Ejemplo n.º 1
0
    def handle(self, **options):
        """
        Prints out a count of number of tweets by screen_name for occupydenver.
        """
        # Get current datetime as a string to append to the file.
        # This should allow us to know when it was run and keep from overwriting.
        now = datetime.datetime.now().strftime('%Y%m%d%H%M')

        # Create a CSV fiile counting tweets by users
        f = open('ODhashtag_users_%s.txt' % now, 'wb')
	csv_writer = UnicodeWriter(f)
	csv_writer.writerow(['Screen Name', 'Num od Tweets'])

        tweets = Tweet.objects.filter(tags__name='occupydenver').order_by('twitter_user')        

	# Django has problems with aggregation across generic relationships
        # So we'll have to read it into a dict and count it there.
        # https://docs.djangoproject.com/en/dev/ref/contrib/contenttypes/#generic-relations-and-aggregation
	tweeters = defaultdict(int)

        for tweet in tweets:
	    tweeters[tweet.twitter_user.screen_name] += 1

        for user, count in tweeters.items():
            csv_writer.writerow([user.encode('utf-8'), '%s' % count])
Ejemplo n.º 2
0
    def handle(self, **options):
        """
        """
        # Get current datetime as a string to append to the file.
        # This should allow us to know when it was run and keep from overwriting.
        now = datetime.datetime.now().strftime('%Y%m%d%H%M')

        # Create a CSV fiile counting tweets by users
        f = open('userlocations_%s.csv' % now, 'wb')
        writer = UnicodeWriter(f)
        writer.writerow(['Name', 'Location', 'Tweet Count'])

        user_list = TwitterUser.objects.annotate(tweetcount=Count('tweet'))
        for user in user_list:
            if user.location:
                writer.writerow([user.screen_name, user.location, '%s' % user.tweetcount])
Ejemplo n.º 3
0
    def handle(self, **options):
        """
        """
        # Get current datetime as a string to append to the file.
        # This should allow us to know when it was run and keep from overwriting.
        now = datetime.datetime.now().strftime('%Y%m%d%H%M')

        # Create a CSV fiile counting tweets by users
        f = open('usergeo_%s.csv' % now, 'wb')
        writer = UnicodeWriter(f)
        writer.writerow(['Username', 'Tweet', 'DateTweeted', 'Coordinates(lat/long)'])
	geotweets = TwitterGeo.objects.all()

        for geotweet in geotweets:
            tweetdate = geotweet.tweet.created_at.strftime('%Y-%m-%d %H:%M:%S')
	    location = "%s, %s" % (geotweet.latitude, geotweet.longitude)
	    writer.writerow([
		geotweet.tweet.twitter_user.screen_name,
		geotweet.tweet.text,
		tweetdate,
		location,
		])
Ejemplo n.º 4
0
    def handle(self, **options):
        """
        This is helps write some basic aggregation data about users and their tweets.

        It should accomplish 2 things.

            *  Save a list of individual users screen_name, location, # of tweets
            * Save a list of # of tweets, # of users with those number of tweets.

            Sandemous Highschool Football Rules!
        """
        # Get current datetime as a string to append to the file.
        # This should allow us to know when it was run and keep from overwriting.
        now = datetime.datetime.now().strftime("%Y%m%d%H%M")

        # Create a CSV fiile counting tweets by users
        f = open("hashtagcount_%s.csv" % now, "wb")
        writer = UnicodeWriter(f)
        writer.writerow(["HashTag", "Count"])

        tag_list = Tag.objects.annotate(count=Count("taggit_taggeditem_items")).order_by("-count")
        for tag in tag_list:
            writer.writerow([tag.name, "%s" % tag.count])
Ejemplo n.º 5
0
    def handle(self, **options):
        """
        This is helps write some basic aggregation data about users and their tweets.

        It should accomplish 2 things.

            *  Save a list of individual users screen_name, location, # of tweets
            * Save a list of # of tweets, # of users with those number of tweets.

            Sandemous Highschool Football Rules!
        """
        # Get current datetime as a string to append to the file.
        # This should allow us to know when it was run and keep from overwriting.
        now = datetime.datetime.now().strftime("%Y%m%d%H%M")

        # Create a CSV fiile counting tweets by users
        by_user = open("tweetcount_by_user_%s.csv" % now, "wb")
        by_user_writer = UnicodeWriter(by_user)
        by_user_writer.writerow(["screen_name", "location", "total tweets"])

        tweet_counter = {}
        users = TwitterUser.objects.annotate(tweetcount=Count("tweet"))
        for user in users:
            by_user_writer.writerow([user.screen_name, "%s" % user.location, "%s" % user.tweetcount])
            if user.tweetcount in tweet_counter:  # Doing it this way so I don't have to do another DB query.
                tweet_counter[user.tweetcount] += 1  # increment it if it exists
            else:
                tweet_counter[user.tweetcount] = 1  # init it if it doesn't exist
        by_user.close()

        # Create a CSV file aggregating users by their number of tweets
        total_tweets = open("by_total_tweets_%s.csv" % now, "wb")
        total_tweets_writer = UnicodeWriter(total_tweets)
        total_tweets_writer.writerow(["Tweet Count", "Number of Users"])
        for tweetcount, users in tweet_counter.items():
            total_tweets_writer.writerow(["%s" % tweetcount, "%s" % users])
        total_tweets.close()