Forum Discussion
Basic filter in power query editor is not filtering
- 4 years ago
Hi Anand and thank for your reply!
For the true/false-column I use the drop-down to deselect false. Case sensitivity should not be an issue for a boolean I'm guessing?
I have successfully filtered now by creating an index column before the filtering, as this somehow has something to do with something called query folding which I know nothing about (see post in other comment thread for link), but the index column seem to work.
- 4 years ago
Well done on finding a way round it.
I found some posts and this in the ideas forum https://community.powerbi.com/t5/Issues/Bug-Filtering-in-Query-Editor-not-working/idi-p/442857
Along with the posts you linked to, and your troubleshooting, it's an odata source problem.
Query Folding is when the Power Query transforms get rolled up and sent to the data source for processing. We always aim for it because it makes for speedy and efficient processing. Unfortunately with OData, it seems the transforms won't work unless there are certain properties set in OData (that's from the investigations jamesf did in one of those other posts).
So to get round this, we have to break Query Folding (you've done this with an Index, I think). The post I linked to uses Table.Buffer to load the table in memory and that worked too.
Hope this helps (and for anyone else looking for a solution).
Well done on finding a way round it.
I found some posts and this in the ideas forum https://community.powerbi.com/t5/Issues/Bug-Filtering-in-Query-Editor-not-working/idi-p/442857
Along with the posts you linked to, and your troubleshooting, it's an odata source problem.
Query Folding is when the Power Query transforms get rolled up and sent to the data source for processing. We always aim for it because it makes for speedy and efficient processing. Unfortunately with OData, it seems the transforms won't work unless there are certain properties set in OData (that's from the investigations jamesf did in one of those other posts).
So to get round this, we have to break Query Folding (you've done this with an Index, I think). The post I linked to uses Table.Buffer to load the table in memory and that worked too.
Hope this helps (and for anyone else looking for a solution).
I would never have guessed that it's a specific problem with odata but your questions and suggestion highlighted the proper end in which to begin investigation. Thanks a lot for the help and explanation!