django-postgresql-dag 2026.2.2

Directed Acyclic Graph implementation for Django & Postgresql

Django & Postgresql-based Directed Acyclic Graphs

The main distinguishing factor for this project is that it can retrieve entire sections of a graph with far fewer queries than most other packages. The trade off is portability: it uses Postgres' Common Table Expressions (CTE) to achieve this and is therefore not compatible with other databases.

The primary purpose of this package is to build and manipulate DAGs within a Django project. If you are looking for graph analysis or visualization, this may not be the right package.

All core traversal methods (ancestors(), descendants(), path(), connected_graph(), tree methods, roots(), leaves(), edge queries, etc.) use CTE-based queries.

The package also supports CTE filters (disallow_nodes, allow_nodes, disallow_edges, allow_edges, limiting_edges_set_fk) to limit the area of the graph searched, and optional NetworkX/RustworkX/JSON export via the transforms extra. Manager-level methods connected_components() and graph_stats() provide whole-graph analytics. All traversal and predicate methods accept a convenient edge_type parameter as shorthand for limiting_edges_set_fk.

Additional graph algorithms include: topological sort, depth annotation, all-paths enumeration, lowest common ancestor (LCA), weighted shortest path, critical path (longest path), transitive reduction, and graph hashing (via NetworkX Weisfeiler-Lehman).

Demo

Quickstart example | Tutorial | Full documentation

Install

pip install django-postgresql-dag

With optional dependencies for using transformations:

pip install django-postgresql-dag[transforms]

Configuration

You can optionally configure the default maximum traversal depth for all graph queries by adding this to your Django settings:

# settings.py
DJANGO_POSTGRESQL_DAG_MAX_DEPTH = 50  # default is 20

This sets the project-wide default for all graph traversal methods (ancestors(), descendants(), path(), etc.). You can still override it per-call by passing max_depth=N to any method.

ToDo

See the checklists in issues to understand the future goals of this project.

Credits:

1. This excellent blog post
2. django-dag
3. django-dag-postgresql
4. django-treebeard-dag