_ Git - youtube-dl/blob - youtube_dl/extractor/ninegag.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import json
   5
   6 from .common import InfoExtractor
   7 from ..utils import str_to_int
   8
   9
  10 class NineGagIE(InfoExtractor):
  11     IE_NAME = '9gag'
  12     _VALID_URL = r'https?://(?:www\.)?9gag(?:\.com/tv|\.tv)/p/(?P<id>[a-zA-Z0-9]+)(?:/(?P<display_id>[^?#/]+))?'
  13
  14     _TESTS = [{
  15         "url": "http://9gag.com/tv/p/Kk2X5/people-are-awesome-2013-is-absolutely-awesome",
  16         "info_dict": {
  17             "id": "Kk2X5",
  18             "ext": "mp4",
  19             "description": "This 3-minute video will make you smile and then make you feel untalented and insignificant. Anyway, you should share this awesomeness. (Thanks, Dino!)",
  20             "title": "\"People Are Awesome 2013\" Is Absolutely Awesome",
  21             'uploader_id': 'UCdEH6EjDKwtTe-sO2f0_1XA',
  22             'uploader': 'CompilationChannel',
  23             'upload_date': '20131110',
  24             "view_count": int,
  25             "thumbnail": "re:^https?://",
  26         },
  27         'add_ie': ['Youtube']
  28     }, {
  29         'url': 'http://9gag.com/tv/p/KklwM',
  30         'only_matching': True,
  31     }, {
  32         'url': 'http://9gag.tv/p/Kk2X5',
  33         'only_matching': True,
  34     }]
  35     _EXTERNAL_VIDEO_PROVIDER = {
  36         '1': {
  37             'url': '%s',
  38             'ie_key': 'Youtube',
  39         },
  40         '2': {
  41             'url': 'http://player.vimeo.com/video/%s',
  42             'ie_key': 'Vimeo',
  43         },
  44         '3': {
  45             'url': 'http://instagram.com/p/%s',
  46             'ie_key': 'Instagram',
  47         },
  48         '4': {
  49             'url': 'http://vine.co/v/%s',
  50             'ie_key': 'Vine',
  51         },
  52     }
  53
  54     def _real_extract(self, url):
  55         mobj = re.match(self._VALID_URL, url)
  56         video_id = mobj.group('id')
  57         display_id = mobj.group('display_id') or video_id
  58
  59         webpage = self._download_webpage(url, display_id)
  60
  61         post_view = self._parse_json(
  62             self._search_regex(
  63                 r'var\s+postView\s*=\s*new\s+app\.PostView\({\s*post:\s*({.+?})\s*,\s*posts:\s*prefetchedCurrentPost',
  64                 webpage, 'post view'),
  65             display_id)
  66
  67         ie_key = None
  68         source_url = post_view.get('sourceUrl')
  69         if not source_url:
  70             external_video_id = post_view['videoExternalId']
  71             external_video_provider = post_view['videoExternalProvider']
  72             source_url = self._EXTERNAL_VIDEO_PROVIDER[external_video_provider]['url'] % external_video_id
  73             ie_key = self._EXTERNAL_VIDEO_PROVIDER[external_video_provider]['ie_key']
  74         title = post_view['title']
  75         description = post_view.get('description')
  76         view_count = str_to_int(post_view.get('externalView'))
  77         thumbnail = post_view.get('thumbnail_700w') or post_view.get('ogImageUrl') or post_view.get('thumbnail_300w')
  78
  79         return {
  80             '_type': 'url_transparent',
  81             'url': source_url,
  82             'ie_key': ie_key,
  83             'id': video_id,
  84             'display_id': display_id,
  85             'title': title,
  86             'description': description,
  87             'view_count': view_count,
  88             'thumbnail': thumbnail,
  89         }