youtube-dl/youtube_dl/extractor/ninegag.py

88 lines
3.0 KiB
Python
Raw Normal View History

2014-01-28 09:55:06 -08:00
from __future__ import unicode_literals
2013-12-05 05:29:08 -08:00
import re
2014-04-15 05:49:38 -07:00
import json
2013-12-05 05:29:08 -08:00
from .common import InfoExtractor
2014-04-15 05:49:38 -07:00
from ..utils import str_to_int
2013-12-05 05:29:08 -08:00
class NineGagIE(InfoExtractor):
IE_NAME = '9gag'
_VALID_URL = r'https?://(?:www\.)?9gag(?:\.com/tv|\.tv)/p/(?P<id>[a-zA-Z0-9]+)(?:/(?P<display_id>[^?#/]+))?'
2013-12-05 05:29:08 -08:00
2014-04-10 16:25:24 -07:00
_TESTS = [{
"url": "http://9gag.com/tv/p/Kk2X5/people-are-awesome-2013-is-absolutely-awesome",
2014-01-28 09:55:06 -08:00
"info_dict": {
"id": "Kk2X5",
"ext": "mp4",
2014-01-28 09:55:06 -08:00
"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!)",
2014-03-13 08:40:53 -07:00
"title": "\"People Are Awesome 2013\" Is Absolutely Awesome",
'uploader_id': 'UCdEH6EjDKwtTe-sO2f0_1XA',
'uploader': 'CompilationChannel',
'upload_date': '20131110',
2014-03-13 08:40:53 -07:00
"view_count": int,
"thumbnail": "re:^https?://",
2013-12-05 05:29:08 -08:00
},
2014-01-28 09:55:06 -08:00
'add_ie': ['Youtube']
2014-11-23 12:39:15 -08:00
}, {
2015-09-22 12:41:44 -07:00
'url': 'http://9gag.com/tv/p/KklwM',
'only_matching': True,
}, {
'url': 'http://9gag.tv/p/Kk2X5',
'only_matching': True,
2014-04-10 16:25:24 -07:00
}]
_EXTERNAL_VIDEO_PROVIDER = {
'1': {
'url': '%s',
'ie_key': 'Youtube',
},
'2': {
'url': 'http://player.vimeo.com/video/%s',
'ie_key': 'Vimeo',
},
'3': {
'url': 'http://instagram.com/p/%s',
'ie_key': 'Instagram',
},
'4': {
'url': 'http://vine.co/v/%s',
'ie_key': 'Vine',
},
}
2013-12-05 05:29:08 -08:00
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('id')
2015-09-22 12:40:06 -07:00
display_id = mobj.group('display_id') or video_id
2013-12-05 05:29:08 -08:00
2014-04-10 16:25:24 -07:00
webpage = self._download_webpage(url, display_id)
2013-12-05 05:29:08 -08:00
2014-04-15 05:49:38 -07:00
post_view = json.loads(self._html_search_regex(
r'var\s+postView\s*=\s*new\s+app\.PostView\({\s*post:\s*({.+?})\s*,\s*posts:\s*prefetchedCurrentPost',
webpage, 'post view'))
2014-04-15 05:49:38 -07:00
2015-09-22 11:20:18 -07:00
ie_key = None
source_url = post_view.get('sourceUrl')
2015-09-22 12:28:00 -07:00
if not source_url:
2015-09-22 11:20:18 -07:00
external_video_id = post_view['videoExternalId']
external_video_provider = post_view['videoExternalProvider']
source_url = self._EXTERNAL_VIDEO_PROVIDER[external_video_provider]['url'] % external_video_id
ie_key = self._EXTERNAL_VIDEO_PROVIDER[external_video_provider]['ie_key']
2014-04-15 05:49:38 -07:00
title = post_view['title']
2015-09-22 12:46:40 -07:00
description = post_view.get('description')
view_count = str_to_int(post_view.get('externalView'))
2014-04-15 05:49:38 -07:00
thumbnail = post_view.get('thumbnail_700w') or post_view.get('ogImageUrl') or post_view.get('thumbnail_300w')
2013-12-05 05:29:08 -08:00
return {
'_type': 'url_transparent',
2015-09-22 11:20:18 -07:00
'url': source_url,
'ie_key': ie_key,
2013-12-05 05:29:08 -08:00
'id': video_id,
2014-04-10 16:25:24 -07:00
'display_id': display_id,
'title': title,
2014-03-13 08:40:53 -07:00
'description': description,
'view_count': view_count,
2014-04-15 05:49:38 -07:00
'thumbnail': thumbnail,
2013-12-05 05:29:08 -08:00
}