# scraper/tasks/pipeline.py

from celery import chain
from logbus.publisher import log

from scraper.tasks.download_tasks import download_chapter
from scraper.tasks.parse_tasks import parse_chapter
from scraper.tasks.save_tasks import save_chapter


def build_chapter_pipeline(chapter_number: int, chapter_url: str, base_path: str):
    """
    Build a Celery pipeline for a single chapter:
        download -> parse -> save
    """

    log(f"[PIPELINE] Building chain for chapter {chapter_number}")

    # Important: download returns dict {chapter, url, html}
    # parse accepts html + chapter_url (via s())
    # save accepts chapter_number, text, base_path
    workflow = chain(
        download_chapter.s(chapter_number, chapter_url),
        parse_chapter.s(),                # takes previous result dict
        save_chapter.s(base_path=base_path)
    )

    return workflow