From 5e60d653b4a6d306f6a87c34e5bae46f6f898213 Mon Sep 17 00:00:00 2001 From: n0stal6ic Date: Wed, 8 Apr 2026 13:29:57 -0500 Subject: [PATCH] Update __init__.py --- PBSK/__init__.py | 480 +++++++++++++++++++++++++++++------------------ 1 file changed, 294 insertions(+), 186 deletions(-) diff --git a/PBSK/__init__.py b/PBSK/__init__.py index 967740a..03b53b1 100644 --- a/PBSK/__init__.py +++ b/PBSK/__init__.py @@ -1,232 +1,340 @@ from __future__ import annotations import json import re +import uuid from collections.abc import Generator from http.cookiejar import CookieJar -from typing import Optional, Union +from typing import Any, Optional, Union import click from langcodes import Language -from unshackle.core.constants import AnyTrack from unshackle.core.credential import Credential -from unshackle.core.manifests import DASH, HLS +from unshackle.core.manifests import HLS from unshackle.core.search_result import SearchResult from unshackle.core.service import Service from unshackle.core.titles import Episode, Movie, Movies, Series, Title_T, Titles_T -from unshackle.core.tracks import Chapter, Tracks +from unshackle.core.tracks import Chapters, Tracks -class PBSK(Service): +class PBS(Service): """ - Service code for the PBS Kids streaming service (https://pbskids.org) - + Service code for PBS (https://www.pbs.org) + Author: n0stal6ic - Authorization: None - Security: FHD@L3 + Authorization: Cookies + Geofence: US """ - TITLE_RE = r"^(?:https?://(?:www\.)?pbskids\.org)?(?P/videos/watch/[^?#\s]+)" + TITLE_RE = r"^(?:https?://(?:www\.)?pbs\.org/(?Pvideo|show)/)?(?P[a-zA-Z0-9-]+)" GEOFENCE = ("US",) @staticmethod - @click.command(name="PBSK", short_help="https://pbskids.org") + @click.command(name="PBS", short_help="https://www.pbs.org", help=__doc__) @click.argument("title", type=str) + @click.option( + "-a", "--all", "all_", + is_flag=True, + default=False, + help="Parse input as slug to download all content.", + ) @click.pass_context def cli(ctx, **kwargs): - return PBSK(ctx, **kwargs) + return PBS(ctx, **kwargs) - def __init__(self, ctx, title: str): + def __init__(self, ctx, title: str, all_: bool = False): super().__init__(ctx) self.title = title - self._license_url: Optional[str] = None + self.is_show = all_ - def authenticate(self, cookies: Optional[CookieJar] = None, credential: Optional[Credential] = None) -> None: + m = re.match(self.TITLE_RE, title) + if not m: + raise ValueError(f"Could not parse PBS URL or ID: {title!r}.") + + url_type = m.group("type") + self.title_id = m.group("id") + + if url_type == "show": + self.is_show = True + + self.uid: str = str(uuid.uuid4()) + self.passport: str = "no" + self.callsign: Optional[str] = None + self.station_id: Optional[str] = None + + def authenticate( + self, + cookies: Optional[CookieJar] = None, + credential: Optional[Credential] = None, + ) -> None: super().authenticate(cookies, credential) + if cookies: + uid = next((c.value for c in cookies if c.name == "pbs_uid"), None) + if uid: + self.uid = uid + self.passport = "yes" + self.callsign = next((c.value for c in cookies if c.name == "pbsol.station"), None) + self.station_id = next((c.value for c in cookies if c.name == "pbsol.station_id"), None) + + self.session.headers.update({"User-Agent": self.config["client"]["web"]["user_agent"]}) def search(self) -> Generator[SearchResult, None, None]: return yield def get_titles(self) -> Titles_T: - match = re.match(self.TITLE_RE, self.title.strip()) - if not match: - raise ValueError(f"Unrecognized URL: {self.title!r}") - - path = match.group("path").rstrip("/") - page_url = f"https://pbskids.org{path}" - - self.log.info("Detecting buildId") - resp = self.session.get(page_url) - resp.raise_for_status() - - nd_match = re.search( - r']+id="__NEXT_DATA__"[^>]*>(.*?)', - resp.text, - re.DOTALL, - ) - if not nd_match: - raise RuntimeError("__NEXT_DATA__ not found.") - - next_data = json.loads(nd_match.group(1)) - build_id = next_data.get("buildId") - if not build_id: - raise RuntimeError("buildId not found.") - self.log.debug(f"BuildID: {build_id}") - - api_url = f"https://pbskids.org/_next/data/{build_id}/en-US{path}.json" - self.log.info(f"Fetching JSON from: {api_url}") - resp = self.session.get(api_url) - resp.raise_for_status() - data = resp.json() - - video_data = data["pageProps"]["videoData"] - asset = video_data["mediaManagerAsset"] - - video_id = str(video_data["id"]) - video_type = video_data.get("videoType", "short") - title_text = video_data.get("title") or asset.get("title", "Unknown") - description = asset.get("description_short") or asset.get("description_long", "") - season_number = asset.get("season_number") - episode_number = asset.get("episode_number") - - show_title: Optional[str] = None - for prop in video_data.get("properties", []): - show_title = prop.get("title") - if show_title: - break - - title_data = { - "asset": asset, - "drm_enabled": bool(asset.get("drm_enabled", False)), - } - - if ( - video_type == "fullEpisode" - and show_title - and season_number is not None - and episode_number is not None - ): - return Series( - [ - Episode( - id_=video_id, - service=self.__class__, - title=show_title, - season=season_number, - number=episode_number, - name=title_text, - description=description, - year=None, - language=Language.get("en"), - data=title_data, - ) - ] - ) - else: - return Movies( - [ - Movie( - id_=video_id, - service=self.__class__, - name=title_text, - description=description, - year=None, - language=Language.get("en"), - data=title_data, - ) - ] - ) + if self.is_show: + return self._get_show_titles(self.title_id) + return self._get_video_title(self.title_id) def get_tracks(self, title: Title_T) -> Tracks: - asset = title.data["asset"] - drm_enabled = title.data["drm_enabled"] - videos = asset.get("videos", []) + video_bridge = title.data.get("video_bridge") or self._fetch_video_bridge(title.id) - if drm_enabled: - return self._get_drm_tracks(title, videos) - else: - return self._get_clear_tracks(title, videos) + if video_bridge.get("availability") != "available": + raise ValueError( + f"Video '{title.id}' is unavailable. (Status: {video_bridge.get('availability')!r}). " + "Passport content requires cookies from a logged-in PBS account." + ) - def _resolve_redirect(self, url: str) -> str: - resp = self.session.head(url, allow_redirects=True) - return str(resp.url) + encodings = video_bridge.get("encodings", []) + if not encodings: + raise ValueError(f"No streams found for '{title.id}'.") - @staticmethod - def _get_profile_height(profile: str) -> int: - match = re.search(r"-(\d+)p(?:-|$)", profile) - return int(match.group(1)) if match else -1 + m3u8_url = self._resolve_encoding(encodings[0]) + self.log.debug(f"HLS master: {m3u8_url}") - def _pick_best_stream( - self, - videos: list[dict], - prefix: str, - suffix: str = "", - required_key: Optional[str] = None, - ) -> Optional[dict]: - candidates: list[dict] = [] - seen_urls: set[str] = set() + return HLS.from_url(url=m3u8_url, session=self.session).to_tracks(language=title.language) - for video in videos: - profile = str(video.get("profile") or "") - url = video.get("url", "") - if not profile.startswith(prefix): - continue - if suffix and not profile.endswith(suffix): - continue - if required_key and not video.get(required_key): - continue - if not url or url in seen_urls: - continue - seen_urls.add(url) - candidates.append(video) + def get_chapters(self, title: Title_T) -> Chapters: + return Chapters() - if not candidates: - return None - - return max(candidates, key=lambda v: self._get_profile_height(str(v.get("profile") or ""))) - - def _get_clear_tracks(self, title: Title_T, videos: list[dict]) -> Tracks: - video = self._pick_best_stream(videos, prefix="hls-") - if not video: - raise RuntimeError("No HLS stream found.") - - stream_url = self._resolve_redirect(video["url"]) - self.log.debug(f"HLS stream URL: {stream_url}") - - return HLS.from_url(url=stream_url, session=self.session).to_tracks(language=title.language) - - def _get_drm_tracks(self, title: Title_T, videos: list[dict]) -> Tracks: - video = self._pick_best_stream(videos, prefix="dash-", suffix="-drm", required_key="widevine_license") - if not video: - raise RuntimeError("No Widevine DASH stream found.") - - self._license_url = video.get("widevine_license") - if not self._license_url: - raise RuntimeError("No Widevine license URL detected from DASH stream.") - - stream_url = self._resolve_redirect(video["url"]) - self.log.debug(f"DASH stream URL: {stream_url}") - self.log.debug(f"Widevine license URL: {self._license_url}") - - return DASH.from_url(url=stream_url, session=self.session).to_tracks(language=title.language) - - def get_chapters(self, title: Title_T) -> list[Chapter]: - return [] - - def get_widevine_service_certificate(self, **_: any) -> Optional[str]: - if self.config: - return self.config.get("certificate") + def get_widevine_service_certificate(self, **_: Any) -> None: return None - def get_widevine_license( - self, *, challenge: bytes, title: Title_T, track: AnyTrack - ) -> Optional[Union[bytes, str]]: - if not self._license_url: - raise RuntimeError("Widevine license URL is not set.") + def get_widevine_license(self, *, challenge: bytes, title: Title_T, track: Any) -> None: + return None - resp = self.session.post( - url=self._license_url, - data=challenge, - headers={"Content-Type": "application/octet-stream"}, + def _get_video_title(self, video_slug: str) -> Titles_T: + video_bridge = self._fetch_video_bridge(video_slug) + show_slug = video_bridge.get("program", {}).get("slug") + show_title = video_bridge.get("program", {}).get("title", "Unknown Show") + + episode_data = self._find_episode_in_show(show_slug, video_slug) if show_slug else None + + if episode_data: + parent = episode_data.get("parent", {}) + season_data = parent.get("season", {}) + show_display = season_data.get("show", {}).get("title", show_title) + season_num = int(season_data.get("ordinal") or 0) + ep_num = int(parent.get("ordinal") or 0) + year = self._parse_year(episode_data.get("premiere_date")) + + return Series([ + Episode( + id_=video_slug, + service=self.__class__, + title=show_display, + season=season_num, + number=ep_num, + name=episode_data.get("title", video_slug), + description=episode_data.get("description_short"), + year=year, + language=Language.get("en"), + data={"video_bridge": video_bridge, "episode": episode_data}, + ) + ]) + + self.log.warning("Unable to find episode metadata.") + fallback_name = video_bridge.get("_page_title") or video_slug.replace("-", " ").title() + return Series([ + Episode( + id_=video_slug, + service=self.__class__, + title=show_title, + season=0, + number=0, + name=fallback_name, + language=Language.get("en"), + data={"video_bridge": video_bridge}, + ) + ]) + + def _get_show_titles(self, show_slug: str) -> Titles_T: + self.log.info(f"Fetching season data for {show_slug}.") + seasons = self._fetch_show_seasons(show_slug) + specials = self._fetch_show_specials(show_slug) + show_display = show_slug.replace("-", " ").title() + episodes = [] + + if seasons: + self.log.info(f"Found {len(seasons)} season(s)") + for season_cid, season_ordinal in seasons: + self.log.info(f"Season {season_ordinal}") + for ep in self._fetch_season_episodes(show_slug, season_cid): + parent = ep.get("parent", {}) + season_data = parent.get("season", {}) + show_display = season_data.get("show", {}).get("title") or show_display + season_num = int(season_data.get("ordinal") or season_ordinal) + ep_num = int(parent.get("ordinal") or 0) + year = self._parse_year(ep.get("premiere_date")) + + episodes.append(Episode( + id_=ep["slug"], + service=self.__class__, + title=show_display, + season=season_num, + number=ep_num, + name=ep.get("title", ep["slug"]), + description=ep.get("description_short"), + year=year, + language=Language.get("en"), + data={"episode": ep}, + )) + + if specials: + self.log.info(f"Found {len(specials)} special(s)") + specials.sort(key=lambda x: x.get("premiere_date") or "") + for i, sp in enumerate(specials, start=1): + self.log.info(f"Special {i}") + parent = sp.get("parent", {}) + show_display = parent.get("show", {}).get("title") or show_display + year = self._parse_year(sp.get("premiere_date")) + + episodes.append(Episode( + id_=sp["slug"], + service=self.__class__, + title=show_display, + season=0, + number=i, + name=sp.get("title", sp["slug"]), + description=sp.get("description_short"), + year=year, + language=Language.get("en"), + data={"episode": sp}, + )) + + if not episodes: + raise ValueError(f"No episodes or specials found for {show_slug!r}.") + + return Series(episodes) + + def _fetch_video_bridge(self, video_slug: str) -> dict: + params: dict[str, str] = { + "uid": self.uid, + "userPassportStatus": self.passport, + "autoplay": "true", + "unsafeDisableUpsellHref": "true", + } + if self.callsign: + params["callsign"] = self.callsign + if self.station_id: + params["station_id"] = self.station_id + + r = self.session.get( + self.config["endpoints"]["portalplayer"] + video_slug + "/", + params=params, ) - resp.raise_for_status() - return resp.content + r.raise_for_status() + + idx = r.text.find("window.videoBridge = ") + if idx == -1: + raise ValueError( + f"videoBridge not found in portalplayer response for {video_slug!r}.\n" + "Double check your input URL and make sure it is correct." + ) + + json_start = r.text.index("{", idx) + video_bridge, _ = json.JSONDecoder().raw_decode(r.text, json_start) + title_m = re.search(r"\s*Video:\s*(.*?)\s*\|\s*Watch", r.text, re.DOTALL) + if title_m: + video_bridge["_page_title"] = " ".join(title_m.group(1).split()) + + return video_bridge + + def _resolve_encoding(self, encoding_url: str) -> str: + r = self.session.get(encoding_url, params={"format": "jsonp", "callback": "__jp0"}) + r.raise_for_status() + + m = re.search(r"__jp0\((.+)\)\s*$", r.text.strip()) + if not m: + raise ValueError(f"Unable to parse URS JSONP response: {r.text[:200]!r}") + + data = json.loads(m.group(1)) + url = data.get("url") + if not url: + raise ValueError(f"URS redirect returned no URL: {data}") + return url + + def _fetch_show_seasons(self, show_slug: str) -> list[tuple[str, int]]: + r = self.session.get(f"https://www.pbs.org/show/{show_slug}/", timeout=15) + if not r.ok: + self.log.warning(f"Page returned {r.status_code} for '{show_slug}'.") + return [] + seasons = self._parse_seasons_from_html(r.text) + if not seasons: + self.log.warning(f"Unable to find season data for '{show_slug}': {r.status_code}.") + return seasons + + def _parse_seasons_from_html(self, html: str) -> list[tuple[str, int]]: + UUID_PAT = r'[a-f0-9]{8}-[a-f0-9]{4}-[a-f0-9]{4}-[a-f0-9]{4}-[a-f0-9]{12}' + seasons: list[tuple[str, int]] = [] + seen: set[str] = set() + + for url_m in re.finditer( + r'content\.services\.pbs\.org/v3/pbsorg/screens/shows/[^/]+/seasons/(' + UUID_PAT + r')/', + html, + ): + cid = url_m.group(1) + if cid in seen: + continue + seen.add(cid) + + window = html[max(0, url_m.start() - 800):url_m.start()] + all_ords = re.findall(r'ordinal[\\\"]*\s*:\s*(\d+)', window) + ordinal = int(all_ords[-1]) if all_ords else (len(seasons) + 1) + seasons.append((cid, ordinal)) + + return sorted(seasons, key=lambda x: x[1], reverse=True) + + def _fetch_show_specials(self, show_slug: str) -> list[dict]: + url = self.config["endpoints"]["show_specials"].format(show_slug=show_slug) + r = self.session.get(url) + if not r.ok: + self.log.warning(f"Unable to find specials for '{show_slug}': {r.status_code}.") + return [] + return [ + sp for sp in r.json() + if sp.get("slug") != sp.get("parent", {}).get("slug") + ] + + def _fetch_season_episodes(self, show_slug: str, season_cid: str) -> list[dict]: + url = self.config["endpoints"]["show_episodes"].format( + show_slug=show_slug, + season_cid=season_cid, + ) + r = self.session.get(url) + if not r.ok: + self.log.warning(f"Unable to find episodes for season {season_cid}: {r.status_code}.") + return [] + return r.json() + + def _find_episode_in_show(self, show_slug: str, video_slug: str) -> Optional[dict]: + r = self.session.get(f"https://www.pbs.org/video/{video_slug}/", timeout=15) + seasons = self._parse_seasons_from_html(r.text) if r.ok else [] + if not seasons: + seasons = self._fetch_show_seasons(show_slug) + for season_cid, _ in seasons: + for ep in self._fetch_season_episodes(show_slug, season_cid): + if ep.get("slug") == video_slug: + return ep + for sp in self._fetch_show_specials(show_slug): + if sp.get("slug") == video_slug: + return sp + return None + + @staticmethod + def _parse_year(date_str: Optional[str]) -> Optional[int]: + if not date_str: + return None + try: + return int(date_str[:4]) + except (ValueError, TypeError): + return None