"""Config for Amazon DataPulse Ingestion Workflow."""

import os

from feed_ingestion.conf.config import ENV, SF_PARAMS

feed_name = 'amazon_datapulse'

feed_version = '1.0'
secrets_path = 'amazon_datapulse'

DATE_FORMAT = '%Y-%m-%d'

archive_bucket = os.environ.get('FEED_INGESTION_DATA_BUCKET', 'dev-cucumbers')
archive_s3_full_path_template = 'AmazonDataPulse/{report_name}/full/{date}/'
archive_s3_partitioned_path_template = (
    'AmazonDataPulse/{report_name}/partitioned/{partition}/'
)


def get_staging_raw_database_schema(report):
    """Get staging raw database, schema, and table name for a report."""
    report_config = reports[report]
    if ENV in {'prod', 'qa'}:
        database = report_config['snowflake_database']
        schema = report_config['snowflake_schema']
        if ENV == 'qa':
            database = f'qa_{database}'
    else:
        # for dev and others fallback to defaults specified in ENV
        database = SF_PARAMS['db']
        schema = SF_PARAMS['schema']
    return database, schema, report_config['staging_raw_table']


reports = {
    'fraud_report': {
        'athena_source_table': 'fraud_report',
        'snowflake_database': 'fraud_reporting',
        'snowflake_schema': 'amazon',
        'staging_raw_table': 'staging_raw_amazon_fraud_report',
        'partitions': {
            'cadence': str,
            'region': str,
            'report_date': str,
            'entity_name': str,
            'service': str,
            'music_territory': str,
        },
    },
    'daily_play_events': {
        'athena_source_table': 'daily_play_events',
        'snowflake_database': 'consumer_reporting',
        'snowflake_schema': 'amazon',
        'staging_raw_table': 'staging_raw_amazon_daily_play_events',
        'partitions': {
            'service': str,
            'region': str,
            'marketplace': str,
            'territory': str,
            'year': str,
            'month': str,
            'day': str,
        },
    },
    'daily_customer_metadata': {
        'athena_source_table': 'daily_customer_metadata',
        'snowflake_database': 'consumer_reporting',
        'snowflake_schema': 'amazon',
        'staging_raw_table': 'staging_raw_amazon_daily_customer_metadata',
        'partitions': {
            'region': str,
            'marketplace': str,
            'territory': str,
            'year': str,
            'month': str,
            'day': str,
        },
    },
    'daily_playlist_metadata': {
        'athena_source_table': 'daily_playlist_metadata',
        'snowflake_database': 'consumer_reporting',
        'snowflake_schema': 'amazon',
        'staging_raw_table': 'staging_raw_amazon_daily_playlist_metadata',
        'partitions': {
            'region': str,
            'marketplace': str,
            'territory': str,
            'year': str,
            'month': str,
            'day': str,
        },
    },
}

athena_source_database = os.environ.get(
    'AMAZON_FRAUD_REPORT_ATHENA_SOURCE_DATABASE',
    'amazon_music_reporting_datalake_link',
)
athena_workgroup = os.environ.get(
    'AMAZON_FRAUD_REPORT_ATHENA_WORKGROUP', 'primary'
)


def get_contextified_feed_name(report_name, partition=None):
    """Get feed name with report name context."""
    assert report_name in reports
    parts = [feed_name, report_name]
    if partition:
        # lowercase all partition keys to match report config
        partition = {k.lower(): str(v).lower() for k, v in partition.items()}
        report_config = reports[report_name]
        parts.extend(
            partition[key] for key in report_config['partitions'].keys()
        )
    contextified_feed_name = '_'.join(parts)
    contextified_feed_name = ''.join(
        c if c.isalnum() else '_' for c in contextified_feed_name
    )
    return contextified_feed_name


def get_partition_date(partition):
    """Return YYYY-MM-DD date string derived from partition keys.

    Supports two partition shapes:
      * `report_date` — used as-is.
      * `year` / `month` / `day` — combined into YYYY-MM-DD with zero-padding.
    """
    if 'report_date' in partition:
        return partition['report_date']
    if all(k in partition for k in ('year', 'month', 'day')):
        return '{}-{}-{}'.format(
            partition['year'],
            partition['month'].zfill(2),
            partition['day'].zfill(2),
        )
    return None