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, ODS, Parquet, Delta Lake, Iceberg, SAS, SPSS, Stata
- Geospatial - GeoJSON, Shapefile, GeoPackage, GeoParquet, GeoTIFF, GML, KML, GPX, ESRI File Geodatabase
- Database introspection - PostgreSQL, MySQL, Oracle, SQL Server, SQLite, DuckDB
- Remote and cloud storage - public HTTP(S) URLs, 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
pip install datannurpy# catalog.yml
app_path: ./my-catalog
open_browser: true
add:
- folder: ./data
include: ["*.csv", "*.xlsx", "*.parquet"]
- database: sqlite:///mydb.sqlitepython -m datannurpy catalog.ymlThis command scans the configured sources, generates the catalog files, and opens the datannur app.
No local install needed: the datannur template lets you publish your own catalog on GitHub Pages in a few minutes — point at open data URLs or drop files, and every commit rebuilds the catalog automatically. See the example catalog built from the template.
📖 Full documentation: docs.datannur.com/builder
🗂️ datannur app: github.com/datannur/datannur
🌐 Website: datannur.com
🚀 Demo: dev.datannur.com
For development documentation and contributing guidelines, see CONTRIBUTING.md.
MIT - see LICENSE. All dependencies are MIT/Apache 2.0/BSD compatible.