#!/usr/bin/python3
#
# tweetfetch.py
#
# Fetches tweets since a given timestamp and stores data in JSON files, as
# well as an HTML dump for each in the subdirectory ./tweets.
#
# Usage: ./tweetfetch.py 1159606554041040896
#
# The timestamp of the last tweet fetched is stored in ./tweets/.sentinel
#
# Copyright © 2017–2019 by martin f. krafft <madduck@madduck.net>
# Released under the Artistic Licence 2.0
#

from authdata import *

from twython import Twython
import json
import sys

twitter = Twython(app_key=consumer_key,
                  app_secret=consumer_secret,
                  oauth_token=access_token,
                  oauth_token_secret=access_secret,
                  oauth_version=1)

config = {'include_rts': False,
          'count': 200,
          'trim_user': True,
          'exclude_replies': True,
         }

if len(sys.argv) > 1:
    config['since_id'] = sys.argv[1]
    print("Limiting results to tweets since ID {}".format(config['since_id']),
            file=sys.stderr)

user_timeline = twitter.get_user_timeline(screen_name="martinkrafft",
        **config)

max_id = int(config.get('since_id', 0))

print("Fetched {} tweets, writing them to disk…".format(len(user_timeline)),
        file=sys.stderr)

for tweet in user_timeline:
    with open("tweets/{}.json".format(tweet['id_str']), "wt") as tf:
        print(json.dumps(tweet), file=tf)

    with open("tweets/{}.html".format(tweet['id_str']), "wt") as tf:
        print(Twython.html_for_tweet(tweet, use_expanded_url=True), file=tf)

    print("  wrote tweet ID {}".format(tweet['id_str']),
            file=sys.stderr)

    max_id = max(tweet['id'], max_id)

print("Writing ID {} to sentinel file…".format(max_id), file=sys.stderr)
with open("tweets/.sentinel", "wt") as tf:
    print('{0:d}'.format(max_id), file=tf)