Reputation: 2444
Need to fetch aggregated data from 2 difrent tables.
Elements
element_type tempcolumn
xyz test1
pqr test2
xyz test3
Users:
User_names
auser
buser
cuser
need output in following format
element_type count
xyz 2
pqr 1
users 3
SQL Query Example:
SELECT element_type, count(*)
FROM Elements group by element_type
union
select 'users',count(*) from Users
can we excute same with django orm?
Upvotes: 0
Views: 648
Reputation: 2444
On Django, you can use |
to join two querysets but I'm not using it here.
Because values/annotate actually returns a list of tuples instead of a query_set
you can run raw SQL on Django but raw is for optimization. https://docs.djangoproject.com/en/1.9/topics/db/sql/
object_count = Elements.objects.all().values('element_type').annotate(total=Count('element_type'))
response_data = {}
for q in object_count:
if q['element_type'] == 'xyz':
response_data['total_ xyz'] = q['total']
if q['element_type'] == 'pqr':
response_data['total_ pqr'] = q['total']
response_data['users_count'] = MyUser.objects.all().count()
Upvotes: 1