Mirroring
/
yt-dlp
mirror of https://github.com/yt-dlp/yt-dlp
1
0
Fork 0
Code Issues Releases Wiki Activity

[common] _parse_dash_manifest() from youtube.py

Browse Source
pull/8/head
Yen Chi Hsuan 9 years ago
parent 05d0d131a7
commit 17b598d30c
2 changed files with 52 additions and 52 deletions
Show Stats Download Patch File Download Diff File

52
youtube_dl/extractor/common.py
Unescape Escape View File

@ -1330,6 +1330,58 @@ class InfoExtractor(object):
})
return entries
def _parse_dash_manifest(self, video_id, dash_doc, fatal=True):
formats = []
for a in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'):
mime_type = a.attrib.get('mimeType')
for r in a.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
if mime_type == 'text/vtt':
# TODO implement WebVTT downloading
pass
elif mime_type.startswith('audio/') or mime_type.startswith('video/'):
segment_list = r.find('{urn:mpeg:DASH:schema:MPD:2011}SegmentList')
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')),
'height': int_or_none(r.attrib.get('height')),
'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')),
}
if segment_list is not None:
f.update({
'initialization_url': segment_list.find('{urn:mpeg:DASH:schema:MPD:2011}Initialization').attrib['sourceURL'],
'segment_urls': [segment.attrib.get('media') for segment in segment_list.findall('{urn:mpeg:DASH:schema:MPD:2011}SegmentURL')],
'protocol': 'http_dash_segments',
})
try:
existing_format = next(
fo for fo in formats
if fo['format_id'] == format_id)
except StopIteration:
full_info = self._formats.get(format_id, {}).copy()
full_info.update(f)
codecs = r.attrib.get('codecs')
if codecs:
if full_info.get('acodec') == 'none':
full_info['vcodec'] = codecs
elif full_info.get('vcodec') == 'none':
full_info['acodec'] = codecs
formats.append(full_info)
else:
existing_format.update(f)
else:
self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
return formats
def _live_title(self, name):
""" Generate the title for a live video """
now = datetime.datetime.now()

52
youtube_dl/extractor/youtube.py
Unescape Escape View File

@ -1035,58 +1035,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
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.')
def _parse_dash_manifest(self, video_id, dash_doc, fatal=True):
formats = []
for a in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'):
mime_type = a.attrib.get('mimeType')
for r in a.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
if mime_type == 'text/vtt':
# TODO implement WebVTT downloading
pass
elif mime_type.startswith('audio/') or mime_type.startswith('video/'):
segment_list = r.find('{urn:mpeg:DASH:schema:MPD:2011}SegmentList')
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')),
'height': int_or_none(r.attrib.get('height')),
'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')),
}
if segment_list is not None:
f.update({
'initialization_url': segment_list.find('{urn:mpeg:DASH:schema:MPD:2011}Initialization').attrib['sourceURL'],
'segment_urls': [segment.attrib.get('media') for segment in segment_list.findall('{urn:mpeg:DASH:schema:MPD:2011}SegmentURL')],
'protocol': 'http_dash_segments',
})
try:
existing_format = next(
fo for fo in formats
if fo['format_id'] == format_id)
except StopIteration:
full_info = self._formats.get(format_id, {}).copy()
full_info.update(f)
codecs = r.attrib.get('codecs')
if codecs:
if full_info.get('acodec') == 'none':
full_info['vcodec'] = codecs
elif full_info.get('vcodec') == 'none':
full_info['acodec'] = codecs
formats.append(full_info)
else:
existing_format.update(f)
else:
self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
return formats
def _real_extract(self, url):
url, smuggled_data = unsmuggle_url(url, {})

Write Preview
Loading…
Cancel
Save