Convert a Goodreads csv export to headings and itemized lists in Markdown

goodreads2md.py

#!/usr/bin/env python

import csv
import re
import sys
from collections import namedtuple

def oxford_join(seq):
    if len(seq) == 1:
        return seq
    elif len(seq) == 2:
        return '%s and %s' % (seq[0], seq[1])
    else:
        return '%s, and %s' % (', '.join(seq[:-1]), seq[-1])

def main(path):
    booklist = {}
    with open(path) as csvfile:
        reader = csv.DictReader(csvfile)
        for row in reader:
            author = row['Author l-f']
            add_authors = row['Additional Authors']
            if add_authors:
                add_authors = re.split('r,\s*', add_authors)
                #author = ', '.join([author] + add_authors)
                author = oxford_join([author] + add_authors)
            title = row['Title']
            shelves = set(re.split(r',\s*', row['Bookshelves']))
            excshelf = row['Exclusive Shelf']
            if excshelf == 'read':
                shelves = set(['~FINISHED'])
            else:
                shelves.remove(excshelf)
            recommended = row['Recommended By']
            book = author, title, recommended
            if not shelves:
                shelves = set(['uncategorized'])
            for shelf in shelves:
                if shelf not in booklist:
                    booklist[shelf] = []
                booklist[shelf].append(book)

    for shelf in sorted(booklist.keys()):
        print
        if isinstance(shelf, basestring):
            cap_shelf = "%s%s" % (shelf[0].upper(), shelf[1:])
        else:
            cap_shelf = shelf
        print "# %s" % cap_shelf
        for author, title, recommended in sorted(booklist[shelf]):
            author = author.rstrip('.')
            if title[-1] not in ('.', '?', '!'):
                title = '%s.' % title
            if recommended:
                recommended = ' (Recommended by %s.)' % recommended
            print "- %s. %s%s" % (author, title, recommended)

if __name__ == '__main__':
    main(sys.argv[1])

Incidentally, I've written a GoodReads CSV-Markdown compiler using Pandoc; you can see the results here.