Gijs,

You're running into the flat data issue...basically you've got one
long list of points as opposed to being expressed as rows and
columns.  This is something that I believe David is working on at the
moment.

Your solution is actually fairly well done imho, although it makes a
number of assumptions right now which, depending on what you actually
want to do with this, might be good or bad.  It would seam to me that
you'd rather have the result be at least polylines instead of
individual segments.  This could be resolved by using the join
component.  However, if you wanted the curve to actually be on the
surface you would be in a situation where you'd have to use the
equivalent of interpolate curve on surface, which would not be able to
work with the flat data issue which would try to make 1 curve as
opposed to individual ones.  Another approach (and possibly a better
one right now) would be to project the curve(s) to the surface, which
would work as long as there was a distinct direction for the
projection.  Pulling the curve to the surface might be better, but
there's no out of the box solution for that.

I think even with a solution to the data flattening issue there will
still be a relatively manual component to creating something like
this. Since this organization of the uv points is not really 1 to 1
with its natural organization, there would have to be some means of
transferring into that format, but it would seam that once you have
the organization of the points worked out that the GH would be able to
handle it from there.  Since I'm don't know exactly how the data
flattening solution that David's coming up with actually works I can't
really say this for sure, but from the screen cap or two that he's
posted it should be at least more achievable.

Your other option at the moment is to sort all of this out in a
scripting component.  There you could organize the data without having
to resort to culling.  You'd have to create your lines (whether
polylines or interp curves on the surface) within the component to
prevent it from getting flattened again once it got out out of the
scripting component.  You also could pull the curve to the surface
within a scripting component which would be easier to code.

HTH,
Damien

On Feb 7, 8:52 am, Gijs <[email protected]> wrote:
> Hi Group,
>
> I just posted this 
> file:http://grasshopper3d.googlegroups.com/web/stripes+on+surface.ghx
>
> I'd appreciate if some experienced grasshoppers can take a look at it
> and give me some feedback if the way I am trying to handle things make
> sense.
>
> The file shows a surface that is being divided. I then tried and
> managed to connect points that are oriented in a diagonal order to one
> another. But I do have the feeling that it should be easier to
> acchieve. For example, there are quite some difficult culling tricks I
> used to cancel out lines I did not want.
>
> Basically what I would like to know, is if and how you can divide a
> surface into points, and then access individual points by their U/V
> index. I can imagine that a surface that is divided consists of two
> arrays: the U array and the V array and that I can then find a
> specific point, something like this: P (3) (6) giving me access to the
> fourth point in the U and 7th in V direction. Is that possible?
>
> thanks in advance,
> Gijs

Reply via email to