[Django]-Django dump data for a single model?

As noted, you can’t do this through a manage.py command in Django 1.0. However you could use a script to export the JSON file, and load it using loaddata:

from django.core import serializers
from myproject.myapp import models
data = serializers.serialize("json", models.MyModel.objects.all())
out = open("mymodel.json", "w")
out.write(data)
out.close()

A specific model to a specific file:

python manage.py dumpdata app_label.ModelName > specific_file.json

and to load it to another app: first move or copy the file to the app where you want process it, then:

python manage.py loaddata specific_file.json

Take all data into json format from django model.

Syntax:

python manage.py dumpdata app_name.model_name

For example dumping data from group_permission model which reside in default auth app in django.

python manage.py dumpdata auth.group_permission

For output take a look on console.

I think you had the solution in your question. You can dump an individual model like this:

./manage.py dumpdata myapp.my_model

For success I had to say it twice, and specify the model two times, like:

./manage.py dumpdata myapp2.my_model myapp2.my_model

If I only said

./manage.py dumpdata myapp2 myapp2.my_model

I got flooded with all the models in myapp2, despite the fact that I specified my_model.

For newbies in Django and Python like me, this might be useful:

if you want to only dump a single row (of, obviously, a single table) and you have for example the app "merchant" and the model is also named "Merchant", and you usually import it using a fully qualified name like this: merchant.models.Merchant; do not even try to use this name. The syntax is as follows:

# only dumps the merchant with id = 123, pretty printed
python manage.py dumpdata merchant.Merchant --pks 123 --indent 2

To write it on specific file:

python manage.py dumpdata app_label.ModelName app_label.ModelName2 > fixtures/specic.json

If you want to dump only the specified objects of a model, you can use the –pks optional argument for the dumpdata command.

–pks PRIMARY_KEYS Outputs only the objects specified by a comma separated list of primary keys. This is only available when dumping
one model. By default, all the records of the model are output.

As a workaround you could make another app and copy the model but point it to the existing table with the db_table meta option. Then you could just dump the models you copied into the new app. You existing app wouldn’t be affected.

I’ve created a management command the generate a fixture on a per model basis.
Fixtures can be generated by running:

./manage generate_fixtures app.model.MyModel --file=dump/MyModel.json

code at: https://gist.github.com/2394883

I think the accepted answer is old and needs a refresher in 2023 with the detailed approach.

I had the similar problem where I had to create fixtures for specific models from different apps having specific id’s. Moreover I had to create just one fixture for all models. so that we don’t have to load individual fixtures rather just one fixture for all models.

I did all of this by creating a custom command that calls the dumpdata internally for each specific model with the specific ids like the following:

products/models.py

class Product(models.Model):
    name = models.CharField(max_length=200)
    price = models.DecimalField(max_digits = 5, decimal_places = 2)

reviews/models.py

class Review(models.Model):
    title = models.CharField(max_length=200)        
    text = models.TextField()
    timestamp = models.DateTimeField(auto_now_add=True)
    rating = models.PositiveIntegerField(default=0)
    product = models.ForeignKey(Product, on_delete=models.CASCADE)

    def natural_key(self):
       return self.title, self.timestamp


class Response(models.Model):
    text = models.TextField()
    timestamp = models.DateTimeField(auto_now_add=True)

And following is the command that creates one fixtures for specific models from different apps with specific id’s:

generate_fixture.py

from django.core.management import BaseCommand, call_command
from io import StringIO
import json

class Command(BaseCommand):
    def add_arguments(self, parser):
        parser.add_argument('id', type=int, help='Primary Key Of Product')

    def handle(self, *args, **options):
        id = options.get('id')

        if product := Product.objects.filter(id=id).first():
            product_fixture = self.get_fixture('products.Product', str(id))

            review_ids = list(
                Review.objects.filter(product__id=id).values_list('id',flat=True)
            )
            review_ids = ','.join([str(review_id) for review_id in review_ids])
            review_fixture = self.get_fixture('reviews.Review', review_ids)

            output = [json.loads(product_fixture)]
            output.extend(json.loads(review_fixture))
            with open('model_fixtures.json', "w") as file:
                json.dump(output, file)
        else:
            print(f"Product with id {id} does not exists!")

    def get_fixture(self, label, pks):
        args = ['dumpdata', '--natural-foreign', '--natural-primary', label]
        with StringIO() as buffer:
            call_command(*args, pks=pks, stdout=buffer)
            data = buffer.getvalue()
        return data

Now you can run the command with the id like this:

python manage.py generate_fixture 50

And fixture of the models will be generated in just one fixture file which can be loaded like this:

python manage.py loaddata model_fixtures.json

Note: The code is not tested.

For DUMPING data out of the specific model for specific APP.

If we take EXAMPLE where we have in a MAIN PROJECT (We go call it project) and in this MAIN PROJECT we have two (02) others applications (We go call each app like this APP1 and APP2) we can do like that

1. python manage.py dumpdata APP1.name_of_model >APP1/fixtures/name_of_model .json
2. python manage.py dumpdata APP2.name_of_model >APP2/fixtures/name_of_model .json