youtube_dl/extractor/googleplus.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import codecs
   6
   7 from .common import InfoExtractor
   8 from ..utils import unified_strdate
   9
  10
  11 class GooglePlusIE(InfoExtractor):
  12     IE_DESC = 'Google Plus'
  13     _VALID_URL = r'https?://plus\.google\.com/(?:[^/]+/)*?posts/(?P<id>\w+)'
  14     IE_NAME = 'plus.google'
  15     _TEST = {
  16         'url': 'https://plus.google.com/u/0/108897254135232129896/posts/ZButuJc6CtH',
  17         'info_dict': {
  18             'id': 'ZButuJc6CtH',
  19             'ext': 'flv',
  20             'title': '嘆きの天使 降臨',
  21             'upload_date': '20120613',
  22             'uploader': '井上ヨシマサ',
  23         }
  24     }
  25
  26     def _real_extract(self, url):
  27         video_id = self._match_id(url)
  28
  29         # Step 1, Retrieve post webpage to extract further information
  30         webpage = self._download_webpage(url, video_id, 'Downloading entry webpage')
  31
  32         title = self._og_search_description(webpage).splitlines()[0]
  33         upload_date = unified_strdate(self._html_search_regex(
  34             r'''(?x)<a.+?class="o-U-s\s[^"]+"\s+style="display:\s*none"\s*>
  35                     ([0-9]{4}-[0-9]{2}-[0-9]{2})</a>''',
  36             webpage, 'upload date', fatal=False, flags=re.VERBOSE))
  37         uploader = self._html_search_regex(
  38             r'rel="author".*?>(.*?)</a>', webpage, 'uploader', fatal=False)
  39
  40         # Step 2, Simulate clicking the image box to launch video
  41         DOMAIN = 'https://plus.google.com/'
  42         video_page = self._search_regex(
  43             r'<a href="((?:%s)?photos/.*?)"' % re.escape(DOMAIN),
  44             webpage, 'video page URL')
  45         if not video_page.startswith(DOMAIN):
  46             video_page = DOMAIN + video_page
  47
  48         webpage = self._download_webpage(video_page, video_id, 'Downloading video page')
  49
  50         def unicode_escape(s):
  51             decoder = codecs.getdecoder('unicode_escape')
  52             return re.sub(
  53                 r'\\u[0-9a-fA-F]{4,}',
  54                 lambda m: decoder(m.group(0))[0],
  55                 s)
  56
  57         # Extract video links all sizes
  58         formats = [{
  59             'url': unicode_escape(video_url),
  60             'ext': 'flv',
  61             'width': int(width),
  62             'height': int(height),
  63         } for width, height, video_url in re.findall(
  64             r'\d+,(\d+),(\d+),"(https?://[^.]+\.googleusercontent\.com.*?)"', webpage)]
  65         self._sort_formats(formats)
  66
  67         return {
  68             'id': video_id,
  69             'title': title,
  70             'uploader': uploader,
  71             'upload_date': upload_date,
  72             'formats': formats,
  73         }