A Jupyter content provider for data.world.

This content provider allows data.world users to store and manage their notebooks and files directly on data.world using Jupyter Notebook or Jupyter Lab.

Once enabled, the content provider will allow you to browse and access your entire data.world library, including datasets and projects that you have created, contribute to or have bookmarked.

You can install it using pip directly from PyPI:

pip install dwcontents

Find or create a file named jupyter_notebook_config.py under your Jupyter folder (~/.jupyter).

Update it to define two configuration parameters: - NotebookApp.contents_manager_class: Must be set to dwcontents.DwContents - DwContents.dw_auth_token: Must be your data.world API token (obtained at https://data.world/settings/advanced)

For example:

import dwcontents
c = get_config()
c.NotebookApp.contents_manager_class = dwcontents.DwContents
c.DwContents.dw_auth_token = 'YOUR TOKEN GOES HERE'

You can also use this to access data.world alongside your local files using dwcontents.HybridContents.

In that case, this is what your jupyter_notebook_config.py might look like:

import dwcontents
from notebook.services.contents.filemanager import FileContentsManager
c = get_config()
c.NotebookApp.contents_manager_class = dwcontents.HybridContents
c.HybridContents.manager_classes = {
    # Associate the root directory with data.world
    '': dwcontents.DwContents,
    # Associate /~local with your working directory, completely disconnected from data.world
    '~local': FileContentsManager
}
c.HybridContents.manager_kwargs = {
    '': {
        'dw_auth_token': 'YOUR TOKEN GOES HERE'
    }
}

Once installation and configuration are complete, run Jupyter Notebook or Labs like you normally would.

For example:

jupyter notebook

• Jupyter supports a wide variety of file operations, whereas support for directories on data.world is limited. For a better experience, try to keep a flat file structure under your datasets and projects.