def parse_tweets(result_dict):
results = result_dict['results']
messages = []
for result in results:
# update timezone of created_at
created_at_str = result['created_at'] + ' UTC' # +0000 = UTC
created_at = datetime.strptime(created_at_str, '%a, %d %b %Y %H:%M:%S +0000 %Z')
result.update({'created_at': created_at})
# save into database
try:
tweet_json = {}
tweet_json['text'] = result['text']
tweet_json['created_at'] = result['created_at']
tweet_json['id_str'] = result['id_str']
tweet_json['profile_image_url'] = result['profile_image_url']
tweet_json['from_user'] = result['from_user']
if result['geo']:
tweet_json['lat'] = result['geo']['coordinates'][0]
tweet_json['long'] = result['geo']['coordinates'][1]
tweet = Tweet(**tweet_json)
tweet.save()
# download photos in the tweet
urls = find_url_in_tweet(tweet.text)
image_urls = extract_urls_from_tweet(urls)
photos = download_all(image_urls)
for photo_name in photos:
photo = Photo(name = photo_name, tweet = tweet)
photo.save()
except (IntegrityError, DatabaseError):
pass # tweet already saved
messages.append(result['text'])
return messages
def test_download_all_returns_list_of_image_paths(self):
image_urls = ['http://p.twimg.com/AZh5jT2CEAAXyR-.jpg:large',
'http://distillery.s3.amazonaws.com/media/2011/09/17/bf5bb54387ca4dd98573128bb02f5c0b_7.jpg',
'http://s3.amazonaws.com/twitpic/photos/full/400298248.jpg?AWSAccessKeyId=AKIAJF3XCCKACR3QDMOA&Expires=1317580915&Signature=MByJjnEyjvdQZnP40RjbrPHDUOM%3D']
image_paths = download_all(image_urls)
expected = [u'httpp.twimg.comAZh5jT2CEAAXyR-.jpglarge',
u'httpdistillery.s3.amazonaws.commedia20110917bf5bb54387ca4dd98573128bb02f5c0b_7.jpg',
u'https3.amazonaws.comtwitpicphotosfull400298248.jpg']
self.assertEqual(expected, image_paths)
