Finally implement the search

memory.py

@@ -1,41 +1,63 @@
 import csv
 import json
+from math import sqrt
+import re
 
-WORD_THRESHOLD = 4
+WORD_LENGTH_THRESHOLD = 4
 
-def build_db(inputCSV, outputJSON):
+class RowDecoder(json.JSONDecoder):
+    def decode(self, s):
+        db = json.JSONDecoder.decode(self, s)
+        return [{**obj, 'index': set(obj['index'])} for obj in db]
+
+class RowEncoder(json.JSONEncoder):
+    def default(self, obj):
+        if isinstance(obj, set):
+            return list(obj)
+        else:
+            return json.JSONEncoder.default(obj)
+
+def keepOnlyAlphaChars(word):
+    return ''.join([c for c in word if c.isalpha()])
+
+def index(text):
+    words = re.split('\s', text)
+    normalized_words = [keepOnlyAlphaChars(word).lower() for word in words]
+    important_words = set([w for w in normalized_words
+                             if len(w) >= WORD_LENGTH_THRESHOLD])
+    return important_words
+
+def insert(db, row):
+    db.append({'title': row[0], 'quote': row[1], 'index': index(row[1])})
+
+def build_db(inputCSV):
     db = []
     with open(inputCSV, 'r') as file:
         csv_reader = csv.reader(file, delimiter=',')
         data = False
         for row in csv_reader:
             if data:
-                db.append(row + (index(row[1]),))
+                insert(db, row)
             else:
                 data = True
-    with open(outputJSON, 'w') as file:
-        json.dump(serialize(db), file)
     return db
 
-def serialize(db):
-    return list(map(lambda row: (row[0], row[1], list(row[2])), db))
-
-def unserialize(db):
-    return list(map(lambda row: (row[0], row[1], set(row[2])), db))
+def save_db(db, outputJSON):
+    with open(outputJSON, 'w') as file:
+        json.dump(db, file, cls=RowEncoder)
 
 def open_db(filePath):
     with open(filePath, 'r') as file:
-        return unserialize(json.load(file))
+        return json.load(file, cls=RowDecoder)
 
+"""
+We define a similarity measure on sets which counts the number of elements
+they have in common
+"""
 def scalar(a, b):
-    return len(a.intersection(b))
+    sizeProduct = len(a)*len(b)
+    return len(a.intersection(b))/sqrt(sizeProduct) if sizeProduct > 0 else 0
 
-def find_best_quote(db, indexed_input)
-    max_score = None
-    for entry in db:
-
-
-def index(text):
-    words = map(lambda w: ''.join([c for c in w if c.isalpha()]), text.split(' '))
-    important_words = set([w for w in words if len(w) > WORD_THRESHOLD])
-    return important_words
+def find_best_quote(db, user_input):
+    indexed_input = index(user_input)
+    return max(db, key=lambda row: scalar(indexed_input, row['index']))

quotes.csv

@@ -0,0 +1,7 @@
+book,quote
+The Color of Magic,"Tourist, Rincewind decided, meant ""idiot""."
+The Colour of Magic,"""Let's just say that if complete and utter chaos was lightning, he'd be the sort to stand on a hilltop in a thunderstorm wearing wet copper armour and shouting 'All gods are bastards'."""
+The Light Fantastic,"""DID YOU SAY HUMANS PLAY IT FOR FUN?""
+""Some of them get to be very good at it, yes. I'm only an amateur,
+I'm afraid""
+""BUT THEY ONLY LIVE EIGHTY OR NINETY YEARS!"""