Notebook Splitting
==================

``split`` splits Jupyter notebooks
by writing each cell type (markdown, code, and raw)
to its own notebook.

Unless in quiet mode,
``split`` shows cell counts: markdown, code, raw, and total.


Result File Names
-----------------

The results of splitting notebook ``nb.ipynb`` are named:

* ``nb-markdown.ipynb``
* ``nb-code.ipynb``
* ``nb-raw.ipynb``

When a particular type of cell does not occur,
that result notebook is not written.

The result name additions can be adjusted in the configuration file
by setting ``split_markdown_result_name``, ``split_code_result_name``, and ``split_raw_result_name``;
see :doc:`configuration`.


Options
-------

The following options are supported by ``split``:

.. code-block:: none

  -t TYPES, --split-cell-types TYPES
                        comma-separated list of cell types to split (default:
                        'markdown,code,raw')


JSON Output
-----------

See :ref:`output-json` for general information about JSON output.

``split`` produces the following members in the JSON output:

  +------------------+----------------------------------------------------+
  | Name             | Value                                              |
  +==================+====================================================+
  | ``"CT"``         | count of cells with type CT                        |
  +------------------+----------------------------------------------------+

Cell types distinguished: ``markdown``, ``code``, ``raw``


Examples
--------

Split notebook ``test.ipynb`` in verbose mode without writing the results:

.. code-block:: bash

  $ nbtb split test.ipynb -v -n
  Dry run (no files written)
  Options for nbsplit:
    (None)

  ::::::::::::::
  test.ipynb
  ::::::::::::::
  Split cell statistics:
            12 code
             7 markdown
             0 raw
            19 total
  No files written

  Notebooks processed: 1