[funnyordie] Copy formats' metadata from hls and sort formats

This commit is contained in:
Sergey M․ 2016-11-21 23:46:55 +07:00
parent 4eece8ba57
commit 6334794f2a
No known key found for this signature in database
GPG Key ID: 2C393E0F18A9236D
1 changed files with 37 additions and 10 deletions

View File

@ -11,7 +11,7 @@ class FunnyOrDieIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?funnyordie\.com/(?P<type>embed|articles|videos)/(?P<id>[0-9a-f]+)(?:$|[?#/])' _VALID_URL = r'https?://(?:www\.)?funnyordie\.com/(?P<type>embed|articles|videos)/(?P<id>[0-9a-f]+)(?:$|[?#/])'
_TESTS = [{ _TESTS = [{
'url': 'http://www.funnyordie.com/videos/0732f586d7/heart-shaped-box-literal-video-version', 'url': 'http://www.funnyordie.com/videos/0732f586d7/heart-shaped-box-literal-video-version',
'md5': 'c26b9ee0e1ca138c12071f59572ba9c7', 'md5': 'bcd81e0c4f26189ee09be362ad6e6ba9',
'info_dict': { 'info_dict': {
'id': '0732f586d7', 'id': '0732f586d7',
'ext': 'mp4', 'ext': 'mp4',
@ -28,6 +28,9 @@ class FunnyOrDieIE(InfoExtractor):
'description': 'Please use this to sell something. www.jonlajoie.com', 'description': 'Please use this to sell something. www.jonlajoie.com',
'thumbnail': 're:^http:.*\.jpg$', 'thumbnail': 're:^http:.*\.jpg$',
}, },
'params': {
'skip_download': True,
},
}, { }, {
'url': 'http://www.funnyordie.com/articles/ebf5e34fc8/10-hours-of-walking-in-nyc-as-a-man', 'url': 'http://www.funnyordie.com/articles/ebf5e34fc8/10-hours-of-walking-in-nyc-as-a-man',
'only_matching': True, 'only_matching': True,
@ -51,21 +54,45 @@ class FunnyOrDieIE(InfoExtractor):
formats = [] formats = []
m3u8_formats = self._extract_m3u8_formats(
m3u8_url, video_id, 'mp4', 'm3u8_native',
m3u8_id='hls', fatal=False)
source_formats = list(filter(
lambda f: f.get('vcodec') != 'none' and f.get('resolution') != 'multiple',
m3u8_formats))
bitrates = [int(bitrate) for bitrate in re.findall(r'[,/]v(\d+)(?=[,/])', m3u8_url)] bitrates = [int(bitrate) for bitrate in re.findall(r'[,/]v(\d+)(?=[,/])', m3u8_url)]
bitrates.sort() bitrates.sort()
for bitrate in bitrates: if source_formats:
for link in links: self._sort_formats(source_formats)
formats.append({
'url': self._proto_relative_url('%s%d.%s' % (link[0], bitrate, link[1])),
'format_id': '%s-%d' % (link[1], bitrate),
'vbr': bitrate,
})
for bitrate, f in zip(bitrates, source_formats or [{}] * len(bitrates)):
for path, ext in links:
ff = f.copy()
if ff:
if ext != 'mp4':
ff = dict(
[(k, v) for k, v in ff.items()
if k in ('height', 'width', 'format_id')])
ff.update({
'format_id': ff['format_id'].replace('hls', ext),
'ext': ext,
'protocol': 'http',
})
else:
ff.update({
'format_id': '%s-%d' % (ext, bitrate),
'vbr': bitrate,
})
ff['url'] = self._proto_relative_url(
'%s%d.%s' % (path, bitrate, ext))
formats.append(ff)
self._check_formats(formats, video_id) self._check_formats(formats, video_id)
formats.extend(self._extract_m3u8_formats( formats.extend(m3u8_formats)
m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)) self._sort_formats(
formats, field_preference=('height', 'width', 'tbr', 'format_id'))
subtitles = {} subtitles = {}
for src, src_lang in re.findall(r'<track kind="captions" src="([^"]+)" srclang="([^"]+)"', webpage): for src, src_lang in re.findall(r'<track kind="captions" src="([^"]+)" srclang="([^"]+)"', webpage):