"""Image Asset Migration from S3 Bucket to S3 Bucket.

Command line script to identify all images in an S3 bucket an copy to another.

Example Usage:

$ python image-migration-script.py
"""

# !/usr/bin/env python
# -*- coding: utf-8 -*-

import argparse
import datetime
import hashlib
import logging
import os
from os.path import dirname
from os.path import join
import re
import sys
import time
import uuid

import boto3
from dotenv import load_dotenv
import requests


dotenv_path = join(dirname(__file__), '.env')
load_dotenv(dotenv_path)

OWS_PRODUCT_URL = os.environ.get('OWS_PRODUCT_URL')
SOURCE_BUCKET_NAME = os.environ.get('SOURCE_BUCKET_NAME')
SOURCE_SUB_FOLDER = os.environ.get('SOURCE_SUB_FOLDER')
DESTINATION_BUCKET_NAME = os.environ.get('DESTINATION_BUCKET_NAME')
DESTINATION_SUB_FOLDER = os.environ.get('DESTINATION_SUB_FOLDER')


def fetch_product_id_by_upc(upc, correlation_id):
    """Return a product id by making a request to ows-product.

    Args:
        upc (str): upc for the product to be retrieved
        correlation_id (uuid): unique identifier for requests made
    Returns:
        product_id (str) or None: the product_id of the upc or None if error
    """
    request_url = '{}/product/upc/{}'.format(OWS_PRODUCT_URL, upc)
    headers = {
        'Content-Type': 'application/json',
        'Correlation-Id': correlation_id}
    product_response = requests.get(
        request_url,
        headers=headers)

    if product_response.status_code != 200:
        logging.warning('failed to find upc {}, status was {}.\n'.format(
            upc, product_response.status_code))
        return None

    return str(product_response.json().get('product_id'))


def save_asset_to_destination_bucket(
        s3_object, filename, product_id, file_extension, src_bucket_name,
        destination_bucket_name, destination_sub_folder):
    """Save an asset to the specified destination bucket.

    Args:
        s3_object (class): AWS S3 object to be copied to new destination
        filename (str): filename of asset object to be copied
        product_id (str): product_id of asset to be copied
        file_extension (str): file extension of original file
        src_bucket_name (str): name of the source bucket
        destination_bucket_name (str): name of the destination bucket
        destination_sub_folder (str): folder where copied image will be placed
    """
    image_type = re.search('(?<=\/)(\w+)', s3_object.key).group(0)
    md5_product_id = hashlib.md5(product_id.encode('utf-8')).hexdigest()
    destination_filename = '{}{}'.format(md5_product_id, file_extension)
    copy_source = {
        'Bucket': src_bucket_name,
        'Key': s3_object.key
    }
    s3.meta.client.copy(
            copy_source,
            destination_bucket_name,
            '{destination_sub_folder}/{image_type}/{filename}'.format(
                destination_sub_folder=destination_sub_folder,
                image_type=image_type,
                filename=destination_filename))


def step_through_images_from_s3_bucket_and_save(
        source_bucket, source_bucket_name, source_sub_folder,
        destination_bucket_name, destination_sub_folder, max_objects):
    """Loop through objects in an AWS S3 bucket and save to another bucket.

    Args:
        source_bucket (class): AWS S3 Bucket to copy objects from
        source_bucket_name (str): name of source bucket
        source_sub_folder (str): name of folder to search within
        destination_bucket_name (str): name of destination bucket
        destination_sub_folder (str): name of folder to save objects to
        max_objects (int): number of objects to copy

    """
    start_time = time.time()
    count = 0
    correlation_id = str(uuid.uuid4())
    logging.warning('Date of execution: {}'.format(datetime.datetime.now()))
    logging.warning('Now running migration with Correlation-Id: {}'.format(
        correlation_id))
    for obj in source_bucket.objects.filter(Prefix=source_sub_folder):
        # Look for S3 objects that end with .jpg or .jpeg
        file_match = re.search('[^\/]*\..+$', obj.key, re.IGNORECASE)
        filename = file_match.group(0)
        file_extension = re.search('\..+$', filename).group(0)
        if file_match:
            upc = re.search('\d+', filename).group(0)
            # Fetch the product_id based on the upc from ows-product
            product_id = fetch_product_id_by_upc(upc, correlation_id)
            # If no product is found, continue looping through the bucket
            if product_id is None:
                continue
            save_asset_to_destination_bucket(
                obj, filename, product_id, file_extension, source_bucket_name,
                destination_bucket_name, destination_sub_folder)
            count += 1
            if count == max_objects:
                break

    run_time = time.time() - start_time
    logging.warning('copied {} objects in {} seconds'.format(count, run_time))
    sys.exit(0)


if __name__ == '__main__':
    parser = argparse.ArgumentParser()
    parser.add_argument(
        'max_objects', nargs='?', default=0, type=int,
        help='the max number of objects copied')
    args = parser.parse_args()
    s3 = boto3.resource('s3')
    source_bucket = s3.Bucket(SOURCE_BUCKET_NAME)
    logging.basicConfig(
        filename='ows-product-calls.log', level=logging.WARNING)
    step_through_images_from_s3_bucket_and_save(
        source_bucket, SOURCE_BUCKET_NAME, SOURCE_SUB_FOLDER,
        DESTINATION_BUCKET_NAME, DESTINATION_SUB_FOLDER, args.max_objects)