added database and userlearning

.gitignore

@@ -1,5 +1,6 @@
 images/**
 output/**
 src/index.json
+src/database.json
 .DS_Store
 src/**.wav

src/build_database.py

@@ -2,16 +2,30 @@ import json
 import argparse
 import sys
 from nltk.tokenize import sent_tokenize, word_tokenize
+from nltk.corpus import stopwords
+import nltk
+import numpy as np
 from rake_nltk import Rake
-
+from textblob import TextBlob
+stop_words = set(stopwords.words('english'))
 r= Rake()
 
 ap = argparse.ArgumentParser("JSON Dumper")
 ap.add_argument("text", nargs="+", help="text sources")
 args=ap.parse_args()
 
+def cleanedText(text):
+    word_tokens = word_tokenize(text)
+    word_tokens = [word.lower() for word in word_tokens]
+    word_tokens = [word for word in word_tokens if word.isalpha()]
+    filtered_sentence = [w for w in word_tokens if not w in stop_words]
+    text = " ".join(filtered_sentence)
+    print(text)
+    return text
+
+#### INDEX (DEPR)
 
-with open('src/index.json') as f:
+with open('src/index.json', 'w') as f:
     try:
         index = json.load(f)
     except:
@@ -36,3 +50,86 @@ for n in args.text:
 
 with open('src/index.json', 'w') as outfile:
     json.dump(index, outfile)
+
+
+
+######## DATABASE
+
+with open('src/database.json', 'w') as f:
+    try:
+        index = json.load(f)
+    except:
+        index={}
+
+
+nouns = []
+verbs = []
+adverbs = []
+pronouns = ["I", "you", "we", "they"]
+adjectives = []
+keywords = []
+keywordspersentence = {}
+alltext = ""
+allcharacters = []
+
+for n in args.text:
+    text = open(n).read()
+    rawtext = text.replace("\n", " ")
+    #get pos
+    for letter in list(rawtext):
+        if letter not in allcharacters and not letter.isalpha():
+            allcharacters.append(letter)
+
+    alltext += rawtext
+    text = cleanedText(text)
+    thistext = TextBlob(text)
+    words = word_tokenize(text)
+    listWithPos = nltk.pos_tag(words)
+    print(thistext.tags)
+    for key, tag in thistext.tags:
+        if(tag == "VBP"):
+            verbs.append(key)
+        if(tag == "NN" or tag == "NNS"):
+            nouns.append(key)
+        if(tag == "RB"):
+            adverbs.append(key)
+        if(tag == "JJ"):
+            adjectives.append(key)
+
+    print("Verbs: {}".format(verbs))
+    print("Nouns: {}".format(nouns))
+    print("Adverbs: {}".format(adverbs))
+    print("Adjectives: {}".format(adjectives))
+    #keywords
+    r.extract_keywords_from_text(rawtext)
+    phrases = r.get_ranked_phrases_with_scores()
+    for key, phrase in phrases:
+        if key > 2:
+            keywords.append(phrase)
+
+    print("Keywords: {}".format(keywords))
+
+    # keywordsofsentences
+    sentences = sent_tokenize(text)
+    for sentence in sentences:
+
+        r.extract_keywords_from_text(sentence)
+        keys = r.get_ranked_phrases()
+        for key in keys:
+            if key not in keywordspersentence:
+                keywordspersentence[key] = []
+            keywordspersentence[key].append({'filename': n, 'sentence': sentence, 'key': key})
+
+print(allcharacters)
+index["nouns"]=nouns
+index["verbs"]=verbs
+index["pronouns"]=pronouns
+index["adverbs"]=adverbs
+index["adjectives"]=adjectives
+index["keywords"]=keywords
+index["keywordspersentence"]=keywordspersentence
+index["rawtext"]=alltext
+index["chars"] = allcharacters
+
+with open('src/database.json', 'w') as outfile:
+    json.dump(index, outfile)

src/chatbook.py

@@ -69,9 +69,10 @@ if __name__ == "__main__":
         try:
             index = json.load(f)
         except:
-            index={}
+            print("I can't work with no knowledge")
+            sys.exit()
+
 
-    #print(index)
     myhost = os.uname()[1]
 
 

src/oulibot.py

@@ -27,6 +27,7 @@ from textblob import TextBlob
 from textblob_aptagger import PerceptronTagger
 import string
 import re
+import json
 import time
 import sys; from PIL import Image; import numpy as np
 chars = np.asarray(list(' .,:;irsXA253hMHGS#9B&@'))
@@ -158,7 +159,7 @@ class HelloBot(irc.bot.SingleServerIRCBot):
         self.poem = []
         self.keys = list(map(lambda x:re.compile(x[0], re.IGNORECASE),basicDialog))
         self.values = list(map(lambda x:x[1],basicDialog))
-        self.POSSIBLE_STRUCTURES=[["adjective", "noun", "adverb", "verb", "adjective", "noun"], ["pronoun", "verb", "noun"]]
+        self.POSSIBLE_STRUCTURES=[["adjective", "noun", "adverb", "verb", "adjective", "noun"], ["pronoun", "verb", "adverb"]]
         self.structure = random.choice(self.POSSIBLE_STRUCTURES)
 
     def translate(self,str,dict):
@@ -455,10 +456,15 @@ class HelloBot(irc.bot.SingleServerIRCBot):
             else:
                 self.sentence.append(incoming_msg)
                 self.lastPOS +=1
+                print(self.waitingfor + " " + incoming_msg)
+                self.index[self.waitingfor + "s"].append(incoming_msg)
+                with open('src/database.json', 'w') as outfile:
+                    json.dump(self.index, outfile)
                 self.generateSentence(c,e)
 
 
 
+
 def cleanedText(text):
     word_tokens = word_tokenize(text)
     word_tokens = [word.lower() for word in word_tokens]
@@ -482,77 +488,13 @@ if __name__ == "__main__":
     ap.add_argument("--text", nargs="+", help="text sources")
     args=ap.parse_args()
 
-    # build the index
+    with open("src/database.json") as f:
-    index = {}
+        try:
-
+            index = json.load(f)
-    nouns = []
+        except:
-    verbs = []
+            print("I can't work with no knowledge")
-    adverbs = []
+            sys.exit()
-    pronouns = ["I", "you", "we", "they"]
-    adjectives = []
-    keywords = []
-    keywordspersentence = {}
-    alltext = ""
-    allcharacters = []
-
-    n = "ocr/output.txt"
-    text = open(n).read()
-    rawtext = text.replace("\n", " ")
-    #get pos
-    for letter in list(rawtext):
-        if letter not in allcharacters and not letter.isalpha():
-            allcharacters.append(letter)
-
-    alltext += rawtext
-    text = cleanedText(text)
-    thistext = TextBlob(text)
-    words = word_tokenize(text)
-    listWithPos = nltk.pos_tag(words)
-    print(thistext.tags)
-    for key, tag in thistext.tags:
-        if(tag == "VBP"):
-            verbs.append(key)
-        if(tag == "NN" or tag == "NNS"):
-            nouns.append(key)
-        if(tag == "RB"):
-            adverbs.append(key)
-        if(tag == "JJ"):
-            adjectives.append(key)
-
-    print("Verbs: {}".format(verbs))
-    print("Nouns: {}".format(nouns))
-    print("Adverbs: {}".format(adverbs))
-    print("Adjectives: {}".format(adjectives))
-    #keywords
-    r.extract_keywords_from_text(rawtext)
-    phrases = r.get_ranked_phrases_with_scores()
-    for key, phrase in phrases:
-        if key > 2:
-            keywords.append(phrase)
-
-    print("Keywords: {}".format(keywords))
-
-    # keywordsofsentences
-    sentences = sent_tokenize(text)
-    for sentence in sentences:
-
-        r.extract_keywords_from_text(sentence)
-        keys = r.get_ranked_phrases()
-        for key in keys:
-            if key not in keywordspersentence:
-                keywordspersentence[key] = []
-            keywordspersentence[key].append({'filename': n, 'sentence': sentence, 'key': key})
-
-    print(allcharacters)
-    index["nouns"]=nouns
-    index["verbs"]=verbs
-    index["pronouns"]=pronouns
-    index["adverbs"]=adverbs
-    index["adjectives"]=adjectives
-    index["keywords"]=keywords
-    index["keywordspersentence"]=keywordspersentence
-    index["rawtext"]=alltext
-    chars = np.asarray(allcharacters)
 
+    chars = np.asarray(index["chars"])
     bot = HelloBot(args.channel, args.nickname, args.server, args.port, index)
     bot.start()