Profile 1d export #1013
Profile 1d export #1013
Conversation
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## fenicsx #1013 +/- ##
===========================================
+ Coverage 95.30% 95.37% +0.07%
===========================================
Files 45 46 +1
Lines 2876 2920 +44
===========================================
+ Hits 2741 2785 +44
Misses 135 135 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
jhdark
left a comment
jhdark
left a comment
There was a problem hiding this comment.
Just some additions to the docs, and don't know if we can move some of the lines to the export class instead? Maybe not necessary
| if export._dofs is None: | ||
| index = self.species.index(export.field) | ||
| V0, export._dofs = self.u.function_space.sub(index).collapse() | ||
| coords = V0.tabulate_dof_coordinates()[:, 0] | ||
| export._sort_coords = np.argsort(coords) | ||
| x = coords[export._sort_coords] | ||
| export.x = x | ||
|
|
||
| c = self.u.x.array[export._dofs][export._sort_coords] | ||
|
|
||
| export.data.append(c) |
There was a problem hiding this comment.
Is this something that could be moved into the Profile1DExport class? Into say a function with the args index and u? Granted it wouldn't save that many lines, maybe not necessary.
There was a problem hiding this comment.
I was thinking of something like that but the trick is that this is fairly minimal code and it would be hard to make a generic function since it also stores data in export.
It could be a method of ProfileExport but then it would be different for the discontinuous case .... Happy to leave that for later
Co-authored-by: James Dark <[email protected]>
Co-authored-by: James Dark <[email protected]>
Proposed changes
Closes #685
Needs #1006
Types of changes
What types of changes does your code introduce to FESTIM?
Checklist