#!/usr/bin/env python
import argparse
import markdown
import os.path
from bs4 import BeautifulSoup


def read_code_from_file(mdfile):
    """returns a string of the code sections in the Markdown file"""
    origfilepos = mdfile.tell()
    mdfile.seek(0)
    html = markdown.markdown(mdfile.read())
    soup = BeautifulSoup(html)
    codestring = ''
    for codeblock in soup.find_all('code'):
        if len(codestring) > 0:
            codestring += '\n'
        codestring += codeblock.text
    mdfile.seek(origfilepos)
    return codestring


def write_file_from_code(originalfilename, extension, code):
    """write a new file of source code"""
    basename = os.path.basename(originalfilename)
    dirname = os.path.dirname(originalfilename)
    newname = basename.split('.')[0] + '.' + extension
    newfile = open(os.path.join(dirname, newname), 'w')
    newfile.write(code)
    newfile.close()


if __name__ == '__main__':
    parser = argparse.ArgumentParser(
        description='Strip the code segments from a Markdown file.')
    parser.add_argument(
        'files', metavar='F', type=file, nargs='+', 
        help='a markdown file to strip')
    parser.add_argument(
        '--ext', dest='extension', required=True,
        help='the extension to use for stripped files')
                       
    args = parser.parse_args()
    for mdfile in args.files:
        code = read_code_from_file(mdfile)
        write_file_from_code(mdfile.name, args.extension, code)