SCons nbconvert tool

About

This tool can be used to convert jupyter pages with nbconvert.

Install

Installing it, requires you to copy (or, even better: checkout) the contents of the package's nbconvert folder to

• /path_to_your_project/site_scons/site_tools/nbconvert", if you need the nbconvert Tool in one project only, or
• /user/share/scons/site_scons/site_tools/nbconvert, for a system-wide installation.

For more infos about this, please refer to

• the SCons User's Guide, sect. "Where to put your custom Builders and Tools"

Activate the tool

Create a build environment with the nbconvert builder

import os
import sconstool_nbconvert

env = Environment(ENV = {'PATH' : os.environ['PATH']},
                  tools=['default', sconstool_nbconvert.generate],
                  NBCONVERT_ENVIRONMENT_VARS={'flags': ['execute', 'no-input'],
                  'to': 'html',
                  'log-level': 'CRITICAL', 
                  'ExecutePreprocessor.kernel_name': 'python3',
                  'HTMLExporter.exclude_anchor_links': True,
                  'template': 'classic'})

The NBCONVERT_ENVIRONMENT_VARS must contain a list of flags. Allowed flags are: [execute] [allow-errors] [inplace] [clear-output] [no-prompt] [no-input] [show-input]

It can also contain a list of build options. For a list of options visit the nbconvert documentation

Call the target

The nbconvert target can be called with one source and target file.

env.nbconvert('rendered_templ.html', os.path.join('test.ipynb'))

Templates

There is a template folder in this repository. It can be used to convert the notebooks to hugo content pages. It will allow metadata in the jupyter notebook for the FrontMatter creation. Jupyterlab is needed to edit this data. The FrontMatter section can be added before the jupyter metadata starting with kernelspec.

{
    "FrontMatter": {
        "author": "spielhuus",
        "category": "article",
        "date": "2021-01-10",
        "tag": "grundlage",
        "title": "cmos",
        "draft": "true"
    },
    
    "kernelspec": {
    },
    "language_info": {
    }
}