"""The module contains activities' tasks."""

__all__ = [
    'bootstrap',
    'check_feed_status',
    'copy_blob_to_archive',
]

from datetime import date as date_module, datetime

from botocore.exceptions import ClientError
from dateutil.relativedelta import relativedelta
from garcon import task
from garcon_contrib.dynamo_feed_status import garcon_feed_status

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

_STOP_RESPONSE = {'stop': True}


@task.decorate(timeout=config.default_task_timeout)
def check_feed_status(activity, date, reload):
    """Check and reset feed status if it is needed.

    Args:
        activity (ActivityWorker): The Garcon activity worker.
        date (str): Reporting date (YYYY-MM-DD).
        reload (Optional[str]): If 'True' delete all feed statuses in DynamoDB.

    Returns:
        dict: {} or STOP_RESPONSE from check_ingested_status decorator
    """
    workflow_day = date_module.fromisoformat(date).day
    if workflow_day != config.workflow_month_day:
        raise ValueError(
            f'Current workflow month day is {workflow_day}, '
            f'but only {config.workflow_month_day} is permitted '
            f'not to initiate SCD type 2 process.'
        )

    if reload == 'True':  # noqa: F821
        activity.logger.info(
            'Delete status for feed: {} {} '.format(config.feed_name, date)
        )
        garcon_feed_status.delete_status(config.feed_name, date)
    else:
        overall_status = garcon_feed_status.get_overall_status(
            config.feed_name,
            date,
        )

        if overall_status == garcon_feed_status.STATUS_INGESTED:
            return _STOP_RESPONSE
    return {'feed_name': config.feed_name}


@task.decorate(timeout=config.default_task_timeout)
def bootstrap(
    activity,
    reload,
    date,
):
    """Bootstrap workflow by injecting initial context from config.

    Args:
        activity (ActivityWorker): The Garcon activity worker.
        date (str): Reporting date (YYYY-MM-DD).
        reload (str or None): If 'True' delete all feed statuses in DynamoDB.
    Returns:
        dict: Initial context for the workflow.
    """
    current_datetime = datetime.fromisoformat(date) if date else datetime.now()

    if reload == 'True':
        garcon_feed_status.delete_status(config.feed_name, date)

    previous_month_datetime = current_datetime - relativedelta(months=1)

    first_day_of_the_month_str = '{current_month_id}01'.format(
        current_month_id=previous_month_datetime.strftime('%Y%m'),
    )
    last_day_of_the_month_str = (
        previous_month_datetime + relativedelta(day=31)
    ).strftime('%Y%m%d')

    blob_file_name_wildcard = (
        f'.*{config.blob_name_prefix}'
        f'-{first_day_of_the_month_str}'
        f'-{last_day_of_the_month_str}'
        f'.*.{config.blob_name_extension}'
    )

    drop_blob_bucket = config.s3_blobs_prefixes['drop_blob_bucket']
    drop_blob_path = config.s3_blobs_prefixes['drop_blob_path']
    full_drop_blob_names = (
        f's3://{drop_blob_bucket}/{drop_blob_path}/{blob_file_name_wildcard}'
    )

    archive_blob_prefix = '{archive_blob_prefix}/{date}'.format(
        archive_blob_prefix=config.s3_blobs_prefixes['archive_blob_prefix'],
        date=date
    )

    return {
        'archive_blob_prefix': archive_blob_prefix,
        'date': current_datetime.date().isoformat(),
        'drop_bucket_name': drop_blob_bucket,
        'drop_blob_path': drop_blob_path,
        'drop_blob_wildcard': blob_file_name_wildcard,
        'feed_name': config.feed_name,
        'secrets_path': config.secrets_path,
        'staging_raw_load_kwargs': {
            'blob_path': full_drop_blob_names,
        },
        'temp_staging_raw_load_kwargs': {
            'file_pattern': blob_file_name_wildcard,
            'snowflake_error_limit': config.snowflake_error_limit,
        },
        'temp_staging_raw_name': config.temp_staging_table_name.format(
            current_datetime.strftime('%Y%m%d'),
        ),
        'snowflake_error_limit': config.snowflake_error_limit,
        'staging_raw_table': config.staging_raw_table_name,
    }


@task.decorate(timeout=36000)
@check_status(feed_name=config.feed_name, task_id='copy_blob_to_archive')
def copy_blob_to_archive(
    activity,
    date,
    archive_blob_prefix,
    drop_bucket_name,
    drop_blob_path,
    drop_blob_wildcard,
    reload,
):
    """Upload files to archive location.

    Args:
        activity (ActivityWorker): The activity worker.
        date (str): Reporting date (YYYY-MM-DD).
        reload (Optional[str]): if 'True' the flow run is reload of data.
        archive_blob_prefix (str): prefix for blobs to copy to.
        drop_bucket_name (str): name of the bucket with dropped objects.
        drop_blob_path (str): path to dropped objects.
        drop_blob_wildcard (str): dropped objects names wildcard.
    """
    try:
        if reload == 'True':
            s3_tasks.remove_files_from_path(
                activity=activity,
                path=archive_blob_prefix,
                return_deleted_files=False,
            )
        files = s3_tasks.source_files(
            activity=activity,
            s3_bucket=drop_bucket_name,
            s3_path=drop_blob_path,
            file_pattern=drop_blob_wildcard,
        )
        s3_tasks.copy_files(
            activity=activity,
            s3_archive_path=archive_blob_prefix,
            s3_download_path=f's3://{drop_bucket_name}/{drop_blob_path}',
            source_files_dict=files['source_files_dict'],
        )
    except ClientError as e:
        activity.logger.info(
            f'blobs {drop_blob_wildcard} '
            f'have not been copied to {archive_blob_prefix} '
            f'because of {e}.'
        )
        return _STOP_RESPONSE

    garcon_feed_status.set_overall_status(
        config.feed_name,
        date,
        garcon_feed_status.STATUS_DOWNLOADED,
    )