7π
If you use the template functionality of django-mptt to generate the JSON data you should be able to do something like the following:
var json =
{% recursetree nodes %}
{
id: "{{ node.id }}",
name: "{{ node.name }}",
data: {},
children: [{{ children }}]
},
{% endrecursetree %}
The children tag is brilliant, basically calling recursetree for each child of the node. However there is a little messiness generated around commas with this solution as the mptt example is around list elements where such things arenβt a problem.
A slightly larger bit of code solves this:
var json =
{
id: "{{ root.id }}",
name: "{{ root.name }}",
data: {},
children: [{% recursetree root.children %}
{
id: "{{ node.id }}",
name: "{{ node.name }}",
data: {},
children: [{{ children }}]
}
{% endrecursetree %}]
}
By making a distinction between the root node (presuming there is only one) which does not reside within an array but is instead assigned to a variable, compared to other nodes which live within the children of another node, the var x = y,
issue is avoided.
There is still a problem with children: [x,y,z,]
having a trailing comma. If that final comma is raising errors then in the view that has called the template you can always do a quick string replace to replace ,]
with ]
Alternatively you could alter the {{ children }}
call of mptt in some way to take a join string, but that would involve much more work.
2π
Create JSON without trailing comma problem:
{% full_tree_for_model YOUR_APP.YOUR_MODEL as nodes %}
var json = {% for genre,structure in nodes|tree_info %}{% if structure.new_level %} [{ {% else %} },{ {% endif %}
id: "{{ genre.id }}",
children: {% if genre.is_leaf_node %}[]{% endif %}
{% for level in structure.closed_levels %}}]{% endfor %}{% endfor %}
2π
If someone is looking for a solution based on templates with the recursetree
tag, it is possible to use get_next_sibling
to control commas. Adapting the example provided by Chris:
{% load mptt_tags %}
[
{% recursetree nodes %}
{
"id": "{{ node.id }}",
"name": "{{ node.name }}",
"data": {},
"children": [{{ children }}]
}{% if node.get_next_sibling %},{% endif %}
{% endrecursetree %}
]
0π
If you want to write your own serializer or do this from a view you can do:
from mptt.templatetags.mptt_tags import cache_tree_children
queryset = Foo.objects.all().select_related('bar')
queryset = cache_tree_children(queryset)
Now the queryset is βcachedβ, meaning you can use model API methods like get_children()
without it hitting the database again.
0π
I found a great solution to rendering children with sub-data in this thread: fastest way to create JSON to reflect a tree structure in Python / Django using mptt (See the accepted answer by craigds). His method also caches results and uses the built in json.dumps to create the structure.
- How do I use request.META.get('HTTP_REFERER') within template?
- Django cache framework. What is the difference between TIMEOUT and CACHE_MIDDLEWARE_SECONDS?
- What became available_attrs on Django 3?
- Django Admin, sort with custom function
0π
In Django view
from django.core import serializers
import json
json_str = serializers.serialize("json",Categories.objects.all())
json_data = json.loads(json_str)
- Django Queryset __in with None value in list
- Django loading data from fixture after backward migration / loaddata is using model schema not database schema
- Django : Static content not found