X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;ds=inline;f=youtube_dl%2Fextractor%2Fcondenast.py;h=d3463b8747a2109e8efef6b93d70843e542b4590;hb=f7923a4c399e0ce8e6cd230db92aefbfcff297c3;hp=e8f2b5a07591410c16fe6fe096678a12006abe48;hpb=d6712378e73951bede475569c887a1ac73f660a9;p=youtube-dl
diff --git a/youtube_dl/extractor/condenast.py b/youtube_dl/extractor/condenast.py
index e8f2b5a07..d3463b874 100644
--- a/youtube_dl/extractor/condenast.py
+++ b/youtube_dl/extractor/condenast.py
@@ -5,12 +5,17 @@ import re
from .common import InfoExtractor
from ..compat import (
- compat_urllib_parse_urlencode,
compat_urllib_parse_urlparse,
compat_urlparse,
)
from ..utils import (
+ determine_ext,
+ extract_attributes,
+ int_or_none,
+ js_to_json,
+ mimetype2ext,
orderedSet,
+ parse_iso8601,
remove_end,
)
@@ -58,7 +63,20 @@ class CondeNastIE(InfoExtractor):
'ext': 'mp4',
'title': '3D Printed Speakers Lit With LED',
'description': 'Check out these beautiful 3D printed LED speakers. You can\'t actually buy them, but LumiGeek is working on a board that will let you make you\'re own.',
+ 'uploader': 'wired',
+ 'upload_date': '20130314',
+ 'timestamp': 1363219200,
}
+ }, {
+ 'url': 'http://video.gq.com/watch/the-closer-with-keith-olbermann-the-only-true-surprise-trump-s-an-idiot?c=series',
+ 'info_dict': {
+ 'id': '58d1865bfd2e6126e2000015',
+ 'ext': 'mp4',
+ 'title': 'The Only True Surprise? Trumpâs an Idiot',
+ 'uploader': 'gq',
+ 'upload_date': '20170321',
+ 'timestamp': 1490126427,
+ },
}, {
# JS embed
'url': 'http://player.cnevids.com/embedjs/55f9cf8b61646d1acf00000c/5511d76261646d5566020000.js',
@@ -67,70 +85,100 @@ class CondeNastIE(InfoExtractor):
'id': '55f9cf8b61646d1acf00000c',
'ext': 'mp4',
'title': '3D printed TSA Travel Sentry keys really do open TSA locks',
+ 'uploader': 'arstechnica',
+ 'upload_date': '20150916',
+ 'timestamp': 1442434955,
}
}]
def _extract_series(self, url, webpage):
- title = self._html_search_regex(r'
.*?
(.+?)
',
- webpage, 'series title', flags=re.DOTALL)
+ title = self._html_search_regex(
+ r'(?s)
.*?
(.+?)
',
+ webpage, 'series title')
url_object = compat_urllib_parse_urlparse(url)
base_url = '%s://%s' % (url_object.scheme, url_object.netloc)
- m_paths = re.finditer(r'
.*?.*?(.+?)
',
- r'
(.+?)
',
- ],
- webpage, 'description', fatal=False, flags=re.DOTALL)
+ query = {}
+ params = self._search_regex(
+ r'(?s)var params = {(.+?)}[;,]', webpage, 'player params', default=None)
+ if params:
+ query.update({
+ 'videoId': self._search_regex(r'videoId: [\'"](.+?)[\'"]', params, 'video id'),
+ 'playerId': self._search_regex(r'playerId: [\'"](.+?)[\'"]', params, 'player id'),
+ 'target': self._search_regex(r'target: [\'"](.+?)[\'"]', params, 'target'),
+ })
else:
- description = None
- params = self._search_regex(r'var params = {(.+?)}[;,]', webpage,
- 'player params', flags=re.DOTALL)
- video_id = self._search_regex(r'videoId: [\'"](.+?)[\'"]', params, 'video id')
- player_id = self._search_regex(r'playerId: [\'"](.+?)[\'"]', params, 'player id')
- target = self._search_regex(r'target: [\'"](.+?)[\'"]', params, 'target')
- data = compat_urllib_parse_urlencode({'videoId': video_id,
- 'playerId': player_id,
- 'target': target,
- })
- base_info_url = self._search_regex(r'url = [\'"](.+?)[\'"][,;]',
- webpage, 'base info url',
- default='http://player.cnevids.com/player/loader.js?')
- info_url = base_info_url + data
- info_page = self._download_webpage(info_url, video_id,
- 'Downloading video info')
- video_info = self._search_regex(r'var\s+video\s*=\s*({.+?});', info_page, 'video info')
- video_info = self._parse_json(video_info, video_id)
-
- formats = [{
- 'format_id': '%s-%s' % (fdata['type'].split('/')[-1], fdata['quality']),
- 'url': fdata['src'],
- 'ext': fdata['type'].split('/')[-1],
- 'quality': 1 if fdata['quality'] == 'high' else 0,
- } for fdata in video_info['sources'][0]]
+ params = extract_attributes(self._search_regex(
+ r'(<[^>]+data-js="video-player"[^>]+>)',
+ webpage, 'player params element'))
+ query.update({
+ 'videoId': params['data-video'],
+ 'playerId': params['data-player'],
+ 'target': params['id'],
+ })
+ video_id = query['videoId']
+ video_info = None
+ info_page = self._download_json(
+ 'http://player.cnevids.com/player/video.js',
+ video_id, 'Downloading video info', fatal=False, query=query)
+ if info_page:
+ video_info = info_page.get('video')
+ if not video_info:
+ info_page = self._download_webpage(
+ 'http://player.cnevids.com/player/loader.js',
+ video_id, 'Downloading loader info', query=query)
+ video_info = self._parse_json(
+ self._search_regex(
+ r'(?s)var\s+config\s*=\s*({.+?});', info_page, 'config'),
+ video_id, transform_source=js_to_json)['video']
+
+ title = video_info['title']
+
+ formats = []
+ for fdata in video_info['sources']:
+ src = fdata.get('src')
+ if not src:
+ continue
+ ext = mimetype2ext(fdata.get('type')) or determine_ext(src)
+ if ext == 'm3u8':
+ formats.extend(self._extract_m3u8_formats(
+ src, video_id, 'mp4', entry_protocol='m3u8_native',
+ m3u8_id='hls', fatal=False))
+ continue
+ quality = fdata.get('quality')
+ formats.append({
+ 'format_id': ext + ('-%s' % quality if quality else ''),
+ 'url': src,
+ 'ext': ext,
+ 'quality': 1 if quality == 'high' else 0,
+ })
self._sort_formats(formats)
- return {
+ info = self._search_json_ld(
+ webpage, video_id, fatal=False) if url_type != 'embed' else {}
+ info.update({
'id': video_id,
'formats': formats,
- 'title': video_info['title'],
- 'thumbnail': video_info['poster_frame'],
- 'description': description,
- }
+ 'title': title,
+ 'thumbnail': video_info.get('poster_frame'),
+ 'uploader': video_info.get('brand'),
+ 'duration': int_or_none(video_info.get('duration')),
+ 'tags': video_info.get('tags'),
+ 'series': video_info.get('series_title'),
+ 'season': video_info.get('season_title'),
+ 'timestamp': parse_iso8601(video_info.get('premiere_date')),
+ })
+ return info
def _real_extract(self, url):
- mobj = re.match(self._VALID_URL, url)
- site = mobj.group('site')
- url_type = mobj.group('type')
- item_id = mobj.group('id')
+ site, url_type, item_id = re.match(self._VALID_URL, url).groups()
# Convert JS embed to regular embed
if url_type == 'embedjs':
@@ -139,7 +187,6 @@ class CondeNastIE(InfoExtractor):
path=remove_end(parsed_url.path, '.js').replace('/embedjs/', '/embed/')))
url_type = 'embed'
- self.to_screen('Extracting from %s with the Condé Nast extractor' % self._SITES[site])
webpage = self._download_webpage(url, item_id)
if url_type == 'series':