search

hnarayanan / femtable

Web rendition of the periodic table of the finite elements.
http://www-users.math.umn.edu/~arnold/femtable/
7 stars 5 forks source link

P and dP elements from P and Pm families aren't differentiated by the popup generator #18

Closed hnarayanan closed 10 years ago

hnarayanan commented 10 years ago

Because the internal code doesn't differentiate beyond element names such as P1_tetrahedron, the generated popups for the corresponding elements in the P and Pm families aren't differentiated.

dnarnold commented 10 years ago

It seems to me that the most natural way to organize the elements is in terms of family (P-, P, Q-, S), and n (1, 2, 3), k (0, 1, ..., n), and r (1, 2, 3). This gives exactly the 108 boxes. Some attributes of these elements are shared. E.g., the symbol of (P-, 3, 3, 2) is the same as the symbol of (P, 3, 3, 1), namely dP1. But that doesn't have to be a problem. For example, for there could be 108 image files, with names like "P-_3_3_2.png" with some of these identical (or linked to a common file).

hnarayanan commented 10 years ago

I have fixed this.

dnarnold commented 10 years ago

Not quite. The DOF formula for the P family is in terms of the P- family and vice versa. Thus if you go to the element (family=P-,n=2,k=0,r=0) the DOF formula on its popup.

3 x P_0 Lambda^0(Delta_0)

which is correct. However when you go to (family=P,n=2,k=0,r=0) the formula should be

3 x P_0^- Lambda^0(Delta_0).

There is a duality here: the spaces entering the DOF formulas for the P- family are all from the P family and vice versa. The Q- family is self-dual in this sense. And the S family uses the cubic P family.

dnarnold commented 10 years ago

There is still a problem, which I commented on even though the issue is closed. Can you reopen?

On 10/13/2014 07:20 PM, Harish Narayanan wrote:

I have fixed this.

— Reply to this email directly or view it on GitHub https://github.com/hnarayanan/fe-table/issues/18#issuecomment-58973674.

dnarnold commented 10 years ago

Looking more carefully, this issue is unrelated to this bug. I'll open a new issue.

On 10/13/2014 07:30 PM, Douglas N Arnold wrote:

There is still a problem, which I commented on even though the issue is closed. Can you reopen?

On 10/13/2014 07:20 PM, Harish Narayanan wrote:

I have fixed this.

— Reply to this email directly or view it on GitHub https://github.com/hnarayanan/fe-table/issues/18#issuecomment-58973674.