/ linki, dooplay
# filtruje cast w tej sekcji. Jaccard może łapać dubel ale dedup robimy w
# resolverze (po performer_id).
performers = [
RawPerformer(
external_id=f"{source_name}:{m.group(1)}",
name=_decode_html(m.group(2).strip()),
)
for m in _PORNSTAR_RE.finditer(html)
]
tags = [
RawTag(
external_id=f"{source_name}:{m.group(1)}",
name=_decode_html(m.group(2).strip()),
slug=m.group(1),
)
for m in _GENRE_TAG_RE.finditer(html)
]
if page_url is None:
page_url = f"{base_url}/movies/{slug}/"
# Playback sources: każdy host (Doodstream/Lulu/RPM/...) jako osobny entry.
# Dedup po href żeby ten sam host nie wpadł 2x. Raw landing page (origin=
# source_name, bez :host) appendujemy TYLKO gdy nie ma żadnych sub-hosters —
# inaczej myli usera (otwiera WebView z reklamami zamiast video; bug-report
# 2026-05-16: "mangoporn przekierowuje do strony, reklama full screen").
playback_sources: list[RawPlaybackSource] = []
seen_hrefs: set[str] = set()
# Hostery file-download (non-streamable) + malware. Mobile player nie potrafi
# ich odtworzyć — rapidgator/nitroflare/frdl serwują .zip/.rar/.mp4 do download
# (premium login required), streamtape ma malware drive-by .reg. Skipujemy
# przy ingest żeby nie zaśmiecać UI martwym contentem (bug-report 2026-05-18).
SKIP_HOSTERS = {"rapidgator", "nitroflare", "nitro", "frdl", "streamtape"}
def _emit_host_entry(href: str, source: str | None) -> None:
href = href.strip()
if not href or href in seen_hrefs:
return
seen_hrefs.add(href)
try:
from urllib.parse import urlparse
host = urlparse(href).hostname or "unknown"
host_short = host.split(".")[-2] if host.count(".") >= 1 else host
except Exception:
host_short = "unknown"
if host_short.lower() in SKIP_HOSTERS:
return
playback_sources.append(
RawPlaybackSource(
origin=f"{source_name}:{host_short}",
page_url=href,
embed_url=source or href,
thumbnail_url=poster_url,
duration_sec=duration_sec,
)
)
# Stary `` markup (mangoporn).
for m in _PLAYER_OPTION_RE.finditer(html):
_emit_host_entry(m.group("href") or "", (m.group("source") or "").strip() or None)
# Nowy `` markup (pandamovies od ~2026-04 + nowe
# streamporn instances). data-fl-source jest opcjonalny — capturujemy CAŁY
# opening tag w group(1), data-fl-source extract z TEGO tagu (nie z window
# lookback po HTMLu, bo to mogło pickować poprzedni cell).
for m in _PLAYER_OPTION_DIV_RE.finditer(html):
href = m.group("href") or ""
opening_tag = m.group(1)
src_match = _DATA_FL_SOURCE_RE.search(opening_tag)
source = (src_match.group(1).strip() if src_match else "") or None
_emit_host_entry(href, source)
if not playback_sources:
# Brak sub-hosters znalezionych — fallback do landing page (mobile otworzy
# w WebView). Robimy to TYLKO gdy nie ma alternatyw, inaczej landing jest
# niepotrzebnym ad-pageiem.
playback_sources.append(
RawPlaybackSource(
origin=source_name,
page_url=page_url,
thumbnail_url=poster_url,
)
)
return RawMovie(
external_id=slug,
title=title,
description=description,
release_year=release_year,
release_date=release_date,
duration_sec=duration_sec,
rating=rating,
poster_url=poster_url,
url=page_url,
studio=studio,
performers=performers,
tags=tags,
playback_sources=playback_sources,
raw={"slug": slug, "html_len": len(html)},
)
# ---- per-site instances ----------------------------------------------------
class StreampornConnector(DooplayConnector):
name = "streamporn"
base_url = "https://streamporn.nl"
class PandamoviesConnector(DooplayConnector):
name = "pandamovies"
base_url = "https://pandamovies.pw"
class MangopornConnector(DooplayConnector):
name = "mangoporn"
base_url = "https://mangoporn.net"
# ---------------------------------------------------------------------------
# Helpers (zduplikowane z paradisehill.py — celowo, żeby connectory były niezależne)
# ---------------------------------------------------------------------------
_TAG_RE = re.compile(r"<[^>]+>")
def _strip_tags(s: str) -> str:
return _TAG_RE.sub("", s)
_HTML_ENTITIES = {
"&": "&", "<": "<", ">": ">", """: '"', "'": "'",
"'": "'", " ": " ", "’": "'", "‘": "'",
"”": '"', "“": '"', "…": "...", "—": "—", "–": "–",
}
def _decode_html(s: str) -> str:
for k, v in _HTML_ENTITIES.items():
s = s.replace(k, v)
s = re.sub(r"&#(\d+);", lambda m: chr(int(m.group(1))), s)
s = re.sub(r"&#x([0-9a-fA-F]+);", lambda m: chr(int(m.group(1), 16)), s)
return s