import omdb

import os

import csv

import pandas as pd

# omdb is a pythom wrapper for OMDB written by a third party

# run pip install omdb if omdb is not already in your python environment

# Define movie lists

basics_path = os.path.join('.', 'title.akas.tsv')

ratings_path = os.path.join('.', 'title.ratings.tsv')

basics_df = pd.read_csv(basics_path, sep='\t')

eq = basics_df.iloc[3]['region']

basics_df.drop(basics_df[basics_df['region'] != eq].index, inplace=True)

col_to_drop = [x for x in basics_df.columns if x not in ['titleId', 'title']]

basics_df = basics_df.drop(columns=col_to_drop, axis=1)

ratings_df = pd.read_csv(ratings_path, sep='\t')

indexNames = ratings_df[(ratings_df['averageRating'] < 9.0) & (ratings_df['numVotes'] <= 500)].index

ratings_df.drop(indexNames, inplace=True)

usable_df = pd.merge(basics_df, ratings_df, on=['titleId'])

og_csv = os.path.join('.', 'movie_data.csv')

output_path = os.path.join('.', 'rom_act.csv')

movies = usable_df['title'].tolist()

movies_to_try = movies[948:]

# api key is user specifc. Micaela has one (limit 1000 a day)

key = "60f61a52"

omdb.set_default('apikey', key)

with open(og_csv) as already_done:

dr = csv.DictReader(already_done)

already_have = [x['title'] for x in dr]

already_have = set(already_have)

movies_to_try = [x for x in movies_to_try if not x in already_have]

counter = 1

for m in movies_to_try:

print(f'movie: {m}')

movie_dict = {}

try:

res = omdb.title(m, fullplot=True)

movie_dict['title'] = res['title']

movie_dict['genre'] = res['genre']

movie_dict['plot'] = res['plot']

df = pd.DataFrame([movie_dict], columns=movie_dict.keys())

output_path = os.path.join('.', 'movie_data.csv')

genres = [x.lower() if not x.endswith(',') else x[:-1].lower() for x in movie_dict['genre'].split()]

plot = movie_dict['plot'].split()

#OG FILTER

if ('adult' in genres or 'biography' in genres or 'documentary' in genres) and len(plot) <= 30:

continue

# 11/17 filter -- action and romance only

if (not 'action' in genres) and (not 'romance' in genres):

continue

df = pd.DataFrame([movie_dict], columns=movie_dict.keys())

with open(output_path, 'a', newline='') as out_file:

writer = csv.DictWriter(out_file, fieldnames=["title", "genre", "plot"])

if counter == 1:

writer.writeheader()

writer.writerow(movie_dict)

print(f"added movie {counter}")

counter += 1

except KeyError as e:

print(f'response = {res}')