youtube_dl/extractor/googleplus.py

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