equialgo / nbserve

Runs, exports, and then serves iPython notebooks as read-only web pages as a WSGI app or a standalone server.

nbserve

nbserve is a simple script that lets you serve iPython notebooks as read-only HTML files over the web.

One of the key differences between nbserve and other options (like nbviewer) is that it lets you run the files as they're being served, allowing you to create dynamic, always-up-to-date 'reports'.

As nbserve matures, I hope to add features like:

• Stripping of input cells.
• Caching, with some simple defaults and features to minimize the amount of configuration required.
• Convenient/appropriate output templates.
• Password protection
• An example of how to deploy to Heroku or similar.

Usage

1. Install:

   pip install nbserve
   # Or, check out the git repo and setup.py develop

2. Run it:

   $ nbserve -h
   
     usage: nbserve [-h] [-v] [-t {strip-input,collapse-input,full}] [-p PORT] [-r]
                    [-d]
                    [working_directory]
   
     nbserve is a simple script that lets you serve iPython notebooks as read-only
     HTML files over the web. It has some default options that make the notebooks
     appropriate for generating reports, such as hiding input cells.
   
     positional arguments:
       working_directory
   
     optional arguments:
       -h, --help            show this help message and exit
       -v, --version         show program's version number and exit
       -t {strip-input,collapse-input,full}
                             Whether to strip, collapse, or show input code cells.
       -p PORT
       -r                    Whether to run scripts each time they're loaded.
       -d                    Debug mode
   
   $ nbserve
   
       Ctrl-C to stop server.
        * Running on http://127.0.0.1:5000/
        * Restarting with reloader
       Ctrl-C to stop server.
   

Target Use Cases

• A user has run analyses in iPython Notebook and wants to easily make the results available on the web. The results are already contained in the notebook

• A user has analyses that need to be run regularly or on-demand, and the served notebooks are a way to easily generate that dynamic content.

• A user develops notebooks with extensive code, and needs a way to make them accessible to people who will be scared away or distracted by the code.

Links / Inspiration

• https://github.com/joerns/ipython-notebook-tools
• https://github.com/paulgb/runipy
• http://www.plankandwhittle.com/packaging-a-flask-web-app/

Packaging / Deployment Strategy.

• It's a flask app.

• It looks for notebooks (and/or a config file) in the current directory.

• You install it with pip.

• You can run it three ways:

  • gunicorn nbserve:flask_app
  • nbserve
  • add it as a uswgi container to nginx, apache, etc, just like any other flask app.

Development

Releasing

# python setup.py register -r pypi
python setup.py sdist upload -r pypi

Build Status

• Master:
• Develop: