[extractor] Update manifest_urls after redirect (#3575)

Authored by: elyse0

yt_dlp/extractor/common.py

@@ -1982,17 +1982,19 @@ class InfoExtractor:
     def _extract_f4m_formats(self, manifest_url, video_id, preference=None, quality=None, f4m_id=None,
                              transform_source=lambda s: fix_xml_ampersands(s).strip(),
                              fatal=True, m3u8_id=None, data=None, headers={}, query={}):
-        manifest = self._download_xml(
+        res = self._download_xml_handle(
             manifest_url, video_id, 'Downloading f4m manifest',
             'Unable to download f4m manifest',
             # Some manifests may be malformed, e.g. prosiebensat1 generated manifests
             # (see https://github.com/ytdl-org/youtube-dl/issues/6215#issuecomment-121704244)
             transform_source=transform_source,
             fatal=fatal, data=data, headers=headers, query=query)
-
-        if manifest is False:
+        if res is False:
             return []
 
+        manifest, urlh = res
+        manifest_url = urlh.geturl()
+
         return self._parse_f4m_formats(
             manifest, manifest_url, video_id, preference=preference, quality=quality, f4m_id=f4m_id,
             transform_source=transform_source, fatal=fatal, m3u8_id=m3u8_id)
@@ -2400,12 +2402,14 @@ class InfoExtractor:
         return '/'.join(out)
 
     def _extract_smil_formats_and_subtitles(self, smil_url, video_id, fatal=True, f4m_params=None, transform_source=None):
-        smil = self._download_smil(smil_url, video_id, fatal=fatal, transform_source=transform_source)
-
-        if smil is False:
+        res = self._download_smil(smil_url, video_id, fatal=fatal, transform_source=transform_source)
+        if res is False:
             assert not fatal
             return [], {}
 
+        smil, urlh = res
+        smil_url = urlh.geturl()
+
         namespace = self._parse_smil_namespace(smil)
 
         fmts = self._parse_smil_formats(
@@ -2422,13 +2426,17 @@ class InfoExtractor:
         return fmts
 
     def _extract_smil_info(self, smil_url, video_id, fatal=True, f4m_params=None):
-        smil = self._download_smil(smil_url, video_id, fatal=fatal)
-        if smil is False:
+        res = self._download_smil(smil_url, video_id, fatal=fatal)
+        if res is False:
             return {}
+
+        smil, urlh = res
+        smil_url = urlh.geturl()
+
         return self._parse_smil(smil, smil_url, video_id, f4m_params=f4m_params)
 
     def _download_smil(self, smil_url, video_id, fatal=True, transform_source=None):
-        return self._download_xml(
+        return self._download_xml_handle(
             smil_url, video_id, 'Downloading SMIL file',
             'Unable to download SMIL file', fatal=fatal, transform_source=transform_source)
 
@@ -2607,11 +2615,15 @@ class InfoExtractor:
         return subtitles
 
     def _extract_xspf_playlist(self, xspf_url, playlist_id, fatal=True):
-        xspf = self._download_xml(
+        res = self._download_xml_handle(
             xspf_url, playlist_id, 'Downloading xpsf playlist',
             'Unable to download xspf manifest', fatal=fatal)
-        if xspf is False:
+        if res is False:
             return []
+
+        xspf, urlh = res
+        xspf_url = urlh.geturl()
+
         return self._parse_xspf(
             xspf, playlist_id, xspf_url=xspf_url,
             xspf_base_url=base_url(xspf_url))
@@ -2676,7 +2688,10 @@ class InfoExtractor:
         mpd_doc, urlh = res
         if mpd_doc is None:
             return [], {}
-        mpd_base_url = base_url(urlh.geturl())
+
+        # We could have been redirected to a new url when we retrieved our mpd file.
+        mpd_url = urlh.geturl()
+        mpd_base_url = base_url(mpd_url)
 
         return self._parse_mpd_formats_and_subtitles(
             mpd_doc, mpd_id, mpd_base_url, mpd_url)

yt_dlp/extractor/infoq.py

@@ -4,8 +4,10 @@ from ..compat import (
     compat_urlparse,
 )
 from ..utils import (
+    ExtractorError,
     determine_ext,
     update_url_query,
+    traverse_obj,
 )
 from .bokecc import BokeCCBaseIE
 
@@ -34,6 +36,7 @@ class InfoQIE(BokeCCBaseIE):
             'ext': 'flv',
             'description': 'md5:308d981fb28fa42f49f9568322c683ff',
         },
+        'skip': 'Sorry, the page you visited does not exist',
     }, {
         'url': 'https://www.infoq.com/presentations/Simple-Made-Easy',
         'md5': '0e34642d4d9ef44bf86f66f6399672db',
@@ -86,8 +89,10 @@ class InfoQIE(BokeCCBaseIE):
         }]
 
     def _extract_http_audio(self, webpage, video_id):
-        fields = self._form_hidden_inputs('mp3Form', webpage)
-        http_audio_url = fields.get('filename')
+        try:
+            http_audio_url = traverse_obj(self._form_hidden_inputs('mp3Form', webpage), 'filename')
+        except ExtractorError:
+            http_audio_url = None
         if not http_audio_url:
             return []