"""
Takes the given csv bulk data and transforms it to json for processing by the
existing DB PR https://github.com/theorchard/database/pull/9252
"""
import csv
import json
import sys


# added this to avoid an error when processing a large amount of data
field_size_limit = sys.maxsize
while True:
    try:
        csv.field_size_limit(field_size_limit)
        break
    except OverflowError:
        field_size_limit = int(field_size_limit / 10)


def read_file(filename):
    with open(filename) as f:
        reader = csv.reader(f, delimiter='\t', quoting=csv.QUOTE_NONE)
        next(reader)
        for line in reader:
            yield line


def file_details_from_acr_filename(filename):
    parts = filename.split('_')
    assert len(parts) == 6
    return {
        'upc': parts[0],
        'cd': parts[1],
        'track_id': parts[2],
        'tuid': parts[3],
        'isrc': parts[4],
        'identifier_timestamp': int(parts[5])
    }


def file_details_to_dd_filename(details):
    return f"{details['upc']}_{details['cd']}_{details['track_id']}"


def main():
    input_filename = sys.argv[1]
    output_filename = sys.argv[2]

    tuid_to_latest_delivery = {}

    # Find the latest delivery time for each tuid
    for input in read_file(input_filename):
        file_list_str = input[1]
        file_list = file_list_str.strip(']""[').split('","')
        for asset in file_list:
            details = file_details_from_acr_filename(asset)

            current_timestamp = tuid_to_latest_delivery.get(details['tuid'])
            if current_timestamp is not None:
                if current_timestamp >= details['identifier_timestamp']:
                    continue
                assert current_timestamp != details['identifier_timestamp']

            tuid_to_latest_delivery[details['tuid']] = details['identifier_timestamp']

    # Write the file out with only the latest deliveries, and with DD-style filenames
    with open(output_filename, 'w') as output_file:
        for input in read_file(input_filename):
            file_list_str = input[1]
            file_list = file_list_str.strip(']""[').split('","')
            output = {
                'acr_id': input[0],
                'assets': [],
            }
            for asset in file_list:
                details = file_details_from_acr_filename(asset)
                if tuid_to_latest_delivery[details['tuid']] != details['identifier_timestamp']:
                    continue
                output['assets'].append({
                    'filename': file_details_to_dd_filename(details),
                    'tuid': details['tuid'],
                    #'md5': asset['md5'],
                })
            output_file.write(json.dumps(output) + '\n')


if __name__ == "__main__":
    main()