(Extremely basic) Python script to create jupyter notebooks from markdown

md2ipynb.py

#! /usr/bin/env python3

############################################################################
#                                                                          #
# (Extremely basic) Python script to create jupyter notebook from markdown #
# Bruno Grenet, 02.02.2017                                                 #
# License: Public Domain                                                   #
#                                                                          #
############################################################################

# Input format:
# ------------
# Code in fenced blocks (delimited by "```" or "~~~") is put into code cells
# in the notebook. The rest is put in markdown cells. In particular, indented
# code blocks are put into markdown cells.

import re, sys, os, getopt

begincode = """  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
"""
beginmd = """  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
"""
endjson = """ ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}"""

def process(line):
    s = line[:-1]
    s = s.replace('\\', '\\\\')
    s = s.replace('"', '\\"')
    return '    "' + s + '\\n"'

def md2ipynb(inputfile, outputfile):
    if not os.path.isfile(inputfile):
        print("{} is not a valid input file.".format(inputfile))
        return 1

    if os.path.exists(outputfile):
        if not os.path.isfile(outputfile):
            print("{} already exists and is not a file.".format(outputfile))
            return 1
        
        answer = input("The file {} already exists. Overwrite? [Y/n] ".format(outputfile))
        if answer in ['n', 'N', 'no', 'NO', 'No']:
            return 0

    f = open(inputfile, 'r')
    o = open(outputfile, 'w')
    
    md = False
    code = False
    first = True
    first_of_code = False
    width = 0
    codex = "^[~`]{3,}"
    endblock = '\n   ]\n  }'

    o.write('{\n "cells": [\n')
    
    for line in f:
        m = re.search(codex, line)
        if m is not None:
            w = len(m.group(0))
            if code and w >= width:
                o.write(endblock)
                code = False
            elif code:
                o.write(',\n')
                o.write(process(line))
            else:
                if md: 
                    o.write(endblock)
                    o.write(',\n')
                    md = False
                o.write(begincode)
                code = True
                code_width = w
                first_of_code = True
                
        else:
            if not (code or md):
                if not first:
                    o.write(',\n')
                o.write(beginmd)
                md = True
            else:
                if first_of_code:
                    first_of_code = False
                else:
                    o.write(',\n')
            o.write(process(line))

        first = False

    if md or code:
        o.write(endblock)
        md = code = False
    
    o.write(endjson)
    f.close()
    o.close()
    return 0
 
def main():
    usage = "{} [-o outfile|--output=outfile] inputfile".format(sys.argv[0])
    try:
        opts, args = getopt.getopt(sys.argv[1:], "ho:", ["help", "output="])
    except getopt.GetoptError as err:
        print(err)
        print(usage)
        sys.exit(2)

    outputfile = '' 
    for o, a in opts:
        if o in ('-h', '--help'):
            print(usage)
            sys.exit()
        elif o in ('-o', '--output'):
            outputfile = a
        else:
            print(usage)
            sys.exit(1)

    if len(args) != 1:
        print(usage)
        sys.exit(1)

    inputfile = args[0]

    if outputfile == '':
        outputfile = os.path.splitext(inputfile)[0] + '.ipynb'

    ret = md2ipynb(inputfile, outputfile)
    if ret == 0:
        print("{} converted into {} successfully".format(inputfile, outputfile))
        sys.exit()
    else:
        print("Conversion failed")
        sys.exit(1)


if __name__ == '__main__':
    main()