import json
from pathlib import Path
from urllib.parse import unquote, urljoin

import httpx
from selectolax.parser import HTMLParser

from .utils import LOGOS, get_base, get_logger, load_cache, now, safe_process_event

log = get_logger(__name__)

urls: dict[str, dict[str, str]] = {}

MIRRORS = [
    "https://fstv.zip",
    "https://fstv.space",
    "https://fstv.online",
    "https://fstv.us",
]

CACHE_FILE = Path(__file__).parent / "caches" / "fstv.json"


async def get_events(
    client: httpx.AsyncClient,
    base_url: str,
    cached_hrefs: set[str],
) -> list[dict[str, str]]:
    log.info(f'Scraping from "{base_url}"')

    try:
        r = await client.get(base_url)
        r.raise_for_status()
    except Exception as e:
        log.error(f'Failed to fetch "{base_url}"\n{e}')

        return []

    soup = HTMLParser(r.text)

    events = []

    for wrpr in soup.css("div.fixtures-live-wrapper"):
        for games in wrpr.css(".match-table-item"):

            league_name = games.css_first(".league-info a.league-name")

            league_match = games.css_first(".common-table-row a[href*='/match/']")

            if league_name and league_match:
                full_text = league_name.text(strip=True)

                if "]" in full_text:
                    event_name = full_text.split("]", 1)[1].strip()
                else:
                    event_name = full_text

                href = league_match.attributes.get("href")

                link = urljoin(base_url, href)

                if cached_hrefs & {href}:
                    continue

                events.append(
                    {
                        "sport": event_name,
                        "link": link,
                        "logo": LOGOS.get(
                            event_name,
                            "https://i.gyazo.com/ec27417a9644ae517196494afa72d2b9.png",
                        ),
                        "href": href,
                    }
                )

    return events


async def process_event(
    client: httpx.AsyncClient,
    url: str,
    url_num: int,
) -> tuple[str, str]:
    try:
        r = await client.get(url)
        r.raise_for_status()
    except Exception as e:
        log.error(f'URL {url_num}) Failed to fetch "{url}"\n{e}')

        return "", ""

    soup = HTMLParser(r.text)

    if category_links := soup.css(".common-list-category .category-item a"):
        match_name = category_links[-1].text(strip=True)
    else:
        match_name = None

    if not match_name or match_name.lower() == "vs":
        if og_title := soup.css_first("meta[property='og:title']"):
            match_name = (
                og_title.attributes.get("content", "").split(" start on")[0].strip()
            )

    if ifr := soup.css_first("iframe"):
        if src := ifr.attributes.get("src", ""):
            log.info(f"URL {url_num}) Captured M3U8")
            return match_name, unquote(src).split("link=")[-1]

    log.info(f"URL {url_num}) No M3U8 found")
    return "", ""


async def main(client: httpx.AsyncClient) -> None:
    cached_urls = load_cache(CACHE_FILE, exp=14400)
    cached_hrefs = {entry["href"] for entry in cached_urls.values()}
    cached_count = len(cached_urls)
    urls.update(cached_urls)

    log.info(f"Collected {cached_count} event(s) from cache")

    if not (base_url := await get_base(client, MIRRORS)):
        log.warning("No working FSTV mirrors")
        return

    events = await get_events(
        client,
        base_url,
        cached_hrefs,
    )

    log.info(f"Processing {len(events)} new URL(s)")

    for i, ev in enumerate(events, start=1):
        match_name, url = await safe_process_event(
            lambda: process_event(
                client,
                ev["link"],
                url_num=i,
            ),
            url_num=i,
            log=log,
        )

        if url:
            key = (
                f"[{ev['sport']}] {match_name} (FSTV)"
                if match_name
                else f"[{ev['sport']}] (FSTV)"
            )

            entry = {
                "url": url,
                "logo": ev["logo"],
                "base": base_url,
                "timestamp": now.timestamp(),
                "href": ev["href"],
            }

            urls[key] = cached_urls[key] = entry

    if new_count := len(cached_urls) - cached_count:
        log.info(f"Collected and cached {new_count} new event(s)")
    else:
        log.info("No new events found")

    CACHE_FILE.write_text(json.dumps(cached_urls, indent=2), encoding="utf-8")