yt-dlp/youtube_dl/extractor/twitter.py

167 lines
5.9 KiB
Python
Raw Normal View History

2015-10-18 11:07:48 +02:00
# coding: utf-8
2015-06-27 21:22:25 +02:00
from __future__ import unicode_literals
import re
from .common import InfoExtractor
from ..compat import compat_urllib_request
from ..utils import (
float_or_none,
2015-10-18 11:11:55 +02:00
xpath_text,
2015-10-18 12:04:13 +02:00
remove_end,
2015-06-27 21:22:25 +02:00
)
class TwitterCardIE(InfoExtractor):
2015-10-18 11:13:58 +02:00
IE_NAME = 'twitter:card'
2015-06-27 21:22:25 +02:00
_VALID_URL = r'https?://(?:www\.)?twitter\.com/i/cards/tfw/v1/(?P<id>\d+)'
2015-07-21 23:45:36 +02:00
_TESTS = [
{
'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889',
2015-10-18 11:07:48 +02:00
'md5': '7d2f6b4d2eb841a7ccc893d479bfceb4',
2015-07-21 23:45:36 +02:00
'info_dict': {
'id': '560070183650213889',
'ext': 'mp4',
'title': 'TwitterCard',
'thumbnail': 're:^https?://.*\.jpg$',
'duration': 30.033,
}
2015-06-27 21:22:25 +02:00
},
2015-07-21 23:45:36 +02:00
{
'url': 'https://twitter.com/i/cards/tfw/v1/623160978427936768',
'md5': '7ee2a553b63d1bccba97fbed97d9e1c8',
'info_dict': {
'id': '623160978427936768',
'ext': 'mp4',
'title': 'TwitterCard',
'thumbnail': 're:^https?://.*\.jpg',
'duration': 80.155,
},
2015-10-18 13:07:37 +02:00
},
{
'url': 'https://twitter.com/i/cards/tfw/v1/654001591733886977',
'md5': 'b6f35e8b08a0bec6c8af77a2f4b3a814',
'info_dict': {
'id': 'dq4Oj5quskI',
'ext': 'mp4',
'title': 'Ubuntu 11.10 Overview',
'description': 'Take a quick peek at what\'s new and improved in Ubuntu 11.10.\n\nOnce installed take a look at 10 Things to Do After Installing: http://www.omgubuntu.co.uk/2011/10/10-things-to-do-after-installing-ubuntu-11-10/',
'upload_date': '20111013',
'uploader': 'OMG! Ubuntu!',
'uploader_id': 'omgubuntu',
},
2015-07-21 23:45:36 +02:00
}
]
2015-06-27 21:22:25 +02:00
def _real_extract(self, url):
video_id = self._match_id(url)
# Different formats served for different User-Agents
USER_AGENTS = [
'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20150101 Firefox/20.0 (Chrome)', # mp4
'Mozilla/5.0 (Windows NT 5.2; WOW64; rv:38.0) Gecko/20100101 Firefox/38.0', # webm
]
config = None
formats = []
for user_agent in USER_AGENTS:
request = compat_urllib_request.Request(url)
request.add_header('User-Agent', user_agent)
webpage = self._download_webpage(request, video_id)
2015-10-18 13:07:37 +02:00
youtube_url = self._html_search_regex(
r'<iframe[^>]+src="((?:https?:)?//www.youtube.com/embed/[^"]+)"',
webpage, 'youtube iframe', default=None)
if youtube_url:
return self.url_result(youtube_url, 'Youtube')
2015-10-18 12:08:24 +02:00
config = self._parse_json(self._html_search_regex(
r'data-player-config="([^"]+)"', webpage, 'data player config'),
2015-06-27 21:22:25 +02:00
video_id)
2015-07-21 23:45:36 +02:00
if 'playlist' not in config:
if 'vmapUrl' in config:
2015-10-18 11:11:55 +02:00
vmap_data = self._download_xml(config['vmapUrl'], video_id)
video_url = xpath_text(vmap_data, './/MediaFile').strip()
2015-10-18 11:15:47 +02:00
formats.append({
2015-07-21 23:45:36 +02:00
'url': video_url,
2015-10-18 11:15:47 +02:00
})
2015-07-21 23:45:36 +02:00
break # same video regardless of UA
continue
2015-06-27 21:22:25 +02:00
video_url = config['playlist'][0]['source']
f = {
'url': video_url,
}
m = re.search(r'/(?P<width>\d+)x(?P<height>\d+)/', video_url)
if m:
f.update({
'width': int(m.group('width')),
'height': int(m.group('height')),
})
formats.append(f)
self._sort_formats(formats)
thumbnail = config.get('posterImageUrl')
duration = float_or_none(config.get('duration'))
return {
'id': video_id,
'title': 'TwitterCard',
'thumbnail': thumbnail,
'duration': duration,
'formats': formats,
}
class TwitterIE(InfoExtractor):
2015-10-18 11:13:58 +02:00
IE_NAME = 'twitter'
2015-10-18 12:04:13 +02:00
_VALID_URL = r'https?://(?:www\.|m\.|mobile\.)?twitter\.com/(?P<user_id>[^/]+)/status/(?P<id>\d+)'
_TEMPLATE_URL = 'https://twitter.com/%s/status/%s'
_TEST = {
2015-10-18 11:07:48 +02:00
'url': 'https://twitter.com/freethenipple/status/643211948184596480',
'md5': '31cd83a116fc41f99ae3d909d4caf6a0',
'info_dict': {
2015-10-18 11:07:48 +02:00
'id': '643211948184596480',
'ext': 'mp4',
2015-10-18 12:04:13 +02:00
'title': 'FREE THE NIPPLE - FTN supporters on Hollywood Blvd today!',
'thumbnail': 're:^https?://.*\.jpg',
2015-10-18 11:07:48 +02:00
'duration': 12.922,
'description': 'FREE THE NIPPLE on Twitter: "FTN supporters on Hollywood Blvd today! http://t.co/c7jHH749xJ"',
'uploader': 'FREE THE NIPPLE',
'uploader_id': 'freethenipple',
},
}
def _real_extract(self, url):
2015-10-18 12:04:13 +02:00
mobj = re.match(self._VALID_URL, url)
user_id = mobj.group('user_id')
twid = mobj.group('id')
webpage = self._download_webpage(self._TEMPLATE_URL % (user_id, twid), twid)
username = remove_end(self._og_search_title(webpage), ' on Twitter')
title = self._og_search_description(webpage).strip('').replace('\n', ' ')
# strip 'https -_t.co_BJYgOjSeGA' junk from filenames
2015-10-18 12:23:56 +02:00
mobj = re.match(r'“(.*)\s+(https?://[^ ]+)”', title)
2015-10-18 12:04:13 +02:00
title, short_url = mobj.groups()
card_id = self._search_regex(
r'["\']/i/cards/tfw/v1/(\d+)', webpage, 'twitter card url')
card_url = 'https://twitter.com/i/cards/tfw/v1/' + card_id
2015-10-18 12:04:13 +02:00
return {
'_type': 'url_transparent',
'ie_key': 'TwitterCard',
2015-10-18 12:04:13 +02:00
'uploader_id': user_id,
'uploader': username,
'url': card_url,
'webpage_url': url,
2015-10-18 12:04:13 +02:00
'description': '%s on Twitter: "%s %s"' % (username, title, short_url),
'title': username + ' - ' + title,
}