osf-downloader 0.1.0

Download files from osf.io

osf-downloader

Download files (or a ZIP of an entire project) from the Open Science Framework (OSF) from the command line. Useful for pulling OSF-hosted datasets into local workflows and CI pipelines.

Features

• Download an OSF project/component by ID (as a ZIP)
• Download a single file by its path inside OSF storage
• Simple CLI suitable for scripting

Install

Option A: From source (recommended for development)

git clone https://github.com/aselimc/osf-downloader.git
cd osf-downloader
# editable install
pip install -e .

Option B: Python package (if published)

pip install osf-downloader

Requirements:

• Python 3.9+

Usage

The installed CLI entrypoint is osf-download.

Replace <OSF_ID> with your OSF project or component id (the short code in the OSF URL).

osf-download download <OSF_ID> ./data

Download a single file by path inside OSF storage:

osf-download download <OSF_ID> ./data/myfile.csv path/inside/osf/myfile.csv

Notes:

• Command shape: osf-download download PROJECT_ID SAVE_PATH [FILE_PATH].
• file_path is the path inside osfstorage (case-sensitive; use / separators).
• save_path is treated as an output file path if it already has a suffix/extension.
• If save_path has no suffix/extension, the tool appends one automatically:
  • project download → .zip
  • file download → uses the extension from file_path (e.g. .csv)

Examples

Download an OSF project ZIP to ./datasets/osf.zip:

osf-download download abcd1 ./datasets/osf.zip

Download an OSF project ZIP where the .zip extension is chosen automatically:

osf-download download abcd1 ./datasets/osf

Download a single file into ./datasets/:

osf-download download abcd1 ./datasets results/data.csv

Tip: to avoid ambiguity, prefer providing a full output filename (e.g. ./datasets/results.csv).

Troubleshooting

• 404 / not found: verify the OSF id and that the project/component is public.
• 401 / unauthorized: this tool currently does not handle OSF authentication; use public resources.
• Path not found: if downloading a single file, ensure file_path matches the OSF storage path (case-sensitive).

Caveats

• Project downloads create a ZIP by fetching all files; for very large projects this may take time and memory.
• Only the osfstorage provider is supported.

Contributing

Issues and pull requests are welcome. Include:

• A clear description of the OSF resource and expected behavior
• Minimal reproduction steps
• Logs/trace output if available

License

MIT. See LICENSE.