[infoq] Fix extractor

This commit is contained in:
Kai Weber 2014-04-20 01:01:37 +02:00
parent e6c6d10d99
commit 282cb9c7ba
1 changed files with 23 additions and 24 deletions

View File

@ -11,16 +11,16 @@ from ..utils import (
class InfoQIE(InfoExtractor): class InfoQIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?infoq\.com/[^/]+/(?P<id>[^/]+)$' _VALID_URL = r'https?://(?:www\.)?infoq\.com/[^/]+/(?P<id>[^/]+)$'
_TEST = { _TEST = {
"name": "InfoQ", u'name': u'InfoQ',
"url": "http://www.infoq.com/presentations/A-Few-of-My-Favorite-Python-Things", u'url': u'http://www.infoq.com/presentations/A-Few-of-My-Favorite-Python-Things',
"file": "12-jan-pythonthings.mp4", u'md5': u'fcaa3d995e04080dcb9465d86b5eef62',
"info_dict": { u'info_dict': {
"description": "Mike Pirnat presents some tips and tricks, standard libraries and third party packages that make programming in Python a richer experience.", u'id': u'12-jan-pythonthings',
"title": "A Few of My Favorite [Python] Things", u'ext': u'mp4',
}, u'description': u'Mike Pirnat presents some tips and tricks, standard libraries and third party packages that make programming in Python a richer experience.',
"params": { u'title': u'A Few of My Favorite [Python] Things',
"skip_download": True,
}, },
} }
@ -30,26 +30,25 @@ class InfoQIE(InfoExtractor):
webpage = self._download_webpage(url, video_id) webpage = self._download_webpage(url, video_id)
# Extract video URL self.report_extraction(video_id)
encoded_id = self._search_regex(r"jsclassref ?= ?'([^']*)'", webpage, 'encoded id')
real_id = compat_urllib_parse.unquote(base64.b64decode(encoded_id.encode('ascii')).decode('utf-8'))
video_url = 'rtmpe://video.infoq.com/cfx/st/' + real_id
# Extract title video_title = self._html_search_regex(r'<title>(.*?)</title>', webpage, 'title')
video_title = self._search_regex(r'contentTitle = "(.*?)";', video_description = self._html_search_meta('description', webpage, 'description')
webpage, 'title')
# Extract description video_url = 'rtmpe://video.infoq.com/cfx/st/'
video_description = self._html_search_regex(r'<meta name="description" content="(.*)"(?:\s*/)?>', base64playpath = self._search_regex(r"jsclassref = '([^']*)'", webpage, 'jsclassref')
webpage, 'description', fatal=False) playpath = 'mp4:' + base64.b64decode(base64playpath).decode('utf-8')
video_filename = video_url.split('/')[-1] video_filename = playpath.split('/')[-1]
video_id, extension = video_filename.split('.') video_id, extension = video_filename.split('.')
return { return [{
'id': video_id, 'id': video_id,
'url': video_url,
'title': video_title, 'title': video_title,
'ext': extension, # Extension is always(?) mp4, but seems to be flv
'description': video_description, 'description': video_description,
} 'formats': [{
'url': video_url,
'ext': extension,
'play_path': playpath,
}],
}]