john-bodley commented on a change in pull request #15648:
URL: https://github.com/apache/superset/pull/15648#discussion_r668382831
##########
File path: superset/models/dashboard.py
##########
@@ -246,12 +247,27 @@ def data(self) -> Dict[str, Any]:
unless=lambda: not is_feature_enabled("DASHBOARD_CACHE"),
)
def datasets_trimmed_for_slices(self) -> List[Dict[str, Any]]:
- datasource_slices = utils.indexed(self.slices, "datasource")
+
+ # Efficient manner of processing datasources associated with a
dashboard. Rather
+ # than keying by the convenient `Slice.datasource` model (which
results in a
+ # per slice database query), we lazily request datasource after
deduping given
+ # that the same datasource could appear multiple times within a
dashboard.
+ datasource_slices: Dict[
+ Tuple(Type["BaseDatasource"], int), Slice
+ ] = defaultdict(set)
+
+ for slc in self.slices:
+ datasource_slices[(slc.cls_model, slc.datasource_id)].add(slc)
+
return [
# Filter out unneeded fields from the datasource payload
datasource.data_for_slices(slices)
- for datasource, slices in datasource_slices.items()
- if datasource
+ for (cls_model, datasource_id), slices in datasource_slices.items()
+ if (
+ datasource := db.session.query(cls_model)
Review comment:
Loving the walrus. Sadly I later found out we needed to support Python
3.7.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
