parser.py

gistfile1.txt

import re
import codecs
from pprint import pprint

EOR = u"=========="


def records(file_path):
    clip_file = codecs.open(file_path)
    clip_file.seek(3)  # skip magic cookie

    record = list()
    for line in clip_file:
        line = line.decode('utf-8')
        if line.strip() == EOR:
            assert record[2] == '', "Non-blank line expected separating the header from the body of the clipping:%s" % record[2]
            clip = dict()
            match = re.match(r'(.*?)\((.*)\)', record[0])
            clip['title'], clip['attribution'] = match.groups()
            clip['attribution'] = clip['attribution'].split(') (')

            match = re.match(r'- (\w+) Loc. ([^|]+)\| Added on (\w+), (\w+ \d+, \d+), (\d+:\d+ \w\w)', record[1])
            clip['type'], clip['location'], clip['dow'], clip['date'], clip['time'] = match.groups()

            clip['body'] = "\n".join(record[3:])

            # a little tidying
            clip['title'] = clip['title'].strip()
            clip['location'] = clip['location'].strip()

            # yield and reset for next record
            yield clip
            record = list()
        else:
            record.append(line.strip())

    clip_file.close()


if __name__ == '__main__':
    from sys import argv
    for n, r in enumerate(records(argv[1])):  # 'My Clippings.txt') ):
        print n,
        pprint(r)