[Django]-Prefetching unrelated model on arbitrary field equivalence

by

0👍

✅

On a different note, why don’t you just read the whole querysets (for MyModel and MyOtherModel) into memory and then compare? Would be efficient in terms of number of queries to db

m_rows = MyModel.objects.filter(...)
m = m_rows.order_by('myField').values_list('myField')
m_set = set([x for x in m])
mo_rows = MyOtherModel.objects.all()
mo = mo_rows.values_list('myOtherField')
mo_set = set([x for x in mo])
common = list(m_set.intersection(mo_set))
mo_result = MyOtherModel.objects.filter(myOtherField__in=common).order_by('myOtherField')
for i in range(len(mo_result)):
    print("{:s}{:s}".format(m_rows[i], mo_rows[i]))

I believe this is overly complicated and there should be a better way to do this.
But this saves you n queries to db.

👤Gaurav Dhameeja

0👍

There are some points:
1- If these queries will run in an operational system with large data by a user request, you need to redesign your ER model or consider other solutions.
2-If you don’t have a lack of memory problem, you can load your whole objects of the models in memory with two queries.

MyOtherModelAll=MyOtherModel.objects.all()
for m in MyModel.objects.filter(...):
   for mo in MyOtherModelAll:
   if(mo.myOtherField=m.myField)
        print("{:s}{:s}".format(m, mo))

3. Annotate method tells Django to add an extra attribute to every object returned.The following sample me be usefull for you: 

Book.objects.annotate(search=searchVector  ('title','abstract'),).filter(search='django')
👤Azdy

Leave a comment