We Keep Coding

using F expression to access a field from one to many relationship not using the queryset resulted from prefetch_related method

hey guys let's say I have these models
class Object1:
.....
class Object2:
user = models.ForeignKey(User)
rel = models.ForeignKey(Object1, related_name = 'objects')
isCOmpleted = models.BooleanField()
and I wanted to perform this query:
Object1.objects.all().prefetch_related(Prefetch('objects', Object2.objects.filter(user = specific_user))).annotate(is_completed=F('objects__isCompleted'))
and the user is only related to one object from the Object2, but I got duplicate objects in my query, and I know the reason is that I have two objects from Object2 in my database for two different users, but the problem is that F expression didn't look in my prefetched query using prefetch_related method, I tried the exact same thing in the shell and it's giving me the results that I have expected but not in the view, so what's the problem here exactly any help would be really appreciated

Referencing twice related field for django model query

So I am using Django to construct a Query and I have 3 models as defined:
class Book(models.Model):
...
class Upload(models.Model):
...
book = models.ForeignKey(Book, on_delete=models.CASCADE)
class Region(models.Model):
...
page = models.ForeignKey(Upload, on_delete=models.CASCADE)
Given these 3 models I wanted a query that lists all the books and annotate them with a segmented_pages variable that contains the count of all the Upload that have non-zero number of regions.
Basically, counting the number of uploads per book that have atleast one region.
I am assuming the basic structure of the query would look like this and mainly the logic inside filter needs to be modified as there is no convenient count lookup.
Book.objects.annotate(segmented_pages=Count('upload', filter=Q(upload__region__count__gt=0))
Can someone please help me with the logic of the filter and a simple explanation of how to go about designing these types of queries using django models?

You can rewrite "non-zero number of regions" as "In the join produced by the query, the region for any upload must not be null", hence you can simply use isnull [Django docs]:
from django.db.models import Count, Q
Book.objects.annotate(
segmented_pages=Count(
'upload',
filter=Q(upload__region__isnull=False),
distinct=True
)
)

Django - Aggregate equal values of fields in model

I have activity logs for user activities, basically structured like this:
class ActivityLog(TimeStampedModel):
user = models.ForeignKey(User, on_delete=models.CASCADE)
action_type = models.CharField(max_length=25)
object_raw = models.CharField(max_length=500)
I want to aggregate all the entries where object_raw matches, i.e. so if a user searched for 'foo' on 4 different occasions, I get back one entry for 'foo', with count=4. I'm having trouble doing this right now. I know how to do it in sql, but don't understand that django syntax. I've been reading through the docs but I still don't get it. If anyone could help, it would be much appreciated!

To get one object with 'foo' and how many objects have object_raw='foo' you can do:
activity_logs = ActivityLog.objects.filter(object_raw='foo')
if activity_logs.exists():
activity_logs.first() # get one object
activity_logs.count() # get number of objects
If you just want how many objects have object_raw='foo', you can use conditional expressions with aggregates:
from django.db.models import IntegerField, Sum
Client.objects.aggregate(
num_object_raw=Sum(
Case(
When(object_raw='foo', then=1),
output_field=IntegerField()
)
)
)

Django: order by count of filtered foreign key

I have two models in Django. First model is User, the second model is Action. Action has foreign key to the User. Basically what I want to do is get users ordered by how many likes (actions) they did.
I would like to not change the models, because I have another functionality build upon it.
class User(models.Model):
is_superuser = models.BooleanField(default=False)
...
class Action(models.Model):
TYPES = (
("like", "User liked something"),
("comment", "User commeted on something"),
)
user = models.ForeignKey('User')
action = models.CharField(max_lenght=20, choices=TYPES)
This is how I was thinking to do this:
I need to filter the User model with some condition, this condition is quite complicated, but for the sake of an example say I want only superusers
then I would like to get only like actions from Action model for each user, that is only actions which have type equals to the "like". I don't want to filter users here, just filter the actions.
I want to group the actions by user_id and get their count for each user
then I would like to use the count to order the users
This is what I've tried:
users = User.objects.filter(is_superuser=True)
actions = Action.objects.filter(action="like", user_id__in=users).values('user') \
.annotate(action_count=models.Count('user'))
No idea what to do next. Maybe something like:
users.extra(select={'user_actions': actions}).order_by('user_actions')
But this doesn't work and results in ProgrammingError: syntax error at or near "[".
EDIT:
To make it more clear this is equivalent SQL of what I want to do:
SELECT * FROM user
INNER JOIN (SELECT action.user_id, COUNT(action.user_id) as action_count FROM action
WHERE action.action = "like" GROUP BY action.user_id) as ac
ON user.id = ac.user_id WHERE user.is_superuser = 1 ORDER BY ac.action_count DESC;
However I don't want to use SQL, because my condition on the User model is extremly complex and includes other joins as well, so it's better to do it using the ORM.

Get foreign key objects in a single query

I have 2 models in my Django code:
class ModelA(models.Model):
name = models.CharField(max_length=255)
description = models.CharField(max_length=255)
created_by = models.ForeignKey(User)
class ModelB(models.Model):
category = models.CharField(max_length=255)
modela_link = models.ForeignKey(ModelA, 'modelb_link')
functions = models.CharField(max_length=255)
created_by = models.ForeignKey(User)
Say ModelA has 100 records, all of which may or may not have links to ModelB
Now say I want to get a list of every ModelA record along with the data from ModelB
I would do:
list_a = ModelA.objects.all()
Then to get the data for ModelB I would have to do
for i in list_a:
i.additional_data = i.modelb_link.all()
However, this runs a query on every instance of i. Thus making 101 queries to run.
Is there any way of running this all in just 1 query? Or at least less than the 101 queries.
I've tried putting in ModelA.objects.select_related().all() but this didn't seem to have any effect.

As Ofri says, select_related only works on forwards relations, not reverse ones.
There's no built-in way to automatically follow reverse relations in Django, but see my blog post for a technique to do it reasonably efficiently. The basic idea is to get all the related objects for every item at once, then associate them manually with their related item - so you can do it in 2 queries rather than n+1.

Django ORM is a good thing but some some things is better to do manually.
You may import connection cursor and execute raw sql in single query.
from django.db import connection
cur=connection.cursor()
cur.execute(query)
rows = cur.fetchall()
your query should look like (for MySQL)
SELECT * FROM appname_modela INNER JOIN appname_modelb ON appname_modela.id=appname_modelb.modela_link_id

The reason .select_related() didn't work, is that .select_related() is used to follow foreign keys. Your ModelA doesn't have a foreign key to ModelB. Its ModelB that has a foreign key to ModelA. (so a ModelA instance can have multiple ModelB instances related to it).
You could use this to do it in 2 queries, and a bit of python code:
list_b = ModelB.objects.all()
list_a = ModelA.objects.all()
for a in list_a:
a.additional_data = [b for b in list_b if b.modela_link_id==a.id]

from django.db.models import OuterRef, Subquery
newest = ModelB.objects.filter(modela_link=OuterRef('pk'))
ModelA.objects.annotate(newest=Subquery(newest))
https://docs.djangoproject.com/en/3.2/ref/models/expressions/#subquery-expressions