"""Tasks of the GfK Physical Ingestion Workflow."""

import datetime
import os
import tempfile
import zipfile

import boto3
from garcon import task
from garcon_contrib.dynamo_feed_status import garcon_feed_status
import paramiko

from feed_ingestion.flows.gfk_physical import config
from feed_ingestion.tasks import check_status


@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_obj = (
        datetime.datetime.strptime(date, '%Y-%m-%d').date()
        if date
        else datetime.date.today()
    )
    date = date_obj.strftime('%Y-%m-%d')

    if reload == 'True':
        activity.logger.info(
            f'Delete status for feed: {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(f'Feed already ingested for {date}')
        return {
            'stop': True,
            'message': f'{config.feed_name} is already ingested for {date}',
        }

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

    yy = date_obj.strftime('%y')
    mm = date_obj.strftime('%m')
    dd = date_obj.strftime('%d')

    filename = config.file_template.format(yy=yy, mm=mm, dd=dd)
    archive_dir = config.s3['archive_path'].format(date=date)
    processed_dir = config.s3['processed_path'].format(date=date)
    processed_filename = filename.replace('.ZIP', '.csv')
    s3_dir_path = f's3://{config.data_bucket}/{processed_dir}'

    return dict(
        feed_name=config.feed_name,
        date=date,
        filename=filename,
        archive_path=f'{archive_dir}{filename}',
        processed_path=f'{processed_dir}{processed_filename}',
        staging_raw_table=config.staging_raw_table,
        s3_dir_path=s3_dir_path,
    )


def _get_sftp_connection(sftp_creds):
    """Establish an SFTP connection using paramiko.

    Args:
        sftp_creds (dict): SFTP credentials with keys host, username, password,
            and optionally port.

    Returns:
        paramiko.SFTPClient: Connected SFTP client.
    """
    paramiko.sftp_file.SFTPFile.MAX_REQUEST_SIZE = 4194304
    transport = paramiko.Transport(
        (sftp_creds['host'], sftp_creds.get('port', 22))
    )
    transport.connect(
        username=sftp_creds['username'],
        password=sftp_creds['password'],
    )
    return paramiko.SFTPClient.from_transport(transport)


@task.decorate(timeout=1800)
@check_status()
def fetch_from_sftp(
    activity, feed_name, date, filename, archive_path, processed_path
):
    """Download ZIP from SFTP, extract CSV, and upload both to S3.

    Args:
        activity (ActivityWorker): The activity worker.
        feed_name (str): The name of the feed.
        date (str): Reporting date (YYYY-MM-DD).
        filename (str): Expected ZIP file name on the SFTP server.
        archive_path (str): S3 key for the raw ZIP archive.
        processed_path (str): S3 key for the extracted CSV file.
    """
    activity.logger.info(f'Connecting to SFTP to download {filename}')
    sftp = _get_sftp_connection(config.sftp)
    remote_dir = config.sftp.get('path', '/FEED').rstrip('/')
    remote_file_path = f'{remote_dir}/{filename}' if remote_dir else filename

    try:
        sftp.stat(remote_file_path)
    except FileNotFoundError:
        activity.logger.info(f'File not available on SFTP: {filename}')
        garcon_feed_status.set_overall_status(
            feed_name, date, garcon_feed_status.STATUS_NOT_AVAILABLE
        )
        sftp.close()
        return {'stop': True, 'message': f'Missing file {filename}'}

    local_zip_path = os.path.join(tempfile.gettempdir(), filename)
    sftp.get(remote_file_path, local_zip_path)
    sftp.close()
    activity.logger.info(f'Downloaded {filename} to {local_zip_path}')

    s3_client = boto3.client('s3')

    activity.logger.info(f'Uploading ZIP to S3 archive: {archive_path}')
    s3_client.upload_file(local_zip_path, config.data_bucket, archive_path)

    activity.logger.info(f'Extracting CSV from ZIP: {filename}')
    with zipfile.ZipFile(local_zip_path) as zf:
        csv_names = [n for n in zf.namelist() if n.upper().endswith('.CSV')]
        if not csv_names:
            raise ValueError(f'No CSV file found inside ZIP: {filename}')
        csv_data = zf.read(csv_names[0])

    os.remove(local_zip_path)

    activity.logger.info(f'Uploading extracted CSV to S3: {processed_path}')
    s3_client.put_object(
        Bucket=config.data_bucket,
        Key=processed_path,
        Body=csv_data,
    )

    garcon_feed_status.set_overall_status(
        feed_name, date, garcon_feed_status.STATUS_DOWNLOADED
    )
    activity.logger.info(f'Successfully processed {filename}')