youtube_dl/extractor/googleplus.py

   1 # coding: utf-8
   2
   3 import datetime
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     ExtractorError,
   9 )
  10
  11
  12 class GooglePlusIE(InfoExtractor):
  13     IE_DESC = u'Google Plus'
  14     _VALID_URL = r'(?:https://)?plus\.google\.com/(?:[^/]+/)*?posts/(\w+)'
  15     IE_NAME = u'plus.google'
  16     _TEST = {
  17         u"url": u"https://plus.google.com/u/0/108897254135232129896/posts/ZButuJc6CtH",
  18         u"file": u"ZButuJc6CtH.flv",
  19         u"info_dict": {
  20             u"upload_date": u"20120613",
  21             u"uploader": u"井上ヨシマサ",
  22             u"title": u"嘆きの天使 降臨"
  23         }
  24     }
  25
  26     def _real_extract(self, url):
  27         # Extract id from URL
  28         mobj = re.match(self._VALID_URL, url)
  29         if mobj is None:
  30             raise ExtractorError(u'Invalid URL: %s' % url)
  31
  32         post_url = mobj.group(0)
  33         video_id = mobj.group(1)
  34
  35         video_extension = 'flv'
  36
  37         # Step 1, Retrieve post webpage to extract further information
  38         webpage = self._download_webpage(post_url, video_id, u'Downloading entry webpage')
  39
  40         self.report_extraction(video_id)
  41
  42         # Extract update date
  43         upload_date = self._html_search_regex(
  44             r'''(?x)<a.+?class="o-U-s\s[^"]+"\s+style="display:\s*none"\s*>
  45                     ([0-9]{4}-[0-9]{2}-[0-9]{2})</a>''',
  46             webpage, u'upload date', fatal=False, flags=re.VERBOSE)
  47         if upload_date:
  48             # Convert timestring to a format suitable for filename
  49             upload_date = datetime.datetime.strptime(upload_date, "%Y-%m-%d")
  50             upload_date = upload_date.strftime('%Y%m%d')
  51
  52         # Extract uploader
  53         uploader = self._html_search_regex(r'rel\="author".*?>(.*?)</a>',
  54             webpage, u'uploader', fatal=False)
  55
  56         # Extract title
  57         # Get the first line for title
  58         video_title = self._html_search_regex(r'<meta name\=\"Description\" content\=\"(.*?)[\n<"]',
  59             webpage, 'title', default=u'NA')
  60
  61         # Step 2, Simulate clicking the image box to launch video
  62         DOMAIN = 'https://plus.google.com/'
  63         video_page = self._search_regex(r'<a href="((?:%s)?photos/.*?)"' % re.escape(DOMAIN),
  64             webpage, u'video page URL')
  65         if not video_page.startswith(DOMAIN):
  66             video_page = DOMAIN + video_page
  67
  68         webpage = self._download_webpage(video_page, video_id, u'Downloading video page')
  69
  70         # Extract video links on video page
  71         """Extract video links of all sizes"""
  72         pattern = r'\d+,\d+,(\d+),"(http\://redirector\.googlevideo\.com.*?)"'
  73         mobj = re.findall(pattern, webpage)
  74         if len(mobj) == 0:
  75             raise ExtractorError(u'Unable to extract video links')
  76
  77         # Sort in resolution
  78         links = sorted(mobj)
  79
  80         # Choose the lowest of the sort, i.e. highest resolution
  81         video_url = links[-1]
  82         # Only get the url. The resolution part in the tuple has no use anymore
  83         video_url = video_url[-1]
  84         # Treat escaped \u0026 style hex
  85         try:
  86             video_url = video_url.decode("unicode_escape")
  87         except AttributeError: # Python 3
  88             video_url = bytes(video_url, 'ascii').decode('unicode-escape')
  89
  90
  91         return [{
  92             'id':       video_id,
  93             'url':      video_url,
  94             'uploader': uploader,
  95             'upload_date':  upload_date,
  96             'title':    video_title,
  97             'ext':      video_extension,
  98         }]