"""Tasks of the Sme latam Ingestion Workflow."""
import datetime
import os

from garcon import task
from garcon_contrib.dynamo_feed_status import garcon_feed_status
import pandas as pd

from feed_ingestion.flows.sme_latam import config
from feed_ingestion.tasks import check_status
from feed_ingestion.tasks import s3_tasks

STOP_RESPONSE = {'stop': True}


@task.decorate(timeout=700)
def bootstrap(activity, date, reload):
    """Bootstrap workflow by getting the correct configurations.

    Args:
        activity (ActivityWorker): The activity worker.
        date (str): Reporting date (YYYY-MM-DD).
        reload (str or None): If 'True' reload status.

    Returns:
        dict: Context.
    """
    date = datetime.datetime.strptime(
        date, '%Y-%m-%d').date() if date else datetime.date.today()

    # date for DynamoDB status
    date_obj = date + datetime.timedelta(days=-date.weekday())
    date = date_obj.strftime('%Y-%m-%d')

    if reload == 'True':
        activity.logger.info('Delete status for feed: {} {} '.format(
            config.feed_name, date))
        garcon_feed_status.delete_status(config.feed_name, date)

    elif garcon_feed_status.get_overall_status(
            config.feed_name, date) == garcon_feed_status.STATUS_INGESTED:
        activity.logger.info('Feed already ingested for {}'.format(date))
        return {
            'stop': True,
            'message':
                '{feed_name} is already ingested for {date:%Y-%m-%d}'.format(
                    feed_name=config.feed_name,
                    date=date_obj)}

    activity.logger.info('Bootstrap flow for {}'.format(date))

    week_number = date_obj.isocalendar()[1]
    week_year = date_obj.isocalendar()[0]

    activity.logger.info(f'Week number {week_number} Year: {week_year}')

    format_args = dict(
        week_number=week_number,
        week_year=week_year
    )

    archive_path = config.s3['archive_path'].format(**format_args)
    drop_path = config.s3['drop_path'].format(**format_args)

    filename = config.file_template.format(**format_args)
    processed_path = config.s3['processed_path'].format(**format_args)
    processed_filename = filename.replace('.xlsx', '.csv')

    return dict(
        feed_name=config.feed_name,
        date=date,
        secrets_path=config.secrets_path,
        archive_path=f'{archive_path}{filename}',
        drop_path=f'{drop_path}{filename}',
        processed_path=f'{processed_path}{processed_filename}',
        staging_raw_table=config.staging_raw_table)


@task.decorate(timeout=7000)
@check_status()
def fetch_from_drop_location(
        activity, feed_name, date, drop_path, archive_path):
    """Download a feed file from the drop location into the archive folder.

    Args:
        activity (ActivityWorker): The activity worker.
        feed_name (str): The name of the feed.
        date (str): Reporting date (YYYY-MM-DD).
        drop_path (str): Source S3 path to the archive location.
        archive_path (str): Destination S3 path to the archive location.
    """
    # copy zip archive
    result = s3_tasks.copy_file(
        activity,
        source_bucket_name=config.drop_bucket,
        source_key_name=drop_path,
        destination_bucket_name=config.data_bucket,
        destination_key_name=archive_path,
        replace=True)

    # file is not downloaded
    if not result.get(os.path.basename(drop_path)):
        garcon_feed_status.set_overall_status(
            feed_name, date, garcon_feed_status.STATUS_NOT_AVAILABLE)
        return {'stop': True, 'message': f'Missing file {drop_path}'}
    else:
        garcon_feed_status.set_overall_status(
            feed_name, date, garcon_feed_status.STATUS_DOWNLOADED)
        return result


@task.decorate(timeout=1000)
def process_drop_files(
        activity, feed_name, date, archive_path, processed_path):
    """Process xlsx file to csv.

    Args:
        activity (ActivityWorker): The activity worker.
        feed_name (str): The name of the feed.
        date (str): Reporting date (YYYY-MM-DD).
        archive_path (str): Destination S3 path to the archive location.
        processed_path (str): S3 path to save csv files.
    Returns:
        dict: s3_dir_path & source_files_dict for loading data into snowflake.
    """
    df = pd.read_excel(
        f's3://{config.data_bucket}/{archive_path}',
        engine='openpyxl', skiprows=4)
    activity.logger.info(f'Loading initial file {archive_path}')

    s3_processed_path = f's3://{config.data_bucket}/{processed_path}'
    df.to_csv(s3_processed_path, index=None, header=False)
    activity.logger.info('xlsx file has been processed to csv.')

    source_files_dict = {'files': [{
        'file_name': os.path.basename(s3_processed_path),
        'file_size': 1,
        'found': True
    }]}

    return dict(
        s3_dir_path=s3_processed_path,
        source_files_dict=source_files_dict
    )