ad0284585b
Goon — self-hosted aggregator for adult-content scene metadata. Indexes scenes from TPDB, StashDB, and 30+ public adult tube sites. Cross-source deduplication via perceptual hash + Levenshtein distance. FastAPI backend + APScheduler worker + React Native (Expo) mobile client. FOSS, ad-free, donation-funded. See README for details.
78 lines
2.4 KiB
Python
78 lines
2.4 KiB
Python
"""SxyLandScraper — direct HTML scrape sxyland.com search.
|
|
|
|
Search: `https://sxyland.com/?s=<query>` zwraca wyniki w formacie
|
|
`https://sxyland.com/<numeric_id>/<slug>/`. Filtrujemy linki bez numeric ID
|
|
(legal pages typu /18-u-s-c-2257/).
|
|
"""
|
|
from __future__ import annotations
|
|
|
|
import logging
|
|
import re
|
|
import urllib.parse
|
|
from collections.abc import Iterator
|
|
|
|
from app.connectors.base import RawPerformer, RawPlaybackSource, RawScene
|
|
from app.connectors.direct_scrapers.base import BaseDirectTubeScraper
|
|
from app.extractors import browser_get
|
|
|
|
log = logging.getLogger(__name__)
|
|
|
|
|
|
_SCENE_URL_RE = re.compile(r'href="(https://sxyland\.com/(\d+)/([^"/]+))/?"')
|
|
|
|
|
|
class SxyLandScraper(BaseDirectTubeScraper):
|
|
sitetag = "sxylandcom"
|
|
|
|
def search(
|
|
self,
|
|
query: str,
|
|
*,
|
|
page: int = 1,
|
|
limit: int | None = None,
|
|
) -> Iterator[RawScene]:
|
|
q = urllib.parse.quote_plus(query.strip())
|
|
url = f"https://sxyland.com/page/{page}/?s={q}"
|
|
try:
|
|
r = browser_get(url, timeout=30)
|
|
except Exception as e:
|
|
log.warning("sxyland search fetch failed: %s", e)
|
|
return
|
|
if r.status_code != 200:
|
|
return
|
|
|
|
query_tokens = {tok for tok in query.lower().split() if len(tok) >= 3}
|
|
|
|
seen: set[str] = set()
|
|
yielded = 0
|
|
for m in _SCENE_URL_RE.finditer(r.text):
|
|
scene_url = m.group(1) + "/"
|
|
slug = m.group(3)
|
|
if scene_url in seen:
|
|
continue
|
|
seen.add(scene_url)
|
|
|
|
slug_lower = slug.lower()
|
|
if query_tokens and not any(tok in slug_lower for tok in query_tokens):
|
|
continue
|
|
|
|
title = slug.replace("-", " ").strip()
|
|
|
|
yield RawScene(
|
|
external_id=f"sxylandcom:{scene_url}",
|
|
title=title,
|
|
url=scene_url,
|
|
playback_sources=[
|
|
RawPlaybackSource(origin="tube:sxylandcom", page_url=scene_url)
|
|
],
|
|
performers=[RawPerformer(name=query.strip())],
|
|
raw={
|
|
"source": "direct_scraper:sxyland",
|
|
"query": query,
|
|
"page": page,
|
|
"url": scene_url,
|
|
},
|
|
)
|
|
yielded += 1
|
|
if limit is not None and yielded >= limit:
|
|
return
|