Dealing with nested relationships
Closed this issue · 5 comments
Hello,
I apologize if this isn't the right place to ask this question, but I am having a rough time trying to get this package working with nested relationships. I have tried using DRFs built in solution, DRF-Flex-Fields, and Dynamic Rest. None of these options seem to be able to expand my sub-serializers and apply the DRF-Datatables formatting appropriately.
Instead, what is happening is that the nested relationships are being stripped away and only the first layer is being presented within the data:[...] layer that is fed to the datatable.
Hello,
Sorry I can't help here, AFAIK Datatables expects a data array and does not know how to deal with nested structures.
At best it can do this but this is not currently supported by drf-datatables.
Hi,
The nested objects were the type of nested structures I was talking about. My API is nested in that fashion.
It seems that drf-datatables strips out nested layers during the _filter_unused_fields() method. I'm going to hack around with it to see if I can get it to filter unused fields while still maintaining the information from columns that have nested data.
I don't have much time to work on this, but the two quick and super-dirty options I've figured out so far are:
-
disabling
_filter_unused_fields()entirely and allowing a more robust API field filtering package to take over (e.g. DRF-Flex-Fields). -
adding
col.split('.').pop(0)to line 68 ofrenderers.pyas seen in my fork (warning: I haven't written a unit test for this yet). This will search the top level of the data tree and maintain all subsequent serializers. It's quick, it's dirty, and I hate it, but I don't really have time to think of a better way to recursively iterate through the data dict and filtering the appropriate fields at the appropriate level (i.e. recreating the Flex-Fields functionality).