datannurpy

datannurpy is the Python builder for datannur. It scans files and databases, extracts metadata and statistics, then generates a ready-to-use catalog bundled with the datannur app.

Key features:

• Broad format support - CSV, Excel, Parquet, Delta Lake, Iceberg, SAS, SPSS, Stata
• Database introspection - PostgreSQL, MySQL, Oracle, SQL Server, SQLite, DuckDB
• Remote and cloud storage - SFTP, S3, Azure Blob, GCS via fsspec
• Metadata extraction - Schemas, statistics, frequencies, enumerations, auto-tagging
• Incremental scans - Only rescan what changed between runs
• YAML or Python API - Declarative configuration or programmatic control

Quick start

pip install datannurpy

# catalog.yml
app_path: ./my-catalog
open_browser: true

add:
  - folder: ./data
    include: ["*.csv", "*.xlsx", "*.parquet"]

  - database: sqlite:///mydb.sqlite

python -m datannurpy catalog.yml

This command scans the configured sources, generates the catalog files, and opens the datannur app.

Documentation

📖 Full documentation: docs.datannur.com/builder

🗂️ datannur app: github.com/datannur/datannur

🌐 Website: datannur.com

🚀 Demo: dev.datannur.com

Contributing

For development documentation and contributing guidelines, see CONTRIBUTING.md.

License

MIT - see LICENSE. All dependencies are MIT/Apache 2.0/BSD compatible.