"""Ingest Logic."""

import re
from typing import Any

import boto3
from sqlalchemy import func

from video import config
from video.constants.ingest import DEFAULT_BATCH_SIZE, INGEST_INCOMING_PATH
from video.logic import job as job_logic
from video.models.sql.classes import product_video


def ingest(data: dict[str, Any]) -> list[dict[str, Any]]:
    """Check if there are files to ingest and if so setup the workflows.

    Args:
        data (dict): Parameters to configure the ingest.
        data['batch_size'] (int): The number of products to ingest.

    Returns:
        list: the files that will be ingested.
    """
    batch_size = data.get("batch_size", DEFAULT_BATCH_SIZE)

    files = _filter_files_with_product(_get_files_to_ingest())
    files = files[:batch_size]

    for file in files:
        jobs = job_logic.setup_ingest_from_s3_workflow(file)
        file["latest_pipeline_run_id"] = jobs[0]["parent_id"]
        product_video.upsert({**file, "migrated_asset_at": func.now()})

    return files


def _get_files_to_ingest() -> list[dict[str, Any]]:
    """Get the list of files to ingest from S3.

    Returns:
        [dict]: containing the list of files to ingest.
    """
    result = boto3.client("s3").list_objects_v2(
        Bucket=config.VIDEO_BUCKET_NAME, Prefix=INGEST_INCOMING_PATH
    )

    files = []

    for item in result.get("Contents", []):
        key = item["Key"]
        found_filename = re.search(r"[^/]+\.mov$", key)

        if found_filename:
            files.append({"path": key, "filename": found_filename.group(0)})

    return files


def _filter_files_with_product(files: list[dict[str, Any]]) -> list[dict[str, Any]]:
    """Filter the files with an existing product_video from a list of files.

    Args:
        files ([dict]): The list of files.

    Returns:
        [dict]: containing the filtered list.
    """
    files_with_product = []

    for file in files:
        product_response = product_video.get_by_ingest_filename(file["filename"])

        # Filter files that don't have a row in product_video
        if product_response:
            # Filter products that have already been through the pipeline
            if not product_response.get("latest_pipeline_run_id"):
                file["release_id"] = product_response["release_id"]
                files_with_product.append(file)

    return files_with_product