I know this has been answered before, but I have been through every answer and nothing either makes sense or I feel like my code is already doing what is being said in the answers.
So I have a model for terms and links, which has a many-to-many relationship.
class ProjectTerms(models.Model):
id = models.IntegerField(primary_key=True) # AutoField?
terms = models.CharField(max_length=100)
class Meta:
db_table = 'project_terms'
class ProjectLinks(models.Model):
id = models.IntegerField(primary_key=True) # AutoField?
links = models.CharField(max_length=100)
relations = models.ManyToManyField(ProjectTerms)
class Meta:
db_table = 'project_links'
class ProjectLinksRelations(models.Model):
id = models.IntegerField(primary_key=True) # AutoField?
project_terms = models.ForeignKey('ProjectTerms')
project_links = models.ForeignKey(ProjectLinks)
Script:
def create_models(my_term, my_link):
saved_term = ProjectTerms.objects.update_or_create(terms = my_term)
saved_link = ProjectLinks.objects.update_or_create(links = my_link)
ProjectLinksRelations.objects.update_or_create(project_terms=saved_term, project_links=saved_link)
A lot of places say that the model (both term and link here) has to be saved before being added to the connecting database.
I've also tried creating each term and link separately and saving them like this:
def create_models(my_term, my_link):
saved_term = ProjectTerms(terms = my_term)
saved_term.save()
saved_link = ProjectLinks(links = my_link)
saved_link.save()
relation = ProjectLinksRelations.objects.update_or_create(project_terms=saved_term, project_links=saved_link)
relation.save()
All you need for this is:
class ProjectTerms(models.Model):
terms = models.CharField(max_length=100)
class ProjectLinks(models.Model):
links = models.CharField(max_length=100)
relations = models.ManyToManyField(ProjectTerms)
The ID fields and the table for the ManyToManyField will be automatically created.
To set the ManyToMany relationship, use add:
def create_models(my_term, my_link):
(saved_term, created) = ProjectTerms.objects.update_or_create(terms = my_term)
(saved_link, created) = ProjectLinks.objects.update_or_create(links = my_link)
saved_link.relations.add(saved_term)
saved_link.save()
Related
In the following model...
class Question(models.Model):
question_text = models.CharField(max_length=200)
likes = models.IntegerField(default=0)
dislikes = models.IntegerField(default=0)
pub_at = models.DateTimeField(auto_now_add=True)
category = models.ForeignKey(
Category, on_delete=models.SET_NULL, null=True)
def __str__(self):
return f"{self.question_text}"
def validity(self):
total_likes = self.likes + self.dislikes
if total_likes != 0:
return (self.likes / total_likes) * 100
else:
return 100
I want to be able to access Question.objects.get(pk=1).validity() assuming that pk=1 exists in this case. In python shell I can do this easily. But how do I do this using React. I am able to get all my questions and the fields in React without a problem but I don't think I have a way to access the validity method I created.
In this case I would suggest the following. First, remove the property from the model:
# models.py
class Question(models.Model):
question_text = models.CharField(max_length=200)
likes = models.IntegerField(default=0)
dislikes = models.IntegerField(default=0)
pub_at = models.DateTimeField(auto_now_add=True)
category = models.ForeignKey(
Category, on_delete=models.SET_NULL, null=True)
def __str__(self):
return f"{self.question_text}"
Then add a SerializerMethodField (docs) to your serializer. It is read-only and can be used to pass computed values to your views:
# serializers.py
class QuestionSerializer(serializers.ModelSerializer):
validity = serializers.SerializerMethodField()
class Meta:
model = Question
fields = ['question_text', 'likes', 'dislikes', 'pub_at', 'category', 'validity']
def get_validity(self, instance):
total_likes = instance.likes + instance.dislikes
# Your approach is not wrong. This is a more explicit way of dealing with that particular error type
try:
return (instance.likes / total_likes) * 100
except ZeroDivisionError:
return 100
Bear in mind that the Foreign Key category will be serialized as its database unique id value (Primary Key) in this case.
You might want to use the #property decorator so that you can access the value the same way you would access any of the other fields on your Question model:
class Question(models.Model):
question_text = models.CharField(max_length=200)
likes = models.IntegerField(default=0)
dislikes = models.IntegerField(default=0)
pub_at = models.DateTimeField(auto_now_add=True)
category = models.ForeignKey(
Category, on_delete=models.SET_NULL, null=True)
def __str__(self):
return f"{self.question_text}"
#property
def validity(self):
total_likes = self.likes + self.dislikes
percentage = (self.likes / total_likes) * 100
return percentage
Explanations can be found in the docs or here. Keep in mind that it will not be saved like the other attributes as columns on the database when you run migrations.
I am answering my own question here I found a solution to. Although, #property does work when rendering using a simple Django template when using React and rendering json responses validity is still not available.
In my serializers.py file I did the following...
class QuestionSerializer(serializers.ModelSerializer):
validity = serializers.ReadOnlyField()
class Meta:
model = Question
fields = '__all__'
Take away the #property from the models as it is no longer needed. This has worked for me and you can go to the Django rest_framework or test it in your React application to see if you have access to this.
I would like to know if there are any issues doing this and/or a better way. I was also trying to do validity = serializers.Field() instead of validity = serializers.ReadOnlyField() but got an error saying I needed a Field.to_representation() that takes in self, value as positional arguments.
What arguments exactly do I pass in here. I tried self, Question.validity and did not work. I am not sure what I am doing here.
As an update the method in the model I updated to...
def validity(self):
total_likes = self.likes + self.dislikes
if total_likes != 0:
return (self.likes / total_likes) * 100
else:
return 100
I did not notice before and does not really matter for the question but division by zero is not allowed being that by default division by zero will always occur.
I am trying create a new database entry using a custom Django model I created. However, when I try to create the model and save it, the id does not increment. Instead, the previous database entry is overwritten whose id == 1. I have tried setting force_insert=True inside the save() function, but it results in a runtime error where the primary key already exists. I don't set any primary values in the creation of the object, so I'm not sure why the id is not being incremented. I am running the test code in the manage.py shell. All the models have been migrated properly.
The model:
class RoadWayData(models.Model):
blocked_lanes = models.PositiveIntegerField()
city = models.CharField(max_length=255)
county = models.CharField(max_length=255)
direction = models.CharField(max_length=255, blank=True, null=True, default=None)
eto = models.CharField(max_length=255)
incident_type = models.ForeignKey(ContentType, on_delete=models.CASCADE)
incident_object = GenericForeignKey('incident_type', 'id')
injuries = models.PositiveIntegerField()
postmile = models.CharField(max_length=255, blank=True, null=True, default=None)
queue = models.CharField(max_length=255, default="NONE: Freeflow Conditions")
route = models.CharField(max_length=255, blank=True, null=True, default=None)
street = models.CharField(max_length=255, blank=True, null=True, default=None)
timestamp = models.DateTimeField(auto_now=True)
update = models.PositiveIntegerField()
maintenance = models.CharField(max_length=255)
tow = models.CharField(max_length=255)
weather = models.CharField(max_length=255)
vehicles_involved = models.PositiveIntegerField()
The test code:
from incident.models import *
import datetime
x = IncidentIndex.objects.get(id=1)
y = CHPIncident.objects.get(id=x.incident_object.id)
print("ID already exists in DB: {}".format(RoadWayData.objects.get(id=1).id))
z = RoadWayData(
blocked_lanes=0,
city="testCity",
county="testCounty",
direction="NB",
eto="Unknown",
highway_accident=True,
incident_object=y,
injuries=0,
postmile="New Postmile",
route="new Route",
update = 2,
maintenance= "Not Requested",
tow="Not Requested",
weather="Clear Skies",
vehicles_involved=0,
)
z.save()
print("New Data Object ID: {}".format(z.id))
Shell Output:
ID already exists in DB: 1
New Data Object ID: 1
Edit #1:
I am using a mySQL database and have not overridden the save() function. The mySQL console shows only one entry in the table(the model that was most recently saved).
Edit #2
I commented out the RoadWayData model and migrated the changes to wipe the table. Afterwards, I un-commented the model and migrated the changes to add it back to the database. The issue still persists.
Edit #3
I was able to manually insert a new entry into the table using the mySQL console. The ID incremented correctly. Perhaps it is a Django bug?
Edit #4
I've pinpointed the source of the problem. The problem stems from the contenttypes library. More specifically, the GenericForeignKey. For some reason when an the content object is assigned, the model inherits the content object's id.
Code with problem isolated:
x = IncidentIndex.objects.get(id=1)
y = CHPIncident.objects.get(id=x.incident_object.id)
r = RoadWayData(
...
incident_object = None, # Do not assign the generic foreign key
...
)
r.save()
print(r) # Shows <RoadWayData object> with CORRECT id
r.incident_object = y # Assign the general object
print(r) # Shows <RoadWayData object> with the id of y. INCORRECT
The easiest fix would be to create a variable to keep track of the Model's id BEFORE assigning the content_object (incident_object in my case).
FIX:
... initialization from code above ...
r.save()
r_id = r.id # SAVE THE CORRECT ID BEFORE ASSIGNING GENERIC FOREIGN KEY
r.incident_object = y # ASSIGN THE GENERIC FOREIGN OBJECT
r.id = r_id # OVERWRITE THE ID WITH THE CORRECT OLD ID
r.save()
The incident_object field in the RoadWayData model, has the reference id (the second parameter) set to its own id. So, when model assigns incident_object , it overwrites the id of the model.
To fix it, create a new PostiveIntegerField (like incident_id) and replace
incident_object = GenericForeignKey('incident_type', 'id')
with
incident_id = models.PostiveIntegerField(null=True)
incident_object = GenericForeignKey('incident_type', 'incident_id')
I am having a model structure like:
class user(models.Model):
name = models.CharField(max_length=100)
tasks = models.IntegerField(default=0)
class project(models.Model):
worker = models.ForeignKey(user, on_delete=models.CASCADE)
project_name = models.CharField(max_length=100)
class task(models.Model):
project = models.ForeignKey(project, on_delete=models.CASCADE)
task_name = models.CharField(max_length=150)
expected_date = models.DateField(auto_now=False,auto_now_add=False,)
actual_date = models.DateField(auto_now=False,auto_now_add=False,blank=True,null=True,)
I want to traverse through the task list and if actual date field is not null i.e. task completed then to update the tasks field in user class by 1. I have written the following code:
a = task.objects.filter(actual_date__isnull=False)
for x in a:
x.project.worker.tasks+=1
However this is not giving the desired result. What should I do?
You are not saving your object after modifying it - simply modifying the value doesn't write it to the database. Try this instead:
a = task.objects.filter(actual_date__isnull=False)
for x in a:
worker = x.project.worker
worker.tasks += 1
worker.save()
On a separate note you should consider following PEP8 conventions and using CamelCase for your class names. As it is currently you can very easily mix up classes with objects.
I know that OneToOneField makes one to one correspondence, and ForeignKey makes one-to-many correspondence.
Is there a more efficient way than ForeignKey to store one-to-two correspondence?
Thank you.
Here is an example with two OneToOneField keys:
from django.db import models
class Meaning(models.Model):
definition = models.TextField(null=True)
class Word(models.Model):
masculine_meaning = models.OneToOneField(Meaning, null=True, related_name='mword')
feminine_meaning = models.OneToOneField(Meaning, null=True, related_name='fword')
pronunciation = models.CharField(max_length=64, null=True, blank=True)
Example based on Spanish (my Italian is peggiore)
meaning1 = Meaning()
meaning1.meaning = 'Hand'
meaning1.save()
meaning2 = Meaning()
meaning2.meaning = 'Lever'
meaning2.save()
word1 = Word()
word1.pronunciation = 'braccio'
word1.masculine_meaning = meaning1
word1.feminine_meaning = meaning2
word1.save()
Query for hand
qs = Meaning.objects.filter(definition='Hand').select_related('mword', 'fword')
for x in qs:
if x.mword: print x.mword.pronounciation
if x.fword: print x.fword.pronounciation
after adding depth = 1 doesn't work properly
=> models.py file
class State(models.Model):
state_name = models.CharField(max_length = 30, unique=True)
def __unicode__(self):
return str(self.state_name)
class City(models.Model):
state = models.ForeignKey(State, related_name='state_city')
city_name = models.CharField(max_length = 30)
def __unicode__(self):
return str(self.city_name)
class Meta:
ordering = ('city_name',)
unique_together = ('state', 'city_name',)
class Snippet(models.Model):
created = models.DateTimeField(auto_now_add=True)
title = models.CharField(max_length=100, blank=True, default='')
code = models.TextField()
linenos = models.BooleanField(default=False)
owner = models.ForeignKey('auth.User', related_name='snippets')
state = models.ForeignKey(State,blank=True,null=True)
city = models.ManyToManyField(City)
=> serializers.py file
class StateSerializer(serializers.ModelSerializer):
class Meta:
model = State
class CitySerializer(serializers.ModelSerializer):
state_name = serializers.CharField(source='state.state_name', read_only=True)
class Meta:
model = City
class SnippetSerializer(serializers.ModelSerializer):
owner = serializers.ReadOnlyField(source='owner.username', read_only=True)
class Meta:
model = Snippet
fields = ('id', 'title', 'code', 'linenos', 'owner', 'state', 'city')
depth = 1
I have added ForeignKey and ManyToManyField fields in state and city respectively. It doesn't save values in SnippetSerializer while added depth = 1 in Meta Class (it saves Null value in state and city fields). When I add depth = 1 JSON showing related fields as it should be but it doesn't work properly while add new Snippet. Without depth = 1 it works fine.
I have complex database where tables has many ManyToMany and ForeignKey related fields. Please give me suggestion so I can get related data in JSON.
I have djangorestframework-3.1.2 version. I have used latest version too but same problem. please give me solution and thanks in advance.
I faced the same problem and managed to solve it. Since the problem is with the depth, I just change the depth value in the init method.
class SnippetSerializer(serializers.ModelSerializer):
owner = serializers.ReadOnlyField(source='owner.username', read_only=True)
class Meta:
model = Snippet
fields = ('id', 'title', 'code', 'linenos', 'owner', 'state', 'city')
depth = 1
def __init__(self, *args, **kwargs):
super(SnippetSerializer, self).__init__(*args, **kwargs)
request = self.context.get('request')
if request and request.method=='POST':
self.Meta.depth = 0
else:
self.Meta.depth = 1
In the code above, I changed the depth dynamically according to what type of request that I made.
But, this is the workaround that I found myself, I'm not sure if this is the best practice but it solve the problem with just little modification.
depth is only for representation (http://www.django-rest-framework.org/api-guide/serializers/#specifying-nested-serialization). If you want to create/update the related fields too you have to overwrite the create/update methods in the serializer (http://www.django-rest-framework.org/api-guide/relations/#writable-nested-serializers).
Greetings.