yt-dlp/youtube_dl/extractor/smotri.py

405 lines
17 KiB
Python
Raw Normal View History

2013-12-02 14:10:19 +01:00
# encoding: utf-8
from __future__ import unicode_literals
2013-12-02 14:10:19 +01:00
import os.path
2013-12-02 14:10:19 +01:00
import re
import json
import hashlib
import uuid
2013-12-02 14:10:19 +01:00
from .common import InfoExtractor
from ..utils import (
compat_urllib_parse,
compat_urllib_request,
ExtractorError,
url_basename,
int_or_none,
2013-12-02 14:10:19 +01:00
)
class SmotriIE(InfoExtractor):
IE_DESC = 'Smotri.com'
IE_NAME = 'smotri'
_VALID_URL = r'^https?://(?:www\.)?(?:smotri\.com/video/view/\?id=|pics\.smotri\.com/(?:player|scrubber_custom8)\.swf\?file=)(?P<videoid>v(?P<realvideoid>[0-9]+)[a-z0-9]{4})'
_NETRC_MACHINE = 'smotri'
2013-12-02 17:08:17 +01:00
2013-12-02 14:10:19 +01:00
_TESTS = [
# real video id 2610366
{
'url': 'http://smotri.com/video/view/?id=v261036632ab',
'md5': '2a7b08249e6f5636557579c368040eb9',
'info_dict': {
'id': 'v261036632ab',
'ext': 'mp4',
'title': 'катастрофа с камер видеонаблюдения',
'uploader': 'rbc2008',
'uploader_id': 'rbc08',
'upload_date': '20131118',
'description': 'катастрофа с камер видеонаблюдения, видео катастрофа с камер видеонаблюдения',
'thumbnail': 'http://frame6.loadup.ru/8b/a9/2610366.3.3.jpg',
2013-12-02 14:10:19 +01:00
},
},
# real video id 57591
{
'url': 'http://smotri.com/video/view/?id=v57591cb20',
'md5': '830266dfc21f077eac5afd1883091bcd',
'info_dict': {
'id': 'v57591cb20',
'ext': 'flv',
'title': 'test',
'uploader': 'Support Photofile@photofile',
'uploader_id': 'support-photofile',
'upload_date': '20070704',
'description': 'test, видео test',
'thumbnail': 'http://frame4.loadup.ru/03/ed/57591.2.3.jpg',
2013-12-02 17:08:17 +01:00
},
2013-12-02 14:10:19 +01:00
},
# video-password
{
'url': 'http://smotri.com/video/view/?id=v1390466a13c',
'md5': 'f6331cef33cad65a0815ee482a54440b',
'info_dict': {
'id': 'v1390466a13c',
'ext': 'mp4',
'title': 'TOCCA_A_NOI_-_LE_COSE_NON_VANNO_CAMBIAMOLE_ORA-1',
'uploader': 'timoxa40',
'uploader_id': 'timoxa40',
'upload_date': '20100404',
'thumbnail': 'http://frame7.loadup.ru/af/3f/1390466.3.3.jpg',
'description': 'TOCCA_A_NOI_-_LE_COSE_NON_VANNO_CAMBIAMOLE_ORA-1, видео TOCCA_A_NOI_-_LE_COSE_NON_VANNO_CAMBIAMOLE_ORA-1',
2013-12-02 14:10:19 +01:00
},
'params': {
'videopassword': 'qwerty',
2013-12-02 14:10:19 +01:00
},
},
# age limit + video-password
{
'url': 'http://smotri.com/video/view/?id=v15408898bcf',
'md5': '91e909c9f0521adf5ee86fbe073aad70',
'info_dict': {
'id': 'v15408898bcf',
'ext': 'flv',
'title': 'этот ролик не покажут по ТВ',
'uploader': 'zzxxx',
'uploader_id': 'ueggb',
'upload_date': '20101001',
'thumbnail': 'http://frame3.loadup.ru/75/75/1540889.1.3.jpg',
'age_limit': 18,
'description': 'этот ролик не покажут по ТВ, видео этот ролик не покажут по ТВ',
2013-12-02 17:08:17 +01:00
},
'params': {
'videopassword': '333'
2013-12-02 14:10:19 +01:00
}
},
# swf player
{
'url': 'http://pics.smotri.com/scrubber_custom8.swf?file=v9188090500',
'md5': '4d47034979d9390d14acdf59c4935bc2',
'info_dict': {
'id': 'v9188090500',
'ext': 'mp4',
'title': 'Shakira - Don\'t Bother',
'uploader': 'HannahL',
'uploader_id': 'lisaha95',
'upload_date': '20090331',
'description': 'Shakira - Don\'t Bother, видео Shakira - Don\'t Bother',
'thumbnail': 'http://frame8.loadup.ru/44/0b/918809.7.3.jpg',
},
},
2013-12-02 14:10:19 +01:00
]
2013-12-02 14:10:19 +01:00
_SUCCESS = 0
_PASSWORD_NOT_VERIFIED = 1
_PASSWORD_DETECTED = 2
_VIDEO_NOT_FOUND = 3
2013-12-02 17:08:17 +01:00
@classmethod
def _extract_url(cls, webpage):
mobj = re.search(
r'<embed[^>]src=(["\'])(?P<url>http://pics\.smotri\.com/(?:player|scrubber_custom8)\.swf\?file=v.+?\1)',
webpage)
if mobj is not None:
return mobj.group('url')
mobj = re.search(
r'''(?x)<div\s+class="video_file">http://smotri\.com/video/download/file/[^<]+</div>\s*
<div\s+class="video_image">[^<]+</div>\s*
<div\s+class="video_id">(?P<id>[^<]+)</div>''', webpage)
if mobj is not None:
return 'http://smotri.com/video/view/?id=%s' % mobj.group('id')
2013-12-02 14:10:19 +01:00
def _search_meta(self, name, html, display_name=None):
if display_name is None:
display_name = name
return self._html_search_regex(
r'<meta itemprop="%s" content="([^"]+)" />' % re.escape(name),
html, display_name, fatal=False)
2013-12-02 17:08:17 +01:00
return self._html_search_meta(name, html, display_name)
2013-12-02 14:10:19 +01:00
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('videoid')
real_video_id = mobj.group('realvideoid')
# Download video JSON data
video_json_url = 'http://smotri.com/vt.php?id=%s' % real_video_id
video_json_page = self._download_webpage(video_json_url, video_id, 'Downloading video JSON')
2013-12-02 14:10:19 +01:00
video_json = json.loads(video_json_page)
2013-12-02 14:10:19 +01:00
status = video_json['status']
if status == self._VIDEO_NOT_FOUND:
raise ExtractorError('Video %s does not exist' % video_id, expected=True)
elif status == self._PASSWORD_DETECTED: # The video is protected by a password, retry with
2013-12-02 14:10:19 +01:00
# video-password set
video_password = self._downloader.params.get('videopassword', None)
if not video_password:
raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True)
2013-12-02 17:08:17 +01:00
video_json_url += '&md5pass=%s' % hashlib.md5(video_password.encode('utf-8')).hexdigest()
video_json_page = self._download_webpage(video_json_url, video_id, 'Downloading video JSON (video-password set)')
2013-12-02 14:10:19 +01:00
video_json = json.loads(video_json_page)
status = video_json['status']
if status == self._PASSWORD_NOT_VERIFIED:
raise ExtractorError('Video password is invalid', expected=True)
2013-12-02 14:10:19 +01:00
if status != self._SUCCESS:
raise ExtractorError('Unexpected status value %s' % status)
2013-12-02 14:10:19 +01:00
# Extract the URL of the video
video_url = video_json['file_data']
2013-12-02 14:10:19 +01:00
# Video JSON does not provide enough meta data
# We will extract some from the video web page instead
video_page_url = 'http://smotri.com/video/view/?id=%s' % video_id
video_page = self._download_webpage(video_page_url, video_id, 'Downloading video page')
# Warning if video is unavailable
warning = self._html_search_regex(
r'<div class="videoUnModer">(.*?)</div>', video_page,
'warning message', default=None)
if warning is not None:
self._downloader.report_warning(
'Video %s may not be available; smotri said: %s ' %
(video_id, warning))
2013-12-02 14:10:19 +01:00
# Adult content
if re.search('EroConfirmText">', video_page) is not None:
2013-12-02 14:10:19 +01:00
self.report_age_confirmation()
confirm_string = self._html_search_regex(
2013-12-02 17:08:17 +01:00
r'<a href="/video/view/\?id=%s&confirm=([^"]+)" title="[^"]+">' % video_id,
video_page, 'confirm string')
2013-12-02 14:10:19 +01:00
confirm_url = video_page_url + '&confirm=%s' % confirm_string
video_page = self._download_webpage(confirm_url, video_id, 'Downloading video page (age confirmed)')
2013-12-02 14:10:19 +01:00
adult_content = True
else:
adult_content = False
2013-12-02 14:10:19 +01:00
# Extract the rest of meta data
video_title = self._search_meta('name', video_page, 'title')
2013-12-02 14:10:19 +01:00
if not video_title:
video_title = os.path.splitext(url_basename(video_url))[0]
2013-12-02 17:08:17 +01:00
video_description = self._search_meta('description', video_page)
END_TEXT = ' на сайте Smotri.com'
if video_description and video_description.endswith(END_TEXT):
2013-12-02 17:08:17 +01:00
video_description = video_description[:-len(END_TEXT)]
START_TEXT = 'Смотреть онлайн ролик '
if video_description and video_description.startswith(START_TEXT):
2013-12-02 17:08:17 +01:00
video_description = video_description[len(START_TEXT):]
video_thumbnail = self._search_meta('thumbnail', video_page)
2013-12-02 17:08:17 +01:00
upload_date_str = self._search_meta('uploadDate', video_page, 'upload date')
if upload_date_str:
upload_date_m = re.search(r'(?P<year>\d{4})\.(?P<month>\d{2})\.(?P<day>\d{2})T', upload_date_str)
video_upload_date = (
(
upload_date_m.group('year') +
upload_date_m.group('month') +
upload_date_m.group('day')
)
if upload_date_m else None
2013-12-02 14:10:19 +01:00
)
else:
video_upload_date = None
duration_str = self._search_meta('duration', video_page)
if duration_str:
duration_m = re.search(r'T(?P<hours>[0-9]{2})H(?P<minutes>[0-9]{2})M(?P<seconds>[0-9]{2})S', duration_str)
video_duration = (
(
(int(duration_m.group('hours')) * 60 * 60) +
(int(duration_m.group('minutes')) * 60) +
int(duration_m.group('seconds'))
)
if duration_m else None
2013-12-02 14:10:19 +01:00
)
else:
video_duration = None
2013-12-02 14:10:19 +01:00
video_uploader = self._html_search_regex(
'<div class="DescrUser"><div>Автор.*?onmouseover="popup_user_info[^"]+">(.*?)</a>',
video_page, 'uploader', fatal=False, flags=re.MULTILINE|re.DOTALL)
2013-12-02 14:10:19 +01:00
video_uploader_id = self._html_search_regex(
'<div class="DescrUser"><div>Автор.*?onmouseover="popup_user_info\\(.*?\'([^\']+)\'\\);">',
video_page, 'uploader id', fatal=False, flags=re.MULTILINE|re.DOTALL)
2013-12-02 14:10:19 +01:00
video_view_count = self._html_search_regex(
'Общее количество просмотров.*?<span class="Number">(\\d+)</span>',
video_page, 'view count', fatal=False, flags=re.MULTILINE|re.DOTALL)
2013-12-02 14:10:19 +01:00
return {
'id': video_id,
'url': video_url,
'title': video_title,
'thumbnail': video_thumbnail,
'description': video_description,
'uploader': video_uploader,
'upload_date': video_upload_date,
'uploader_id': video_uploader_id,
2013-12-18 23:56:48 +01:00
'duration': video_duration,
'view_count': int_or_none(video_view_count),
2013-12-02 14:10:19 +01:00
'age_limit': 18 if adult_content else 0,
'video_page_url': video_page_url
}
2013-12-02 17:08:17 +01:00
2013-12-02 14:10:19 +01:00
class SmotriCommunityIE(InfoExtractor):
IE_DESC = 'Smotri.com community videos'
IE_NAME = 'smotri:community'
2013-12-02 17:08:17 +01:00
_VALID_URL = r'^https?://(?:www\.)?smotri\.com/community/video/(?P<communityid>[0-9A-Za-z_\'-]+)'
2013-12-02 14:10:19 +01:00
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
community_id = mobj.group('communityid')
2013-12-02 17:08:17 +01:00
2013-12-02 14:10:19 +01:00
url = 'http://smotri.com/export/rss/video/by/community/-/%s/video.xml' % community_id
rss = self._download_xml(url, community_id, 'Downloading community RSS')
2013-12-02 17:08:17 +01:00
2013-12-02 14:10:19 +01:00
entries = [self.url_result(video_url.text, 'Smotri')
for video_url in rss.findall('./channel/item/link')]
2013-12-02 17:08:17 +01:00
description_text = rss.find('./channel/description').text
2013-12-02 14:10:19 +01:00
community_title = self._html_search_regex(
'^Видео сообщества "([^"]+)"$', description_text, 'community title')
2013-12-02 14:10:19 +01:00
return self.playlist_result(entries, community_id, community_title)
2013-12-02 17:08:17 +01:00
2013-12-02 14:10:19 +01:00
class SmotriUserIE(InfoExtractor):
IE_DESC = 'Smotri.com user videos'
IE_NAME = 'smotri:user'
2013-12-02 17:08:17 +01:00
_VALID_URL = r'^https?://(?:www\.)?smotri\.com/user/(?P<userid>[0-9A-Za-z_\'-]+)'
2013-12-02 14:10:19 +01:00
def _real_extract(self, url):
2013-12-02 17:08:17 +01:00
mobj = re.match(self._VALID_URL, url)
2013-12-02 14:10:19 +01:00
user_id = mobj.group('userid')
2013-12-02 17:08:17 +01:00
2013-12-02 14:10:19 +01:00
url = 'http://smotri.com/export/rss/user/video/-/%s/video.xml' % user_id
rss = self._download_xml(url, user_id, 'Downloading user RSS')
2013-12-02 17:08:17 +01:00
2013-12-02 14:10:19 +01:00
entries = [self.url_result(video_url.text, 'Smotri')
for video_url in rss.findall('./channel/item/link')]
2013-12-02 17:08:17 +01:00
description_text = rss.find('./channel/description').text
2013-12-02 14:10:19 +01:00
user_nickname = self._html_search_regex(
'^Видео режиссера (.*)$', description_text,
'user nickname')
2013-12-02 14:10:19 +01:00
return self.playlist_result(entries, user_id, user_nickname)
class SmotriBroadcastIE(InfoExtractor):
IE_DESC = 'Smotri.com broadcasts'
IE_NAME = 'smotri:broadcast'
_VALID_URL = r'^https?://(?:www\.)?(?P<url>smotri\.com/live/(?P<broadcastid>[^/]+))/?.*'
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
broadcast_id = mobj.group('broadcastid')
broadcast_url = 'http://' + mobj.group('url')
broadcast_page = self._download_webpage(broadcast_url, broadcast_id, 'Downloading broadcast page')
if re.search('>Режиссер с логином <br/>"%s"<br/> <span>не существует<' % broadcast_id, broadcast_page) is not None:
raise ExtractorError('Broadcast %s does not exist' % broadcast_id, expected=True)
# Adult content
if re.search('EroConfirmText">', broadcast_page) is not None:
(username, password) = self._get_login_info()
if username is None:
raise ExtractorError('Erotic broadcasts allowed only for registered users, '
'use --username and --password options to provide account credentials.', expected=True)
login_form = {
'login-hint53': '1',
'confirm_erotic': '1',
'login': username,
'password': password,
}
request = compat_urllib_request.Request(broadcast_url + '/?no_redirect=1', compat_urllib_parse.urlencode(login_form))
request.add_header('Content-Type', 'application/x-www-form-urlencoded')
broadcast_page = self._download_webpage(request, broadcast_id, 'Logging in and confirming age')
if re.search('>Неверный логин или пароль<', broadcast_page) is not None:
raise ExtractorError('Unable to log in: bad username or password', expected=True)
adult_content = True
else:
adult_content = False
ticket = self._html_search_regex(
'window\.broadcast_control\.addFlashVar\\(\'file\', \'([^\']+)\'\\);',
broadcast_page, 'broadcast ticket')
url = 'http://smotri.com/broadcast/view/url/?ticket=%s' % ticket
broadcast_password = self._downloader.params.get('videopassword', None)
if broadcast_password:
url += '&pass=%s' % hashlib.md5(broadcast_password.encode('utf-8')).hexdigest()
broadcast_json_page = self._download_webpage(url, broadcast_id, 'Downloading broadcast JSON')
try:
broadcast_json = json.loads(broadcast_json_page)
protected_broadcast = broadcast_json['_pass_protected'] == 1
if protected_broadcast and not broadcast_password:
raise ExtractorError('This broadcast is protected by a password, use the --video-password option', expected=True)
broadcast_offline = broadcast_json['is_play'] == 0
if broadcast_offline:
raise ExtractorError('Broadcast %s is offline' % broadcast_id, expected=True)
rtmp_url = broadcast_json['_server']
if not rtmp_url.startswith('rtmp://'):
raise ExtractorError('Unexpected broadcast rtmp URL')
broadcast_playpath = broadcast_json['_streamName']
broadcast_thumbnail = broadcast_json['_imgURL']
broadcast_title = broadcast_json['title']
broadcast_description = broadcast_json['description']
broadcaster_nick = broadcast_json['nick']
broadcaster_login = broadcast_json['login']
rtmp_conn = 'S:%s' % uuid.uuid4().hex
except KeyError:
if protected_broadcast:
raise ExtractorError('Bad broadcast password', expected=True)
raise ExtractorError('Unexpected broadcast JSON')
return {
'id': broadcast_id,
'url': rtmp_url,
'title': broadcast_title,
'thumbnail': broadcast_thumbnail,
'description': broadcast_description,
'uploader': broadcaster_nick,
'uploader_id': broadcaster_login,
'age_limit': 18 if adult_content else 0,
'ext': 'flv',
'play_path': broadcast_playpath,
'rtmp_live': True,
'rtmp_conn': rtmp_conn
}