Merge pull request #4110 from nemunaire/channel9-fix

[channel9] Fix extraction
This commit is contained in:
Jaime Marquínez Ferrándiz 2014-11-05 19:03:24 +01:00
commit cde9b380e6

View file

@ -27,7 +27,7 @@ class Channel9IE(InfoExtractor):
'title': 'Developer Kick-Off Session: Stuff We Love',
'description': 'md5:c08d72240b7c87fcecafe2692f80e35f',
'duration': 4576,
'thumbnail': 'http://media.ch9.ms/ch9/9d51/03902f2d-fc97-4d3c-b195-0bfe15a19d51/KOS002_220.jpg',
'thumbnail': 'http://video.ch9.ms/ch9/9d51/03902f2d-fc97-4d3c-b195-0bfe15a19d51/KOS002_220.jpg',
'session_code': 'KOS002',
'session_day': 'Day 1',
'session_room': 'Arena 1A',
@ -43,7 +43,7 @@ class Channel9IE(InfoExtractor):
'title': 'Self-service BI with Power BI - nuclear testing',
'description': 'md5:d1e6ecaafa7fb52a2cacdf9599829f5b',
'duration': 1540,
'thumbnail': 'http://media.ch9.ms/ch9/87e1/0300391f-a455-4c72-bec3-4422f19287e1/selfservicenuk_512.jpg',
'thumbnail': 'http://video.ch9.ms/ch9/87e1/0300391f-a455-4c72-bec3-4422f19287e1/selfservicenuk_512.jpg',
'authors': [ 'Mike Wilmot' ],
},
}
@ -94,7 +94,7 @@ def _formats_from_html(self, html):
def _extract_title(self, html):
title = self._html_search_meta('title', html, 'title')
if title is None:
if title is None:
title = self._og_search_title(html)
TITLE_SUFFIX = ' (Channel 9)'
if title is not None and title.endswith(TITLE_SUFFIX):
@ -115,7 +115,7 @@ def _extract_description(self, html):
return self._html_search_meta('description', html, 'description')
def _extract_duration(self, html):
m = re.search(r'data-video_duration="(?P<hours>\d{2}):(?P<minutes>\d{2}):(?P<seconds>\d{2})"', html)
m = re.search(r'"length": *"(?P<hours>\d{2}):(?P<minutes>\d{2}):(?P<seconds>\d{2})"', html)
return ((int(m.group('hours')) * 60 * 60) + (int(m.group('minutes')) * 60) + int(m.group('seconds'))) if m else None
def _extract_slides(self, html):
@ -167,7 +167,7 @@ def _extract_session_speakers(self, html):
return re.findall(r'<a href="/Events/Speakers/[^"]+">([^<]+)</a>', html)
def _extract_content(self, html, content_path):
# Look for downloadable content
# Look for downloadable content
formats = self._formats_from_html(html)
slides = self._extract_slides(html)
zip_ = self._extract_zip(html)
@ -258,16 +258,17 @@ def _real_extract(self, url):
webpage = self._download_webpage(url, content_path, 'Downloading web page')
page_type_m = re.search(r'<meta name="Search.PageType" content="(?P<pagetype>[^"]+)"/>', webpage)
if page_type_m is None:
raise ExtractorError('Search.PageType not found, don\'t know how to process this page', expected=True)
page_type_m = re.search(r'<meta name="WT.entryid" content="(?P<pagetype>[^:]+)[^"]+"/>', webpage)
if page_type_m is not None:
page_type = page_type_m.group('pagetype')
if page_type == 'Entry': # Any 'item'-like page, may contain downloadable content
return self._extract_entry_item(webpage, content_path)
elif page_type == 'Session': # Event session page, may contain downloadable content
return self._extract_session(webpage, content_path)
elif page_type == 'Event':
return self._extract_list(content_path)
else:
raise ExtractorError('Unexpected WT.entryid %s' % page_type, expected=True)
page_type = page_type_m.group('pagetype')
if page_type == 'List': # List page, may contain list of 'item'-like objects
else: # Assuming list
return self._extract_list(content_path)
elif page_type == 'Entry.Item': # Any 'item'-like page, may contain downloadable content
return self._extract_entry_item(webpage, content_path)
elif page_type == 'Session': # Event session page, may contain downloadable content
return self._extract_session(webpage, content_path)
else:
raise ExtractorError('Unexpected Search.PageType %s' % page_type, expected=True)