179 lines
4.6 KiB
Python
179 lines
4.6 KiB
Python
import asyncio
|
|
from functools import partial
|
|
|
|
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("webcast.json", exp=3_600)
|
|
|
|
HTML_CACHE = Cache("webcast-html.json", exp=28_800)
|
|
|
|
BASE_URLS = {"NFL": "https://nflwebcast.com", "NHL": "https://slapstreams.com"}
|
|
|
|
TAG = "WEBCST"
|
|
|
|
|
|
def fix_event(s: str) -> str:
|
|
return " vs ".join(s.split("@"))
|
|
|
|
|
|
async def refresh_html_cache(
|
|
client: httpx.AsyncClient, url: str
|
|
) -> dict[str, dict[str, str | float]]:
|
|
log.info("Refreshing HTML cache")
|
|
|
|
try:
|
|
r = await client.get(url)
|
|
r.raise_for_status()
|
|
except Exception as e:
|
|
log.error(f'Failed to fetch "{url}": {e}')
|
|
|
|
return {}
|
|
|
|
now = Time.now()
|
|
|
|
soup = HTMLParser(r.content)
|
|
|
|
events = {}
|
|
|
|
title = soup.css_first("title").text(strip=True)
|
|
|
|
sport = "NFL" if "NFL" in title else "NHL"
|
|
|
|
date_text = now.strftime("%B %d, %Y")
|
|
|
|
if date_row := soup.css_first("tr.mdatetitle"):
|
|
if mtdate_span := date_row.css_first("span.mtdate"):
|
|
date_text = mtdate_span.text(strip=True)
|
|
|
|
for row in soup.css("tr.singele_match_date"):
|
|
if not (time_node := row.css_first("td.matchtime")):
|
|
continue
|
|
|
|
time = time_node.text(strip=True)
|
|
|
|
if not (vs_node := row.css_first("td.teamvs a")):
|
|
continue
|
|
|
|
event_name = vs_node.text(strip=True)
|
|
|
|
for span in vs_node.css("span.mtdate"):
|
|
date = span.text(strip=True)
|
|
|
|
event_name = event_name.replace(date, "").strip()
|
|
|
|
if not (href := vs_node.attributes.get("href")):
|
|
continue
|
|
|
|
event_dt = Time.from_str(f"{date_text} {time} PM", timezone="EST")
|
|
|
|
event = fix_event(event_name)
|
|
|
|
key = f"[{sport}] {event} ({TAG})"
|
|
|
|
events[key] = {
|
|
"sport": sport,
|
|
"event": event,
|
|
"link": href,
|
|
"event_ts": event_dt.timestamp(),
|
|
"timestamp": now.timestamp(),
|
|
}
|
|
|
|
return events
|
|
|
|
|
|
async def get_events(
|
|
client: httpx.AsyncClient, cached_keys: set[str]
|
|
) -> list[dict[str, str]]:
|
|
now = Time.clean(Time.now())
|
|
|
|
if not (events := HTML_CACHE.load()):
|
|
tasks = [refresh_html_cache(client, url) for url in BASE_URLS.values()]
|
|
|
|
results = await asyncio.gather(*tasks)
|
|
|
|
events = {k: v for data in results for k, v in data.items()}
|
|
|
|
HTML_CACHE.write(events)
|
|
|
|
live = []
|
|
|
|
start_ts = now.delta(minutes=-30).timestamp()
|
|
end_ts = now.delta(minutes=30).timestamp()
|
|
|
|
for k, v in events.items():
|
|
if cached_keys & {k}:
|
|
continue
|
|
|
|
if not start_ts <= v["event_ts"] <= end_ts:
|
|
continue
|
|
|
|
live.append({**v})
|
|
|
|
return live
|
|
|
|
|
|
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 "{' & '.join(BASE_URLS.values())}"')
|
|
|
|
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["event_ts"]
|
|
|
|
key = f"[{sport}] {event} ({TAG})"
|
|
|
|
tvg_id, logo = leagues.get_tvg_info(sport, event)
|
|
|
|
entry = {
|
|
"url": url,
|
|
"logo": logo,
|
|
"base": BASE_URLS[sport],
|
|
"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)
|