This script gets urls from tabs in a firefox session.

firefox_url_extractor.py

#!/usr/bin/env python3
import json

def get_session_from_file(filename):
    with open(filename, 'r') as fp:
        session = json.load(fp)
    return session

def get_tabs_from_session(session):
    return session['windows'][0]['tabs']

def get_group_info_from_session(session):
    groups = []
    group_data = json.loads(session['windows'][0]['extData']['tabview-group'])
    for id, data in group_data.items():
        groups.append((id, data['title']))
    return groups



def get_urls_from_tabs(tabs):
    return map(lambda t: t['entries'][0]['url'], tabs)

def get_urls_from_group_tab(tabs, group_id):
    return get_urls_from_tabs(filter(lambda t:
        t['extData']['tabview-tab'].find('{{"groupID":{}'.format(group_id)) != -1, tabs))


def print_usage_and_exit():
    sys.stdout.write(
        "usage: {} [-g] <path_to_session_file>\n".format(sys.argv[0]))
    sys.exit(1)

if __name__ == "__main__":
    import sys
    import pprint

    # I should use argparse, but who cares?
    if len(sys.argv) < 2:
        print_usage_and_exit()
    elif '-g' in sys.argv and len(sys.argv) <3: # fuck pep8 for the sake of ASCII
        print_usage_and_exit()

    path = sys.argv[-1]

    session = get_session_from_file(path)
    tabs = get_tabs_from_session(session)

    if '-g' in sys.argv:
        groups = get_group_info_from_session(session)
        for group_id, group_name in groups:
            group_urls = get_urls_from_group_tab(tabs, group_id)

            print('[{}]'.format(group_name if group_name else 'unnamed group'))
            list(map(print, group_urls))

    else:
        urls = get_urls_from_tabs(tabs)
        list(map(print, urls))