Skip to content

Latest commit

 

History

History
108 lines (77 loc) · 2.9 KB

README.md

File metadata and controls

108 lines (77 loc) · 2.9 KB
Raw

Schemaviz

Visualize dependencies between the tables in your database.

Prerequisites

  1. The Node Package Manager (NPM)
  2. Python (3.9 or newer)
  3. Poetry (Python package manager)

Installing

First, clone the repo locally:

$ git clone https://github.com/shashank025/schemaviz.git

Next, browse to the frontend/ folder, and run npm install:

$ cd schemaviz/frontend
$ npm install

Using this tool

In one terminal, start the Python backend:

$ cd schemaviz/backend
$ poetry run uvicorn schemaviz_backend.main:app --reload

In another terminal, start the frontend:

$ cd schemaviz/frontend
$ npm start

This will start a local webserver at http://localhost:8080/.

Visualizing your schema

  1. Navigate to the local webserver in your browser.
  2. Enter the connection string for your database, and
  3. Click "Visualize"

The web application should shortly display a directed graph showing the tables in your database and the relationships between them.

For example, when I point this tool at the following publically available database:

postgres://reader:[email protected]:5432/pfmegrnargs

here's how the resulting visualization looks:

screenshot showing schemaviz interface on local webserver

Architecture

This tool consists of a frontend, and a backend:

  • The backend is a Python-based API service that currently exposes just one API that lets you specify a database to connect to. This service connects to that database, queries its schema, and assembles dependency info to the API caller.

  • The frontend is a Node-based web application that queries the backend and renders the database schema visually.

Visualization Approach

We use d3's force directed layout scheme to produce an SVG image where each relation (table) in your schema is rendered as a node. A directed edge is added from node A to node B if some columns in A refer to some columns in B, as expressed via a foreign key constraint:

    A ---> B

Improvements (for large graphs)

A good visualization is particularly useful when the underlying database schema is pretty large (1000+ tables). But it is not straightforward to render the force directed layout for a graph that large. Here are some ideas to make the visualization more useful:

  1. Ability to search for and highlight a given table in the schema.
  2. Expose the parameters of the force directed layout: height, width, linkDistance, charge, gravity, etc.
  3. Vertical/horizontal scrolling through the rendered SVG.
  4. Fisheye distortion to focus on a particular area of the graph.