import multiprocessing
import traceback
import sys

import boto3
import botocore

import mysql
import query
import config

BUCKET = 'dev-orcd-raw-assets'
PAGE_SIZE = 100
THREADS = 12

s3_client = boto3.client('s3')

paginator_config = {
    'PageSize': PAGE_SIZE
}


def check_file(arguments):
    k = arguments[0]
    e = True
    try:
        s3_client.head_object(Bucket=BUCKET, Key=k)
    except botocore.exceptions.ClientError:
        e = False
    return k, e


if __name__ == '__main__':
    with mysql.mysql_connection(**config.DB_CONFIG) as conn:
        with conn.cursor() as cursor:
            cursor.execute(query.SELECT_SONGS)
            songs = list(cursor)
            files = []
            missing = []
            for song in songs:
                files.append('{}{}{}'.format(song['path'], song['song_name'], song['extension']))

            pool = multiprocessing.Pool(THREADS)
            iterator = pool.imap_unordered(check_file, list(zip(files)))

            while True:
                try:
                    key, exists = next(iterator)
                except multiprocessing.TimeoutError:
                    continue
                except StopIteration:
                    break
                except Exception:
                    # Print traceback because we can't reraise it here
                    traceback.print_exc(file=sys.stdout)
                else:
                    if not exists:
                        print('missing {}'.format(key))
                        missing.append(key)
            pool.close()
            pool.join()

    with open('missings.csv', 'w') as txt_file:
        for file in missing:
            txt_file.write('prod-orcd-raw-assets,' + file + '\n')

    # paginator = s3_client.get_paginator('list_objects')
    # page_iterator = paginator.paginate(
    #     Bucket=BUCKET,
    #     Prefix=PREFIX,
    #     PaginationConfig=paginator_config
    # )
    # files = []
    # for page in page_iterator:
    #     items = page['Contents']  # One page of image files.
    #     for item in items:
    #         key = item['Key']
    #         files.append('dev-orcd-raw-asset,{}'.format(key))
    # with open('files2.csv', 'w') as txt_file:
    #     for file in files:
    #         txt_file.write(file + '\n')