That sounds good. I think this would lend well to then creating super fields in the future which contain a mapping to their components. It would be cool to be able to set up your velocity field then apply arbitrary vector field calculations on it, without rewriting them specifically for velocity. Anyways, +1. On Fri, Apr 20, 2012 at 7:59 AM, david collins wrote:

That sounds great to me.

d.

On Fri, Apr 20, 2012 at 7:51 AM, Matthew Turk wrote:

...

Okay -- hearing no other suggestions (which is cool, this isn't that widely used or known of an ugly spot!) here's my solution, which I think I'll implement ASAP (definitely before another tagged release).

1) Remove all fields where vector_field = True 2) Split vector fields into components with X, Y, Z suffices. For 3.0 these will follow the new naming convention and be _x, _y, _z, but for now they should stick to what we have already for L components.

-Matt

On Wed, Apr 18, 2012 at 11:41 AM, Matthew Turk wrote:

...

Hi all,

I got an email from someone identifying a problem in profiles. This person was dealing with vector fields, which were giving bad results. I wanted to open up discussion here, as I know Jeff and Sam and others have thought about this. I'll contribute some thoughts after giving the mic over to someone else.

In an ideal world, how should a vector field -- like angular momentum, in particular -- be presented and dealt with in yt?

-Matt

---

yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org

-- Sent from my computer. _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org