Creating an IPython Notebook programatically

ProgrammaticNotebook.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "# Creating an IPython Notebook programatically\n",
        "\nThe `nbformat` package gives us the necessary tools to create a new Jupyter Notebook without having to know the specifics of the file format, JSON schema, etc."
      ],
      "metadata": {}
    },
    {
      "cell_type": "code",
      "source": [
        "import nbformat as nbf"
      ],
      "outputs": [],
      "execution_count": 1,
      "metadata": {
        "collapsed": false
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Now we create a new notebook object, that we can then populate with cells, metadata, etc:"
      ],
      "metadata": {}
    },
    {
      "cell_type": "code",
      "source": [
        "nb = nbf.v4.new_notebook()"
      ],
      "outputs": [],
      "execution_count": 2,
      "metadata": {
        "collapsed": false
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Our simple text notebook will only have a text cell and a code cell:"
      ],
      "metadata": {}
    },
    {
      "cell_type": "code",
      "source": [
        "text = \"\"\"\\\n",
        "# My first automatic Jupyter Notebook\n",
        "This is an auto-generated notebook.\"\"\"\n",
        "\n",
        "code = \"\"\"\\\n",
        "%pylab inline\n",
        "hist(normal(size=2000), bins=50);\"\"\"\n",
        "\n",
        "nb['cells'] = [nbf.v4.new_markdown_cell(text),\n",
        "               nbf.v4.new_code_cell(code) ]"
      ],
      "outputs": [],
      "execution_count": 10,
      "metadata": {
        "collapsed": false
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Next, we write it to a file on disk that we can then open as a new notebook.\n",
        "\n",
        "Note: This should be as easy as: `nbf.write(nb, fname)`, but the current api is a little more verbose and needs a real file-like\n",
        "object."
      ],
      "metadata": {}
    },
    {
      "cell_type": "code",
      "source": [
        "fname = 'test.ipynb'\n",
        "\n",
        "with open(fname, 'w') as f:\n",
        "    nbf.write(nb, f)"
      ],
      "outputs": [],
      "execution_count": 11,
      "metadata": {
        "collapsed": false
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "This notebook can be run at the command line with:\n",
        "\n",
        "    jupyter nbconvert --execute --inplace test.ipynb\n",
        "\nOr you can open it [as a live notebook](test.ipynb)."
      ],
      "metadata": {}
    }
  ],
  "metadata": {
    "anaconda-cloud": {},
    "kernelspec": {
      "name": "python3",
      "language": "python",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python",
      "version": "3.6.0",
      "mimetype": "text/x-python",
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "pygments_lexer": "ipython3",
      "nbconvert_exporter": "python",
      "file_extension": ".py"
    },
    "kernel_info": {
      "name": "python3"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 1
}