|
@ -417,6 +417,22 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): |
|
|
'upload_date': '20140605', |
|
|
'upload_date': '20140605', |
|
|
}, |
|
|
}, |
|
|
}, |
|
|
}, |
|
|
|
|
|
# video_info is None (https://github.com/rg3/youtube-dl/issues/4421) |
|
|
|
|
|
{ |
|
|
|
|
|
'url': '__2ABJjxzNo', |
|
|
|
|
|
'info_dict': { |
|
|
|
|
|
'id': '__2ABJjxzNo', |
|
|
|
|
|
'ext': 'mp4', |
|
|
|
|
|
'upload_date': '20100430', |
|
|
|
|
|
'uploader_id': 'deadmau5', |
|
|
|
|
|
'description': 'md5:12c56784b8032162bb936a5f76d55360', |
|
|
|
|
|
'uploader': 'deadmau5', |
|
|
|
|
|
'title': 'Deadmau5 - Some Chords (HD)', |
|
|
|
|
|
}, |
|
|
|
|
|
'expected_warnings': [ |
|
|
|
|
|
'DASH manifest missing', |
|
|
|
|
|
] |
|
|
|
|
|
} |
|
|
] |
|
|
] |
|
|
|
|
|
|
|
|
def __init__(self, *args, **kwargs): |
|
|
def __init__(self, *args, **kwargs): |
|
@ -666,6 +682,45 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): |
|
|
url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id |
|
|
url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id |
|
|
return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.') |
|
|
return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.') |
|
|
|
|
|
|
|
|
|
|
|
def _parse_dash_manifest(self, video_id, dash_manifest_url): |
|
|
|
|
|
def decrypt_sig(mobj): |
|
|
|
|
|
s = mobj.group(1) |
|
|
|
|
|
dec_s = self._decrypt_signature(s, video_id, player_url, age_gate) |
|
|
|
|
|
return '/signature/%s' % dec_s |
|
|
|
|
|
dash_manifest_url = re.sub(r'/s/([\w\.]+)', decrypt_sig, dash_manifest_url) |
|
|
|
|
|
dash_doc = self._download_xml( |
|
|
|
|
|
dash_manifest_url, video_id, |
|
|
|
|
|
note='Downloading DASH manifest', |
|
|
|
|
|
errnote='Could not download DASH manifest') |
|
|
|
|
|
|
|
|
|
|
|
formats = [] |
|
|
|
|
|
for r in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}Representation'): |
|
|
|
|
|
url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL') |
|
|
|
|
|
if url_el is None: |
|
|
|
|
|
continue |
|
|
|
|
|
format_id = r.attrib['id'] |
|
|
|
|
|
video_url = url_el.text |
|
|
|
|
|
filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength')) |
|
|
|
|
|
f = { |
|
|
|
|
|
'format_id': format_id, |
|
|
|
|
|
'url': video_url, |
|
|
|
|
|
'width': int_or_none(r.attrib.get('width')), |
|
|
|
|
|
'tbr': int_or_none(r.attrib.get('bandwidth'), 1000), |
|
|
|
|
|
'asr': int_or_none(r.attrib.get('audioSamplingRate')), |
|
|
|
|
|
'filesize': filesize, |
|
|
|
|
|
'fps': int_or_none(r.attrib.get('frameRate')), |
|
|
|
|
|
} |
|
|
|
|
|
try: |
|
|
|
|
|
existing_format = next( |
|
|
|
|
|
fo for fo in formats |
|
|
|
|
|
if fo['format_id'] == format_id) |
|
|
|
|
|
except StopIteration: |
|
|
|
|
|
f.update(self._formats.get(format_id, {})) |
|
|
|
|
|
formats.append(f) |
|
|
|
|
|
else: |
|
|
|
|
|
existing_format.update(f) |
|
|
|
|
|
return formats |
|
|
|
|
|
|
|
|
def _real_extract(self, url): |
|
|
def _real_extract(self, url): |
|
|
proto = ( |
|
|
proto = ( |
|
|
'http' if self._downloader.params.get('prefer_insecure', False) |
|
|
'http' if self._downloader.params.get('prefer_insecure', False) |
|
@ -943,51 +998,19 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): |
|
|
|
|
|
|
|
|
# Look for the DASH manifest |
|
|
# Look for the DASH manifest |
|
|
if self._downloader.params.get('youtube_include_dash_manifest', True): |
|
|
if self._downloader.params.get('youtube_include_dash_manifest', True): |
|
|
try: |
|
|
|
|
|
# The DASH manifest used needs to be the one from the original video_webpage. |
|
|
|
|
|
# The one found in get_video_info seems to be using different signatures. |
|
|
|
|
|
# However, in the case of an age restriction there won't be any embedded dashmpd in the video_webpage. |
|
|
|
|
|
# Luckily, it seems, this case uses some kind of default signature (len == 86), so the |
|
|
|
|
|
# combination of get_video_info and the _static_decrypt_signature() decryption fallback will work here. |
|
|
|
|
|
dash_manifest_url = video_info.get('dashmpd')[0] |
|
|
|
|
|
|
|
|
|
|
|
def decrypt_sig(mobj): |
|
|
|
|
|
s = mobj.group(1) |
|
|
|
|
|
dec_s = self._decrypt_signature(s, video_id, player_url, age_gate) |
|
|
|
|
|
return '/signature/%s' % dec_s |
|
|
|
|
|
dash_manifest_url = re.sub(r'/s/([\w\.]+)', decrypt_sig, dash_manifest_url) |
|
|
|
|
|
dash_doc = self._download_xml( |
|
|
|
|
|
dash_manifest_url, video_id, |
|
|
|
|
|
note='Downloading DASH manifest', |
|
|
|
|
|
errnote='Could not download DASH manifest') |
|
|
|
|
|
for r in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}Representation'): |
|
|
|
|
|
url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL') |
|
|
|
|
|
if url_el is None: |
|
|
|
|
|
continue |
|
|
|
|
|
format_id = r.attrib['id'] |
|
|
|
|
|
video_url = url_el.text |
|
|
|
|
|
filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength')) |
|
|
|
|
|
f = { |
|
|
|
|
|
'format_id': format_id, |
|
|
|
|
|
'url': video_url, |
|
|
|
|
|
'width': int_or_none(r.attrib.get('width')), |
|
|
|
|
|
'tbr': int_or_none(r.attrib.get('bandwidth'), 1000), |
|
|
|
|
|
'asr': int_or_none(r.attrib.get('audioSamplingRate')), |
|
|
|
|
|
'filesize': filesize, |
|
|
|
|
|
'fps': int_or_none(r.attrib.get('frameRate')), |
|
|
|
|
|
} |
|
|
|
|
|
try: |
|
|
|
|
|
existing_format = next( |
|
|
|
|
|
fo for fo in formats |
|
|
|
|
|
if fo['format_id'] == format_id) |
|
|
|
|
|
except StopIteration: |
|
|
|
|
|
f.update(self._formats.get(format_id, {})) |
|
|
|
|
|
formats.append(f) |
|
|
|
|
|
else: |
|
|
|
|
|
existing_format.update(f) |
|
|
|
|
|
|
|
|
|
|
|
except (ExtractorError, KeyError) as e: |
|
|
|
|
|
self.report_warning('Skipping DASH manifest: %r' % e, video_id) |
|
|
|
|
|
|
|
|
dash_mpd = video_info.get('dashmpd') |
|
|
|
|
|
if not dash_mpd: |
|
|
|
|
|
self.report_warning('%s: DASH manifest missing' % video_id) |
|
|
|
|
|
else: |
|
|
|
|
|
dash_manifest_url = dash_mpd[0] |
|
|
|
|
|
try: |
|
|
|
|
|
dash_formats = self._parse_dash_manifest( |
|
|
|
|
|
video_id, dash_manifest_url) |
|
|
|
|
|
except (ExtractorError, KeyError) as e: |
|
|
|
|
|
self.report_warning( |
|
|
|
|
|
'Skipping DASH manifest: %r' % e, video_id) |
|
|
|
|
|
else: |
|
|
|
|
|
formats.extend(dash_formats) |
|
|
|
|
|
|
|
|
self._sort_formats(formats) |
|
|
self._sort_formats(formats) |
|
|
|
|
|
|
|
|