[youtube] Rewrite Extractor

- improve format sorting
- remove unused code(swf parsing, ...)
- fix series metadata extraction
- fix trailer video extraction
- improve error reporting
- extract video location
pull/26990/head
Remita Amine 4 years ago
parent 8bf9591a70
commit 67299f23d8

@ -2064,7 +2064,7 @@ class InfoExtractor(object):
})
return entries
def _extract_mpd_formats(self, mpd_url, video_id, mpd_id=None, note=None, errnote=None, fatal=True, formats_dict={}, data=None, headers={}, query={}):
def _extract_mpd_formats(self, mpd_url, video_id, mpd_id=None, note=None, errnote=None, fatal=True, data=None, headers={}, query={}):
res = self._download_xml_handle(
mpd_url, video_id,
note=note or 'Downloading MPD manifest',
@ -2078,10 +2078,9 @@ class InfoExtractor(object):
mpd_base_url = base_url(urlh.geturl())
return self._parse_mpd_formats(
mpd_doc, mpd_id=mpd_id, mpd_base_url=mpd_base_url,
formats_dict=formats_dict, mpd_url=mpd_url)
mpd_doc, mpd_id, mpd_base_url, mpd_url)
def _parse_mpd_formats(self, mpd_doc, mpd_id=None, mpd_base_url='', formats_dict={}, mpd_url=None):
def _parse_mpd_formats(self, mpd_doc, mpd_id=None, mpd_base_url='', mpd_url=None):
"""
Parse formats from MPD manifest.
References:
@ -2359,15 +2358,7 @@ class InfoExtractor(object):
else:
# Assuming direct URL to unfragmented media.
f['url'] = base_url
# According to [1, 5.3.5.2, Table 7, page 35] @id of Representation
# is not necessarily unique within a Period thus formats with
# the same `format_id` are quite possible. There are numerous examples
# of such manifests (see https://github.com/ytdl-org/youtube-dl/issues/15111,
# https://github.com/ytdl-org/youtube-dl/issues/13919)
full_info = formats_dict.get(representation_id, {}).copy()
full_info.update(f)
formats.append(full_info)
formats.append(f)
else:
self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
return formats

File diff suppressed because it is too large Load Diff
Loading…
Cancel
Save