Installation

Datacrafter requires Python 3.10 or newer.

The PyPI distribution is datacrafter-etl (the datacrafter name on PyPI belongs to an unrelated project). The Python module and the CLI command stay datacrafter.

Using pip (recommended)

python3 -m venv .venv
source .venv/bin/activate   # Windows: .venv\Scripts\activate
pip install datacrafter-etl
datacrafter version

From source

git clone https://github.com/apicrafter/datacrafter.git
cd datacrafter
pip install -e .

For tests and linters, install the development extras:

pip install -r requirements-dev.txt
pip install -e .

Optional destinations

File JSONL, BSON, and CSV destinations ship with the core package. Some destinations need extra Python packages:

Destination / feature	Package
file-parquet	pip install "datacrafter[parquet]" (pyarrow)
.zst compression (sources and destinations)	pip install "datacrafter[compression]" (zstandard)
mongodb	pymongo
arangodb	python-arango
couchdb	pycouchdb
meilisearch	meilisearch

APIBackuper extractors also need the apibackuper CLI (>=1.0.4) on PATH.

datacrafter check reports missing optional packages for the destination you configured.

Requirements

• Python 3.10 or greater (CI tests 3.10–3.13)
• A writable project directory for current/, output/, and state.json

Next steps

• Quick start
• When to use Datacrafter
• Configuration schema