add support for episodes from mobile site

auth is a bit hmm at the minute, i haven't implemented the multi-area
process yet
but, it works!

a

1 files changed, 84 insertions, 0 deletions

diff --git a/yt_dlp_plugins/extractor/radiko_mobile.py b/yt_dlp_plugins/extractor/radiko_mobile.py
new file mode 100644
index 0000000..d3793d0
--- /dev/null
+++ b/yt_dlp_plugins/extractor/radiko_mobile.py
@@ -0,0 +1,84 @@
+import re
+
+from yt_dlp.utils import (
+	clean_html,
+	join_nonempty,
+	traverse_obj,
+	update_url_query
+)
+import yt_dlp_plugins.extractor.radiko_time as rtime
+from yt_dlp_plugins.extractor.radiko import _RadikoBaseIE, RadikoTimeFreeIE
+
+
+class _RadikoMobileBaseIE(_RadikoBaseIE):
+	_MOBILE_USER_AGENT = 'Mozilla/5.0 (Linux; Android 10; Pixel 4 XL) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/80.0.3987.87 Mobile Safari/537.36'
+
+	def _get_nextjs_data(self, url, video_id):
+		html = self._download_webpage(url, video_id, headers={"User-Agent": self._MOBILE_USER_AGENT})
+		data = self._search_nextjs_data(html, video_id)
+		return traverse_obj(data, ("props", "pageProps")), data
+
+	def _get_station_region(self, station):
+		api_url = update_url_query("https://radiko.jp/api/stations/batchGetStations", {"stationId": station})
+		data = self._download_json(api_url, station, note="Getting station regions")
+		return traverse_obj(data, ("stationList", lambda _, v: v["id"] == station, "prefecturesList"), get_all=False)
+
+	_extract_chapters = RadikoTimeFreeIE._extract_chapters
+
+
+class RadikoMobileEventIE(_RadikoMobileBaseIE):
+	_VALID_URL = r"https?://(?:www\.)?radiko\.jp/mobile/events/(?P<id>[0-9-]+$)"
+
+	def _get_programme_meta(self, program, actors=None):
+		if actors is not None:
+			cast = []
+			for actor_id in program.get("actorIdsList"):
+				cast.append(traverse_obj(actors, ("actorsList", lambda _, v: v["key"] == actor_id, "name"), get_all=None))
+		else:
+			cast = traverse_obj(program, (
+				'performer', {lambda x: re.split(r'[/／、　,，]', x)}, ..., {str.strip})) or None
+
+		start = traverse_obj(program, ("startAt", "seconds"))
+		end = traverse_obj(program, ("endAt", "seconds"))
+
+		return {
+			**traverse_obj(program, {
+				"id": "id",
+				"title": "title",
+				"series": "rSeasonName",
+				"tags": "tagsList",
+				"thumbnail": "imageUrl",
+
+				"channel": "stationName",
+				"channel_id": "stationId",
+				"uploader": "stationName",
+				"uploader_id": "stationId",
+			}),
+			"description": clean_html(join_nonempty("summary", "description", from_dict=program, delim="\n")),
+			"cast": cast,
+			"timestamp": start,
+			"release_timestamp": end,
+			"duration": end - start,
+		}
+
+	def _real_extract(self, url):
+		event_id = self._match_id(url)
+		pageProps, _ = self._get_nextjs_data(url, event_id)
+
+		meta = self._get_programme_meta(pageProps.get("program"), pageProps.get("actors"))
+		station = meta.get("channel_id")
+		start = rtime.RadikoTime.fromtimestamp(meta.get("timestamp"), tz=rtime.JST)
+		end = rtime.RadikoTime.fromtimestamp(meta.get("release_timestamp"), tz=rtime.JST)
+
+		chapters = self._extract_chapters(station, start, end, video_id=meta["id"])
+		area = self._get_station_region(station)[0]
+		auth_data = self._auth(area)
+		formats = self._get_station_formats(station, True, auth_data, start_at=start, end_at=end)
+
+		return {
+			**meta,
+			"chapters": chapters,
+			"formats": formats,
+			"live_status": "was_live",
+			"container": "m4a_dash",  # force fixup, AAC-only HLS
+		}