X-Git-Url: http://git.cielonegro.org/gitweb.cgi?a=blobdiff_plain;ds=inline;f=youtube_dl%2Fextractor%2Fdailymotion.py;h=1ea449ca824bbf100edf9dc851a3cd74d1dcd266;hb=54fda45bace9d37135a4d29a03130b22ed272ce5;hp=3297a8549e38e7ddb5be2bda2053964bf70802e7;hpb=6f5ac90cf314e8e1a7c5c416fb51e3a7b8483e9c;p=youtube-dl.git
diff --git a/youtube_dl/extractor/dailymotion.py b/youtube_dl/extractor/dailymotion.py
index 3297a8549..1ea449ca8 100644
--- a/youtube_dl/extractor/dailymotion.py
+++ b/youtube_dl/extractor/dailymotion.py
@@ -1,12 +1,14 @@
import re
+import json
+import itertools
from .common import InfoExtractor
from ..utils import (
compat_urllib_request,
- compat_urllib_parse,
+ get_element_by_attribute,
+ get_element_by_id,
ExtractorError,
- unescapeHTML,
)
class DailymotionIE(InfoExtractor):
@@ -19,7 +21,7 @@ class DailymotionIE(InfoExtractor):
u'file': u'x33vw9.mp4',
u'md5': u'392c4b85a60a90dc4792da41ce3144eb',
u'info_dict': {
- u"uploader": u"Alex and Van .",
+ u"uploader": u"Amphora Alex and Van .",
u"title": u"Tutoriel de Youtubeur\"DL DES VIDEO DE YOUTUBE\""
}
}
@@ -39,33 +41,7 @@ class DailymotionIE(InfoExtractor):
# Extract URL, uploader and title from webpage
self.report_extraction(video_id)
- mobj = re.search(r'\s*var flashvars = (.*)', webpage)
- if mobj is None:
- raise ExtractorError(u'Unable to extract media URL')
- flashvars = compat_urllib_parse.unquote(mobj.group(1))
- for key in ['hd1080URL', 'hd720URL', 'hqURL', 'sdURL', 'ldURL', 'video_url']:
- if key in flashvars:
- max_quality = key
- self.to_screen(u'Using %s' % key)
- break
- else:
- raise ExtractorError(u'Unable to extract video URL')
-
- mobj = re.search(r'"' + max_quality + r'":"(.+?)"', flashvars)
- if mobj is None:
- raise ExtractorError(u'Unable to extract video URL')
-
- video_url = compat_urllib_parse.unquote(mobj.group(1)).replace('\\/', '/')
-
- # TODO: support choosing qualities
-
- mobj = re.search(r'', webpage)
- if mobj is None:
- raise ExtractorError(u'Unable to extract title')
- video_title = unescapeHTML(mobj.group('title'))
-
- video_uploader = None
video_uploader = self._search_regex([r'(?im)[^<]+?]+?>([^<]+?)',
# Looking for official user
r'<(?:span|a) .*?rel="author".*?>([^<]+?)'],
@@ -76,11 +52,59 @@ class DailymotionIE(InfoExtractor):
if mobj is not None:
video_upload_date = mobj.group(3) + mobj.group(2) + mobj.group(1)
+ embed_url = 'http://www.dailymotion.com/embed/video/%s' % video_id
+ embed_page = self._download_webpage(embed_url, video_id,
+ u'Downloading embed page')
+ info = self._search_regex(r'var info = ({.*?}),', embed_page, 'video info')
+ info = json.loads(info)
+
+ # TODO: support choosing qualities
+
+ for key in ['stream_h264_hd1080_url','stream_h264_hd_url',
+ 'stream_h264_hq_url','stream_h264_url',
+ 'stream_h264_ld_url']:
+ if info.get(key):#key in info and info[key]:
+ max_quality = key
+ self.to_screen(u'Using %s' % key)
+ break
+ else:
+ raise ExtractorError(u'Unable to extract video URL')
+ video_url = info[max_quality]
+
return [{
'id': video_id,
'url': video_url,
'uploader': video_uploader,
'upload_date': video_upload_date,
- 'title': video_title,
+ 'title': self._og_search_title(webpage),
'ext': video_extension,
+ 'thumbnail': info['thumbnail_url']
}]
+
+
+class DailymotionPlaylistIE(InfoExtractor):
+ _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P.+?)/'
+ _MORE_PAGES_INDICATOR = r''
+
+ def _real_extract(self, url):
+ mobj = re.match(self._VALID_URL, url)
+ playlist_id = mobj.group('id')
+ video_ids = []
+
+ for pagenum in itertools.count(1):
+ webpage = self._download_webpage('https://www.dailymotion.com/playlist/%s/%s' % (playlist_id, pagenum),
+ playlist_id, u'Downloading page %s' % pagenum)
+
+ playlist_el = get_element_by_attribute(u'class', u'video_list', webpage)
+ video_ids.extend(re.findall(r'data-id="(.+?)" data-ext-id', playlist_el))
+
+ if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
+ break
+
+ entries = [self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
+ for video_id in video_ids]
+ return {'_type': 'playlist',
+ 'id': playlist_id,
+ 'title': get_element_by_id(u'playlist_name', webpage),
+ 'entries': entries,
+ }