Database Condenser

A config-driven database subsetting tool for PostgreSQL and MySQL, forked from Tonic's Condenser.

Major changes from the original Condenser implementation:

• Supports subsetting databases larger than 10 GB, which the original implementation struggled to process
• Designed to run against read-only instances and replicas, reducing load on the primary database
• Supports incremental top-ups, allowing existing subsets to grow without rebuilding the destination database
• Handles dense and complex FK relationship graphs
• Uses concurrent worker pools throughout the subsetting process to improve performance
• Postgres COPY protocol for fast row transfer
• Automatic sequence reset after subsetting
• Built with psycopg3 and managed with Astral's uv

Subsetting data is the process of taking a representative sample of your data in a manner that preserves the integrity of your database, e.g., give me 5% of my users. If you do this naively, e.g., just grab 5% of all the tables in your database, most likely, your database will break foreign key constraints. At best, you’ll end up with a statistically non-representative data sample.

One common use-case is to scale down a production database to a more reasonable size so that it can be used in staging, test, and development environments. This can be done to save costs and, when used in tandem with PII removal, can be quite powerful as a productivity enhancer. Another example is copying specific rows from one database and placing them into another while maintaining referential integrity.

You can find more about the original Condenser details here and here.

Requirements

• Python 3.10+
• Postgres 14+

Installation

Six steps to set up from source:

1. Install astral-uv

2. Install Postgres and/or MySQL database tools. For Postgres we need pg_dump and psql tools; they need to be on your $PATH or point to them with $POSTGRES_PATH. For MySQL we need mysqldump and mysql, they can be on your $PATH or point to them with $MYSQL_PATH.

3. Clone this project locally.

4. Install the project with uv sync --frozen

5. Setup your configuration and save it in config.json. The provided config.json.example has the skeleton of what you need to provide: source and destination database connection details, as well as subsetting goals in initial_targets. Here's an example that will collect 10% of a table named public.target_table.

   "initial_targets": [
       {
           "table": "public.target_table",
           "percent": 10
       }
   ]
   

   There may be more required configuration depending on your database, but simple databases should be easy. Run subset --help-config for the full configuration reference (also in src/db_condenser/CONFIG.md), and subset --example-config for all of the options in a single config file.

6. Run! $ uv run subset

Or, from PyPI:

pip install db-condenser

pipx install db-condenser

Running

Almost all the configuration is in the config.json file, so running it is as simple as

uv run subset

If you installed from PyPI:

subset --config config.local.json

These command-line arguments are supported:

-v: Verbose output. Useful for performance debugging. Lists almost every query made, and its speed.

--no-constraints: For Postgres this will not add constraints found in the source database to the destination database. This option has no effect for MySQL.

-y: Skip confirmation on subsetting to destination host if it isn't localhost or 127.0.0.1

--config <file>: Use a custom JSON config file instead of the default config.json.

--help-config: Print the full configuration reference and exit.

--example-config: Print an example config.json with all options and exit. Useful as a starting point: subset --example-config > config.json.