]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/googleplus.py
8895ad2897f8abd30471ba3f7ef07963c44e7b57
[youtubedl] / 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 ['title="Timestamp">(.*?)</a>', r'<a.+?class="g-M.+?>(.+?)</a>'],
45 webpage, u'upload date', fatal=False)
46 if upload_date:
47 # Convert timestring to a format suitable for filename
48 upload_date = datetime.datetime.strptime(upload_date, "%Y-%m-%d")
49 upload_date = upload_date.strftime('%Y%m%d')
50
51 # Extract uploader
52 uploader = self._html_search_regex(r'rel\="author".*?>(.*?)</a>',
53 webpage, u'uploader', fatal=False)
54
55 # Extract title
56 # Get the first line for title
57 video_title = self._html_search_regex(r'<meta name\=\"Description\" content\=\"(.*?)[\n<"]',
58 webpage, 'title', default=u'NA')
59
60 # Step 2, Simulate clicking the image box to launch video
61 DOMAIN = 'https://plus.google.com/'
62 video_page = self._search_regex(r'<a href="((?:%s)?photos/.*?)"' % re.escape(DOMAIN),
63 webpage, u'video page URL')
64 if not video_page.startswith(DOMAIN):
65 video_page = DOMAIN + video_page
66
67 webpage = self._download_webpage(video_page, video_id, u'Downloading video page')
68
69 # Extract video links on video page
70 """Extract video links of all sizes"""
71 pattern = r'\d+,\d+,(\d+),"(http\://redirector\.googlevideo\.com.*?)"'
72 mobj = re.findall(pattern, webpage)
73 if len(mobj) == 0:
74 raise ExtractorError(u'Unable to extract video links')
75
76 # Sort in resolution
77 links = sorted(mobj)
78
79 # Choose the lowest of the sort, i.e. highest resolution
80 video_url = links[-1]
81 # Only get the url. The resolution part in the tuple has no use anymore
82 video_url = video_url[-1]
83 # Treat escaped \u0026 style hex
84 try:
85 video_url = video_url.decode("unicode_escape")
86 except AttributeError: # Python 3
87 video_url = bytes(video_url, 'ascii').decode('unicode-escape')
88
89
90 return [{
91 'id': video_id,
92 'url': video_url,
93 'uploader': uploader,
94 'upload_date': upload_date,
95 'title': video_title,
96 'ext': video_extension,
97 }]