"""
Takes the given csv bulk file and splits up the file details into separate columns
to make it easier to load into a relational db
"""
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 file_details_from_acr_filename(filename):
    parts = filename.split('_')
    assert len(parts) == 6
    return [
        parts[0],
        parts[1],
        parts[2],
        parts[3],
        parts[4].upper(),
        int(parts[5])
    ]


def main():
    input_filename = sys.argv[1]
    output_filename = sys.argv[2]
    tracks = dict()
    with open(input_filename) as read_file:
        reader = csv.reader(read_file, delimiter='\t', quoting=csv.QUOTE_NONE)
        next(reader)  # skip header
        for line in reader:
            acr_id = line[0]
            file_list_str = line[1]
            file_list = json.loads(file_list_str)
            assert type(file_list) == list
            for filename in file_list:
                sr_row = []
                sr_row.append(acr_id)
                file_details = file_details_from_acr_filename(filename)
                for file_detail in file_details:
                    sr_row.append(file_detail)
                tuid = int(sr_row[4])
                if tuid == 14908799:
                    continue
                if tuid not in tracks:
                    tracks[tuid] = sr_row
                else:
                    if tracks[tuid][-1] > sr_row[-1]:
                        tracks[tuid] = sr_row
    with open(output_filename, 'w', newline='') as write_file:
        writer = csv.writer(write_file, delimiter=',')
        for _, final_row in tracks.items():
            writer.writerow(final_row)


if __name__ == "__main__":
    main()