Welcome to the jupyter_fsspec
repo, a Jupyter extension for the fsspec
Python library.
Browse your fsspec filesystems in Jupyter, copy files and paths and otherwise interact with
your data in the Jupyter interface and inside your notebooks.
Read the documentation at https://jupyter-fsspec.readthedocs.io/en/latest/
- JupyterLab >= 4.0.0
To install the extension, execute:
pip install jupyter_fsspec
To remove the extension, execute:
pip uninstall jupyter_fsspec
jupyter_fsspec
was created to provide seamless integration between Jupyter
environments and the powerful fsspec
ecosystem. While the Jupyter ecosystem
already had file browsing capabilities, there was a need for a dedicated extension
that could leverage the full power of fsspec
's unified interface to various
filesystems.
The creation of jupyter_fsspec
was inspired by several ideas we wanted to explore:
-
Direct fsspec Integration: We wanted to build a bridge directly to the
fsspec
ecosystem, making it simple to access all filesystems supported byfsspec
with minimal layers. -
Unified Interface:
fsspec
offers a consistent API across different storage backends (local, cloud, and remote filesystems). We thought it would be valuable to bring this unified approach to the Jupyter interface. -
Complementary Approach: We designed the UI with certain workflows in mind, focusing on unified browsing, easy file path copying, and seamless notebook integration.
jupyter-fs
is also a great project
with similar goals but different design choices that make it excellent for certain
use cases:
jupyter-fs
leverages PyFilesystem, providing robust filesystem abstractions with its own set of advantages- The separate panel per filesystem in
jupyter-fs
offers a clean separation that some users may prefer for their workflows - Both projects enhance the Jupyter experience by bringing better filesystem access to users
We see jupyter_fsspec
and jupyter-fs
as complementary tools in the ecosystem.
We have also been discussing trying to merge the projects and jupyter-fs
recently
added a fsspec
backend as well.
If you're deciding between the two, consider your specific needs:
- If you're already using
fsspec
in your Python code,jupyter_fsspec
provides a consistent experience - If you prefer separate filesystem panels or are already using PyFilesystem,
jupyter-fs
might be more aligned with your workflow
We welcome contributions and feedback from the community as we continue developing this extension. We're also open to collaborations with other filesystem projects in the Jupyter ecosystem to collectively improve the experience for data scientists and researchers working with diverse data sources.