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

[common] Modify _parse_dash_manifest for use in Facebook

This commit is contained in:
Yen Chi Hsuan 2016-01-30 21:27:43 +08:00
parent 17b598d30c
commit b323e1707d
1 changed files with 17 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

29
youtube_dl/extractor/common.py
View File

@ -1330,22 +1330,24 @@ class InfoExtractor(object):
}) })
return entries return entries
def _parse_dash_manifest(self, video_id, dash_doc, fatal=True): def _parse_dash_manifest(self, video_id, dash_doc, default_ns='urn:mpeg:DASH:schema:MPD:2011', formats_dict={}, fatal=True):
def _add_ns(tag):
return '{%s}%s' % (default_ns, tag)
formats = [] formats = []
for a in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'): for a in dash_doc.findall('.//' + _add_ns('AdaptationSet')):
mime_type = a.attrib.get('mimeType') mime_type = a.attrib.get('mimeType')
for r in a.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'): for r in a.findall(_add_ns('Representation')):
url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL') mime_type = r.attrib.get('mimeType') or mime_type
if url_el is None: url_el = r.find(_add_ns('BaseURL'))
continue
if mime_type == 'text/vtt': if mime_type == 'text/vtt':
# TODO implement WebVTT downloading # TODO implement WebVTT downloading
pass pass
elif mime_type.startswith('audio/') or mime_type.startswith('video/'): elif mime_type.startswith('audio/') or mime_type.startswith('video/'):
segment_list = r.find('{urn:mpeg:DASH:schema:MPD:2011}SegmentList') segment_list = r.find(_add_ns('SegmentList'))
format_id = r.attrib['id'] format_id = r.attrib['id']
video_url = url_el.text video_url = url_el.text if url_el else None
filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength')) filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength') if url_el else None)
f = { f = {
'format_id': format_id, 'format_id': format_id,
'url': video_url, 'url': video_url,
@ -1357,17 +1359,20 @@ class InfoExtractor(object):
'fps': int_or_none(r.attrib.get('frameRate')), 'fps': int_or_none(r.attrib.get('frameRate')),
} }
if segment_list is not None: if segment_list is not None:
initialization_url = segment_list.find(_add_ns('Initialization')).attrib['sourceURL']
f.update({ f.update({
'initialization_url': segment_list.find('{urn:mpeg:DASH:schema:MPD:2011}Initialization').attrib['sourceURL'], 'initialization_url': initialization_url,
'segment_urls': [segment.attrib.get('media') for segment in segment_list.findall('{urn:mpeg:DASH:schema:MPD:2011}SegmentURL')], 'segment_urls': [segment.attrib.get('media') for segment in segment_list.findall(_add_ns('SegmentURL'))],
'protocol': 'http_dash_segments', 'protocol': 'http_dash_segments',
}) })
if not f.get('url'):
f['url'] = initialization_url
try: try:
existing_format = next( existing_format = next(
fo for fo in formats fo for fo in formats
if fo['format_id'] == format_id) if fo['format_id'] == format_id)
except StopIteration: except StopIteration:
full_info = self._formats.get(format_id, {}).copy() full_info = formats_dict.get(format_id, {}).copy()
full_info.update(f) full_info.update(f)
codecs = r.attrib.get('codecs') codecs = r.attrib.get('codecs')
if codecs: if codecs: