Why Py-Auto-Migrate?

Migrating data between different database systems is often a tedious and error-prone task. Py-Auto-Migrate is here to change that. It's a powerful, flexible, and easy-to-use Python tool that automates the entire process.

• Universal Connector: Supports a vast range of databases, from SQL to NoSQL.
• Zero Configuration: Point to your source and target, and let the tool handle schema detection, data type mapping, and destination creation.
• Blazing Fast: Optimized for performance, even with large datasets.
• Safe & Reliable: Built-in checks ensure data integrity throughout the migration.

Installation

Get started in seconds with pip.

pip install py-auto-migrate

Prefer a Standalone Shell?

Don't have Python? No problem! Download the dedicated PAM-Shell for your OS from the Releases page. It's a ready-to-run executable with the same powerful features.

Quick Start

Using Py-Auto-Migrate is as simple as running one command.

Basic Command Structure

py-auto-migrate migrate --source <SOURCE_URI> --target <TARGET_URI>

Examples

1. Migrate an entire database from MongoDB to MySQL:

py-auto-migrate migrate \
  --source "mongodb://user:pass@localhost:27017/source_db" \
  --target "mysql://user:pass@localhost:3306/target_db" \
  --pgbar \
  --skip-existing

--pgbar prints row-level progress while the command runs. --skip-existing checks destination rows and submits only source rows that are not already there; it is disabled by default. See the note below for how duplicate source rows are handled.

2. Migrate a single PostgreSQL table to a new MongoDB collection:

py-auto-migrate migrate \
  --source "postgresql://user:pass@localhost:5432/mydb" \
  --target "mongodb://user:pass@localhost:27017/mydb" \
  --table users

The Magic: If the target database or table doesn't exist, Py-Auto-Migrate automatically creates it for you! It intelligently maps source data types to the appropriate target schema.

By default, migration appends every source row, including duplicate rows. Use --skip-existing to skip only copies already present in the target. This is a row-content comparison, so it may read the existing target table into memory. Duplicate rows in the source are preserved according to their count: if the source has two identical rows and the target has one, one row is added. Without --skip-existing, running the same migration again appends the rows again.

3. Query using AI (Target must be a relational database):

export OPENAI_API_KEY="your-api-key"

Before using AI, make sure the OPENAI_API_KEY has been exported.

py-auto-migrate migrate \
  --source "oracle://user:password@host:port/service_name" \
  --target "mssql://user:password@host:port/database" \ 
  --ai-ask "Please select and insert only names that start with 'S', then sort them." /
  --ai-model "gpt-3.5-turbo"

4. Migrate table forigen key dependency (Source must be a relational database):

py-auto-migrate migrate \
  --source "postgresql://user:pass@localhost:5432/mydb" \
  --target "mongodb://user:pass@localhost:27017/mydb" \
  --table orders \
  --dep

--dep automatically adds all tables referenced by foreign key relationships from the source table to the insertion list.

Dashboard

For an even easier and more visual migration experience, Py-Auto-Migrate includes a dedicated web dashboard.

Usage:

py-auto-migrate dashboard

Default Dashboard URL:

http://localhost:8123

or :

py-auto-migrate dashboard --host "localhost" --port 8000

Features:

• Visual Interface – No need to memorize commands or URIs
• Connection Builder – Easily build source/target URIs with form inputs
• Table Selection – Choose specific tables or migrate all with checkboxes
• Real-time Logs – Watch migration progress and errors live
• AI Query UI – Use natural language queries with a simple text box
• Job Management – Run, monitor, and manage multiple migration jobs
• One-Click Migrate – Start migrations with a single button click

The dashboard skips rows already present in the target by default and writes each table in one call to avoid the overhead of progress-driven write batches. Uncheck the option to append every source row. Progress updates advance after each table write completes.

Supported Databases

We support a wide and growing range of databases.

Category	Databases
Relational
NoSQL
Analysis

Have a feature request? Open an issue and let us know!