Handle Areena redirecting to /audio/

Refs aajanki#261

yledl/extractors.py

@@ -14,11 +14,18 @@
 from .streamprobe import FullHDFlavorProber
 from .timestamp import parse_areena_timestamp
 from .subtitles import Subtitle
+from .http import Redirected
 
 
 logger = logging.getLogger('yledl')
 
 
+class PlaylistRedirected(Exception):
+    def __init__(self, message: str, suggested_url: str) -> None:
+        super().__init__(message)
+        self.suggested_url = suggested_url
+
+
 def extractor_factory(url, filters, language_chooser, httpclient):
     if re.match(r'^https?://yle\.fi/aihe/', url) or \
        re.match(r'^https?://svenska\.yle\.fi/artikel/', url) or \
@@ -205,10 +212,21 @@ def __init__(self, httpclient):
         self.httpclient = httpclient
 
     def extract(self, url, latest_only, title_formatter, ffprobe):
-        playlist = self.get_playlist(url, latest_only)
+        playlist = self.guarded_get_playlist(url, latest_only)
         return (self.extract_clip(clipurl, title_formatter, ffprobe)
                 for clipurl in playlist)
 
+    def guarded_get_playlist(self, url, latest_only=False):
+        # get_playlist, but with a guard to turn a http Redirected error to a PlaylistRedirected
+        # for upstream handling
+        try:
+            return self.get_playlist(url, latest_only)
+        except Redirected as r:
+            raise PlaylistRedirected(
+                f"Redirected while retrieving playlist: please try {r.response.url!r} instead",
+                suggested_url=r.response.url,
+            ) from r
+
     def get_playlist(self, url, latest_only=False):
         raise NotImplementedError("get_playlist must be overridden")
 

yledl/yledl.py

@@ -277,7 +277,7 @@ def clips():
                                  title_formatter, io.ffprobe())
 
     if action == StreamAction.PRINT_EPISODE_PAGES:
-        print_lines(extractor.get_playlist(url))
+        print_lines(extractor.guarded_get_playlist(url))
         return RD_SUCCESS
     elif action == StreamAction.PRINT_STREAM_URL:
         print_lines(dl.get_urls(clips(), stream_filters))