with this, need to join only those intersecting on the domain frontier.

Permalink Reply by Yoann Mescam (Systemiq) on March 22, 2011 at 11:46pm

Sloppy piece of code, and it returns Breps instead of trimmed surfaces.

Attachments: