Django Model Map

Stop guessing your query optimizations.

django-model-map is a simple management command that inspects your Django models and outputs a JSON map of relationships. It explicitly categorizes relations into select_related and prefetch_related candidates, helping you avoid N+1 problems and write optimized QuerySets faster.

Features

• Automatic Classification: Distinguishes between select_related (ForeignKey, OneToOne) and prefetch_related (ManyToMany, Reverse FK).
• Deep Inspection: Supports configurable nesting levels for mapping nested relationships (e.g., prefetch_related('lvl1__lvl2__lvl3')).
• Reverse Relation Discovery: Finds standard _set accessors and custom related_name attributes.
• Recursion Detection: Identifies self-referencing models.
• GenericForeignKey: Detects GFK fields and marks them in the output.
• JSON / YAML Output: Easy to read, parse, or integrate into other tools.

Installation

pip install django-model-map

# With YAML output support (optional)
pip install django-model-map[yaml]

Add to INSTALLED_APPS:

INSTALLED_APPS = [
    ...
    'django_model_map',
    ...
]

Usage

CLI

# Inspect all installed apps
python manage.py modelmap

# Inspect a specific app
python manage.py modelmap [app_name]

# Named alternative to positional
python manage.py modelmap --app [app_name]

# Specify inspection nesting level (default: 1)
python manage.py modelmap [app_name] --depth 2

# Exclude models or entire apps (repeatable)
python manage.py modelmap --exclude auth.Permission --exclude auth.Group

# Save to file
python manage.py modelmap [app_name] --output relations.json

# YAML output (requires pip install django-model-map[yaml])
python manage.py modelmap [app_name] --format yaml

Python API

from django_model_map import get_model_map

# Inspect a specific app, depth 2, exclude some models
model_map = get_model_map(app_label='myapp', depth=2, exclude=['myapp.OldModel'])

for model_name, data in model_map.items():
    print(data['queryset_snippet'])

Example Output

{
    "blog.Post": {
        "queryset_snippet": "Post.objects.select_related('author', 'category').prefetch_related('tags', 'comments')",
        "select_related_fields": [
            "author",
            "category"
        ],
        "prefetch_related_fields": [
            "tags",
            "comments"
        ],
        "details": {
            "select_related": [
                {
                    "field_name": "author",
                    "target_model": "users.User",
                    "is_recursive": false
                },
                 ...
            ],
            "prefetch_related": [
                {
                    "field_name": "tags",
                    "target_model": "blog.Tag",
                    "is_recursive": false
                },
                {
                    "field_name": "content_object",
                    "target_model": "Generic",
                    "is_recursive": false,
                    "type": "generic"
                },
                ...
            ]
        }
    }
}

How it helps

When writing a view, instead of opening models.py and mentally parsing the relationships, just look at the output. With the --depth argument, you can automatically discover deeply nested relationships that need optimization:

• Copy fields from "queryset_snippet" -> paste into project.
• Copy fields from "select_related_fields" -> paste into .select_related(...).
• Copy fields from "prefetch_related_fields" -> paste into .prefetch_related(...).

Contributing

Pull requests are welcome! For major changes, please open an issue first.

License

MIT