paradizelost/youtube-dl
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Implement @dstftw review comments

Browse Source
This commit is contained in:
dirkf 2021-04-18 15:37:14 +00:00 committed by df
parent 13d2684263
commit 97c1053ba8
1 changed files with 43 additions and 29 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

72
youtube_dl/extractor/bbc.py
View File

@ -1099,35 +1099,49 @@ class BBCIE(BBCCoUkIE):
media = simorgh_data['pageData']['promo'] media = simorgh_data['pageData']['promo']
if media['media'].get('format') == 'video': if media['media'].get('format') == 'video':
media.update(media['media']) media.update(media['media'])
formats = [] title = (dict_get(media.get('headlines') or {},
keys = {'url', 'format', 'format_id', 'language', 'quality', 'tbr', 'resolution'} ('shortHeadline', 'headline'))
for format in playlist: or playlist_title),
if not (format.get('url') and format.get('format')): programme_id = media.get('id')
continue if programme_id and title:
bitrate = format.pop('bitrate') formats = []
if bitrate: keys = {'url', 'format', 'format_id', 'language', 'quality', 'tbr', 'resolution'}
bitrate = re.sub(r'000\s*$', 'kbps', bitrate) for format in playlist:
format['tbr'] = parse_bitrate(bitrate) if not (format.get('url') and format.get('format')):
format['language'] = media.get('language') continue
# format id: penultimate item from the url split on _ and . bitrate = format.pop('bitrate')
(fmt,) = re.split('[_.]', format['url'])[-2:][:1] format['tbr'] = int_or_none(bitrate, scale=1000) or parse_bitrate(bitrate)
format['format_id'] = '%s_%s' % (format['format'], fmt) format['language'] = media.get('language')
if not format.get('resolution'): # format id: penultimate item from the url split on _ and .
format['resolution'] = fmt (fmt,) = re.split('[_.]', format['url'])[-2:][:1]
format['quality'] = -1 format['format_id'] = '%s_%s' % (format['format'], fmt)
formats.append(dict((k, format[k]) for k in keys)) # try to set resolution using any available data
self._sort_formats(formats) aspect_ratio = re.split(r'[xX:]', media.get('aspectRatio') or '')
return { if len(aspect_ratio) != 2:
'id': media.get('id'), aspect_ratio = None
'title': (dict_get(media.get('headlines'), else:
('shortHeadline', 'headline')) aspect_ratio = float_or_none(aspect_ratio[0], scale=aspect_ratio[1])
or playlist_title), # these may not be present, but try anyway
'description': media.get('summary') or playlist_description, width = int_or_none(format.get('width'))
'formats': formats, height = int_or_none(format.get('height'))
'subtitles': None, if (not height) and aspect_ratio:
'thumbnail': try_get(media, lambda x: x['image']['href']), height = int(width / aspect_ratio)
'timestamp': int_or_none(media.get('timestamp'), scale=1000) elif (not width) and aspect_ratio:
} width = int(height * aspect_ratio)
format['resolution'] = ('%dx%d' % (width, height) if width and height
else dict_get(format, ('resolution', 'res'), default=fmt))
format['quality'] = -1
formats.append(dict((k, format[k]) for k in keys))
self._sort_formats(formats)
return {
'id': programme_id,
'title': title,
'description': media.get('summary') or playlist_description,
'formats': formats,
'subtitles': None,
'thumbnail': try_get(media, lambda x: x['image']['href']),
'timestamp': int_or_none(media.get('timestamp'), scale=1000)
}
# general case: media nested in content object # general case: media nested in content object
# test: https://www.bbc.co.uk/scotland/articles/cm49v4x1r9lo # test: https://www.bbc.co.uk/scotland/articles/cm49v4x1r9lo