188 lines
4.5 KiB
Python
188 lines
4.5 KiB
Python
import json
|
|
import re
|
|
from functools import partial
|
|
from urllib.parse import urljoin
|
|
|
|
import httpx
|
|
from playwright.async_api import async_playwright
|
|
from selectolax.parser import HTMLParser
|
|
|
|
from .utils import Cache, Time, get_logger, leagues, network
|
|
|
|
log = get_logger(__name__)
|
|
|
|
urls: dict[str, dict[str, str | float]] = {}
|
|
|
|
CACHE_FILE = Cache("vuen.json", exp=3_600)
|
|
|
|
API_FILE = Cache("vuen-html.json", exp=28_800)
|
|
|
|
BASE_URL = "https://vuen.link"
|
|
|
|
|
|
async def refresh_api_cache(
|
|
client: httpx.AsyncClient,
|
|
url: str,
|
|
) -> dict[str, list[dict]]:
|
|
log.info("Refreshing API cache")
|
|
|
|
data = {}
|
|
|
|
try:
|
|
r = await client.get(url)
|
|
r.raise_for_status()
|
|
except Exception as e:
|
|
log.error(f'Failed to fetch "{url}": {e}')
|
|
|
|
return data
|
|
|
|
soup = HTMLParser(r.text)
|
|
|
|
for script in soup.css("script"):
|
|
if not script:
|
|
continue
|
|
|
|
content = script.text(strip=True)
|
|
|
|
if not (
|
|
match := re.search(
|
|
r"window\.matches\s*=\s*JSON\.parse\(`(.*?)`\)",
|
|
content,
|
|
re.DOTALL,
|
|
)
|
|
):
|
|
continue
|
|
|
|
data["matches"] = json.loads(match[1])
|
|
|
|
data["timestamp"] = Time.now().timestamp()
|
|
|
|
break
|
|
|
|
return data
|
|
|
|
|
|
async def get_events(
|
|
client: httpx.AsyncClient,
|
|
cached_keys: set[str],
|
|
) -> list[dict[str, str]]:
|
|
if not (api_data := API_FILE.load(per_entry=False)):
|
|
api_data = await refresh_api_cache(client, BASE_URL)
|
|
|
|
API_FILE.write(api_data)
|
|
|
|
events = []
|
|
|
|
now = Time.clean(Time.now())
|
|
start_dt = now.delta(minutes=-30)
|
|
end_dt = now.delta(minutes=30)
|
|
|
|
for match in api_data.get("matches", []):
|
|
if not (ts := match.get("startTimestamp")):
|
|
continue
|
|
|
|
start_ts = int(f"{ts}"[:-3])
|
|
|
|
event_dt = Time.from_ts(start_ts)
|
|
|
|
if not start_dt <= event_dt <= end_dt:
|
|
continue
|
|
|
|
event_name = match["matchstr"]
|
|
|
|
sport = match["league"]
|
|
|
|
channels: list[dict[str, str | list[str]]] = match.get("channels", [])
|
|
|
|
event_link = None
|
|
|
|
if not channels:
|
|
event_link = urljoin(BASE_URL, match["slug"])
|
|
|
|
for channel in channels:
|
|
event_link = (channel.get("oldLinks") or channel.get("links") or [None])[0]
|
|
|
|
if event_link:
|
|
break
|
|
|
|
if not event_link:
|
|
continue
|
|
|
|
key = f"[{sport}] {event_name} (VUEN)"
|
|
|
|
if cached_keys & {key}:
|
|
continue
|
|
|
|
events.append(
|
|
{
|
|
"sport": sport,
|
|
"event": event_name,
|
|
"link": event_link,
|
|
"timestamp": now.timestamp(),
|
|
}
|
|
)
|
|
|
|
return events
|
|
|
|
|
|
async def scrape(client: httpx.AsyncClient) -> None:
|
|
cached_urls = CACHE_FILE.load()
|
|
cached_count = len(cached_urls)
|
|
urls.update(cached_urls)
|
|
|
|
log.info(f"Loaded {cached_count} event(s) from cache")
|
|
|
|
log.info(f'Scraping from "{BASE_URL}"')
|
|
|
|
events = await get_events(client, set(cached_urls.keys()))
|
|
|
|
log.info(f"Processing {len(events)} new URL(s)")
|
|
|
|
if events:
|
|
async with async_playwright() as p:
|
|
browser, context = await network.browser(p)
|
|
|
|
for i, ev in enumerate(events, start=1):
|
|
handler = partial(
|
|
network.process_event,
|
|
url=ev["link"],
|
|
url_num=i,
|
|
context=context,
|
|
log=log,
|
|
)
|
|
|
|
url = await network.safe_process(
|
|
handler,
|
|
url_num=i,
|
|
log=log,
|
|
)
|
|
|
|
if url:
|
|
sport, event, ts = (
|
|
ev["sport"],
|
|
ev["event"],
|
|
ev["timestamp"],
|
|
)
|
|
|
|
key = f"[{sport}] {event} (VUEN)"
|
|
|
|
tvg_id, logo = leagues.get_tvg_info(sport, event)
|
|
|
|
entry = {
|
|
"url": url,
|
|
"logo": logo,
|
|
"base": "https://vividmosaica.com/",
|
|
"timestamp": ts,
|
|
"id": tvg_id or "Live.Event.us",
|
|
}
|
|
|
|
urls[key] = cached_urls[key] = entry
|
|
|
|
await browser.close()
|
|
|
|
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(cached_urls)
|