def _find_video_id(self, url):
m = _url_re.match(url)
if m.group("video_id"):
log.debug("Video ID from URL")
return m.group("video_id")
res = self.session.http.get(url)
datam = _ytdata_re.search(res.text)
if datam:
data = parse_json(datam.group(1))
# find the videoRenderer object, where there is a LVE NOW badge
for vid_ep in search_dict(data, 'currentVideoEndpoint'):
video_id = vid_ep.get("watchEndpoint", {}).get("videoId")
if video_id:
log.debug("Video ID from currentVideoEndpoint")
return video_id
for x in search_dict(data, 'videoRenderer'):
for bstyle in search_dict(x.get("badges", {}), "style"):
if bstyle == "BADGE_STYLE_TYPE_LIVE_NOW":
if x.get("videoId"):
log.debug("Video ID from videoRenderer (live)")
return x["videoId"]
raise PluginError("Could not find a video on this page")
def _find_video_id(self, url):
m = _url_re.match(url)
if m.group("video_id"):
log.debug("Video ID from URL")
return m.group("video_id")
res = self.session.http.get(url)
datam = _ytdata_re.search(res.text)
if datam:
data = parse_json(datam.group(1))
# find the videoRenderer object, where there is a LVE NOW badge
for vid_ep in search_dict(data, 'currentVideoEndpoint'):
video_id = vid_ep.get("watchEndpoint", {}).get("videoId")
if video_id:
log.debug("Video ID from currentVideoEndpoint")
return video_id
for x in search_dict(data, 'videoRenderer'):
for bstyle in search_dict(x.get("badges", {}), "style"):
if bstyle == "BADGE_STYLE_TYPE_LIVE_NOW":
if x.get("videoId"):
log.debug("Video ID from videoRenderer (live)")
return x["videoId"]
if "/embed/live_stream" in url:
for link in itertags(res.text, "link"):
if link.attributes.get("rel") == "canonical":
canon_link = link.attributes.get("href")
if canon_link != url:
log.debug("Re-directing to canonical URL: {0}".format(canon_link))
return self._find_video_id(canon_link)
raise PluginError("Could not find a video on this page")
def _find_video_id(self, url):
m = _url_re.match(url)
video_id = m.group("video_id") or m.group("video_id_short")
if video_id:
log.debug("Video ID from URL")
return video_id
res = self.session.http.get(url)
if urlparse(res.url).netloc == "consent.youtube.com":
c_data = {}
for _i in itertags(res.text, "input"):
if _i.attributes.get("type") == "hidden":
c_data[_i.attributes.get("name")] = _i.attributes.get(
"value")
log.debug(f"c_data_keys: {', '.join(c_data.keys())}")
res = self.session.http.post("https://consent.youtube.com/s",
data=c_data)
datam = _ytdata_re.search(res.text)
if datam:
data = parse_json(datam.group(1))
# find the videoRenderer object, where there is a LVE NOW badge
for vid_ep in search_dict(data, 'currentVideoEndpoint'):
video_id = vid_ep.get("watchEndpoint", {}).get("videoId")
if video_id:
log.debug("Video ID from currentVideoEndpoint")
return video_id
for x in search_dict(data, 'videoRenderer'):
if x.get("viewCountText", {}).get("runs"):
if x.get("videoId"):
log.debug("Video ID from videoRenderer (live)")
return x["videoId"]
for bstyle in search_dict(x.get("badges", {}), "style"):
if bstyle == "BADGE_STYLE_TYPE_LIVE_NOW":
if x.get("videoId"):
log.debug("Video ID from videoRenderer (live)")
return x["videoId"]
if urlparse(url).path.endswith(("/embed/live_stream", "/live")):
for link in itertags(res.text, "link"):
if link.attributes.get("rel") == "canonical":
canon_link = link.attributes.get("href")
if canon_link != url:
if canon_link.endswith("v=live_stream"):
log.debug("The video is not available")
break
else:
log.debug(
"Re-directing to canonical URL: {0}".format(
canon_link))
return self._find_video_id(canon_link)
raise PluginError("Could not find a video on this page")
def test_search_dict(self):
self.assertSequenceEqual(
list(search_dict(["one", "two"], "one")),
[]
)
self.assertSequenceEqual(
list(search_dict({"two": "test2"}, "one")),
[]
)
self.assertSequenceEqual(
list(search_dict({"one": "test1", "two": "test2"}, "one")),
["test1"]
)
self.assertSequenceEqual(
list(search_dict({"one": {"inner": "test1"}, "two": "test2"}, "inner")),
["test1"]
)
self.assertSequenceEqual(
list(search_dict({"one": [{"inner": "test1"}], "two": "test2"}, "inner")),
["test1"]
)
self.assertSequenceEqual(
list(sorted(search_dict({"one": [{"inner": "test1"}], "two": {"inner": "test2"}}, "inner"))),
list(sorted(["test1", "test2"]))
)
def test_search_dict(self):
self.assertSequenceEqual(
list(search_dict(["one", "two"], "one")),
[]
)
self.assertSequenceEqual(
list(search_dict({"two": "test2"}, "one")),
[]
)
self.assertSequenceEqual(
list(search_dict({"one": "test1", "two": "test2"}, "one")),
["test1"]
)
self.assertSequenceEqual(
list(search_dict({"one": {"inner": "test1"}, "two": "test2"}, "inner")),
["test1"]
)
self.assertSequenceEqual(
list(search_dict({"one": [{"inner": "test1"}], "two": "test2"}, "inner")),
["test1"]
)
self.assertSequenceEqual(
list(sorted(search_dict({"one": [{"inner": "test1"}], "two": {"inner": "test2"}}, "inner"))),
list(sorted(["test1", "test2"]))
)
def _find_channel_video(self):
res = http.get(self.url)
datam = _ytdata_re.search(res.text)
if datam:
data = parse_json(datam.group(1))
# find the videoRenderer object, where there is a LVE NOW badge
for x in search_dict(data, 'videoRenderer'):
for bstyle in search_dict(x.get("badges", {}), "style"):
if bstyle == "BADGE_STYLE_TYPE_LIVE_NOW":
if x.get("videoId"):
self.logger.debug("Found channel video ID via HTML: {0}", x["videoId"])
return x["videoId"]
else:
# fall back on API
self.logger.debug("No channel data, falling back to API")
match = _channelid_re.search(res.text)
if not match:
return
channel_id = match.group(1)
self.logger.debug("Found channel_id: {0}".format(channel_id))
return self._get_channel_video(channel_id)
def _get_streams(self):
data = self.session.http.get(self.url, schema=self._data_schema)
if data:
for streaming_info in search_dict(data, "streamingInfo"):
video_assets = streaming_info.get("videoAssets")
if video_assets.get("hls"):
log.debug("Found HLS stream: {}".format(
video_assets.get("hls")))
for s in HLSStream.parse_variant_playlist(
self.session, video_assets.get("hls")).items():
yield s
if video_assets.get("mpeg"):
for mpeg_stream in video_assets.get("mpeg"):
q = mpeg_stream.get("renditionValue").strip(
"_") or "{}k".format(
mpeg_stream.get("bitrate") // 1000) or "vod"
yield q, HTTPStream(self.session, mpeg_stream["url"])
def _data_video_id(data):
if data:
for videoRenderer in search_dict(data, "videoRenderer"):
videoId = videoRenderer.get("videoId")
if videoId is not None:
return videoId
