from musicbrainz Url Schema to extract some useful infomation and export csv , with parse to json. there is a amusing way to decide source due to the url type was various.

musicbrainzUrlSchemaParseCsvToJSON.py

import csv
import json
import os
from urlparse import urlparse

csvfile = open('url.csv', 'r')
jsonfile = open('test.json', 'w')

fieldnames = ("@id","sourceUrl")
reader = csv.DictReader( csvfile, fieldnames)
for row in reader:
   # print 'row.sourceUrl is ' + row['sourceUrl']
    url = urlparse(row['sourceUrl'])
    if len(url.query) != 0:
        row.update({"idAtSource":url.query})
    else:
        # there also antoher problem, if the url contain the '/', the basename will be none , but it's shoudn't be that
        row.update({"idAtSource":os.path.basename(url.path)})

# URL domain was complex can't be parse, but this way was amusing to resolve it.  
    domain = url.netloc.split('.')
    if len(domain) == 2:
        row.update({'sourceName':domain[0]})
    else:
        row.update({'sourceName':domain[1]})
        
#     elif len(domain) == 3 :
#         row.update({'sourceName':domain[1]})
#     elif len(domain) == 4 :
#         row.update({'sourceName':domain[1]})
#     else:
#         row.update({'sourceName':domain[1]})

    row.update({"sourceDomain":url.netloc})
    row.update({"@type":"ExternalLink"})
    json.dump(row,jsonfile)
    jsonfile.write('\n')