mirror of
https://github.com/ytdl-org/youtube-dl
synced 2024-11-30 21:12:54 +01:00
[revision3] add support pages of type embed
This commit is contained in:
parent
d1b29d1342
commit
18b0b23992
@ -52,6 +52,20 @@ class Revision3IE(InfoExtractor):
|
|||||||
'uploader': 'DNews',
|
'uploader': 'DNews',
|
||||||
'uploader_id': 'dnews',
|
'uploader_id': 'dnews',
|
||||||
},
|
},
|
||||||
|
}, {
|
||||||
|
'url': 'http://testtube.com/tt-editors-picks/the-israel-palestine-conflict-explained-in-ten-min',
|
||||||
|
'info_dict': {
|
||||||
|
'id': '73573',
|
||||||
|
'ext': 'mp4',
|
||||||
|
'display_id': 'tt-editors-picks/the-israel-palestine-conflict-explained-in-ten-min',
|
||||||
|
'title': 'The Israel-Palestine Conflict Explained in Ten Minutes',
|
||||||
|
'description': 'If you\'d like to learn about the struggle between Israelis and Palestinians, this video is a great place to start',
|
||||||
|
'uploader': 'Editors\' Picks',
|
||||||
|
'uploader_id': 'tt-editors-picks',
|
||||||
|
'timestamp': 1453309200,
|
||||||
|
'upload_date': '20160120',
|
||||||
|
},
|
||||||
|
'add_ie': ['Youtube'],
|
||||||
}]
|
}]
|
||||||
_PAGE_DATA_TEMPLATE = 'http://www.%s/apiProxy/ddn/%s?domain=%s'
|
_PAGE_DATA_TEMPLATE = 'http://www.%s/apiProxy/ddn/%s?domain=%s'
|
||||||
_API_KEY = 'ba9c741bce1b9d8e3defcc22193f3651b8867e62'
|
_API_KEY = 'ba9c741bce1b9d8e3defcc22193f3651b8867e62'
|
||||||
@ -61,9 +75,38 @@ class Revision3IE(InfoExtractor):
|
|||||||
page_info = self._download_json(
|
page_info = self._download_json(
|
||||||
self._PAGE_DATA_TEMPLATE % (domain, display_id, domain), display_id)
|
self._PAGE_DATA_TEMPLATE % (domain, display_id, domain), display_id)
|
||||||
|
|
||||||
if page_info['data']['type'] == 'episode':
|
page_data = page_info['data']
|
||||||
episode_data = page_info['data']
|
page_type = page_data['type']
|
||||||
video_id = compat_str(episode_data['video']['data']['id'])
|
if page_type == 'episode' or page_type == 'embed':
|
||||||
|
show_data = page_data['show']['data']
|
||||||
|
video_id = compat_str(page_data['video']['data']['id'])
|
||||||
|
|
||||||
|
preference = qualities(['mini', 'small', 'medium', 'large'])
|
||||||
|
thumbnails = [{
|
||||||
|
'url': image_url,
|
||||||
|
'id': image_id,
|
||||||
|
'preference': preference(image_id)
|
||||||
|
} for image_id, image_url in page_data.get('images', {}).items()]
|
||||||
|
|
||||||
|
info = {
|
||||||
|
'id': video_id,
|
||||||
|
'display_id': display_id,
|
||||||
|
'title': unescapeHTML(page_data['name']),
|
||||||
|
'description': unescapeHTML(page_data.get('summary')),
|
||||||
|
'timestamp': parse_iso8601(page_data.get('publishTime'), ' '),
|
||||||
|
'author': page_data.get('author'),
|
||||||
|
'uploader': show_data.get('name'),
|
||||||
|
'uploader_id': show_data.get('slug'),
|
||||||
|
'thumbnails': thumbnails,
|
||||||
|
}
|
||||||
|
|
||||||
|
if page_type == 'embed':
|
||||||
|
info.update({
|
||||||
|
'_type': 'url_transparent',
|
||||||
|
'url': page_data['video']['data']['embed'],
|
||||||
|
})
|
||||||
|
return info
|
||||||
|
|
||||||
video_data = self._download_json(
|
video_data = self._download_json(
|
||||||
'http://revision3.com/api/getPlaylist.json?api_key=%s&codecs=h264,vp8,theora&video_id=%s' % (self._API_KEY, video_id),
|
'http://revision3.com/api/getPlaylist.json?api_key=%s&codecs=h264,vp8,theora&video_id=%s' % (self._API_KEY, video_id),
|
||||||
video_id)['items'][0]
|
video_id)['items'][0]
|
||||||
@ -84,26 +127,15 @@ class Revision3IE(InfoExtractor):
|
|||||||
})
|
})
|
||||||
self._sort_formats(formats)
|
self._sort_formats(formats)
|
||||||
|
|
||||||
preference = qualities(['mini', 'small', 'medium', 'large'])
|
info.update({
|
||||||
thumbnails = [{
|
|
||||||
'url': image_url,
|
|
||||||
'id': image_id,
|
|
||||||
'preference': preference(image_id)
|
|
||||||
} for image_id, image_url in video_data.get('images', {}).items()]
|
|
||||||
|
|
||||||
return {
|
|
||||||
'id': video_id,
|
|
||||||
'display_id': display_id,
|
|
||||||
'title': unescapeHTML(video_data['title']),
|
'title': unescapeHTML(video_data['title']),
|
||||||
'description': unescapeHTML(video_data.get('summary')),
|
'description': unescapeHTML(video_data.get('summary')),
|
||||||
'timestamp': parse_iso8601(episode_data.get('publishTime'), ' '),
|
|
||||||
'author': episode_data.get('author'),
|
|
||||||
'uploader': video_data.get('show', {}).get('name'),
|
'uploader': video_data.get('show', {}).get('name'),
|
||||||
'uploader_id': video_data.get('show', {}).get('slug'),
|
'uploader_id': video_data.get('show', {}).get('slug'),
|
||||||
'duration': int_or_none(video_data.get('duration')),
|
'duration': int_or_none(video_data.get('duration')),
|
||||||
'thumbnails': thumbnails,
|
|
||||||
'formats': formats,
|
'formats': formats,
|
||||||
}
|
})
|
||||||
|
return info
|
||||||
else:
|
else:
|
||||||
show_data = page_info['show']['data']
|
show_data = page_info['show']['data']
|
||||||
episodes_data = page_info['episodes']['data']
|
episodes_data = page_info['episodes']['data']
|
||||||
|
Loading…
Reference in New Issue
Block a user