natschil

What's the status of this pull-request? I'm encountering some non-convex triangulations in practice, which is causing some code I have to fail.

I think there is still a bug in `isexternal`, as here `ranges` is taken to be a `ranges::NTuple{4,Float64} `, but only its first two values are used. Also, there should...

It also seems like `isexternal` doesn't do what I expect it to do, it seems like triangles that are not part of the triangulation don't end up with a sufficiently...

This still doesn't seem to be working, I'm still getting a non-convex triangulation when using the nodes (0.0,0.0),(0.0,1.0),(1.0,0.0),(1.0,1.0),(0.99,0.5). 

My plotting function wasn't specifying `ranges` to `iterate()` (i.e had some code that used a `for t in tess` construct) which was causing the erroneous results shown above. I agree...

I'm not sure this is the correct way to go about things on github, but I've now created a pull-request to MiriamHinzen:master with my changes.

An issue with this pull-request currently is that it results in some *very* large ranges if the data has some regularity. Here is an example (inspired by a real-life use...

To comment on my previous comment: perhaps the `scaleShiftPoints` function can be used to scale (without the cirumcircle checks) the points into a suitable subset of the interval [1,2] x...

I'm currently working on rewriting the triangulation code to (a) return a convex result and (b) be more readable/documented. I'll push the results to a pull-request soon.

I am referring to the part in the documentation where it says "1 + 1 = 2 means the test point is in front of a, exactly on the triangle"....