Ashish/yt-dlp
1
1
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-05 09:47:27 -05:00
Code Issues 7 Projects Releases Wiki Activity

[youtube] Improve format filesize extraction (#16453)

Browse Source
This commit is contained in:
Sergey M․ 2018-05-14 23:27:56 +07:00
parent 84a9fef899
commit c63ca0eef8
No known key found for this signature in database
GPG Key ID: 2C393E0F18A9236D
1 changed files with 10 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
youtube_dl/extractor/youtube.py
View File

@ -1815,6 +1815,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
chapters = self._extract_chapters(description_original, video_duration) chapters = self._extract_chapters(description_original, video_duration)
def _extract_filesize(media_url):
return int_or_none(self._search_regex(
r'\bclen[=/](\d+)', media_url, 'filesize', default=None))
if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'): if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
self.report_rtmp_download() self.report_rtmp_download()
formats = [{ formats = [{
@ -1919,8 +1923,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0]) mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None) width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
filesize = int_or_none(url_data.get(
'clen', [None])[0]) or _extract_filesize(url)
more_fields = { more_fields = {
'filesize': int_or_none(url_data.get('clen', [None])[0]), 'filesize': filesize,
'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000), 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
'width': width, 'width': width,
'height': height, 'height': height,
@ -1994,6 +2001,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
for df in self._extract_mpd_formats( for df in self._extract_mpd_formats(
mpd_url, video_id, fatal=dash_mpd_fatal, mpd_url, video_id, fatal=dash_mpd_fatal,
formats_dict=self._formats): formats_dict=self._formats):
if not df.get('filesize'):
df['filesize'] = _extract_filesize(df['url'])
# Do not overwrite DASH format found in some previous DASH manifest # Do not overwrite DASH format found in some previous DASH manifest
if df['format_id'] not in dash_formats: if df['format_id'] not in dash_formats:
dash_formats[df['format_id']] = df dash_formats[df['format_id']] = df