import csv
import json
from math import sqrt
import re

WORD_THRESHOLD = 4

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) for word in words]
    important_words = set([w for w in normalized_words if len(w) > WORD_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:
                insert(db, row)
            else:
                data = True
    return 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 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))/sqrt(len(a)*len(b))

def find_best_quote(db, user_input):
    indexed_input = index(user_input)
    max_score = None
    for entry in db:
        score = scalar(indexed_input, entry