You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We should get rid of the need to call fetch_atoms at the top of every recipe to transform a dict to Atoms object. The only reason it's there is for workflow codes like Parsl where calling .result() is necessary to pass a key/value pair from a completed task into a forthcoming task in the workflow and we don't want to block. However, fetch_atoms is not ideal because it has to be included in every single recipe.
It's also not the only instance where this comes up. For instance, when passing the bandgap between jobs, you have to pull it from the schema.
What new feature would you like to see?
We should get rid of the need to call
fetch_atoms
at the top of every recipe to transform adict
toAtoms
object. The only reason it's there is for workflow codes like Parsl where calling.result()
is necessary to pass a key/value pair from a completed task into a forthcoming task in the workflow and we don't want to block. However,fetch_atoms
is not ideal because it has to be included in every single recipe.It's also not the only instance where this comes up. For instance, when passing the bandgap between jobs, you have to pull it from the schema.
This hinges on:
[]
and.
operators into AppFutures Parsl/parsl#2904The text was updated successfully, but these errors were encountered: