- pattern = r'<title>(.+?)</title>'
- compiled = re.compile(pattern, re.DOTALL)
- title = self._search_regex(compiled, webpage, u'video title')
- title = clean_html(title).split('-')[0].strip()
- self.to_screen('Title: %s' % title)
- pattern = re.compile(r'var vid="(\d+)"')
- result = re.search(pattern, webpage)
- if not result:
- logging.info('[Sohu] could not get vid')
- return None
- vid = result.group(1)
- logging.info('vid: %s' % vid)
- base_url_1 = 'http://hot.vrs.sohu.com/vrs_flash.action?vid='
- url_1 = base_url_1 + vid
- logging.info('json url: %s' % url_1)
- webpage = self._download_webpage(url_1, vid)
- json_1 = json.loads(webpage)
- # get the highest definition video vid and json infomation.
- vids = []
- qualities = ('oriVid', 'superVid', 'highVid', 'norVid')
- for vid_name in qualities:
- vids.append(json_1['data'][vid_name])
- clearest_vid = 0
- for i, v in enumerate(vids):
- if v != 0:
- clearest_vid = v
- logging.info('quality definition: %s' % qualities[i][:-3])
- break
- if not clearest_vid:
- logging.warning('could not find valid clearest_vid')
- return None
- if vid != clearest_vid:
- url_1 = '%s%d' % (base_url_1, clearest_vid)
- logging.info('highest definition json url: %s' % url_1)
- json_1 = json.loads(urllib2.urlopen(url_1).read())
- allot = json_1['allot']
- prot = json_1['prot']
- clipsURL = json_1['data']['clipsURL']
- su = json_1['data']['su']
- num_of_parts = json_1['data']['totalBlocks']
- logging.info('Total parts: %d' % num_of_parts)
- base_url_3 = 'http://allot/?prot=prot&file=clipsURL[i]&new=su[i]'
- files_info = []
- for i in range(num_of_parts):
- self.to_screen('Geting json infomation of part %s/%s' % (i + 1, num_of_parts))
- middle_url = 'http://%s/?prot=%s&file=%s&new=%s' % (allot, prot, clipsURL[i], su[i])
- logging.info('middle url part %d: %s' % (i, middle_url))
- middle_info = urllib2.urlopen(middle_url).read().split('|')
- middle_part_1 = middle_info[0]
- download_url = '%s%s?key=%s' % (middle_info[0], su[i], middle_info[3])