Entity relation diagrams generator

eralchemy2 generates Entity Relation (ER) diagram (like the one below) from databases or from SQLAlchemy models. Works with SQLAlchemy < 1.4 but also with versions greater than 1.4

Example

Example for NewsMeme

Quick Start

Install

To install eralchemy2, just do:

$ pip install eralchemy2

eralchemy2 requires GraphViz to generate the graphs and Python. Both are available for Windows, Mac and Linux.

For Debian based systems, run:

$ apt install graphviz libgraphviz-dev

before installing eralchemy2.

Install using conda

There is also a packaged version in conda-forge, which directly installs the dependencies:

$ conda install -c conda-forge eralchemy2

Usage from Command Line

From a database

$ eralchemy2 -i sqlite:///relative/path/to/db.db -o erd_from_sqlite.pdf

The database is specified as a SQLAlchemy database url.

From a markdown file.

$ curl 'https://raw.githubusercontent.com/maurerle/eralchemy2/main/example/newsmeme.er' > markdown_file.er
$ eralchemy2 -i 'markdown_file.er' -o erd_from_markdown_file.pdf

From a Postgresql DB to a markdown file excluding tables named temp and audit

$ eralchemy2 -i 'postgresql+psycopg2://username:password@hostname:5432/databasename' -o filtered.er --exclude-tables temp audit

From a Postgresql DB to a markdown file excluding columns named created_at and updated_at from all tables

$ eralchemy2 -i 'postgresql+psycopg2://username:password@hostname:5432/databasename' -o filtered.er --exclude-columns created_at updated_at

From a Postgresql DB to a markdown file for the schemas schema1 and schema2

$ eralchemy2 -i 'postgresql+psycopg2://username:password@hostname:5432/databasename' -s "schema1, schema2"

Specify Output Mode

$ eralchemy2 -i 'markdown_file.er' -o erd_from_markdown_file.md -m mermaid_er

Usage from Python

from eralchemy2 import render_er
## Draw from SQLAlchemy base
render_er(Base, 'erd_from_sqlalchemy.png')

## Or draw from SQLAlchemy metadata
metadata = mapper_registry.metadata
render_er(metadata, 'erd_from_sqlalchemy.png')

## You can also generate Mermaid ER and render it easily in markdown
render_er(metadata, "mermaid_erd.md", mode="mermaid_er")

## Draw from database
render_er("sqlite:///relative/path/to/db.db", 'erd_from_sqlite.png')

Architecture

Thanks to it's modular architecture, it can be connected to other ORMs/ODMs/OGMs/O*Ms.

Contribute

Every feedback is welcome on the GitHub issues.

Development

Install the development dependencies using

$ pip install -e .[ci,dev]

Make sure to run the pre-commit to fix formatting

$ pre-commit run --all

All tested PR are welcome.

Running tests

This project uses the pytest test suite. To run the tests, use : $ pytest or $ tox.

Some tests require having a local PostgreSQL database with a schema named test in a database named test all owned by a user named eralchemy with a password of eralchemy. If docker compose is available, one can use docker compose up -d for this purpose. You can deselct the tests which require a PostgreSQL database using:

$ pytest -m "not external_db"

Publishing a release

$ rm -r dist && python -m build && python3 -m twine upload --repository pypi dist/*

Notes

eralchemy2 is a fork of its predecessor ERAlchemy by @Alexis-benoist, which is not maintained anymore and does not work with SQLAlchemy > 1.4. If it is maintained again, I'd like to push the integrated changes upstream.

ERAlchemy was inspired by erd, though it is able to render the ER diagram directly from the database and not just only from the ER markup language.

Released under an Apache License 2.0

Initial Creator: Alexis Benoist Alexis_Benoist