Select and filter related models - django-models

I have the following models:
class Competition(models.Model):
# ...
class Result(models.Model):
# ...
competition = ForeignKey(Competition, on_delete=models.CASCADE,
related_name='individual_results',
related_query_name='individual_results')
athlete = ForeignKey(Athlete, on_delete=models.CASCADE)
class Athlete(models.Model):
# ...
What I'd like to do is create an overview for each athlete and display the results for that athlete grouped by competition.
Tried something like this:
athlete = Athlete.objects.get(pk=1)
competitions = Competition.objects
.prefetch_related('individual_results')
.filter(individual_results__athlete=athlete)
But when I do competition.indvidual_results.all in a template, it displays all results from the competition and not only the ones for that athlete.

A .filter(..) does not filter the related managers. It can be used to filter when you aggregate, but not items in the prefetch manager.
You can however use a Prefetch object [Django-doc], like:
results = Result.objects.filter(athlete_id=1)
competitions = Competition.objects.prefetch_related(
Prefetch('individual_results', queryset=results, to_attr='athlete_results')
)
Now the individual results for the athlete with id=1, are stored in an attribute athlete_results in the Competitions that arise from this queryset.

Related

Creating object with ManyToMany field via DRF ViewSet's perform_create

I have a simple model:
class Item(models.Model):
user = ForeignKey(User, related_name="user_items")
users = ManyToManyField(User, related_name="users_items")
I want it so that when a user creates an Item via ViewSet, that user is automatically assigned to the user and users fields.
I typically do this for ForeignKey's via the ViewSet's perform_create:
class ItemViewSet(viewsets.ModelViewSet):
...
def perform_create(self, serializer):
if self.request.user.is_authenticated:
serializer.save(user=self.request.user)
else:
serializer.save()
When I try to do it for the ManyToManyField too, I get this error:
{'users': [ErrorDetail(string='This list may not be empty.', code='empty')]}
I've tried the following in perform_create():
# 1
serializer.save(user=self.request.user, users=self.request.user)
# 2
serializer.save(user=self.request.user, users=[self.request.user])
# 2
serializer.save(user=self.request.user, users=[self.request.user.id])
How can I update a ManyToManyField via a ViewSet's perform_create?
Edit:
I guess the following works:
obj = serializer.save(user=self.request.user)
obj.users.add(self.request.user)
Is there no way to have the M2M field when the object is initially created though?
when you want set a list to m2m field one of the things you can do is this:
item_obj.users.set(user_list)
probably you need first get your item_obj.
for this you can get your object id from item_id = serializer.data.get('id') , and then item_obj = Item.objects.get(id = item_id)

Filtering on PageQuerySet (which is retuned by specific()) is returning FieldError in Wagtail

I am using subclassed model of Wagtail Page.
In below code you can see that PhoenixPage is base page which subclasses Wagtail Page model.
PhoenixArticlePage & PhoenixMealPrepPage subclasses PhoenixPage
PhoenixArticleIndexPage subclasses PhoenixBaseIndexPage which in turn subclasses PhoenixPage
Idea is to use PhoenixArticleIndexPage for all other article pages.
Problem is even after using the specific() method on queryset i am unable to use filter or any other operation on the queryset.
i tried using order_by() as well as filter()
Can someone share some insights here ? what might be wrong ?
Here is a model example:
class PhoenixPage(Page):
"""
General use page with caching, templating, and SEO functionality.
All pages should inherit from this.
"""
class Meta:
verbose_name = _("Phoenix Page")
# Do not allow this page type to be created in wagtail admin
is_creatable = False
tags = ClusterTaggableManager(
through=PhoenixBaseTag,
verbose_name="Tags",
blank=True,
related_name="phoenixpage_tags",
)
class PhoenixBaseIndexPage(PaginatedListPageMixin, PhoenixPage):
class meta:
verbose_name = "Phoenix Base Index Page"
app_label = "v1"
index_show_subpages_default = True
is_creatable = False
class PhoenixArticleIndexPage(PhoenixBaseIndexPage):
class Meta:
verbose_name = "Phoenix Article Index Page"
app_label = "v1"
class PhoenixArticlePage(PhoenixPage):
class Meta:
verbose_name = "Phoenix Article Page"
app_label = "v1"
subpage_types = []
parent_page_types = ["v1.PhoenixArticleIndexPage"]
class PhoenixMealPrepPage(PhoenixPage):
class Meta:
verbose_name = "Phoenix Meal Prep Page"
app_label = "v1"
subpage_types = []
parent_page_types = ["v1.PhoenixArticleIndexPage"]
Here are shell queries i tried.
Index page
In [4]: a = PhoenixArticleIndexPage.objects.all()[0]
In [5]: a
Out[5]: <PhoenixArticleIndexPage: articles>
As expected, get_children returning all instances of Wagtail Page.
In [6]: a.get_children()
Out[6]: <PageQuerySet [<Page: article title>, <Page: article title2>, <Page: Our 30-Day Reset Recipes Are So Easy AND Delicious>]>
Getting specific children from the Index page.
In [7]: a.get_children().specific()
Out[7]: <PageQuerySet [<PhoenixArticlePage: article title>, <PhoenixArticlePage: article title2>, <PhoenixMealPrepPage: Our 30-Day Reset Recipes Are So Easy AND Delicious>]>
Get Tag and try to filter the queryset
In [8]: q = a.get_children().specific()
In [12]: m = PhoenixTag.objects.get(slug='meal')
In [16]: k={"tags":m}
In [19]: q.filter(**k)
***FieldError: Cannot resolve keyword 'tags' into field. Choices are ...***
But if i go to particular entry in queryset then i can see tags field on it.
In [15]: q[2]
Out[15]: <PhoenixMealPrepPage: Our 30-Day Reset Recipes Are So Easy AND Delicious>
In [16]: q[2].tags
Out[16]: <modelcluster.contrib.taggit._ClusterTaggableManager at 0x1060832b0>
Could be different question all together but for reference adding it here.
Found the corner case of using difference() and specific() method on a queryset.
In [87]: q = PhoenixPage.objects.child_of(a).live()
In [89]: f = q.filter(featured=True)[:3]
In [91]: l = q.difference(f)
In [93]: l.order_by(a.index_order_by).specific() . <-- does not work
DatabaseError: ORDER BY term does not match any column in the result set.
The specific() method on PageQuerySet works by running the initial query on the basic Page model as normal, then running additional queries - one for each distinct page type found in the results - to retrieve the information from the specific page models. This means it's not possible to use fields from the specific model in filter or order_by clauses, because those have to be part of the initial query, and at that point Django has no way to know which page models are involved.
However, if you know that your query should only ever return pages of one particular type (PhoenixPage in this case) containing the field you want to filter/order on, you can reorganise your query expression so that the query happens on that model instead:
PhoenixPage.objects.child_of(a).filter(tags=m).specific()

Storing lists of words in database - best practice

I'm implementing a user filter system on a website. Users are to be able to select 'categories' and 'packages' of interest to them and have the matching data presented when they log in. Both sets of data will come from HTML select forms eg. Categories: 'null pointers', 'dead code'... and packages 'package_x', 'package_y', 'package_z'...
My question is about the best way to store this list information in a database (I am using Django and PostgresSQL).
My initial thought is to have a table like this:
user_id - one to one field
categories - textfield - store json data
packages - textfield - store json data
Is there a better way to be doing this?
I would go the route of using a user profile with categories and packages being many to many fields.
In models.py
from django.contrib.auth.models import User
class Category(models.Model):
name = models.CharField(max_length=255)
class Package(models.Model):
name = models.CharField(max_length=255)
class UserProfile(models.Model):
user = models.ForiegnKey(User)
categories = models.ManyToManyField(Category)
packages = models.ManyToManyField(Package)
In admin.py
from django.contrib.auth.admin import UserAdmin
from django.contrib.auth.models import User
class CustomUserAdmin(UserAdmin):
inlines = [UserProfile]
#filter_horizontal = ('',) #Makes the selection a bit more friendly
admin.site.unregister(User)
admin.site.register(User, CustomUserAdmin)
In views.py
user_with_profile = User.objects.get(pk=user_id).get_profile()
All that being said. Django 1.5 will replace the user profile with being able to use a configurable user model.

Tastypie filter by minimum value

I have a Django-tastypie resource that represents a banner and has a field called impression that I increment whenever the banner appears on the site.
class BannerResource(ModelResource):
owner = fields.ForeignKey('advertisment.api.AdvertiserResource', 'owner', full=True)
class Meta:
queryset = Banner.objects.all()
resource_name = 'banner'
authorization = Authorization()
I would like to get the banner that has the minimum impression, in the official documentation there is nothing like
filtering = {'impressions': ('min',)}
I'm using BackboneJS in the front end and I could get all the banners with Backbone collection and do the filtering with JavaScript but I'm looking for a quicker way to do it.
Any ideas?
Thanks
If you'd like to retrieve banners with number of impressions greater than X you need to things. For one you need to define possible filtering operations on your resource like so (given your model has impressions field):
class BannerResource(ModelResource):
owner = fields.ForeignKey('advertisment.api.AdvertiserResource', 'owner', full=True)
class Meta:
queryset = Banner.objects.all()
resource_name = 'banner'
authorization = Authorization()
filtering = { 'impressions' : ALL }
for available options take a look at Tastypie's documentation on filtering.
Then if you made the following request:
GET http://<your_host>/v1/banners?impressions__gte=X
you should get what you need.

How to update database in Django via an illegal intermediate state?

I'm trying to change some database entries from one legal state to another, but the intermediate (partially updated) state is not legal. As an example, suppose I'm modeling lectures, each of which is made up of several short topics in some order:
class Lecture(models.Model):
slug = models.TextField(
help_text='Abbreviated name of lecture.'
)
class Topic(models.Model):
slug = models.TextField(
help_text='Abbreviated name of topic.'
)
lecture = models.ForeignKey(
Lecture,
help_text='The lecture this topic is part of.'
)
order = models.IntegerField(
help_text='Impose ordering on topics.'
)
class Meta:
unique_together = (('lecture', 'order'),)
My test case is:
class TestTopicOrder(TestCase):
def test_reordering_topics(self):
# The lecture these topics are part of.
lecture = Lecture(title='Test Lecture', slug='lec')
lecture.save()
# Two topics 'zero' and 'one' in that order.
Topic(lecture=lecture, slug='zero', order=0).save()
Topic(lecture=lecture, slug='one, order=1).save()
# Try to invert the order.
t0 = Topic.objects.get(slug='zero')
t1 = Topic.objects.get(slug='one')
t0.order = 1
t1.order = 0
t0.save()
t1.save()
Essentially, I'm trying to do:
t0.order, t1.order = t1.order, t0.order
and then save, but whichever modified object I save first will have the same 'order' value as the other entry. I could delete and re-make, but when it comes time to re-order a dozen topics at once, that'll be a pain. What's the cleanest way to do this?
A dirty dirty solution... you could drop and recreate the restriction on the database using south api:
from south.db import db
db.delete_unique('app_lecture', ['lecture', 'order'])
# do your stuff
# then reenable the unique constraint...
db.create_unique('app_lecture', ['lecture', 'order'])

Resources