#29 Extract named channels using dynamic endpoint with second__path param instead of /user/

extractors/channel.py

@@ -12,12 +12,12 @@ channel_cache_lock = Lock()
 channel_latest_cache = TTLCache(maxsize=500, ttl=300)
 channel_latest_cache_lock = Lock()
 
-def extract_channel(ucid):
+def extract_channel(ucid, second__path="user"):
 	with channel_cache_lock:
 		if ucid in channel_cache:
 			return channel_cache[ucid]
 
-	channel_type = "channel" if len(ucid) == 24 and ucid[:2] == "UC" else "user"
+	channel_type = "channel" if len(ucid) == 24 and ucid[:2] == "UC" else second__path
 	r = requests.get("https://www.youtube.com/{}/{}/videos?hl=en".format(channel_type, ucid), cookies=eu_consent_cookie())
 	r.raise_for_status()
 	yt_initial_data = extract_yt_initial_data(r.content.decode("utf8"))
@@ -35,7 +35,10 @@ def extract_channel(ucid):
 				"identifier": "ACCOUNT_TERMINATED"
 			}
 		else:
-			print("Seen alert text '{}'".format(alert_text))
+			return {
+				"error": alert_text,
+				"identifier": "UNKNOWN"
+			}
 
 	header = yt_initial_data["header"]["c4TabbedHeaderRenderer"] if "c4TabbedHeaderRenderer" in yt_initial_data["header"] else {}
 	channel_metadata = yt_initial_data["metadata"]["channelMetadataRenderer"]

index.py

@@ -58,7 +58,7 @@ class NewLeaf(object):
 
 	@cherrypy.expose
 	@cherrypy.tools.json_out()
-	def channels(self, *suffix, **kwargs):
+	def channels(self, *suffix, second__path="user", **kwargs):
 		ucid = ""
 		part = ""
 		possible_parts = ("videos", "latest", "playlists")
@@ -74,15 +74,21 @@ class NewLeaf(object):
 					"error": "Two components specified in URL, but neither component was recognised as a part keyword.",
 					"identifier": "PART_KEYWORD_NOT_RECOGNISED"
 				}
+		possible_paths = ("channel", "c", "user")
+		if second__path not in possible_paths:
+			return {
+				"error": "second__path parameter must be one of: " + str(possible_paths),
+				"identifier": "PATH_PARAMETER_NOT_RECOGNISED"
+			}
 
 		if part == "playlists":
 			return []
 		elif part == "latest":
-			return extract_channel_latest(ucid)
+			return extract_channel_latest(ucid, second__path)
 		elif part == "videos":
-			return extract_channel_videos(ucid)
+			return extract_channel_videos(ucid, second__path)
 		else: # part == "", so extract whole channel
-			return extract_channel(ucid)
+			return extract_channel(ucid, second__path)
 
 	@cherrypy.expose
 	@cherrypy.tools.json_out()