#!/usr/bin/env python3
import time, os, sys, logging, glob, datetime, json
from lib                import utils, FileQueue
from lib.sources        import WestAllisSource, WestAllisSalesHistorySource
from lib.parsers        import WestAllisTax, WestAllisSalesHistory
from lib.serializers    import WestAllisSalesHistorySerializer
from selenium.webdriver.common.desired_capabilities import DesiredCapabilities

LOG_FORMAT = '%(levelname)s:%(name)s:%(asctime)-15s: %(message)s'

def wa_ti_scrape(arguments):
    logging.basicConfig(format=LOG_FORMAT)
    logger = logging.getLogger('scrape')
    logger.setLevel(logging.INFO)

    # LOCAL or REMOTE
    DRIVER_TYPE                 = arguments.get('--driver-type')

    # required for REMOTE
    DRIVER_URL                  = arguments.get('--driver-url')

    # WEST_ALLIS, WEST_ALLIS_SALES_HISTORY
    ENDPOINT                    = arguments.get('--endpoint')

    TAX_IDS_FILEPATH            = arguments.get('--tax-ids-filepath')
    CURRENT_INDEX_FILEPATH      = arguments.get('--current-index-filepath')
    SECONDS_BETWEEN_REQUESTS    = int(arguments.get('--seconds-between-requests'))
    OUTPUT_DIRECTORY            = arguments.get('--output-directory')
    SOURCE_TYPES                = {
        'WEST_ALLIS': WestAllisSource,
        'WEST_ALLIS_SALES_HISTORY': WestAllisSalesHistorySource
    }

    driver = utils.get_driver(DRIVER_TYPE, {
        'command_executor': DRIVER_URL,
        'desired_capabilities': DesiredCapabilities.FIREFOX
    })

    q = FileQueue(TAX_IDS_FILEPATH, CURRENT_INDEX_FILEPATH)
    create_source = SOURCE_TYPES[ENDPOINT]
    source = create_source(driver)

    while True:
        next_id = q.peek()
        content = source.fetch(next_id)
        archive_path = '{}/{}_{}.html'.format(OUTPUT_DIRECTORY, int(time.time()), next_id)
        utils.write_archive(archive_path, content)
        logger.info('Recieved content for {} and saving to {}'.format(next_id, archive_path))
        q.pop()
        time.sleep(SECONDS_BETWEEN_REQUESTS)

def wa_ti_json(arguments):
    logging.basicConfig(format=LOG_FORMAT)
    logger = logging.getLogger('json')
    logger.setLevel(logging.INFO)

    ENDPOINT                    = arguments.get('--endpoint')
    HTML_DIR                    = arguments.get('--html-directory')
    SOURCE_TYPES                = {
        'WEST_ALLIS': WestAllisTax,
        'WEST_ALLIS_SALES_HISTORY': WestAllisSalesHistory
    }

    create_parser = SOURCE_TYPES[ENDPOINT]

    files = glob.glob('{}/*.html'.format(HTML_DIR))

    for fp in files:
        contents = None
        with open(fp, 'r') as f:
            contents = f.read()
            f.close()

        if contents == None:
            raise Exception('Contents cannot be none for {}'.format(fp))

        filename = fp.split('/')[-1]
        filename = filename.split('.')[0]


        timestamp = int(filename.split('_')[0])
        created_on = datetime.datetime.fromtimestamp(timestamp)
        parsed = create_parser(contents).parse(created_on=created_on)

        write_fp = '{}/{}.json'.format(HTML_DIR, filename)
        serialized = json.dumps(parsed)
        logger.info('Parsed content for {} and saving to {}'.format(filename, write_fp))

        with open(write_fp, 'w') as f:
            f.write(serialized)
            f.close()

def wa_ti_csv(arguments):
    logging.basicConfig(format=LOG_FORMAT)
    logger = logging.getLogger('json')
    logger.setLevel(logging.INFO)

    ENDPOINT                    = arguments.get('--endpoint')
    JSON_DIR                    = arguments.get('--json-directory')
    SOURCE_TYPES                = {
        'WEST_ALLIS_SALES_HISTORY': WestAllisSalesHistorySerializer
    }

    create_serializer = SOURCE_TYPES[ENDPOINT]
    serializer = create_serializer()
    files = glob.glob('{}/*.json'.format(JSON_DIR))

    csv_data = ''

    for fp in files:
        contents = None
        with open(fp, 'r') as f:
            contents = f.read()
            f.close()

        if contents == None:
            raise Exception('Contents cannot be none for {}'.format(fp))

        filename = fp.split('/')[-1]
        filename = filename.split('.')[0]
        tax_id = filename.split('_')[1]

        csv_data += serializer.serialize(tax_id, contents)


    write_fp = '{}/sales_history.csv'.format(JSON_DIR)
    logger.info('Serialized all json content and saving to {}'.format(write_fp))

    with open(write_fp, 'w') as f:
        f.write(csv_data)
        f.close()

operations = {
    'scrape': wa_ti_scrape,
    'json': wa_ti_json,
    'csv': wa_ti_csv
}

def run():
    sys.argv.pop(0) # pops script name
    operation = sys.argv.pop(0)
    arguments = {}

    while len(sys.argv) > 0:
        arg = sys.argv.pop()
        s = arg.split('=')
        topic = s[0]
        value = s[1]
        arguments[topic] = value
    callback = operations[operation](arguments)

if __name__ == '__main__':
    run()