Mini 5

Auto_tweeter.py

@@ -0,0 +1,75 @@
+import data_finder
+from pickle import dump, load
+import random
+
+class Auto_Tweeter:
+    def __init__ (self):
+        self.data_grab()
+        self.tweet_maker()
+
+    def data_grab(self):
+        # asks if you need to pull new data for this person
+        print ('What is the handle of the person you want to mimic?')
+        self.user_handle = input()
+
+        print ('Have you collected Data? (y/n)')
+        collect_data = input()
+        #pulls new data, by calling Auto_tweeter
+        if collect_data == 'n':
+            self.statuses = data_finder.data_finder(self.user_handle)
+            self.dictionary_maker() # calls the dictionary_maker method
+    def dictionary_maker(self):
+        #sorts data into for markov analysis
+        self.dictionary = {}
+        self.start = []
+        self.end = []
+        for tweet in self.statuses:
+            # collects all the starts of each tweet
+            self.each_start = [[tweet[0],tweet[1]]]
+            self.start = self.start + self.each_start
+        self.dictionary['START']= self.start
+        for tweet in self.statuses:
+            tweet += ['END'] # adding and end flag to the end of the tweet
+            i= len(tweet)
+            for x in range (0,(i-2)):
+                #forming a dictionary linking previos two to the next word
+                if tweet[x]+tweet[x+1] in self.dictionary:
+                    self.dictionary[tweet[x]+tweet[x+1]] += [tweet[x+2]]
+                else:
+                    self.dictionary[tweet[x]+tweet[x+1]] = [tweet[x+2]]
+
+        self.pickle()
+
+    def pickle(self):
+        #saves the data set in a file
+        Data_Set = open(str(self.user_handle)+'.txt','wb')
+        dump(self.dictionary, Data_Set)
+        Data_Set.close()
+
+
+    def tweet_maker(self):
+        Data_Set = open(str(self.user_handle)+'.txt','rb')
+        #loading the data
+        self.dictionary = load(Data_Set)
+        print(len(self.dictionary['START']))
+        self.new_tweet=[]
+        self.new_tweet += random.choice(self.dictionary['START'])
+        #starts the tweet off with a random begining
+        #keeps adding randomly untill we arive at an END
+        while not self.new_tweet[-1] == 'END':
+            self.new_tweet += [random.choice(self.dictionary[self.new_tweet[-2]+self.new_tweet[-1]])]
+        self.final_tweet = ' '
+        #adds the words together and adds spaces
+        for word in self.new_tweet:
+            if not word =='END':
+                self.final_tweet += ' ' + word
+        #prints the new tweet
+        print(str(self.user_handle)+' says:'+ self.final_tweet)
+
+        print('Make another tweet (y/n)')
+        another= input()
+        if another == 'y':
+            self.tweet_maker()
+
+if __name__ == '__main__':
+    tweeter = Auto_Tweeter()

README.md

@@ -1,3 +1,21 @@
+Update 5/9/17
+ to run
+ >> python3 Auto_tweeter.py
+
+and then follow directions
+
+
+
+
+
 # TextMining
 
 This is the base repo for the text mining and analysis project for Software Design at Olin College.
+
+I was told to note the hours that I worked on this assignment and that I was
+instructed to continue prioritizing good code over finished code.
+I spent about 9 hours on this project this week, most of that was spent
+rewriting data_finder over and over again to find a method that i liked....
+
+my write up:
+https://docs.google.com/document/d/1DwI3nODol5V7itbYxsNl4fq1XazYqyo1oBaXSLcysN4/edit?usp=sharing

data_finder.py

@@ -0,0 +1,52 @@
+import twitter
+import pickle
+#who do you want to mimic
+def data_finder(user_handle):
+    while True:
+        #data finder prompst the user to input login material
+        print ('consumer_key:')
+        c_k = input()
+        print ('consumer_secret:')
+        c_s = input()
+        print ('access_token_key:')
+        a_t_k = input()
+        print ('access_token_secret:')
+        a_t_s = input()
+        #data fider uses these to login
+
+        api = twitter.Api(consumer_key=c_k,
+                          consumer_secret=c_s,
+                          access_token_key=a_t_k,
+                          access_token_secret=a_t_s)
+
+        # test to see if this brings an error if it doesnt break! if it does,
+        #try autenticating again
+        try:
+            api.VerifyCredentials()
+            break
+        except twitter.error.TwitterError as e:
+            print(e)
+            data_finder(user_handle)
+
+    # how do i make a list of lists of words from each tweet?
+    # How do i set up a login error that brings people back to the login option and works for more people than just me
+    # How do i save that list so i can open it in another progam
+
+
+    statuses = api.GetUserTimeline(screen_name=user_handle, count=850)
+    statuses = [s.text for s in statuses]
+    #print (statuses)
+    words = []
+    for status in statuses:
+        tweet = [status.split()]
+        words = words +  tweet
+    statuses = words
+    return statuses
+
+
+
+
+if __name__ == '__main__':
+    #import doctest
+    #doctest.testmod(verbose=True)
+    data_finder('pontifex')