def get_tweets_in(location_id=my.HBK_LOCATION_ID):
# get all tweets in hollenbeck area
import sys
sys.path.append("/home/gambit/collector/gambit2/")
from django.core.management import setup_environ
from gambit import settings
setup_environ(settings)
from scraper.models import Location, Tweet
print 'Querying database...'
loc = Location.objects.get(id=location_id)
tweets = Tweet.filter(geo__within=loc.polygon)
if not os.path.exists('data/' + my.DATA_FOLDER):
os.makedirs('data/' + my.DATA_FOLDER)
with open('data/' + my.DATA_FOLDER + my.HBK_TWEET_LOC_FILE, 'wb') as fp:
csv_writer = csv.writer(fp, delimiter=',')
for tweet in tweets:
csv_writer.writerow([tweet.user_id, tweet.geo[0], tweet.geo[1]])
print 'Done! Fetched ' + str(tweets.count()) + ' entries.'
def scrape_target(long_run=False):
try:
latest_tweet = Tweet.objects.latest('twitter_published_date')
latest_tweet_id = latest_tweet.tweet_id
except ObjectDoesNotExist:
latest_tweet_id = None
if not long_run:
timeline = api.user_timeline(id=TARGET_HANDLE, since_id=latest_tweet_id, count=200)
else:
timeline = api.user_timeline(id=TARGET_HANDLE, max_id=latest_tweet_id, count=200)
count = 0
check = 0
while timeline:
for tweet in timeline:
count += 1
tweet_id = tweet['id_str']
try:
match = Tweet.objects.get(tweet_id=tweet_id)
except ObjectDoesNotExist:
pass
else:
# This is a really ugly hack
if long_run:
check += 1
if check > 1:
timeline = []
break
else:
check = 0
continue
else:
timeline = []
break
hashtags = tweet['entities']['hashtags']
user_mentions = tweet['entities']['user_mentions']
pubdate = datetime.strptime(tweet['created_at'], '%a %b %d %H:%M:%S +0000 %Y')
pubdate = pubdate.replace(tzinfo=tz)
date_added = datetime.now(pytz.utc)
tweet_url = 'https://twitter.com/' + TARGET_HANDLE + '/status/' + str(tweet_id)
tweet_text = tweet['text']
text_first_word, tweet_text_clean = tweet_cleaner(tweet)
if text_first_word == 'RT':
continue
retweet_count = tweet['retweet_count']
favourite_count = tweet['favorite_count']
in_reply_to_screen_name = tweet['in_reply_to_screen_name']
in_reply_to_status_id = tweet['in_reply_to_status_id_str']
in_reply_to_user_id = tweet['in_reply_to_user_id_str']
is_quote_status = tweet['is_quote_status']
source = tweet['source']
json_dump = json.dumps(tweet)
if in_reply_to_screen_name:
try:
existing_user = OtherUser.objects.get(handle=in_reply_to_screen_name)
except ObjectDoesNotExist:
new_user = OtherUser(handle=in_reply_to_screen_name, twitter_id=in_reply_to_user_id)
new_user.save()
reply_user = new_user
else:
reply_user = existing_user
is_reply = True
else:
reply_user = None
is_reply = False
if not is_quote_status and text_first_word is not None:
try:
existing_firstword = FirstWord.objects.get(word=text_first_word)
tweet_first_word = existing_firstword
except ObjectDoesNotExist:
new_firstword = FirstWord(word=text_first_word)
new_firstword.save()
tweet_first_word = new_firstword
new_tweet = Tweet(text=tweet_text, clean_text=tweet_text_clean, tweet_id=tweet_id,
date_added=date_added, twitter_published_date=pubdate, tweet_url=tweet_url,
retweet_count=retweet_count, favourite_count=favourite_count, is_reply=is_reply,
reply_user=reply_user, in_reply_to_status_id=in_reply_to_status_id,
is_quote_status=is_quote_status, quoted_status_id=quoted_status_id, first_word=tweet_first_word,
source=source, json_dump=json_dump)
new_tweet.save()
if not long_run:
corpus = Corpus.objects.all()[0]
corpus.content = corpus.content + new_tweet.clean_text + ' '
corpus.save()
if hashtags:
for hashtag in hashtags:
try:
hashtag_obj = HashTag.objects.get(text=hashtag['text'])
except ObjectDoesNotExist:
hashtag_obj = HashTag(text=hashtag['text'])
hashtag_obj.save()
hashtag_obj.related_tweet.add(new_tweet)
if user_mentions:
for user_mention in user_mentions:
try:
user_mention_obj = OtherUser.objects.get(handle=user_mention['screen_name'])
except ObjectDoesNotExist:
user_mention_obj = OtherUser(handle=user_mention['screen_name'], twitter_id=user_mention['id_str'])
user_mention_obj.save()
user_mention_obj.related_tweet.add(new_tweet)
else:
timeline = api.user_timeline(id=TARGET_HANDLE, max_id=tweet_id, count=200)
