4 from .common
import InfoExtractor
7 get_element_by_attribute
,
11 class ImdbIE(InfoExtractor
):
13 IE_DESC
= u
'Internet Movie Database trailers'
14 _VALID_URL
= r
'http://(?:www|m)\.imdb\.com/video/imdb/vi(?P<id>\d+)'
17 u
'url': u
'http://www.imdb.com/video/imdb/vi2524815897',
18 u
'md5': u
'9f34fa777ade3a6e57a054fdbcb3a068',
22 u
'title': u
'Ice Age: Continental Drift Trailer (No. 2) - IMDb',
23 u
'description': u
'md5:9061c2219254e5d14e03c25c98e96a81',
27 def _real_extract(self
, url
):
28 mobj
= re
.match(self
._VALID
_URL
, url
)
29 video_id
= mobj
.group('id')
30 webpage
= self
._download
_webpage
('http://www.imdb.com/video/imdb/vi%s' % video_id
, video_id
)
31 descr
= get_element_by_attribute('itemprop', 'description', webpage
)
32 available_formats
= re
.findall(
33 r
'case \'(?P
<f_id
>.*?
)\' :$\s
+url
= \'(?P
<path
>.*?
)\'', webpage,
36 for f_id, f_path in available_formats:
37 f_path = f_path.strip()
38 format_page = self._download_webpage(
39 compat_urlparse.urljoin(url, f_path),
40 u'Downloading info
for %s format
' % f_id)
41 json_data = self._search_regex(
42 r'<script
[^
>]+class="imdb-player-data"[^
>]*?
>(.*?
)</script
>',
43 format_page, u'json data
', flags=re.DOTALL)
44 info = json.loads(json_data)
45 format_info = info['videoPlayerObject
']['video
']
48 'url
': format_info['url
'],
53 'title
': self._og_search_title(webpage),
56 'thumbnail
': format_info['slate
'],
59 class ImdbListIE(InfoExtractor):
60 IE_NAME = u'imdb
:list'
61 IE_DESC = u'Internet Movie Database lists
'
62 _VALID_URL = r'http
://www\
.imdb\
.com
/list/(?P
<id>[\da
-zA
-Z_
-]{11}
)'
64 def _real_extract(self, url):
65 mobj = re.match(self._VALID_URL, url)
66 list_id = mobj.group('id')
68 # RSS XML is sometimes malformed
69 rss = self._download_webpage('http
://rss
.imdb
.com
/list/%s' % list_id, list_id, u'Downloading
list RSS
')
70 list_title = self._html_search_regex(r'<title
>(.*?
)</title
>', rss, u'list title
')
72 # Export is independent of actual author_id, but returns 404 if no author_id is provided.
73 # However, passing dummy author_id seems to be enough.
74 csv = self._download_webpage('http
://www
.imdb
.com
/list/export?list_id
=%s&author_id
=ur00000000
' % list_id,
75 list_id, u'Downloading
list CSV
')
78 for item in csv.split('\n')[1:]:
79 cols = item.split(',')
82 item_id = cols[1][1:-1]
83 if item_id.startswith('vi
'):
84 entries.append(self.url_result('http
://www
.imdb
.com
/video
/imdb
/%s' % item_id, 'Imdb
'))
86 return self.playlist_result(entries, list_id, list_title)