90 |
MNC_COMP_VTYPE_VAR( ind_vt, ind_fv_ids, ires, myThid ) |
MNC_COMP_VTYPE_VAR( ind_vt, ind_fv_ids, ires, myThid ) |
91 |
|
|
92 |
|
|
93 |
|
|
94 |
To-Do: |
To-Do: |
95 |
====== |
====== |
96 |
|
|
99 |
|
|
100 |
2) UNITS -- as with the names we need to follow conventions |
2) UNITS -- as with the names we need to follow conventions |
101 |
|
|
102 |
3) AM described her "diags" (or "myDiags" or "mDiags") interface |
3) CNH pointed out that grid interpolation needs to be handled |
|
which should use MNC for output. The data storage idea is similar |
|
|
to the MNC tables-of-indicies approach but also includes one huge |
|
|
double-precision "accumulator" to hold all the temporary values |
|
|
(eg. partial sums for averages, current max/mins): |
|
|
|
|
|
vname ( ni ) |
|
|
vlen ( ni ) |
|
|
vind ( ni ) ------+ |
|
|
| |
|
|
|
|
|
vij_diag ( i, j, [...] ) w/ lat/lon indicies |
|
|
vgl_diag ( [...] ) wo/ lat/lon indicies (global) |
|
|
|
|
|
4) CNH pointed out that grid interpolation needs to be handled |
|
103 |
"on-the-fly" since pre-processing would result in overly large |
"on-the-fly" since pre-processing would result in overly large |
104 |
input files. We need an interpolation API... |
input files. We need an interpolation API... |
105 |
|
|
|
5) From the group meeting on 2004/01/21, we need to define |
|
|
"sub-grids" corresponding to: |
|
|
|
|
|
var_name HGRID VGRID TIME |
|
|
=========================================== |
|
|
u U_xy r_c t |
|
|
eta T_xy - t |
|
|
rac T_xy - - |
|
|
|
|
|
And write a convenience wrapper so that users can write variables |
|
|
using just two function calls. JMC and I worked out the |
|
|
following: |
|
|
|
|
106 |
|
|