class flu_tweets:

    def __init__(self):
        self.tweets = []
        self.labels = []


    def load(self, tweets_filename, labels_filename = ""):
        f = open(tweets_filename, 'r')
        for tweet in f:
            if tweet != "\n":
                self.tweets.append(tweet)
        f = open(labels_filename, 'r')
        for label in f.readlines():
            b = label.strip("\n"):
                self.labels.append(b)


    def label_tweets(self):
        for tweets in self.tweets:
            print json.loads(tweet)['text']
        while True:
            l = raw_input("Label as pos or neg")
            if l == "pos" or l == "neg":
                self.labels.append(l)
            break


    def count(self, label):
        count = 0
        for i in self.labels:
            if i == label:
                count += 1
        return count


    def trim(self):
        for i in range(len(self.tweets)):
            if json.loads(self.tweets[i])['lang'] != 'en':
                del self.labels[i]
                del self.tweets[i]

        cnt_pos = 0
        cnt_neg = 0
        for i,j in enumerate(self.labels):
            if j == "pos":
                if cnt_pos < 10:
                    cnt_pos += 1
                else:
                    del self.tweets[i]
                    del j

            if j == "neg":
                if cnt_neg < 90:
                    cnt_neg += 1
                else:
                    del self.tweets[i]
                    del j


    def merge(self, another_flu_tweets):
        self.tweets = self.tweets + another_flu_tweets.tweets
        self.labels = self.labels + another_flu_tweets.labels