1
mirror of https://github.com/yt-dlp/yt-dlp synced 2024-12-04 13:45:17 +01:00
yt-dlp/youtube_dl/InfoExtractors.py

4218 lines
167 KiB
Python
Raw Normal View History

#!/usr/bin/env python
# -*- coding: utf-8 -*-
2012-11-28 03:34:40 +01:00
from __future__ import absolute_import
2013-01-01 21:07:37 +01:00
import base64
import datetime
2013-01-27 03:01:23 +01:00
import itertools
import netrc
import os
import re
import socket
import time
import email.utils
import xml.etree.ElementTree
2012-08-08 20:04:02 +02:00
import random
import math
import operator
2012-11-28 03:34:40 +01:00
from .utils import *
class InfoExtractor(object):
2012-11-28 02:04:46 +01:00
"""Information Extractor class.
2012-11-28 02:04:46 +01:00
Information extractors are the classes that, given a URL, extract
information about the video (or videos) the URL refers to. This
information includes the real video URL, the video title, author and
2012-12-19 15:19:08 +01:00
others. The information is stored in a dictionary which is then
2012-11-28 02:04:46 +01:00
passed to the FileDownloader. The FileDownloader processes this
information possibly downloading the video to the file system, among
other possible outcomes.
2012-11-28 02:04:46 +01:00
The dictionaries must include the following fields:
2012-11-28 02:04:46 +01:00
id: Video identifier.
url: Final video URL.
title: Video title, unescaped.
ext: Video filename extension.
2012-11-28 02:04:46 +01:00
The following fields are optional:
2012-11-28 02:04:46 +01:00
format: The video format, defaults to ext (used for --get-format)
thumbnail: Full URL to a video thumbnail image.
description: One-line video description.
uploader: Full name of the video uploader.
upload_date: Video upload date (YYYYMMDD).
2012-12-20 16:28:16 +01:00
uploader_id: Nickname or id of the video uploader.
2013-01-12 17:34:31 +01:00
location: Physical location of the video.
2012-11-28 02:04:46 +01:00
player_url: SWF Player URL (used for rtmpdump).
subtitles: The subtitle file contents.
2012-11-28 02:04:46 +01:00
urlhandle: [internal] The urlHandle to be used to download the file,
like returned by urllib.request.urlopen
2012-11-28 02:04:46 +01:00
The fields should all be Unicode strings.
2012-11-28 02:04:46 +01:00
Subclasses of this one should re-define the _real_initialize() and
_real_extract() methods and define a _VALID_URL regexp.
Probably, they should also be added to the list of extractors.
2012-11-28 02:04:46 +01:00
_real_extract() must return a *list* of information dictionaries as
described above.
2012-11-28 02:04:46 +01:00
Finally, the _WORKING attribute should be set to False for broken IEs
in order to warn the users and skip the tests.
"""
2012-11-28 02:04:46 +01:00
_ready = False
_downloader = None
_WORKING = True
2012-11-28 02:04:46 +01:00
def __init__(self, downloader=None):
"""Constructor. Receives an optional downloader."""
self._ready = False
self.set_downloader(downloader)
@classmethod
def suitable(cls, url):
2012-11-28 02:04:46 +01:00
"""Receives a URL and returns True if suitable for this IE."""
return re.match(cls._VALID_URL, url) is not None
@classmethod
def working(cls):
2012-11-28 02:04:46 +01:00
"""Getter method for _WORKING."""
return cls._WORKING
2012-11-28 02:04:46 +01:00
def initialize(self):
"""Initializes an instance (authentication, etc)."""
if not self._ready:
self._real_initialize()
self._ready = True
2012-11-28 02:04:46 +01:00
def extract(self, url):
"""Extracts URL information and returns it in list of dicts."""
self.initialize()
return self._real_extract(url)
2012-11-28 02:04:46 +01:00
def set_downloader(self, downloader):
"""Sets the downloader for this IE."""
self._downloader = downloader
2012-11-28 02:04:46 +01:00
def _real_initialize(self):
"""Real initialization process. Redefine in subclasses."""
pass
2012-11-28 02:04:46 +01:00
def _real_extract(self, url):
"""Real extraction process. Redefine in subclasses."""
pass
2012-12-27 01:38:41 +01:00
@property
def IE_NAME(self):
return type(self).__name__[:-2]
def _request_webpage(self, url_or_request, video_id, note=None, errnote=None):
""" Returns the response handle """
2013-01-01 20:43:43 +01:00
if note is None:
self.report_download_webpage(video_id)
elif note is not False:
self.to_screen(u'%s: %s' % (video_id, note))
2013-01-01 20:43:43 +01:00
try:
return compat_urllib_request.urlopen(url_or_request)
2013-01-01 20:43:43 +01:00
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
if errnote is None:
errnote = u'Unable to download webpage'
raise ExtractorError(u'%s: %s' % (errnote, compat_str(err)), sys.exc_info()[2])
2013-01-01 20:43:43 +01:00
2013-05-04 02:53:26 +02:00
def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None):
""" Returns a tuple (page content as string, URL handle) """
urlh = self._request_webpage(url_or_request, video_id, note, errnote)
2013-03-12 01:08:54 +01:00
content_type = urlh.headers.get('Content-Type', '')
m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type)
if m:
encoding = m.group(1)
else:
encoding = 'utf-8'
webpage_bytes = urlh.read()
2013-04-11 18:31:35 +02:00
if self._downloader.params.get('dump_intermediate_pages', False):
try:
url = url_or_request.get_full_url()
except AttributeError:
url = url_or_request
self.to_screen(u'Dumping request to ' + url)
2013-04-11 18:31:35 +02:00
dump = base64.b64encode(webpage_bytes).decode('ascii')
self._downloader.to_screen(dump)
2013-05-04 02:53:26 +02:00
content = webpage_bytes.decode(encoding, 'replace')
return (content, urlh)
def _download_webpage(self, url_or_request, video_id, note=None, errnote=None):
""" Returns the data of the page as a string """
return self._download_webpage_handle(url_or_request, video_id, note, errnote)[0]
def to_screen(self, msg):
"""Print msg to screen, prefixing it with '[ie_name]'"""
self._downloader.to_screen(u'[%s] %s' % (self.IE_NAME, msg))
def report_extraction(self, id_or_name):
"""Report information extraction."""
self.to_screen(u'%s: Extracting information' % id_or_name)
def report_download_webpage(self, video_id):
"""Report webpage download."""
self.to_screen(u'%s: Downloading webpage' % video_id)
def report_age_confirmation(self):
"""Report attempt to confirm age."""
self.to_screen(u'Confirming age')
#Methods for following #608
#They set the correct value of the '_type' key
def video_result(self, video_info):
"""Returns a video"""
video_info['_type'] = 'video'
return video_info
def url_result(self, url, ie=None):
"""Returns a url that points to a page that should be processed"""
#TODO: ie should be the class used for getting the info
video_info = {'_type': 'url',
'url': url,
'ie_key': ie}
return video_info
def playlist_result(self, entries, playlist_id=None, playlist_title=None):
"""Returns a playlist"""
video_info = {'_type': 'playlist',
'entries': entries}
if playlist_id:
video_info['id'] = playlist_id
if playlist_title:
video_info['title'] = playlist_title
return video_info
2013-01-01 20:43:43 +01:00
class YoutubeIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information extractor for youtube.com."""
_VALID_URL = r"""^
(
(?:https?://)? # http(s):// (optional)
(?:youtu\.be/|(?:\w+\.)?youtube(?:-nocookie)?\.com/|
tube\.majestyc\.net/) # the various hostnames, with wildcard subdomains
(?:.*?\#/)? # handle anchor (#/) redirect urls
(?: # the various things that can precede the ID:
(?:(?:v|embed|e)/) # v/ or embed/ or e/
|(?: # or the v= param in all its forms
(?:watch(?:_popup)?(?:\.php)?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
(?:\?|\#!?) # the params delimiter ? or # or #!
(?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
2012-11-28 02:04:46 +01:00
v=
)
)? # optional -> youtube.com/xxxx is OK
)? # all until now is optional -> you can pass the naked ID
([0-9A-Za-z_-]+) # here is it! the YouTube video ID
(?(1).+)? # if we found the ID, everything can follow
$"""
_LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1'
2013-02-06 21:22:53 +01:00
_LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
2012-11-28 02:04:46 +01:00
_AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en'
_NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
_NETRC_MACHINE = 'youtube'
# Listed in order of quality
_available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '17', '13']
_available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '17', '13']
_video_extensions = {
'13': '3gp',
'17': 'mp4',
'18': 'mp4',
'22': 'mp4',
'37': 'mp4',
'38': 'video', # You actually don't know if this will be MOV, AVI or whatever
'43': 'webm',
'44': 'webm',
'45': 'webm',
'46': 'webm',
}
_video_dimensions = {
'5': '240x400',
'6': '???',
'13': '???',
'17': '144x176',
'18': '360x640',
'22': '720x1280',
'34': '360x640',
'35': '480x854',
'37': '1080x1920',
'38': '3072x4096',
'43': '360x640',
'44': '480x854',
'45': '720x1280',
'46': '1080x1920',
2012-12-19 15:19:08 +01:00
}
2012-11-28 02:04:46 +01:00
IE_NAME = u'youtube'
@classmethod
def suitable(cls, url):
2012-11-28 02:04:46 +01:00
"""Receives a URL and returns True if suitable for this IE."""
if YoutubePlaylistIE.suitable(url): return False
return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
2012-11-28 02:04:46 +01:00
def report_lang(self):
"""Report attempt to set language."""
self.to_screen(u'Setting language')
2012-11-28 02:04:46 +01:00
def report_login(self):
"""Report attempt to log in."""
self.to_screen(u'Logging in')
2012-11-28 02:04:46 +01:00
def report_video_webpage_download(self, video_id):
"""Report attempt to download video webpage."""
self.to_screen(u'%s: Downloading video webpage' % video_id)
2012-11-28 02:04:46 +01:00
def report_video_info_webpage_download(self, video_id):
"""Report attempt to download video info webpage."""
self.to_screen(u'%s: Downloading video info webpage' % video_id)
2012-11-28 02:04:46 +01:00
def report_video_subtitles_download(self, video_id):
"""Report attempt to download video info webpage."""
self.to_screen(u'%s: Checking available subtitles' % video_id)
2012-11-28 02:04:46 +01:00
def report_video_subtitles_request(self, video_id, sub_lang, format):
"""Report attempt to download video info webpage."""
self.to_screen(u'%s: Downloading video subtitles for %s.%s' % (video_id, sub_lang, format))
def report_video_subtitles_available(self, video_id, sub_lang_list):
"""Report available subtitles."""
sub_lang = ",".join(list(sub_lang_list.keys()))
self.to_screen(u'%s: Available subtitles for video: %s' % (video_id, sub_lang))
2012-11-28 02:04:46 +01:00
def report_information_extraction(self, video_id):
"""Report attempt to extract video information."""
self.to_screen(u'%s: Extracting video information' % video_id)
2012-11-28 02:04:46 +01:00
def report_unavailable_format(self, video_id, format):
"""Report extracted video URL."""
self.to_screen(u'%s: Format %s not available' % (video_id, format))
2012-11-28 02:04:46 +01:00
def report_rtmp_download(self):
"""Indicate the download will use the RTMP protocol."""
self.to_screen(u'RTMP download detected')
2012-11-28 02:04:46 +01:00
def _get_available_subtitles(self, video_id):
self.report_video_subtitles_download(video_id)
request = compat_urllib_request.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id)
try:
sub_list = compat_urllib_request.urlopen(request).read().decode('utf-8')
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
return (u'unable to download video subtitles: %s' % compat_str(err), None)
sub_lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list)
sub_lang_list = dict((l[1], l[0]) for l in sub_lang_list)
if not sub_lang_list:
return (u'video doesn\'t have subtitles', None)
return sub_lang_list
def _list_available_subtitles(self, video_id):
sub_lang_list = self._get_available_subtitles(video_id)
self.report_video_subtitles_available(video_id, sub_lang_list)
def _request_subtitle(self, sub_lang, sub_name, video_id, format):
"""
Return tuple:
(error_message, sub_lang, sub)
"""
self.report_video_subtitles_request(video_id, sub_lang, format)
params = compat_urllib_parse.urlencode({
'lang': sub_lang,
'name': sub_name,
'v': video_id,
'fmt': format,
})
url = 'http://www.youtube.com/api/timedtext?' + params
try:
sub = compat_urllib_request.urlopen(url).read().decode('utf-8')
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
return (u'unable to download video subtitles: %s' % compat_str(err), None, None)
if not sub:
return (u'Did not fetch video subtitles', None, None)
return (None, sub_lang, sub)
def _extract_subtitle(self, video_id):
2013-03-30 14:11:33 +01:00
"""
Return a list with a tuple:
[(error_message, sub_lang, sub)]
"""
sub_lang_list = self._get_available_subtitles(video_id)
sub_format = self._downloader.params.get('subtitlesformat')
2013-03-30 14:11:33 +01:00
if isinstance(sub_lang_list,tuple): #There was some error, it didn't get the available subtitles
return [(sub_lang_list[0], None, None)]
if self._downloader.params.get('subtitleslang', False):
sub_lang = self._downloader.params.get('subtitleslang')
elif 'en' in sub_lang_list:
sub_lang = 'en'
else:
sub_lang = list(sub_lang_list.keys())[0]
if not sub_lang in sub_lang_list:
return [(u'no closed captions found in the specified language "%s"' % sub_lang, None, None)]
subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format)
return [subtitle]
def _extract_all_subtitles(self, video_id):
sub_lang_list = self._get_available_subtitles(video_id)
sub_format = self._downloader.params.get('subtitlesformat')
if isinstance(sub_lang_list,tuple): #There was some error, it didn't get the available subtitles
return [(sub_lang_list[0], None, None)]
subtitles = []
for sub_lang in sub_lang_list:
subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format)
subtitles.append(subtitle)
return subtitles
2012-11-28 02:04:46 +01:00
def _print_formats(self, formats):
print('Available formats:')
for x in formats:
print('%s\t:\t%s\t[%s]' %(x, self._video_extensions.get(x, 'flv'), self._video_dimensions.get(x, '???')))
def _real_initialize(self):
if self._downloader is None:
return
username = None
password = None
downloader_params = self._downloader.params
# Attempt to use provided username and password or .netrc data
if downloader_params.get('username', None) is not None:
username = downloader_params['username']
password = downloader_params['password']
elif downloader_params.get('usenetrc', False):
try:
info = netrc.netrc().authenticators(self._NETRC_MACHINE)
if info is not None:
username = info[0]
password = info[2]
else:
raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE)
except (IOError, netrc.NetrcParseError) as err:
2013-03-02 11:24:07 +01:00
self._downloader.report_warning(u'parsing .netrc: %s' % compat_str(err))
2012-11-28 02:04:46 +01:00
return
# Set language
request = compat_urllib_request.Request(self._LANG_URL)
try:
self.report_lang()
compat_urllib_request.urlopen(request).read()
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
2013-03-02 11:24:07 +01:00
self._downloader.report_warning(u'unable to set language: %s' % compat_str(err))
2012-11-28 02:04:46 +01:00
return
# No authentication to be performed
if username is None:
return
2013-02-06 21:22:53 +01:00
request = compat_urllib_request.Request(self._LOGIN_URL)
try:
login_page = compat_urllib_request.urlopen(request).read().decode('utf-8')
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
2013-03-02 11:24:07 +01:00
self._downloader.report_warning(u'unable to fetch login page: %s' % compat_str(err))
2013-02-06 21:22:53 +01:00
return
galx = None
dsh = None
match = re.search(re.compile(r'<input.+?name="GALX".+?value="(.+?)"', re.DOTALL), login_page)
if match:
galx = match.group(1)
match = re.search(re.compile(r'<input.+?name="dsh".+?value="(.+?)"', re.DOTALL), login_page)
if match:
dsh = match.group(1)
2012-11-28 02:04:46 +01:00
# Log in
2013-02-06 21:22:53 +01:00
login_form_strs = {
u'continue': u'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
2013-02-06 21:22:53 +01:00
u'Email': username,
u'GALX': galx,
u'Passwd': password,
u'PersistentCookie': u'yes',
u'_utf8': u'',
u'bgresponse': u'js_disabled',
u'checkConnection': u'',
u'checkedDomains': u'youtube',
u'dnConn': u'',
u'dsh': dsh,
u'pstMsg': u'0',
u'rmShown': u'1',
u'secTok': u'',
u'signIn': u'Sign in',
u'timeStmp': u'',
u'service': u'youtube',
u'uilel': u'3',
u'hl': u'en_US',
}
# Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
# chokes on unicode
login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items())
login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
request = compat_urllib_request.Request(self._LOGIN_URL, login_data)
2012-11-28 02:04:46 +01:00
try:
self.report_login()
login_results = compat_urllib_request.urlopen(request).read().decode('utf-8')
2013-02-06 21:22:53 +01:00
if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
2013-03-02 11:24:07 +01:00
self._downloader.report_warning(u'unable to log in: bad username or password')
2012-11-28 02:04:46 +01:00
return
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
2013-03-02 11:24:07 +01:00
self._downloader.report_warning(u'unable to log in: %s' % compat_str(err))
2012-11-28 02:04:46 +01:00
return
# Confirm age
age_form = {
'next_url': '/',
'action_confirm': 'Confirm',
}
request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form))
try:
self.report_age_confirmation()
age_results = compat_urllib_request.urlopen(request).read().decode('utf-8')
2012-11-28 02:04:46 +01:00
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err))
2012-11-28 02:04:46 +01:00
def _extract_id(self, url):
2012-11-28 02:04:46 +01:00
mobj = re.match(self._VALID_URL, url, re.VERBOSE)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
video_id = mobj.group(2)
return video_id
def _real_extract(self, url):
# Extract original video URL from URL with redirection, like age verification, using next_url parameter
mobj = re.search(self._NEXT_URL_RE, url)
if mobj:
url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
video_id = self._extract_id(url)
2012-11-28 02:04:46 +01:00
# Get video webpage
self.report_video_webpage_download(video_id)
url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id
request = compat_urllib_request.Request(url)
2012-11-28 02:04:46 +01:00
try:
video_webpage_bytes = compat_urllib_request.urlopen(request).read()
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
raise ExtractorError(u'Unable to download video webpage: %s' % compat_str(err))
2012-11-28 02:04:46 +01:00
video_webpage = video_webpage_bytes.decode('utf-8', 'ignore')
# Attempt to extract SWF player URL
mobj = re.search(r'swfConfig.*?"(http:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
if mobj is not None:
player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
else:
player_url = None
# Get video info
self.report_video_info_webpage_download(video_id)
for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
2013-04-11 18:18:15 +02:00
video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
2012-11-28 02:04:46 +01:00
% (video_id, el_type))
2013-04-11 18:18:15 +02:00
video_info_webpage = self._download_webpage(video_info_url, video_id,
note=False,
errnote='unable to download video info webpage')
video_info = compat_parse_qs(video_info_webpage)
if 'token' in video_info:
break
2012-11-28 02:04:46 +01:00
if 'token' not in video_info:
if 'reason' in video_info:
raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0])
2012-11-28 02:04:46 +01:00
else:
raise ExtractorError(u'"token" parameter not in video info for unknown reason')
2012-11-28 02:04:46 +01:00
# Check for "rental" videos
if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
raise ExtractorError(u'"rental" videos not supported')
2012-11-28 02:04:46 +01:00
# Start extracting information
self.report_information_extraction(video_id)
# uploader
if 'author' not in video_info:
raise ExtractorError(u'Unable to extract uploader name')
2012-11-28 02:04:46 +01:00
video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
2012-12-20 16:28:16 +01:00
# uploader_id
video_uploader_id = None
mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
2012-12-20 16:28:16 +01:00
if mobj is not None:
video_uploader_id = mobj.group(1)
else:
self._downloader.report_warning(u'unable to extract uploader nickname')
2012-12-20 16:28:16 +01:00
2012-11-28 02:04:46 +01:00
# title
if 'title' not in video_info:
raise ExtractorError(u'Unable to extract video title')
2012-11-28 02:04:46 +01:00
video_title = compat_urllib_parse.unquote_plus(video_info['title'][0])
# thumbnail image
if 'thumbnail_url' not in video_info:
self._downloader.report_warning(u'unable to extract video thumbnail')
2012-11-28 02:04:46 +01:00
video_thumbnail = ''
else: # don't panic if we can't find it
video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
# upload date
upload_date = None
mobj = re.search(r'id="eow-date.*?>(.*?)</span>', video_webpage, re.DOTALL)
if mobj is not None:
upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
upload_date = unified_strdate(upload_date)
2012-11-28 02:04:46 +01:00
# description
video_description = get_element_by_id("eow-description", video_webpage)
if video_description:
video_description = clean_html(video_description)
else:
fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
if fd_mobj:
video_description = unescapeHTML(fd_mobj.group(1))
else:
video_description = u''
2012-11-28 02:04:46 +01:00
# subtitles
2012-11-28 02:04:46 +01:00
video_subtitles = None
2012-11-28 02:04:46 +01:00
if self._downloader.params.get('writesubtitles', False):
video_subtitles = self._extract_subtitle(video_id)
if video_subtitles:
(sub_error, sub_lang, sub) = video_subtitles[0]
if sub_error:
self._downloader.report_error(sub_error)
if self._downloader.params.get('allsubtitles', False):
video_subtitles = self._extract_all_subtitles(video_id)
for video_subtitle in video_subtitles:
(sub_error, sub_lang, sub) = video_subtitle
if sub_error:
self._downloader.report_error(sub_error)
2012-11-28 02:04:46 +01:00
if self._downloader.params.get('listsubtitles', False):
sub_lang_list = self._list_available_subtitles(video_id)
return
2012-11-28 02:04:46 +01:00
if 'length_seconds' not in video_info:
self._downloader.report_warning(u'unable to extract video duration')
2012-11-28 02:04:46 +01:00
video_duration = ''
else:
video_duration = compat_urllib_parse.unquote_plus(video_info['length_seconds'][0])
# token
video_token = compat_urllib_parse.unquote_plus(video_info['token'][0])
# Decide which formats to download
req_format = self._downloader.params.get('format', None)
if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
self.report_rtmp_download()
video_url_list = [(None, video_info['conn'][0])]
elif 'url_encoded_fmt_stream_map' in video_info and len(video_info['url_encoded_fmt_stream_map']) >= 1:
url_data_strs = video_info['url_encoded_fmt_stream_map'][0].split(',')
url_data = [compat_parse_qs(uds) for uds in url_data_strs]
url_data = [ud for ud in url_data if 'itag' in ud and 'url' in ud]
url_map = dict((ud['itag'][0], ud['url'][0] + '&signature=' + ud['sig'][0]) for ud in url_data)
2012-11-28 02:04:46 +01:00
format_limit = self._downloader.params.get('format_limit', None)
available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats
if format_limit is not None and format_limit in available_formats:
format_list = available_formats[available_formats.index(format_limit):]
else:
format_list = available_formats
existing_formats = [x for x in format_list if x in url_map]
if len(existing_formats) == 0:
2013-04-22 19:51:42 +02:00
raise ExtractorError(u'no known formats available for video')
2012-11-28 02:04:46 +01:00
if self._downloader.params.get('listformats', None):
self._print_formats(existing_formats)
return
if req_format is None or req_format == 'best':
video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality
elif req_format == 'worst':
video_url_list = [(existing_formats[len(existing_formats)-1], url_map[existing_formats[len(existing_formats)-1]])] # worst quality
elif req_format in ('-1', 'all'):
video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats
else:
# Specific formats. We pick the first in a slash-delimeted sequence.
# For example, if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'.
req_formats = req_format.split('/')
video_url_list = None
for rf in req_formats:
if rf in url_map:
video_url_list = [(rf, url_map[rf])]
break
if video_url_list is None:
2013-04-22 19:51:42 +02:00
raise ExtractorError(u'requested format not available')
2012-11-28 02:04:46 +01:00
else:
2013-04-22 19:51:42 +02:00
raise ExtractorError(u'no conn or url_encoded_fmt_stream_map information found in video info')
2012-11-28 02:04:46 +01:00
results = []
for format_param, video_real_url in video_url_list:
# Extension
video_extension = self._video_extensions.get(format_param, 'flv')
2012-11-28 11:28:59 +01:00
video_format = '{0} - {1}'.format(format_param if format_param else video_extension,
self._video_dimensions.get(format_param, '???'))
2012-11-28 02:04:46 +01:00
results.append({
'id': video_id,
'url': video_real_url,
'uploader': video_uploader,
2012-12-20 16:28:16 +01:00
'uploader_id': video_uploader_id,
2012-11-28 02:04:46 +01:00
'upload_date': upload_date,
'title': video_title,
'ext': video_extension,
'format': video_format,
'thumbnail': video_thumbnail,
'description': video_description,
'player_url': player_url,
'subtitles': video_subtitles,
'duration': video_duration
})
return results
class MetacafeIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information Extractor for metacafe.com."""
_VALID_URL = r'(?:http://)?(?:www\.)?metacafe\.com/watch/([^/]+)/([^/]+)/.*'
_DISCLAIMER = 'http://www.metacafe.com/family_filter/'
_FILTER_POST = 'http://www.metacafe.com/f/index.php?inputType=filter&controllerGroup=user'
IE_NAME = u'metacafe'
def report_disclaimer(self):
"""Report disclaimer retrieval."""
self.to_screen(u'Retrieving disclaimer')
2012-11-28 02:04:46 +01:00
def _real_initialize(self):
# Retrieve disclaimer
request = compat_urllib_request.Request(self._DISCLAIMER)
try:
self.report_disclaimer()
disclaimer = compat_urllib_request.urlopen(request).read()
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
raise ExtractorError(u'Unable to retrieve disclaimer: %s' % compat_str(err))
2012-11-28 02:04:46 +01:00
# Confirm age
disclaimer_form = {
'filters': '0',
'submit': "Continue - I'm over 18",
}
request = compat_urllib_request.Request(self._FILTER_POST, compat_urllib_parse.urlencode(disclaimer_form))
try:
self.report_age_confirmation()
disclaimer = compat_urllib_request.urlopen(request).read()
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err))
2012-11-28 02:04:46 +01:00
def _real_extract(self, url):
# Extract id and simplified title from URL
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
video_id = mobj.group(1)
# Check if video comes from YouTube
mobj2 = re.match(r'^yt-(.*)$', video_id)
if mobj2 is not None:
return [self.url_result('http://www.youtube.com/watch?v=%s' % mobj2.group(1), 'Youtube')]
2012-11-28 02:04:46 +01:00
# Retrieve video webpage to extract further information
2013-04-20 12:06:58 +02:00
webpage = self._download_webpage('http://www.metacafe.com/watch/%s/' % video_id, video_id)
2012-11-28 02:04:46 +01:00
# Extract URL, uploader and title from webpage
self.report_extraction(video_id)
mobj = re.search(r'(?m)&mediaURL=([^&]+)', webpage)
if mobj is not None:
mediaURL = compat_urllib_parse.unquote(mobj.group(1))
video_extension = mediaURL[-3:]
# Extract gdaKey if available
mobj = re.search(r'(?m)&gdaKey=(.*?)&', webpage)
if mobj is None:
video_url = mediaURL
else:
gdaKey = mobj.group(1)
video_url = '%s?__gda__=%s' % (mediaURL, gdaKey)
else:
mobj = re.search(r' name="flashvars" value="(.*?)"', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract media URL')
2012-11-28 02:04:46 +01:00
vardict = compat_parse_qs(mobj.group(1))
if 'mediaData' not in vardict:
raise ExtractorError(u'Unable to extract media URL')
2013-04-20 12:06:58 +02:00
mobj = re.search(r'"mediaURL":"(?P<mediaURL>http.*?)",(.*?)"key":"(?P<key>.*?)"', vardict['mediaData'][0])
2012-11-28 02:04:46 +01:00
if mobj is None:
raise ExtractorError(u'Unable to extract media URL')
2013-04-20 12:06:58 +02:00
mediaURL = mobj.group('mediaURL').replace('\\/', '/')
2012-11-28 02:04:46 +01:00
video_extension = mediaURL[-3:]
2013-04-20 12:06:58 +02:00
video_url = '%s?__gda__=%s' % (mediaURL, mobj.group('key'))
2012-11-28 02:04:46 +01:00
mobj = re.search(r'(?im)<title>(.*) - Video</title>', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract title')
2012-11-28 02:04:46 +01:00
video_title = mobj.group(1).decode('utf-8')
mobj = re.search(r'submitter=(.*?);', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract uploader nickname')
2012-11-28 02:04:46 +01:00
video_uploader = mobj.group(1)
return [{
'id': video_id.decode('utf-8'),
'url': video_url.decode('utf-8'),
'uploader': video_uploader.decode('utf-8'),
'upload_date': None,
'title': video_title,
'ext': video_extension.decode('utf-8'),
}]
class DailymotionIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information Extractor for Dailymotion"""
_VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/video/([^/]+)'
IE_NAME = u'dailymotion'
def _real_extract(self, url):
# Extract id and simplified title from URL
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
video_id = mobj.group(1).split('_')[0].split('?')[0]
video_extension = 'mp4'
# Retrieve video webpage to extract further information
request = compat_urllib_request.Request(url)
request.add_header('Cookie', 'family_filter=off')
2013-01-01 21:22:30 +01:00
webpage = self._download_webpage(request, video_id)
2012-11-28 02:04:46 +01:00
# Extract URL, uploader and title from webpage
self.report_extraction(video_id)
mobj = re.search(r'\s*var flashvars = (.*)', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract media URL')
2012-11-28 02:04:46 +01:00
flashvars = compat_urllib_parse.unquote(mobj.group(1))
for key in ['hd1080URL', 'hd720URL', 'hqURL', 'sdURL', 'ldURL', 'video_url']:
if key in flashvars:
max_quality = key
self.to_screen(u'Using %s' % key)
2012-11-28 02:04:46 +01:00
break
else:
raise ExtractorError(u'Unable to extract video URL')
2012-11-28 02:04:46 +01:00
mobj = re.search(r'"' + max_quality + r'":"(.+?)"', flashvars)
if mobj is None:
raise ExtractorError(u'Unable to extract video URL')
2012-11-28 02:04:46 +01:00
video_url = compat_urllib_parse.unquote(mobj.group(1)).replace('\\/', '/')
# TODO: support choosing qualities
mobj = re.search(r'<meta property="og:title" content="(?P<title>[^"]*)" />', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract title')
2012-12-15 18:23:17 +01:00
video_title = unescapeHTML(mobj.group('title'))
2012-11-28 02:04:46 +01:00
video_uploader = None
mobj = re.search(r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>', webpage)
if mobj is None:
# lookin for official user
mobj_official = re.search(r'<span rel="author"[^>]+?>([^<]+?)</span>', webpage)
if mobj_official is None:
self._downloader.report_warning(u'unable to extract uploader nickname')
2012-11-28 02:04:46 +01:00
else:
video_uploader = mobj_official.group(1)
else:
video_uploader = mobj.group(1)
video_upload_date = None
mobj = re.search(r'<div class="[^"]*uploaded_cont[^"]*" title="[^"]*">([0-9]{2})-([0-9]{2})-([0-9]{4})</div>', webpage)
if mobj is not None:
video_upload_date = mobj.group(3) + mobj.group(2) + mobj.group(1)
return [{
2012-12-15 18:23:17 +01:00
'id': video_id,
'url': video_url,
'uploader': video_uploader,
2012-11-28 02:04:46 +01:00
'upload_date': video_upload_date,
'title': video_title,
2012-12-15 18:23:17 +01:00
'ext': video_extension,
2012-11-28 02:04:46 +01:00
}]
class PhotobucketIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information extractor for photobucket.com."""
# TODO: the original _VALID_URL was:
# r'(?:http://)?(?:[a-z0-9]+\.)?photobucket\.com/.*[\?\&]current=(.*\.flv)'
# Check if it's necessary to keep the old extracion process
_VALID_URL = r'(?:http://)?(?:[a-z0-9]+\.)?photobucket\.com/.*(([\?\&]current=)|_)(?P<id>.*)\.(?P<ext>(flv)|(mp4))'
2012-11-28 02:04:46 +01:00
IE_NAME = u'photobucket'
def _real_extract(self, url):
# Extract id from URL
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
video_id = mobj.group('id')
2012-11-28 02:04:46 +01:00
video_extension = mobj.group('ext')
2012-11-28 02:04:46 +01:00
# Retrieve video webpage to extract further information
webpage = self._download_webpage(url, video_id)
2012-11-28 02:04:46 +01:00
# Extract URL, uploader, and title from webpage
self.report_extraction(video_id)
# We try first by looking the javascript code:
mobj = re.search(r'Pb\.Data\.Shared\.put\(Pb\.Data\.Shared\.MEDIA, (?P<json>.*?)\);', webpage)
if mobj is not None:
info = json.loads(mobj.group('json'))
return [{
'id': video_id,
'url': info[u'downloadUrl'],
'uploader': info[u'username'],
'upload_date': datetime.date.fromtimestamp(info[u'creationDate']).strftime('%Y%m%d'),
'title': info[u'title'],
'ext': video_extension,
'thumbnail': info[u'thumbUrl'],
}]
# We try looking in other parts of the webpage
2012-11-28 02:04:46 +01:00
mobj = re.search(r'<link rel="video_src" href=".*\?file=([^"]+)" />', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract media URL')
2012-11-28 02:04:46 +01:00
mediaURL = compat_urllib_parse.unquote(mobj.group(1))
video_url = mediaURL
mobj = re.search(r'<title>(.*) video by (.*) - Photobucket</title>', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract title')
2012-11-28 02:04:46 +01:00
video_title = mobj.group(1).decode('utf-8')
video_uploader = mobj.group(2).decode('utf-8')
return [{
'id': video_id.decode('utf-8'),
'url': video_url.decode('utf-8'),
'uploader': video_uploader,
'upload_date': None,
'title': video_title,
'ext': video_extension.decode('utf-8'),
}]
class YahooIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information extractor for video.yahoo.com."""
_WORKING = False
2012-11-28 02:04:46 +01:00
# _VALID_URL matches all Yahoo! Video URLs
# _VPAGE_URL matches only the extractable '/watch/' URLs
_VALID_URL = r'(?:http://)?(?:[a-z]+\.)?video\.yahoo\.com/(?:watch|network)/([0-9]+)(?:/|\?v=)([0-9]+)(?:[#\?].*)?'
_VPAGE_URL = r'(?:http://)?video\.yahoo\.com/watch/([0-9]+)/([0-9]+)(?:[#\?].*)?'
IE_NAME = u'video.yahoo'
def _real_extract(self, url, new_video=True):
# Extract ID from URL
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
video_id = mobj.group(2)
video_extension = 'flv'
# Rewrite valid but non-extractable URLs as
# extractable English language /watch/ URLs
if re.match(self._VPAGE_URL, url) is None:
request = compat_urllib_request.Request(url)
try:
webpage = compat_urllib_request.urlopen(request).read()
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
raise ExtractorError(u'Unable to retrieve video webpage: %s' % compat_str(err))
2012-11-28 02:04:46 +01:00
mobj = re.search(r'\("id", "([0-9]+)"\);', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract id field')
2012-11-28 02:04:46 +01:00
yahoo_id = mobj.group(1)
mobj = re.search(r'\("vid", "([0-9]+)"\);', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract vid field')
2012-11-28 02:04:46 +01:00
yahoo_vid = mobj.group(1)
url = 'http://video.yahoo.com/watch/%s/%s' % (yahoo_vid, yahoo_id)
return self._real_extract(url, new_video=False)
# Retrieve video webpage to extract further information
request = compat_urllib_request.Request(url)
try:
self.report_download_webpage(video_id)
webpage = compat_urllib_request.urlopen(request).read()
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
raise ExtractorError(u'Unable to retrieve video webpage: %s' % compat_str(err))
2012-11-28 02:04:46 +01:00
# Extract uploader and title from webpage
self.report_extraction(video_id)
mobj = re.search(r'<meta name="title" content="(.*)" />', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract video title')
2012-11-28 02:04:46 +01:00
video_title = mobj.group(1).decode('utf-8')
mobj = re.search(r'<h2 class="ti-5"><a href="http://video\.yahoo\.com/(people|profile)/[0-9]+" beacon=".*">(.*)</a></h2>', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract video uploader')
2012-11-28 02:04:46 +01:00
video_uploader = mobj.group(1).decode('utf-8')
# Extract video thumbnail
mobj = re.search(r'<link rel="image_src" href="(.*)" />', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract video thumbnail')
2012-11-28 02:04:46 +01:00
video_thumbnail = mobj.group(1).decode('utf-8')
# Extract video description
mobj = re.search(r'<meta name="description" content="(.*)" />', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract video description')
2012-11-28 02:04:46 +01:00
video_description = mobj.group(1).decode('utf-8')
if not video_description:
video_description = 'No description available.'
# Extract video height and width
mobj = re.search(r'<meta name="video_height" content="([0-9]+)" />', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract video height')
2012-11-28 02:04:46 +01:00
yv_video_height = mobj.group(1)
mobj = re.search(r'<meta name="video_width" content="([0-9]+)" />', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract video width')
2012-11-28 02:04:46 +01:00
yv_video_width = mobj.group(1)
# Retrieve video playlist to extract media URL
# I'm not completely sure what all these options are, but we
# seem to need most of them, otherwise the server sends a 401.
yv_lg = 'R0xx6idZnW2zlrKP8xxAIR' # not sure what this represents
yv_bitrate = '700' # according to Wikipedia this is hard-coded
request = compat_urllib_request.Request('http://cosmos.bcst.yahoo.com/up/yep/process/getPlaylistFOP.php?node_id=' + video_id +
'&tech=flash&mode=playlist&lg=' + yv_lg + '&bitrate=' + yv_bitrate + '&vidH=' + yv_video_height +
'&vidW=' + yv_video_width + '&swf=as3&rd=video.yahoo.com&tk=null&adsupported=v1,v2,&eventid=1301797')
try:
self.report_download_webpage(video_id)
webpage = compat_urllib_request.urlopen(request).read()
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
raise ExtractorError(u'Unable to retrieve video webpage: %s' % compat_str(err))
2012-11-28 02:04:46 +01:00
# Extract media URL from playlist XML
mobj = re.search(r'<STREAM APP="(http://.*)" FULLPATH="/?(/.*\.flv\?[^"]*)"', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract media URL')
2012-11-28 02:04:46 +01:00
video_url = compat_urllib_parse.unquote(mobj.group(1) + mobj.group(2)).decode('utf-8')
video_url = unescapeHTML(video_url)
return [{
'id': video_id.decode('utf-8'),
'url': video_url,
'uploader': video_uploader,
'upload_date': None,
'title': video_title,
'ext': video_extension.decode('utf-8'),
'thumbnail': video_thumbnail.decode('utf-8'),
'description': video_description,
}]
class VimeoIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information extractor for vimeo.com."""
# _VALID_URL matches Vimeo URLs
_VALID_URL = r'(?P<proto>https?://)?(?:(?:www|player)\.)?vimeo\.com/(?:(?:groups|album)/[^/]+/)?(?P<direct_link>play_redirect_hls\?clip_id=)?(?:videos?/)?(?P<id>[0-9]+)'
2012-11-28 02:04:46 +01:00
IE_NAME = u'vimeo'
def _real_extract(self, url, new_video=True):
# Extract ID from URL
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
video_id = mobj.group('id')
if not mobj.group('proto'):
url = 'https://' + url
if mobj.group('direct_link'):
url = 'https://vimeo.com/' + video_id
2012-11-28 02:04:46 +01:00
# Retrieve video webpage to extract further information
request = compat_urllib_request.Request(url, None, std_headers)
webpage = self._download_webpage(request, video_id)
2012-11-28 02:04:46 +01:00
# Now we begin extracting as much information as we can from what we
# retrieved. First we extract the information common to all extractors,
# and latter we extract those that are Vimeo specific.
self.report_extraction(video_id)
# Extract the config JSON
try:
config = webpage.split(' = {config:')[1].split(',assets:')[0]
2012-11-28 02:04:46 +01:00
config = json.loads(config)
except:
if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
raise ExtractorError(u'The author has restricted the access to this video, try with the "--referer" option')
else:
raise ExtractorError(u'Unable to extract info section')
2012-12-19 15:19:08 +01:00
2012-11-28 02:04:46 +01:00
# Extract title
video_title = config["video"]["title"]
2012-12-20 16:28:16 +01:00
# Extract uploader and uploader_id
2012-11-28 02:04:46 +01:00
video_uploader = config["video"]["owner"]["name"]
2012-12-20 16:28:16 +01:00
video_uploader_id = config["video"]["owner"]["url"].split('/')[-1]
2012-11-28 02:04:46 +01:00
# Extract video thumbnail
video_thumbnail = config["video"]["thumbnail"]
# Extract video description
video_description = get_element_by_attribute("itemprop", "description", webpage)
2012-11-28 02:04:46 +01:00
if video_description: video_description = clean_html(video_description)
else: video_description = u''
2012-11-28 02:04:46 +01:00
# Extract upload date
video_upload_date = None
mobj = re.search(r'<meta itemprop="dateCreated" content="(\d{4})-(\d{2})-(\d{2})T', webpage)
2012-11-28 02:04:46 +01:00
if mobj is not None:
video_upload_date = mobj.group(1) + mobj.group(2) + mobj.group(3)
2012-11-28 02:04:46 +01:00
# Vimeo specific: extract request signature and timestamp
sig = config['request']['signature']
timestamp = config['request']['timestamp']
# Vimeo specific: extract video codec and quality information
# First consider quality, then codecs, then take everything
# TODO bind to format param
codecs = [('h264', 'mp4'), ('vp8', 'flv'), ('vp6', 'flv')]
files = { 'hd': [], 'sd': [], 'other': []}
for codec_name, codec_extension in codecs:
if codec_name in config["video"]["files"]:
if 'hd' in config["video"]["files"][codec_name]:
files['hd'].append((codec_name, codec_extension, 'hd'))
elif 'sd' in config["video"]["files"][codec_name]:
files['sd'].append((codec_name, codec_extension, 'sd'))
else:
files['other'].append((codec_name, codec_extension, config["video"]["files"][codec_name][0]))
for quality in ('hd', 'sd', 'other'):
if len(files[quality]) > 0:
video_quality = files[quality][0][2]
video_codec = files[quality][0][0]
video_extension = files[quality][0][1]
self.to_screen(u'%s: Downloading %s file at %s quality' % (video_id, video_codec.upper(), video_quality))
2012-11-28 02:04:46 +01:00
break
else:
raise ExtractorError(u'No known codec found')
2012-11-28 02:04:46 +01:00
video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \
%(video_id, sig, timestamp, video_quality, video_codec.upper())
return [{
'id': video_id,
'url': video_url,
'uploader': video_uploader,
2012-12-20 16:28:16 +01:00
'uploader_id': video_uploader_id,
2012-11-28 02:04:46 +01:00
'upload_date': video_upload_date,
'title': video_title,
'ext': video_extension,
'thumbnail': video_thumbnail,
'description': video_description,
}]
2012-08-26 09:11:19 +02:00
class ArteTvIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""arte.tv information extractor."""
_VALID_URL = r'(?:http://)?videos\.arte\.tv/(?:fr|de)/videos/.*'
_LIVE_URL = r'index-[0-9]+\.html$'
IE_NAME = u'arte.tv'
def fetch_webpage(self, url):
request = compat_urllib_request.Request(url)
try:
self.report_download_webpage(url)
webpage = compat_urllib_request.urlopen(request).read()
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
raise ExtractorError(u'Unable to retrieve video webpage: %s' % compat_str(err))
2012-11-28 02:04:46 +01:00
except ValueError as err:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
return webpage
def grep_webpage(self, url, regex, regexFlags, matchTuples):
page = self.fetch_webpage(url)
mobj = re.search(regex, page, regexFlags)
info = {}
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
for (i, key, err) in matchTuples:
if mobj.group(i) is None:
raise ExtractorError(err)
2012-11-28 02:04:46 +01:00
else:
info[key] = mobj.group(i)
return info
def extractLiveStream(self, url):
video_lang = url.split('/')[-4]
info = self.grep_webpage(
url,
r'src="(.*?/videothek_js.*?\.js)',
0,
[
(1, 'url', u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
]
)
http_host = url.split('/')[2]
next_url = 'http://%s%s' % (http_host, compat_urllib_parse.unquote(info.get('url')))
info = self.grep_webpage(
next_url,
r'(s_artestras_scst_geoFRDE_' + video_lang + '.*?)\'.*?' +
'(http://.*?\.swf).*?' +
'(rtmp://.*?)\'',
re.DOTALL,
[
(1, 'path', u'could not extract video path: %s' % url),
(2, 'player', u'could not extract video player: %s' % url),
(3, 'url', u'could not extract video url: %s' % url)
2012-11-28 02:04:46 +01:00
]
)
video_url = u'%s/%s' % (info.get('url'), info.get('path'))
def extractPlus7Stream(self, url):
video_lang = url.split('/')[-3]
info = self.grep_webpage(
url,
r'param name="movie".*?videorefFileUrl=(http[^\'"&]*)',
0,
[
(1, 'url', u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
]
)
next_url = compat_urllib_parse.unquote(info.get('url'))
info = self.grep_webpage(
next_url,
r'<video lang="%s" ref="(http[^\'"&]*)' % video_lang,
0,
[
(1, 'url', u'Could not find <video> tag: %s' % url)
2012-11-28 02:04:46 +01:00
]
)
next_url = compat_urllib_parse.unquote(info.get('url'))
info = self.grep_webpage(
next_url,
r'<video id="(.*?)".*?>.*?' +
'<name>(.*?)</name>.*?' +
'<dateVideo>(.*?)</dateVideo>.*?' +
'<url quality="hd">(.*?)</url>',
re.DOTALL,
[
(1, 'id', u'could not extract video id: %s' % url),
(2, 'title', u'could not extract video title: %s' % url),
(3, 'date', u'could not extract video date: %s' % url),
(4, 'url', u'could not extract video url: %s' % url)
2012-11-28 02:04:46 +01:00
]
)
return {
'id': info.get('id'),
'url': compat_urllib_parse.unquote(info.get('url')),
'uploader': u'arte.tv',
2013-05-05 11:48:47 +02:00
'upload_date': unified_strdate(info.get('date')),
'title': info.get('title').decode('utf-8'),
2012-11-28 02:04:46 +01:00
'ext': u'mp4',
'format': u'NA',
'player_url': None,
}
def _real_extract(self, url):
video_id = url.split('/')[-1]
self.report_extraction(video_id)
if re.search(self._LIVE_URL, video_id) is not None:
self.extractLiveStream(url)
return
else:
info = self.extractPlus7Stream(url)
return [info]
2012-08-26 09:11:19 +02:00
class GenericIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Generic last-resort information extractor."""
_VALID_URL = r'.*'
IE_NAME = u'generic'
def report_download_webpage(self, video_id):
"""Report webpage download."""
if not self._downloader.params.get('test', False):
self._downloader.report_warning(u'Falling back on generic information extractor.')
super(GenericIE, self).report_download_webpage(video_id)
2012-11-28 02:04:46 +01:00
def report_following_redirect(self, new_url):
"""Report information extraction."""
self._downloader.to_screen(u'[redirect] Following redirect to %s' % new_url)
2012-12-19 15:19:08 +01:00
2012-11-28 02:04:46 +01:00
def _test_redirect(self, url):
"""Check if it is a redirect, like url shorteners, in case return the new url."""
2012-11-28 02:04:46 +01:00
class HeadRequest(compat_urllib_request.Request):
def get_method(self):
return "HEAD"
class HEADRedirectHandler(compat_urllib_request.HTTPRedirectHandler):
"""
2012-12-19 15:19:08 +01:00
Subclass the HTTPRedirectHandler to make it use our
2012-11-28 02:04:46 +01:00
HeadRequest also on the redirected URL
"""
2012-12-19 15:19:08 +01:00
def redirect_request(self, req, fp, code, msg, headers, newurl):
2012-11-28 02:04:46 +01:00
if code in (301, 302, 303, 307):
2012-12-19 15:19:08 +01:00
newurl = newurl.replace(' ', '%20')
2012-11-28 02:04:46 +01:00
newheaders = dict((k,v) for k,v in req.headers.items()
if k.lower() not in ("content-length", "content-type"))
2012-12-19 15:19:08 +01:00
return HeadRequest(newurl,
2012-11-28 02:04:46 +01:00
headers=newheaders,
2012-12-19 15:19:08 +01:00
origin_req_host=req.get_origin_req_host(),
unverifiable=True)
else:
raise compat_urllib_error.HTTPError(req.get_full_url(), code, msg, headers, fp)
2012-11-28 02:04:46 +01:00
class HTTPMethodFallback(compat_urllib_request.BaseHandler):
"""
Fallback to GET if HEAD is not allowed (405 HTTP error)
"""
2012-12-19 15:19:08 +01:00
def http_error_405(self, req, fp, code, msg, headers):
2012-11-28 02:04:46 +01:00
fp.read()
fp.close()
newheaders = dict((k,v) for k,v in req.headers.items()
if k.lower() not in ("content-length", "content-type"))
2012-12-19 15:19:08 +01:00
return self.parent.open(compat_urllib_request.Request(req.get_full_url(),
headers=newheaders,
origin_req_host=req.get_origin_req_host(),
2012-11-28 02:04:46 +01:00
unverifiable=True))
# Build our opener
2012-12-19 15:19:08 +01:00
opener = compat_urllib_request.OpenerDirector()
2012-11-28 02:04:46 +01:00
for handler in [compat_urllib_request.HTTPHandler, compat_urllib_request.HTTPDefaultErrorHandler,
HTTPMethodFallback, HEADRedirectHandler,
compat_urllib_request.HTTPErrorProcessor, compat_urllib_request.HTTPSHandler]:
2012-11-28 02:04:46 +01:00
opener.add_handler(handler())
response = opener.open(HeadRequest(url))
new_url = response.geturl()
if url == new_url:
return False
self.report_following_redirect(new_url)
return new_url
2012-11-28 02:04:46 +01:00
def _real_extract(self, url):
new_url = self._test_redirect(url)
if new_url: return [self.url_result(new_url)]
2012-11-28 02:04:46 +01:00
video_id = url.split('/')[-1]
try:
webpage = self._download_webpage(url, video_id)
2012-11-28 02:04:46 +01:00
except ValueError as err:
# since this is the last-resort InfoExtractor, if
# this error is thrown, it'll be thrown here
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
self.report_extraction(video_id)
# Start with something easy: JW Player in SWFObject
mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
if mobj is None:
# Broaden the search a little bit
mobj = re.search(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage)
if mobj is None:
# Broaden the search a little bit: JWPlayer JS loader
mobj = re.search(r'[^A-Za-z0-9]?file:\s*["\'](http[^\'"&]*)', webpage)
2012-11-28 02:04:46 +01:00
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
# It's possible that one of the regexes
# matched, but returned an empty group:
if mobj.group(1) is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
video_url = compat_urllib_parse.unquote(mobj.group(1))
video_id = os.path.basename(video_url)
# here's a fun little line of code for you:
video_extension = os.path.splitext(video_id)[1][1:]
video_id = os.path.splitext(video_id)[0]
# it's tempting to parse this further, but you would
# have to take into account all the variations like
# Video Title - Site Name
# Site Name | Video Title
# Video Title - Tagline | Site Name
# and so on and so forth; it's just not practical
mobj = re.search(r'<title>(.*)</title>', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract title')
2012-12-15 18:25:00 +01:00
video_title = mobj.group(1)
2012-11-28 02:04:46 +01:00
# video uploader is domain name
mobj = re.match(r'(?:https?://)?([^/]*)/.*', url)
if mobj is None:
raise ExtractorError(u'Unable to extract title')
2012-12-15 18:25:00 +01:00
video_uploader = mobj.group(1)
2012-11-28 02:04:46 +01:00
return [{
2012-12-15 18:25:00 +01:00
'id': video_id,
'url': video_url,
2012-11-28 02:04:46 +01:00
'uploader': video_uploader,
'upload_date': None,
'title': video_title,
2012-12-15 18:25:00 +01:00
'ext': video_extension,
2012-11-28 02:04:46 +01:00
}]
class YoutubeSearchIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information Extractor for YouTube search queries."""
_VALID_URL = r'ytsearch(\d+|all)?:[\s\S]+'
_API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
_max_youtube_results = 1000
IE_NAME = u'youtube:search'
def report_download_page(self, query, pagenum):
"""Report attempt to download search page with given number."""
query = query.decode(preferredencoding())
self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum))
def _real_extract(self, query):
mobj = re.match(self._VALID_URL, query)
if mobj is None:
raise ExtractorError(u'Invalid search query "%s"' % query)
2012-11-28 02:04:46 +01:00
prefix, query = query.split(':')
prefix = prefix[8:]
query = query.encode('utf-8')
if prefix == '':
return self._get_n_results(query, 1)
2012-11-28 02:04:46 +01:00
elif prefix == 'all':
self._get_n_results(query, self._max_youtube_results)
2012-11-28 02:04:46 +01:00
else:
try:
n = int(prefix)
if n <= 0:
raise ExtractorError(u'Invalid download number %s for query "%s"' % (n, query))
2012-11-28 02:04:46 +01:00
elif n > self._max_youtube_results:
2013-03-02 11:24:07 +01:00
self._downloader.report_warning(u'ytsearch returns max %i results (you requested %i)' % (self._max_youtube_results, n))
2012-11-28 02:04:46 +01:00
n = self._max_youtube_results
return self._get_n_results(query, n)
2012-11-28 02:04:46 +01:00
except ValueError: # parsing prefix as integer fails
return self._get_n_results(query, 1)
2012-11-28 02:04:46 +01:00
def _get_n_results(self, query, n):
"""Get a specified number of results for a query"""
2012-11-28 02:04:46 +01:00
video_ids = []
pagenum = 0
limit = n
while (50 * pagenum) < limit:
self.report_download_page(query, pagenum+1)
result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1)
request = compat_urllib_request.Request(result_url)
try:
data = compat_urllib_request.urlopen(request).read().decode('utf-8')
2012-11-28 02:04:46 +01:00
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
raise ExtractorError(u'Unable to download API page: %s' % compat_str(err))
2012-11-28 02:04:46 +01:00
api_response = json.loads(data)['data']
if not 'items' in api_response:
raise ExtractorError(u'[youtube] No video results')
2012-11-28 02:04:46 +01:00
new_ids = list(video['id'] for video in api_response['items'])
video_ids += new_ids
limit = min(n, api_response['totalItems'])
pagenum += 1
if len(video_ids) > n:
video_ids = video_ids[:n]
videos = [self.url_result('http://www.youtube.com/watch?v=%s' % id, 'Youtube') for id in video_ids]
return videos
class GoogleSearchIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information Extractor for Google Video search queries."""
_VALID_URL = r'gvsearch(\d+|all)?:[\s\S]+'
_TEMPLATE_URL = 'http://video.google.com/videosearch?q=%s+site:video.google.com&start=%s&hl=en'
_VIDEO_INDICATOR = r'<a href="http://video\.google\.com/videoplay\?docid=([^"\&]+)'
_MORE_PAGES_INDICATOR = r'class="pn" id="pnnext"'
_max_google_results = 1000
IE_NAME = u'video.google:search'
def report_download_page(self, query, pagenum):
"""Report attempt to download playlist page with given number."""
query = query.decode(preferredencoding())
self.to_screen(u'query "%s": Downloading page %s' % (query, pagenum))
2012-11-28 02:04:46 +01:00
def _real_extract(self, query):
mobj = re.match(self._VALID_URL, query)
if mobj is None:
raise ExtractorError(u'Invalid search query "%s"' % query)
2012-11-28 02:04:46 +01:00
prefix, query = query.split(':')
prefix = prefix[8:]
query = query.encode('utf-8')
if prefix == '':
self._download_n_results(query, 1)
return
elif prefix == 'all':
self._download_n_results(query, self._max_google_results)
return
else:
try:
n = int(prefix)
if n <= 0:
raise ExtractorError(u'Invalid download number %s for query "%s"' % (n, query))
2012-11-28 02:04:46 +01:00
elif n > self._max_google_results:
2013-03-02 11:24:07 +01:00
self._downloader.report_warning(u'gvsearch returns max %i results (you requested %i)' % (self._max_google_results, n))
2012-11-28 02:04:46 +01:00
n = self._max_google_results
self._download_n_results(query, n)
return
except ValueError: # parsing prefix as integer fails
self._download_n_results(query, 1)
return
def _download_n_results(self, query, n):
"""Downloads a specified number of results for a query"""
video_ids = []
pagenum = 0
while True:
self.report_download_page(query, pagenum)
result_url = self._TEMPLATE_URL % (compat_urllib_parse.quote_plus(query), pagenum*10)
request = compat_urllib_request.Request(result_url)
try:
page = compat_urllib_request.urlopen(request).read()
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
raise ExtractorError(u'Unable to download webpage: %s' % compat_str(err))
2012-11-28 02:04:46 +01:00
# Extract video identifiers
for mobj in re.finditer(self._VIDEO_INDICATOR, page):
video_id = mobj.group(1)
if video_id not in video_ids:
video_ids.append(video_id)
if len(video_ids) == n:
# Specified n videos reached
for id in video_ids:
self._downloader.download(['http://video.google.com/videoplay?docid=%s' % id])
return
if re.search(self._MORE_PAGES_INDICATOR, page) is None:
for id in video_ids:
self._downloader.download(['http://video.google.com/videoplay?docid=%s' % id])
return
pagenum = pagenum + 1
class YahooSearchIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information Extractor for Yahoo! Video search queries."""
_WORKING = False
2012-11-28 02:04:46 +01:00
_VALID_URL = r'yvsearch(\d+|all)?:[\s\S]+'
_TEMPLATE_URL = 'http://video.yahoo.com/search/?p=%s&o=%s'
_VIDEO_INDICATOR = r'href="http://video\.yahoo\.com/watch/([0-9]+/[0-9]+)"'
_MORE_PAGES_INDICATOR = r'\s*Next'
_max_yahoo_results = 1000
IE_NAME = u'video.yahoo:search'
def report_download_page(self, query, pagenum):
"""Report attempt to download playlist page with given number."""
query = query.decode(preferredencoding())
self.to_screen(u'query "%s": Downloading page %s' % (query, pagenum))
2012-11-28 02:04:46 +01:00
def _real_extract(self, query):
mobj = re.match(self._VALID_URL, query)
if mobj is None:
raise ExtractorError(u'Invalid search query "%s"' % query)
2012-11-28 02:04:46 +01:00
prefix, query = query.split(':')
prefix = prefix[8:]
query = query.encode('utf-8')
if prefix == '':
self._download_n_results(query, 1)
return
elif prefix == 'all':
self._download_n_results(query, self._max_yahoo_results)
return
else:
try:
n = int(prefix)
if n <= 0:
raise ExtractorError(u'Invalid download number %s for query "%s"' % (n, query))
2012-11-28 02:04:46 +01:00
elif n > self._max_yahoo_results:
2013-03-02 11:24:07 +01:00
self._downloader.report_warning(u'yvsearch returns max %i results (you requested %i)' % (self._max_yahoo_results, n))
2012-11-28 02:04:46 +01:00
n = self._max_yahoo_results
self._download_n_results(query, n)
return
except ValueError: # parsing prefix as integer fails
self._download_n_results(query, 1)
return
def _download_n_results(self, query, n):
"""Downloads a specified number of results for a query"""
video_ids = []
already_seen = set()
pagenum = 1
while True:
self.report_download_page(query, pagenum)
result_url = self._TEMPLATE_URL % (compat_urllib_parse.quote_plus(query), pagenum)
request = compat_urllib_request.Request(result_url)
try:
page = compat_urllib_request.urlopen(request).read()
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
raise ExtractorError(u'Unable to download webpage: %s' % compat_str(err))
2012-11-28 02:04:46 +01:00
# Extract video identifiers
for mobj in re.finditer(self._VIDEO_INDICATOR, page):
video_id = mobj.group(1)
if video_id not in already_seen:
video_ids.append(video_id)
already_seen.add(video_id)
if len(video_ids) == n:
# Specified n videos reached
for id in video_ids:
self._downloader.download(['http://video.yahoo.com/watch/%s' % id])
return
if re.search(self._MORE_PAGES_INDICATOR, page) is None:
for id in video_ids:
self._downloader.download(['http://video.yahoo.com/watch/%s' % id])
return
pagenum = pagenum + 1
class YoutubePlaylistIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information Extractor for YouTube playlists."""
_VALID_URL = r"""(?:
(?:https?://)?
(?:\w+\.)?
youtube\.com/
(?:
(?:course|view_play_list|my_playlists|artist|playlist|watch)
\? (?:.*?&)*? (?:p|a|list)=
| p/
)
((?:PL|EC|UU)?[0-9A-Za-z-_]{10,})
.*
|
((?:PL|EC|UU)[0-9A-Za-z-_]{10,})
)"""
_TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json'
_MAX_RESULTS = 50
2012-11-28 02:04:46 +01:00
IE_NAME = u'youtube:playlist'
@classmethod
def suitable(cls, url):
"""Receives a URL and returns True if suitable for this IE."""
return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
2012-11-28 02:04:46 +01:00
def _real_extract(self, url):
# Extract playlist id
mobj = re.match(self._VALID_URL, url, re.VERBOSE)
2012-11-28 02:04:46 +01:00
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
# Download playlist videos from API
playlist_id = mobj.group(1) or mobj.group(2)
page_num = 1
videos = []
2012-11-28 02:04:46 +01:00
while True:
url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, self._MAX_RESULTS * (page_num - 1) + 1)
page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num)
2012-11-28 02:04:46 +01:00
try:
response = json.loads(page)
except ValueError as err:
raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err))
2012-11-28 02:04:46 +01:00
if 'feed' not in response:
raise ExtractorError(u'Got a malformed response from YouTube API')
playlist_title = response['feed']['title']['$t']
if 'entry' not in response['feed']:
# Number of videos is a multiple of self._MAX_RESULTS
break
videos += [ (entry['yt$position']['$t'], entry['content']['src'])
for entry in response['feed']['entry']
if 'content' in entry ]
if len(response['feed']['entry']) < self._MAX_RESULTS:
2012-11-28 02:04:46 +01:00
break
page_num += 1
2012-11-28 02:04:46 +01:00
2013-02-26 21:21:50 +01:00
videos = [v[1] for v in sorted(videos)]
url_results = [self.url_result(url, 'Youtube') for url in videos]
2013-04-20 18:57:05 +02:00
return [self.playlist_result(url_results, playlist_id, playlist_title)]
2012-10-14 13:48:18 +02:00
class YoutubeChannelIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information Extractor for YouTube channels."""
_VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)"
2012-11-28 02:04:46 +01:00
_TEMPLATE_URL = 'http://www.youtube.com/channel/%s/videos?sort=da&flow=list&view=0&page=%s&gl=US&hl=en'
_MORE_PAGES_INDICATOR = 'yt-uix-load-more'
_MORE_PAGES_URL = 'http://www.youtube.com/channel_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s'
2012-11-28 02:04:46 +01:00
IE_NAME = u'youtube:channel'
def extract_videos_from_page(self, page):
ids_in_page = []
for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page):
if mobj.group(1) not in ids_in_page:
ids_in_page.append(mobj.group(1))
return ids_in_page
2012-11-28 02:04:46 +01:00
def _real_extract(self, url):
# Extract channel id
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
# Download channel page
2012-11-28 02:04:46 +01:00
channel_id = mobj.group(1)
video_ids = []
pagenum = 1
url = self._TEMPLATE_URL % (channel_id, pagenum)
page = self._download_webpage(url, channel_id,
u'Downloading page #%s' % pagenum)
2012-11-28 02:04:46 +01:00
# Extract video identifiers
ids_in_page = self.extract_videos_from_page(page)
video_ids.extend(ids_in_page)
2012-11-28 02:04:46 +01:00
# Download any subsequent channel pages using the json-based channel_ajax query
if self._MORE_PAGES_INDICATOR in page:
while True:
pagenum = pagenum + 1
url = self._MORE_PAGES_URL % (pagenum, channel_id)
page = self._download_webpage(url, channel_id,
u'Downloading page #%s' % pagenum)
page = json.loads(page)
ids_in_page = self.extract_videos_from_page(page['content_html'])
video_ids.extend(ids_in_page)
if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']:
break
2012-11-28 02:04:46 +01:00
self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids)))
urls = ['http://www.youtube.com/watch?v=%s' % id for id in video_ids]
url_entries = [self.url_result(url, 'Youtube') for url in urls]
return [self.playlist_result(url_entries, channel_id)]
2012-10-14 13:48:18 +02:00
class YoutubeUserIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information Extractor for YouTube users."""
2012-11-28 02:04:46 +01:00
_VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/user/)|ytuser:)([A-Za-z0-9_-]+)'
_TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s'
_GDATA_PAGE_SIZE = 50
_GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d'
_VIDEO_INDICATOR = r'/watch\?v=(.+?)[\<&]'
IE_NAME = u'youtube:user'
2012-11-28 02:04:46 +01:00
def _real_extract(self, url):
# Extract username
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
username = mobj.group(1)
2012-11-28 02:04:46 +01:00
# Download video ids using YouTube Data API. Result size per
# query is limited (currently to 50 videos) so we need to query
# page by page until there are no video ids - it means we got
# all of them.
2012-11-28 02:04:46 +01:00
video_ids = []
pagenum = 0
2012-11-28 02:04:46 +01:00
while True:
start_index = pagenum * self._GDATA_PAGE_SIZE + 1
gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index)
page = self._download_webpage(gdata_url, username,
u'Downloading video ids from %d to %d' % (start_index, start_index + self._GDATA_PAGE_SIZE))
2012-11-28 02:04:46 +01:00
# Extract video identifiers
ids_in_page = []
2012-11-28 02:04:46 +01:00
for mobj in re.finditer(self._VIDEO_INDICATOR, page):
if mobj.group(1) not in ids_in_page:
ids_in_page.append(mobj.group(1))
2012-11-28 02:04:46 +01:00
video_ids.extend(ids_in_page)
2012-11-28 02:04:46 +01:00
# A little optimization - if current page is not
# "full", ie. does not contain PAGE_SIZE video ids then
# we can assume that this page is the last one - there
# are no more ids on further pages - no need to query
# again.
2012-11-28 02:04:46 +01:00
if len(ids_in_page) < self._GDATA_PAGE_SIZE:
break
2012-11-28 02:04:46 +01:00
pagenum += 1
urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids]
url_results = [self.url_result(url, 'Youtube') for url in urls]
return [self.playlist_result(url_results, playlist_title = username)]
class BlipTVUserIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information Extractor for blip.tv users."""
2012-11-28 02:04:46 +01:00
_VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?blip\.tv/)|bliptvuser:)([^/]+)/*$'
_PAGE_SIZE = 12
IE_NAME = u'blip.tv:user'
2012-11-28 02:04:46 +01:00
def _real_extract(self, url):
# Extract username
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
username = mobj.group(1)
2012-11-28 02:04:46 +01:00
page_base = 'http://m.blip.tv/pr/show_get_full_episode_list?users_id=%s&lite=0&esi=1'
page = self._download_webpage(url, username, u'Downloading user page')
mobj = re.search(r'data-users-id="([^"]+)"', page)
page_base = page_base % mobj.group(1)
2012-11-28 02:04:46 +01:00
# Download video ids using BlipTV Ajax calls. Result size per
# query is limited (currently to 12 videos) so we need to query
# page by page until there are no video ids - it means we got
# all of them.
2012-11-28 02:04:46 +01:00
video_ids = []
pagenum = 1
2012-11-28 02:04:46 +01:00
while True:
url = page_base + "&page=" + str(pagenum)
page = self._download_webpage(url, username,
u'Downloading video ids from page %d' % pagenum)
2012-11-28 02:04:46 +01:00
# Extract video identifiers
ids_in_page = []
2012-11-28 02:04:46 +01:00
for mobj in re.finditer(r'href="/([^"]+)"', page):
if mobj.group(1) not in ids_in_page:
ids_in_page.append(unescapeHTML(mobj.group(1)))
2012-11-28 02:04:46 +01:00
video_ids.extend(ids_in_page)
2012-11-28 02:04:46 +01:00
# A little optimization - if current page is not
# "full", ie. does not contain PAGE_SIZE video ids then
# we can assume that this page is the last one - there
# are no more ids on further pages - no need to query
# again.
2012-11-28 02:04:46 +01:00
if len(ids_in_page) < self._PAGE_SIZE:
break
2012-11-28 02:04:46 +01:00
pagenum += 1
urls = [u'http://blip.tv/%s' % video_id for video_id in video_ids]
url_entries = [self.url_result(url, 'BlipTV') for url in urls]
return [self.playlist_result(url_entries, playlist_title = username)]
class DepositFilesIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information extractor for depositfiles.com"""
_VALID_URL = r'(?:http://)?(?:\w+\.)?depositfiles\.com/(?:../(?#locale))?files/(.+)'
def _real_extract(self, url):
file_id = url.split('/')[-1]
# Rebuild url in english locale
url = 'http://depositfiles.com/en/files/' + file_id
# Retrieve file webpage with 'Free download' button pressed
free_download_indication = { 'gateway_result' : '1' }
request = compat_urllib_request.Request(url, compat_urllib_parse.urlencode(free_download_indication))
try:
self.report_download_webpage(file_id)
webpage = compat_urllib_request.urlopen(request).read()
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
raise ExtractorError(u'Unable to retrieve file webpage: %s' % compat_str(err))
2012-11-28 02:04:46 +01:00
# Search for the real file URL
mobj = re.search(r'<form action="(http://fileshare.+?)"', webpage)
if (mobj is None) or (mobj.group(1) is None):
# Try to figure out reason of the error.
mobj = re.search(r'<strong>(Attention.*?)</strong>', webpage, re.DOTALL)
if (mobj is not None) and (mobj.group(1) is not None):
restriction_message = re.sub('\s+', ' ', mobj.group(1)).strip()
raise ExtractorError(u'%s' % restriction_message)
2012-11-28 02:04:46 +01:00
else:
raise ExtractorError(u'Unable to extract download URL from: %s' % url)
2012-11-28 02:04:46 +01:00
file_url = mobj.group(1)
file_extension = os.path.splitext(file_url)[1][1:]
# Search for file title
mobj = re.search(r'<b title="(.*?)">', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract title')
2012-11-28 02:04:46 +01:00
file_title = mobj.group(1).decode('utf-8')
return [{
'id': file_id.decode('utf-8'),
'url': file_url.decode('utf-8'),
'uploader': None,
'upload_date': None,
'title': file_title,
'ext': file_extension.decode('utf-8'),
}]
class FacebookIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information Extractor for Facebook"""
_VALID_URL = r'^(?:https?://)?(?:\w+\.)?facebook\.com/(?:video/video|photo)\.php\?(?:.*?)v=(?P<ID>\d+)(?:.*)'
_LOGIN_URL = 'https://login.facebook.com/login.php?m&next=http%3A%2F%2Fm.facebook.com%2Fhome.php&'
_NETRC_MACHINE = 'facebook'
IE_NAME = u'facebook'
def report_login(self):
"""Report attempt to log in."""
self.to_screen(u'Logging in')
2012-11-28 02:04:46 +01:00
def _real_initialize(self):
if self._downloader is None:
return
useremail = None
password = None
downloader_params = self._downloader.params
# Attempt to use provided username and password or .netrc data
if downloader_params.get('username', None) is not None:
useremail = downloader_params['username']
password = downloader_params['password']
elif downloader_params.get('usenetrc', False):
try:
info = netrc.netrc().authenticators(self._NETRC_MACHINE)
if info is not None:
useremail = info[0]
password = info[2]
else:
raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE)
except (IOError, netrc.NetrcParseError) as err:
2013-03-02 11:24:07 +01:00
self._downloader.report_warning(u'parsing .netrc: %s' % compat_str(err))
2012-11-28 02:04:46 +01:00
return
if useremail is None:
return
# Log in
login_form = {
'email': useremail,
'pass': password,
'login': 'Log+In'
}
request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form))
try:
self.report_login()
login_results = compat_urllib_request.urlopen(request).read()
if re.search(r'<form(.*)name="login"(.*)</form>', login_results) is not None:
2013-03-02 11:24:07 +01:00
self._downloader.report_warning(u'unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.')
2012-11-28 02:04:46 +01:00
return
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
2013-03-02 11:24:07 +01:00
self._downloader.report_warning(u'unable to log in: %s' % compat_str(err))
2012-11-28 02:04:46 +01:00
return
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
video_id = mobj.group('ID')
2013-01-25 16:54:48 +01:00
url = 'https://www.facebook.com/video/video.php?v=%s' % video_id
webpage = self._download_webpage(url, video_id)
2013-04-18 04:41:48 +02:00
BEFORE = '{swf.addParam(param[0], param[1]);});\n'
2013-01-25 16:54:48 +01:00
AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});'
m = re.search(re.escape(BEFORE) + '(.*?)' + re.escape(AFTER), webpage)
if not m:
raise ExtractorError(u'Cannot parse data')
data = dict(json.loads(m.group(1)))
2013-02-01 17:56:22 +01:00
params_raw = compat_urllib_parse.unquote(data['params'])
params = json.loads(params_raw)
2013-04-18 04:41:48 +02:00
video_data = params['video_data'][0]
video_url = video_data.get('hd_src')
2013-02-18 23:12:48 +01:00
if not video_url:
2013-04-18 04:41:48 +02:00
video_url = video_data['sd_src']
2013-02-18 23:12:48 +01:00
if not video_url:
raise ExtractorError(u'Cannot find video URL')
2013-04-18 04:41:48 +02:00
video_duration = int(video_data['video_duration'])
thumbnail = video_data['thumbnail_src']
2013-01-25 16:54:48 +01:00
m = re.search('<h2 class="uiHeaderTitle">([^<]+)</h2>', webpage)
if not m:
raise ExtractorError(u'Cannot find title in webpage')
video_title = unescapeHTML(m.group(1))
info = {
'id': video_id,
'title': video_title,
'url': video_url,
'ext': 'mp4',
'duration': video_duration,
2013-04-18 04:41:48 +02:00
'thumbnail': thumbnail,
2013-01-25 16:54:48 +01:00
}
return [info]
2012-11-28 02:04:46 +01:00
class BlipTVIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information extractor for blip.tv"""
_VALID_URL = r'^(?:https?://)?(?:\w+\.)?blip\.tv(/.+)$'
_URL_EXT = r'^.*\.([a-z0-9]+)$'
IE_NAME = u'blip.tv'
def report_direct_download(self, title):
"""Report information extraction."""
self.to_screen(u'%s: Direct download detected' % title)
2012-11-28 02:04:46 +01:00
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
urlp = compat_urllib_parse_urlparse(url)
if urlp.path.startswith('/play/'):
request = compat_urllib_request.Request(url)
response = compat_urllib_request.urlopen(request)
redirecturl = response.geturl()
rurlp = compat_urllib_parse_urlparse(redirecturl)
file_id = compat_parse_qs(rurlp.fragment)['file'][0].rpartition('/')[2]
url = 'http://blip.tv/a/a-' + file_id
return self._real_extract(url)
2012-11-28 02:04:46 +01:00
if '?' in url:
cchar = '&'
else:
cchar = '?'
json_url = url + cchar + 'skin=json&version=2&no_wrap=1'
2012-12-11 17:00:11 +01:00
request = compat_urllib_request.Request(json_url)
request.add_header('User-Agent', 'iTunes/10.6.1')
2012-11-28 02:04:46 +01:00
self.report_extraction(mobj.group(1))
info = None
try:
urlh = compat_urllib_request.urlopen(request)
if urlh.headers.get('Content-Type', '').startswith('video/'): # Direct download
basename = url.split('/')[-1]
title,ext = os.path.splitext(basename)
title = title.decode('UTF-8')
ext = ext.replace('.', '')
self.report_direct_download(title)
info = {
'id': title,
'url': url,
'uploader': None,
'upload_date': None,
'title': title,
'ext': ext,
'urlhandle': urlh
}
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
raise ExtractorError(u'ERROR: unable to download video info webpage: %s' % compat_str(err))
2012-11-28 02:04:46 +01:00
if info is None: # Regular URL
try:
2012-12-11 17:00:11 +01:00
json_code_bytes = urlh.read()
json_code = json_code_bytes.decode('utf-8')
2012-11-28 02:04:46 +01:00
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
raise ExtractorError(u'Unable to read video info webpage: %s' % compat_str(err))
2012-11-28 02:04:46 +01:00
try:
json_data = json.loads(json_code)
if 'Post' in json_data:
data = json_data['Post']
else:
data = json_data
upload_date = datetime.datetime.strptime(data['datestamp'], '%m-%d-%y %H:%M%p').strftime('%Y%m%d')
video_url = data['media']['url']
umobj = re.match(self._URL_EXT, video_url)
if umobj is None:
raise ValueError('Can not determine filename extension')
ext = umobj.group(1)
info = {
'id': data['item_id'],
'url': video_url,
'uploader': data['display_name'],
'upload_date': upload_date,
'title': data['title'],
'ext': ext,
'format': data['media']['mimeType'],
'thumbnail': data['thumbnailUrl'],
'description': data['description'],
'player_url': data['embedUrl'],
'user_agent': 'iTunes/10.6.1',
2012-11-28 02:04:46 +01:00
}
except (ValueError,KeyError) as err:
raise ExtractorError(u'Unable to parse video information: %s' % repr(err))
2012-11-28 02:04:46 +01:00
return [info]
class MyVideoIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information Extractor for myvideo.de."""
_VALID_URL = r'(?:http://)?(?:www\.)?myvideo\.de/watch/([0-9]+)/([^?/]+).*'
IE_NAME = u'myvideo'
def _real_extract(self,url):
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
video_id = mobj.group(1)
# Get video webpage
2013-01-01 20:52:59 +01:00
webpage_url = 'http://www.myvideo.de/watch/%s' % video_id
webpage = self._download_webpage(webpage_url, video_id)
2012-11-28 02:04:46 +01:00
self.report_extraction(video_id)
2013-03-29 15:59:13 +01:00
mobj = re.search(r'<link rel=\'image_src\' href=\'(http://is[0-9].myvideo\.de/de/movie[0-9]+/[a-f0-9]+)/thumbs/.*?\.jpg\'',
2012-11-28 02:04:46 +01:00
webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract media URL')
2012-11-28 02:04:46 +01:00
video_url = mobj.group(1) + ('/%s.flv' % video_id)
mobj = re.search('<title>([^<]+)</title>', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract title')
2012-11-28 02:04:46 +01:00
video_title = mobj.group(1)
return [{
'id': video_id,
'url': video_url,
'uploader': None,
'upload_date': None,
'title': video_title,
'ext': u'flv',
}]
class ComedyCentralIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information extractor for The Daily Show and Colbert Report """
# urls can be abbreviations like :thedailyshow or :colbert
2012-12-19 15:19:08 +01:00
# urls for episodes like:
# or urls for clips like: http://www.thedailyshow.com/watch/mon-december-10-2012/any-given-gun-day
# or: http://www.colbertnation.com/the-colbert-report-videos/421667/november-29-2012/moon-shattering-news
2012-12-19 15:19:08 +01:00
# or: http://www.colbertnation.com/the-colbert-report-collections/422008/festival-of-lights/79524
_VALID_URL = r"""^(:(?P<shortname>tds|thedailyshow|cr|colbert|colbertnation|colbertreport)
|(https?://)?(www\.)?
(?P<showname>thedailyshow|colbertnation)\.com/
(full-episodes/(?P<episode>.*)|
(?P<clip>
(the-colbert-report-(videos|collections)/(?P<clipID>[0-9]+)/[^/]*/(?P<cntitle>.*?))
|(watch/(?P<date>[^/]*)/(?P<tdstitle>.*)))))
2012-12-19 15:19:08 +01:00
$"""
2012-11-28 02:04:46 +01:00
_available_formats = ['3500', '2200', '1700', '1200', '750', '400']
_video_extensions = {
'3500': 'mp4',
'2200': 'mp4',
'1700': 'mp4',
'1200': 'mp4',
'750': 'mp4',
'400': 'mp4',
}
_video_dimensions = {
'3500': '1280x720',
'2200': '960x540',
'1700': '768x432',
'1200': '640x360',
'750': '512x288',
'400': '384x216',
}
@classmethod
def suitable(cls, url):
"""Receives a URL and returns True if suitable for this IE."""
return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
2012-11-28 02:04:46 +01:00
def _print_formats(self, formats):
print('Available formats:')
for x in formats:
print('%s\t:\t%s\t[%s]' %(x, self._video_extensions.get(x, 'mp4'), self._video_dimensions.get(x, '???')))
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url, re.VERBOSE)
2012-11-28 02:04:46 +01:00
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
if mobj.group('shortname'):
if mobj.group('shortname') in ('tds', 'thedailyshow'):
url = u'http://www.thedailyshow.com/full-episodes/'
else:
url = u'http://www.colbertnation.com/full-episodes/'
mobj = re.match(self._VALID_URL, url, re.VERBOSE)
2012-11-28 02:04:46 +01:00
assert mobj is not None
if mobj.group('clip'):
if mobj.group('showname') == 'thedailyshow':
epTitle = mobj.group('tdstitle')
else:
epTitle = mobj.group('cntitle')
dlNewest = False
2012-11-28 02:04:46 +01:00
else:
dlNewest = not mobj.group('episode')
if dlNewest:
epTitle = mobj.group('showname')
else:
epTitle = mobj.group('episode')
2012-11-28 02:04:46 +01:00
self.report_extraction(epTitle)
2013-05-04 02:53:26 +02:00
webpage,htmlHandle = self._download_webpage_handle(url, epTitle)
2012-11-28 02:04:46 +01:00
if dlNewest:
url = htmlHandle.geturl()
mobj = re.match(self._VALID_URL, url, re.VERBOSE)
2012-11-28 02:04:46 +01:00
if mobj is None:
2013-05-04 02:53:26 +02:00
raise ExtractorError(u'Invalid redirected URL: ' + url)
2012-11-28 02:04:46 +01:00
if mobj.group('episode') == '':
2013-05-04 02:53:26 +02:00
raise ExtractorError(u'Redirected URL is still not specific: ' + url)
2012-11-28 02:04:46 +01:00
epTitle = mobj.group('episode')
2013-01-06 21:36:01 +01:00
mMovieParams = re.findall('(?:<param name="movie" value="|var url = ")(http://media.mtvnservices.com/([^"]*(?:episode|video).*?:.*?))"', webpage)
2012-11-28 02:04:46 +01:00
if len(mMovieParams) == 0:
# The Colbert Report embeds the information in a without
# a URL prefix; so extract the alternate reference
# and then add the URL prefix manually.
2013-01-06 21:36:01 +01:00
altMovieParams = re.findall('data-mgid="([^"]*(?:episode|video).*?:.*?)"', webpage)
2012-11-28 02:04:46 +01:00
if len(altMovieParams) == 0:
2013-05-04 02:53:26 +02:00
raise ExtractorError(u'unable to find Flash URL in webpage ' + url)
2012-11-28 02:04:46 +01:00
else:
mMovieParams = [("http://media.mtvnservices.com/" + altMovieParams[0], altMovieParams[0])]
2012-12-19 15:19:08 +01:00
2012-11-28 02:04:46 +01:00
uri = mMovieParams[0][1]
indexUrl = 'http://shadow.comedycentral.com/feeds/video_player/mrss/?' + compat_urllib_parse.urlencode({'uri': uri})
indexXml = self._download_webpage(indexUrl, epTitle,
u'Downloading show index',
u'unable to download episode index')
2012-11-28 02:04:46 +01:00
results = []
idoc = xml.etree.ElementTree.fromstring(indexXml)
itemEls = idoc.findall('.//item')
2013-01-06 21:35:20 +01:00
for partNum,itemEl in enumerate(itemEls):
2012-11-28 02:04:46 +01:00
mediaId = itemEl.findall('./guid')[0].text
shortMediaId = mediaId.split(':')[-1]
showId = mediaId.split(':')[-2].replace('.com', '')
officialTitle = itemEl.findall('./title')[0].text
officialDate = unified_strdate(itemEl.findall('./pubDate')[0].text)
2012-11-28 02:04:46 +01:00
configUrl = ('http://www.comedycentral.com/global/feeds/entertainment/media/mediaGenEntertainment.jhtml?' +
compat_urllib_parse.urlencode({'uri': mediaId}))
configXml = self._download_webpage(configUrl, epTitle,
u'Downloading configuration for %s' % shortMediaId)
2012-11-28 02:04:46 +01:00
cdoc = xml.etree.ElementTree.fromstring(configXml)
turls = []
for rendition in cdoc.findall('.//rendition'):
finfo = (rendition.attrib['bitrate'], rendition.findall('./src')[0].text)
turls.append(finfo)
if len(turls) == 0:
self._downloader.report_error(u'unable to download ' + mediaId + ': No videos found')
2012-11-28 02:04:46 +01:00
continue
2012-12-19 15:19:08 +01:00
2012-11-28 02:04:46 +01:00
if self._downloader.params.get('listformats', None):
self._print_formats([i[0] for i in turls])
return
# For now, just pick the highest bitrate
format,rtmp_video_url = turls[-1]
2012-11-28 02:04:46 +01:00
# Get the format arg from the arg stream
req_format = self._downloader.params.get('format', None)
# Select format if we can find one
for f,v in turls:
if f == req_format:
format, rtmp_video_url = f, v
2012-11-28 02:04:46 +01:00
break
m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp.comedystor/.*)$', rtmp_video_url)
if not m:
raise ExtractorError(u'Cannot transform RTMP url')
base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
video_url = base + m.group('finalid')
2012-11-28 02:04:46 +01:00
2013-01-06 21:35:20 +01:00
effTitle = showId + u'-' + epTitle + u' part ' + compat_str(partNum+1)
2012-11-28 02:04:46 +01:00
info = {
'id': shortMediaId,
'url': video_url,
'uploader': showId,
'upload_date': officialDate,
'title': effTitle,
'ext': 'mp4',
'format': format,
'thumbnail': None,
'description': officialTitle,
}
results.append(info)
2012-12-19 15:19:08 +01:00
2012-11-28 02:04:46 +01:00
return results
class EscapistIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information extractor for The Escapist """
_VALID_URL = r'^(https?://)?(www\.)?escapistmagazine\.com/videos/view/(?P<showname>[^/]+)/(?P<episode>[^/?]+)[/?]?.*$'
IE_NAME = u'escapist'
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
showName = mobj.group('showname')
videoId = mobj.group('episode')
self.report_extraction(showName)
webPage = self._download_webpage(url, showName)
2012-11-28 02:04:46 +01:00
descMatch = re.search('<meta name="description" content="([^"]*)"', webPage)
description = unescapeHTML(descMatch.group(1))
imgMatch = re.search('<meta property="og:image" content="([^"]*)"', webPage)
imgUrl = unescapeHTML(imgMatch.group(1))
playerUrlMatch = re.search('<meta property="og:video" content="([^"]*)"', webPage)
playerUrl = unescapeHTML(playerUrlMatch.group(1))
configUrlMatch = re.search('config=(.*)$', playerUrl)
configUrl = compat_urllib_parse.unquote(configUrlMatch.group(1))
configJSON = self._download_webpage(configUrl, showName,
u'Downloading configuration',
u'unable to download configuration')
2012-11-28 02:04:46 +01:00
# Technically, it's JavaScript, not JSON
configJSON = configJSON.replace("'", '"')
try:
config = json.loads(configJSON)
except (ValueError,) as err:
raise ExtractorError(u'Invalid JSON in configuration file: ' + compat_str(err))
2012-11-28 02:04:46 +01:00
playlist = config['playlist']
videoUrl = playlist[1]['url']
info = {
'id': videoId,
'url': videoUrl,
'uploader': showName,
'upload_date': None,
'title': showName,
'ext': 'mp4',
2012-11-28 02:04:46 +01:00
'thumbnail': imgUrl,
'description': description,
'player_url': playerUrl,
}
return [info]
class CollegeHumorIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information extractor for collegehumor.com"""
_WORKING = False
2012-11-28 02:04:46 +01:00
_VALID_URL = r'^(?:https?://)?(?:www\.)?collegehumor\.com/video/(?P<videoid>[0-9]+)/(?P<shorttitle>.*)$'
IE_NAME = u'collegehumor'
def report_manifest(self, video_id):
2012-11-28 02:04:46 +01:00
"""Report information extraction."""
self.to_screen(u'%s: Downloading XML manifest' % video_id)
2012-11-28 02:04:46 +01:00
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
video_id = mobj.group('videoid')
info = {
'id': video_id,
'uploader': None,
'upload_date': None,
}
self.report_extraction(video_id)
xmlUrl = 'http://www.collegehumor.com/moogaloop/video/' + video_id
2012-11-28 02:04:46 +01:00
try:
metaXml = compat_urllib_request.urlopen(xmlUrl).read()
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
raise ExtractorError(u'Unable to download video info XML: %s' % compat_str(err))
2012-11-28 02:04:46 +01:00
mdoc = xml.etree.ElementTree.fromstring(metaXml)
try:
videoNode = mdoc.findall('./video')[0]
info['description'] = videoNode.findall('./description')[0].text
info['title'] = videoNode.findall('./caption')[0].text
info['thumbnail'] = videoNode.findall('./thumbnail')[0].text
manifest_url = videoNode.findall('./file')[0].text
2012-11-28 02:04:46 +01:00
except IndexError:
raise ExtractorError(u'Invalid metadata XML file')
2012-11-28 02:04:46 +01:00
manifest_url += '?hdcore=2.10.3'
self.report_manifest(video_id)
try:
manifestXml = compat_urllib_request.urlopen(manifest_url).read()
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
raise ExtractorError(u'Unable to download video info XML: %s' % compat_str(err))
adoc = xml.etree.ElementTree.fromstring(manifestXml)
try:
media_node = adoc.findall('./{http://ns.adobe.com/f4m/1.0}media')[0]
node_id = media_node.attrib['url']
video_id = adoc.findall('./{http://ns.adobe.com/f4m/1.0}id')[0].text
except IndexError as err:
raise ExtractorError(u'Invalid manifest file')
url_pr = compat_urllib_parse_urlparse(manifest_url)
url = url_pr.scheme + '://' + url_pr.netloc + '/z' + video_id[:-2] + '/' + node_id + 'Seg1-Frag1'
info['url'] = url
info['ext'] = 'f4f'
2012-11-28 02:04:46 +01:00
return [info]
class XVideosIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information extractor for xvideos.com"""
2012-11-28 02:04:46 +01:00
_VALID_URL = r'^(?:https?://)?(?:www\.)?xvideos\.com/video([0-9]+)(?:.*)'
IE_NAME = u'xvideos'
2012-11-28 02:04:46 +01:00
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-12-15 17:50:45 +01:00
video_id = mobj.group(1)
2013-01-01 20:52:59 +01:00
webpage = self._download_webpage(url, video_id)
2012-11-28 02:04:46 +01:00
self.report_extraction(video_id)
2012-11-28 02:04:46 +01:00
# Extract video URL
mobj = re.search(r'flv_url=(.+?)&', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract video url')
2012-12-15 17:50:45 +01:00
video_url = compat_urllib_parse.unquote(mobj.group(1))
2012-11-28 02:04:46 +01:00
# Extract title
mobj = re.search(r'<title>(.*?)\s+-\s+XVID', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract video title')
2012-12-15 17:50:45 +01:00
video_title = mobj.group(1)
2012-11-28 02:04:46 +01:00
# Extract video thumbnail
mobj = re.search(r'http://(?:img.*?\.)xvideos.com/videos/thumbs/[a-fA-F0-9]+/[a-fA-F0-9]+/[a-fA-F0-9]+/[a-fA-F0-9]+/([a-fA-F0-9.]+jpg)', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract video thumbnail')
2012-12-15 17:50:45 +01:00
video_thumbnail = mobj.group(0)
2012-11-28 02:04:46 +01:00
info = {
'id': video_id,
'url': video_url,
'uploader': None,
'upload_date': None,
'title': video_title,
'ext': 'flv',
'thumbnail': video_thumbnail,
'description': None,
}
2012-11-28 02:04:46 +01:00
return [info]
class SoundcloudIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information extractor for soundcloud.com
To access the media, the uid of the song and a stream token
must be extracted from the page source and the script must make
a request to media.soundcloud.com/crossdomain.xml. Then
the media can be grabbed by requesting from an url composed
of the stream token and uid
"""
_VALID_URL = r'^(?:https?://)?(?:www\.)?soundcloud\.com/([\w\d-]+)/([\w\d-]+)'
IE_NAME = u'soundcloud'
2012-12-07 01:24:51 +01:00
def report_resolve(self, video_id):
2012-11-28 02:04:46 +01:00
"""Report information extraction."""
self.to_screen(u'%s: Resolving id' % video_id)
2012-11-28 02:04:46 +01:00
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
# extract uploader (which is in the url)
2012-11-29 20:40:12 +01:00
uploader = mobj.group(1)
2012-11-28 02:04:46 +01:00
# extract simple title (uploader + slug of song title)
2012-11-29 20:40:12 +01:00
slug_title = mobj.group(2)
2012-11-28 02:04:46 +01:00
simple_title = uploader + u'-' + slug_title
full_title = '%s/%s' % (uploader, slug_title)
2012-11-28 02:04:46 +01:00
self.report_resolve(full_title)
2012-11-28 02:04:46 +01:00
2012-12-07 01:24:51 +01:00
url = 'http://soundcloud.com/%s/%s' % (uploader, slug_title)
resolv_url = 'http://api.soundcloud.com/resolve.json?url=' + url + '&client_id=b45b1aa10f1ac2941910a7f0d10f8e28'
info_json = self._download_webpage(resolv_url, full_title, u'Downloading info JSON')
2012-11-28 02:04:46 +01:00
2012-12-07 01:24:51 +01:00
info = json.loads(info_json)
video_id = info['id']
self.report_extraction(full_title)
2012-11-28 02:04:46 +01:00
2012-12-07 01:24:51 +01:00
streams_url = 'https://api.sndcdn.com/i1/tracks/' + str(video_id) + '/streams?client_id=b45b1aa10f1ac2941910a7f0d10f8e28'
stream_json = self._download_webpage(streams_url, full_title,
u'Downloading stream definitions',
u'unable to download stream definitions')
2012-11-28 02:04:46 +01:00
2012-12-07 01:24:51 +01:00
streams = json.loads(stream_json)
2012-12-07 01:30:03 +01:00
mediaURL = streams['http_mp3_128_url']
upload_date = unified_strdate(info['created_at'])
2012-11-28 02:04:46 +01:00
return [{
2012-12-07 01:30:03 +01:00
'id': info['id'],
2012-11-28 02:04:46 +01:00
'url': mediaURL,
2012-12-07 01:30:03 +01:00
'uploader': info['user']['username'],
'upload_date': upload_date,
2012-12-07 01:30:03 +01:00
'title': info['title'],
2012-11-28 02:04:46 +01:00
'ext': u'mp3',
2012-12-07 01:30:03 +01:00
'description': info['description'],
2012-11-28 02:04:46 +01:00
}]
class SoundcloudSetIE(InfoExtractor):
"""Information extractor for soundcloud.com sets
To access the media, the uid of the song and a stream token
must be extracted from the page source and the script must make
a request to media.soundcloud.com/crossdomain.xml. Then
the media can be grabbed by requesting from an url composed
of the stream token and uid
"""
_VALID_URL = r'^(?:https?://)?(?:www\.)?soundcloud\.com/([\w\d-]+)/sets/([\w\d-]+)'
2013-04-27 20:12:46 +02:00
IE_NAME = u'soundcloud:set'
def report_resolve(self, video_id):
"""Report information extraction."""
self.to_screen(u'%s: Resolving id' % video_id)
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
# extract uploader (which is in the url)
uploader = mobj.group(1)
# extract simple title (uploader + slug of song title)
slug_title = mobj.group(2)
simple_title = uploader + u'-' + slug_title
full_title = '%s/sets/%s' % (uploader, slug_title)
self.report_resolve(full_title)
url = 'http://soundcloud.com/%s/sets/%s' % (uploader, slug_title)
resolv_url = 'http://api.soundcloud.com/resolve.json?url=' + url + '&client_id=b45b1aa10f1ac2941910a7f0d10f8e28'
info_json = self._download_webpage(resolv_url, full_title)
videos = []
info = json.loads(info_json)
if 'errors' in info:
for err in info['errors']:
self._downloader.report_error(u'unable to download video webpage: %s' % compat_str(err['error_message']))
return
self.report_extraction(full_title)
for track in info['tracks']:
video_id = track['id']
streams_url = 'https://api.sndcdn.com/i1/tracks/' + str(video_id) + '/streams?client_id=b45b1aa10f1ac2941910a7f0d10f8e28'
stream_json = self._download_webpage(streams_url, video_id, u'Downloading track info JSON')
self.report_extraction(video_id)
streams = json.loads(stream_json)
mediaURL = streams['http_mp3_128_url']
videos.append({
'id': video_id,
'url': mediaURL,
'uploader': track['user']['username'],
'upload_date': unified_strdate(track['created_at']),
'title': track['title'],
'ext': u'mp3',
'description': track['description'],
})
return videos
class InfoQIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information extractor for infoq.com"""
_VALID_URL = r'^(?:https?://)?(?:www\.)?infoq\.com/[^/]+/[^/]+$'
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
2013-01-01 21:07:37 +01:00
webpage = self._download_webpage(url, video_id=url)
2012-11-28 02:04:46 +01:00
self.report_extraction(url)
# Extract video URL
2013-04-24 21:16:10 +02:00
mobj = re.search(r"jsclassref ?= ?'([^']*)'", webpage)
2012-11-28 02:04:46 +01:00
if mobj is None:
raise ExtractorError(u'Unable to extract video url')
2013-01-01 21:07:37 +01:00
real_id = compat_urllib_parse.unquote(base64.b64decode(mobj.group(1).encode('ascii')).decode('utf-8'))
video_url = 'rtmpe://video.infoq.com/cfx/st/' + real_id
2012-11-28 02:04:46 +01:00
# Extract title
mobj = re.search(r'contentTitle = "(.*?)";', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract video title')
2013-01-01 21:07:37 +01:00
video_title = mobj.group(1)
2012-11-28 02:04:46 +01:00
# Extract description
video_description = u'No description available.'
mobj = re.search(r'<meta name="description" content="(.*)"(?:\s*/)?>', webpage)
if mobj is not None:
2013-01-01 21:07:37 +01:00
video_description = mobj.group(1)
2012-11-28 02:04:46 +01:00
video_filename = video_url.split('/')[-1]
video_id, extension = video_filename.split('.')
info = {
'id': video_id,
'url': video_url,
'uploader': None,
'upload_date': None,
'title': video_title,
'ext': extension, # Extension is always(?) mp4, but seems to be flv
'thumbnail': None,
'description': video_description,
}
return [info]
class MixcloudIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information extractor for www.mixcloud.com"""
_WORKING = False # New API, but it seems good http://www.mixcloud.com/developers/documentation/
2012-11-28 02:04:46 +01:00
_VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/([\w\d-]+)/([\w\d-]+)'
IE_NAME = u'mixcloud'
def report_download_json(self, file_id):
"""Report JSON download."""
self.to_screen(u'Downloading json')
2012-11-28 02:04:46 +01:00
def get_urls(self, jsonData, fmt, bitrate='best'):
"""Get urls from 'audio_formats' section in json"""
file_url = None
try:
bitrate_list = jsonData[fmt]
if bitrate is None or bitrate == 'best' or bitrate not in bitrate_list:
bitrate = max(bitrate_list) # select highest
url_list = jsonData[fmt][bitrate]
except TypeError: # we have no bitrate info.
url_list = jsonData[fmt]
return url_list
def check_urls(self, url_list):
"""Returns 1st active url from list"""
for url in url_list:
try:
compat_urllib_request.urlopen(url)
return url
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
url = None
return None
def _print_formats(self, formats):
print('Available formats:')
for fmt in formats.keys():
for b in formats[fmt]:
try:
ext = formats[fmt][b][0]
print('%s\t%s\t[%s]' % (fmt, b, ext.split('.')[-1]))
except TypeError: # we have no bitrate info
ext = formats[fmt][0]
print('%s\t%s\t[%s]' % (fmt, '??', ext.split('.')[-1]))
break
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
# extract uploader & filename from url
uploader = mobj.group(1).decode('utf-8')
file_id = uploader + "-" + mobj.group(2).decode('utf-8')
# construct API request
file_url = 'http://www.mixcloud.com/api/1/cloudcast/' + '/'.join(url.split('/')[-3:-1]) + '.json'
# retrieve .json file with links to files
request = compat_urllib_request.Request(file_url)
try:
self.report_download_json(file_url)
jsonData = compat_urllib_request.urlopen(request).read()
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
raise ExtractorError(u'Unable to retrieve file: %s' % compat_str(err))
2012-11-28 02:04:46 +01:00
# parse JSON
json_data = json.loads(jsonData)
player_url = json_data['player_swf_url']
formats = dict(json_data['audio_formats'])
req_format = self._downloader.params.get('format', None)
bitrate = None
if self._downloader.params.get('listformats', None):
self._print_formats(formats)
return
if req_format is None or req_format == 'best':
for format_param in formats.keys():
url_list = self.get_urls(formats, format_param)
# check urls
file_url = self.check_urls(url_list)
if file_url is not None:
break # got it!
else:
if req_format not in formats:
raise ExtractorError(u'Format is not available')
2012-11-28 02:04:46 +01:00
url_list = self.get_urls(formats, req_format)
file_url = self.check_urls(url_list)
format_param = req_format
return [{
'id': file_id.decode('utf-8'),
'url': file_url.decode('utf-8'),
'uploader': uploader.decode('utf-8'),
'upload_date': None,
'title': json_data['name'],
'ext': file_url.split('.')[-1].decode('utf-8'),
'format': (format_param is None and u'NA' or format_param.decode('utf-8')),
'thumbnail': json_data['thumbnail_url'],
'description': json_data['description'],
'player_url': player_url.decode('utf-8'),
}]
class StanfordOpenClassroomIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information extractor for Stanford's Open ClassRoom"""
_VALID_URL = r'^(?:https?://)?openclassroom.stanford.edu(?P<path>/?|(/MainFolder/(?:HomePage|CoursePage|VideoPage)\.php([?]course=(?P<course>[^&]+)(&video=(?P<video>[^&]+))?(&.*)?)?))$'
IE_NAME = u'stanfordoc'
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
if mobj is None:
2013-01-27 15:23:26 +01:00
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
if mobj.group('course') and mobj.group('video'): # A specific video
course = mobj.group('course')
video = mobj.group('video')
info = {
'id': course + '_' + video,
'uploader': None,
'upload_date': None,
}
self.report_extraction(info['id'])
baseUrl = 'http://openclassroom.stanford.edu/MainFolder/courses/' + course + '/videos/'
xmlUrl = baseUrl + video + '.xml'
try:
metaXml = compat_urllib_request.urlopen(xmlUrl).read()
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
raise ExtractorError(u'Unable to download video info XML: %s' % compat_str(err))
2012-11-28 02:04:46 +01:00
mdoc = xml.etree.ElementTree.fromstring(metaXml)
try:
info['title'] = mdoc.findall('./title')[0].text
info['url'] = baseUrl + mdoc.findall('./videoFile')[0].text
except IndexError:
raise ExtractorError(u'Invalid metadata XML file')
2012-11-28 02:04:46 +01:00
info['ext'] = info['url'].rpartition('.')[2]
return [info]
elif mobj.group('course'): # A course page
course = mobj.group('course')
info = {
'id': course,
'type': 'playlist',
'uploader': None,
'upload_date': None,
}
2013-01-27 15:23:26 +01:00
coursepage = self._download_webpage(url, info['id'],
note='Downloading course info page',
errnote='Unable to download course info page')
2012-11-28 02:04:46 +01:00
m = re.search('<h1>([^<]+)</h1>', coursepage)
if m:
info['title'] = unescapeHTML(m.group(1))
else:
info['title'] = info['id']
m = re.search('<description>([^<]+)</description>', coursepage)
if m:
info['description'] = unescapeHTML(m.group(1))
links = orderedSet(re.findall('<a href="(VideoPage.php\?[^"]+)">', coursepage))
info['list'] = [
{
'type': 'reference',
'url': 'http://openclassroom.stanford.edu/MainFolder/' + unescapeHTML(vpage),
}
for vpage in links]
results = []
for entry in info['list']:
assert entry['type'] == 'reference'
results += self.extract(entry['url'])
return results
else: # Root page
info = {
'id': 'Stanford OpenClassroom',
'type': 'playlist',
'uploader': None,
'upload_date': None,
}
self.report_download_webpage(info['id'])
rootURL = 'http://openclassroom.stanford.edu/MainFolder/HomePage.php'
try:
rootpage = compat_urllib_request.urlopen(rootURL).read()
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
raise ExtractorError(u'Unable to download course info page: ' + compat_str(err))
2012-11-28 02:04:46 +01:00
info['title'] = info['id']
links = orderedSet(re.findall('<a href="(CoursePage.php\?[^"]+)">', rootpage))
info['list'] = [
{
'type': 'reference',
'url': 'http://openclassroom.stanford.edu/MainFolder/' + unescapeHTML(cpage),
}
for cpage in links]
results = []
for entry in info['list']:
assert entry['type'] == 'reference'
results += self.extract(entry['url'])
return results
class MTVIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information extractor for MTV.com"""
_VALID_URL = r'^(?P<proto>https?://)?(?:www\.)?mtv\.com/videos/[^/]+/(?P<videoid>[0-9]+)/[^/]+$'
IE_NAME = u'mtv'
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
if not mobj.group('proto'):
url = 'http://' + url
video_id = mobj.group('videoid')
2013-01-01 20:52:59 +01:00
webpage = self._download_webpage(url, video_id)
2012-11-28 02:04:46 +01:00
mobj = re.search(r'<meta name="mtv_vt" content="([^"]+)"/>', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract song name')
2012-11-28 02:04:46 +01:00
song_name = unescapeHTML(mobj.group(1).decode('iso-8859-1'))
mobj = re.search(r'<meta name="mtv_an" content="([^"]+)"/>', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract performer')
2012-11-28 02:04:46 +01:00
performer = unescapeHTML(mobj.group(1).decode('iso-8859-1'))
2012-12-19 15:19:08 +01:00
video_title = performer + ' - ' + song_name
2012-11-28 02:04:46 +01:00
mobj = re.search(r'<meta name="mtvn_uri" content="([^"]+)"/>', webpage)
if mobj is None:
raise ExtractorError(u'Unable to mtvn_uri')
2012-11-28 02:04:46 +01:00
mtvn_uri = mobj.group(1)
mobj = re.search(r'MTVN.Player.defaultPlaylistId = ([0-9]+);', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract content id')
2012-11-28 02:04:46 +01:00
content_id = mobj.group(1)
videogen_url = 'http://www.mtv.com/player/includes/mediaGen.jhtml?uri=' + mtvn_uri + '&id=' + content_id + '&vid=' + video_id + '&ref=www.mtvn.com&viewUri=' + mtvn_uri
self.report_extraction(video_id)
request = compat_urllib_request.Request(videogen_url)
try:
metadataXml = compat_urllib_request.urlopen(request).read()
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
raise ExtractorError(u'Unable to download video metadata: %s' % compat_str(err))
2012-11-28 02:04:46 +01:00
mdoc = xml.etree.ElementTree.fromstring(metadataXml)
renditions = mdoc.findall('.//rendition')
# For now, always pick the highest quality.
rendition = renditions[-1]
try:
_,_,ext = rendition.attrib['type'].partition('/')
format = ext + '-' + rendition.attrib['width'] + 'x' + rendition.attrib['height'] + '_' + rendition.attrib['bitrate']
video_url = rendition.find('./src').text
except KeyError:
raise ExtractorError('Invalid rendition field.')
2012-11-28 02:04:46 +01:00
info = {
'id': video_id,
'url': video_url,
'uploader': performer,
'upload_date': None,
'title': video_title,
'ext': ext,
'format': format,
}
return [info]
2012-08-16 01:54:03 +02:00
2012-08-08 20:04:02 +02:00
class YoukuIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
_VALID_URL = r'(?:http://)?v\.youku\.com/v_show/id_(?P<ID>[A-Za-z0-9]+)\.html'
def _gen_sid(self):
nowTime = int(time.time() * 1000)
random1 = random.randint(1000,1998)
random2 = random.randint(1000,9999)
return "%d%d%d" %(nowTime,random1,random2)
def _get_file_ID_mix_string(self, seed):
mixed = []
source = list("abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ/\:._-1234567890")
seed = float(seed)
for i in range(len(source)):
seed = (seed * 211 + 30031 ) % 65536
index = math.floor(seed / 65536 * len(source) )
mixed.append(source[int(index)])
source.remove(source[int(index)])
#return ''.join(mixed)
return mixed
def _get_file_id(self, fileId, seed):
mixed = self._get_file_ID_mix_string(seed)
ids = fileId.split('*')
realId = []
for ch in ids:
if ch:
realId.append(mixed[int(ch)])
return ''.join(realId)
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
video_id = mobj.group('ID')
info_url = 'http://v.youku.com/player/getPlayList/VideoIDS/' + video_id
jsondata = self._download_webpage(info_url, video_id)
2012-11-28 02:04:46 +01:00
self.report_extraction(video_id)
try:
config = json.loads(jsondata)
2012-11-28 02:04:46 +01:00
video_title = config['data'][0]['title']
seed = config['data'][0]['seed']
format = self._downloader.params.get('format', None)
supported_format = list(config['data'][0]['streamfileids'].keys())
2012-11-28 02:04:46 +01:00
if format is None or format == 'best':
if 'hd2' in supported_format:
format = 'hd2'
else:
format = 'flv'
ext = u'flv'
elif format == 'worst':
format = 'mp4'
ext = u'mp4'
else:
format = 'flv'
ext = u'flv'
fileid = config['data'][0]['streamfileids'][format]
keys = [s['k'] for s in config['data'][0]['segs'][format]]
2012-12-15 17:59:09 +01:00
except (UnicodeDecodeError, ValueError, KeyError):
raise ExtractorError(u'Unable to extract info section')
2012-11-28 02:04:46 +01:00
files_info=[]
sid = self._gen_sid()
fileid = self._get_file_id(fileid, seed)
#column 8,9 of fileid represent the segment number
#fileid[7:9] should be changed
for index, key in enumerate(keys):
temp_fileid = '%s%02X%s' % (fileid[0:8], index, fileid[10:])
download_url = 'http://f.youku.com/player/getFlvPath/sid/%s_%02X/st/flv/fileid/%s?k=%s' % (sid, index, temp_fileid, key)
info = {
'id': '%s_part%02d' % (video_id, index),
'url': download_url,
'uploader': None,
'upload_date': None,
'title': video_title,
'ext': ext,
}
files_info.append(info)
return files_info
2012-08-16 01:54:03 +02:00
class XNXXIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information extractor for xnxx.com"""
2013-01-05 22:05:23 +01:00
_VALID_URL = r'^(?:https?://)?video\.xnxx\.com/video([0-9]+)/(.*)'
2012-11-28 02:04:46 +01:00
IE_NAME = u'xnxx'
VIDEO_URL_RE = r'flv_url=(.*?)&amp;'
VIDEO_TITLE_RE = r'<title>(.*?)\s+-\s+XNXX.COM'
VIDEO_THUMB_RE = r'url_bigthumb=(.*?)&amp;'
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-12-15 18:19:25 +01:00
video_id = mobj.group(1)
2012-11-28 02:04:46 +01:00
# Get webpage content
webpage = self._download_webpage(url, video_id)
2012-11-28 02:04:46 +01:00
result = re.search(self.VIDEO_URL_RE, webpage)
if result is None:
raise ExtractorError(u'Unable to extract video url')
2012-12-15 18:19:25 +01:00
video_url = compat_urllib_parse.unquote(result.group(1))
2012-11-28 02:04:46 +01:00
result = re.search(self.VIDEO_TITLE_RE, webpage)
if result is None:
raise ExtractorError(u'Unable to extract video title')
2012-12-15 18:19:25 +01:00
video_title = result.group(1)
2012-11-28 02:04:46 +01:00
result = re.search(self.VIDEO_THUMB_RE, webpage)
if result is None:
raise ExtractorError(u'Unable to extract video thumbnail')
2012-12-15 18:19:25 +01:00
video_thumbnail = result.group(1)
2012-11-28 02:04:46 +01:00
return [{
'id': video_id,
'url': video_url,
'uploader': None,
'upload_date': None,
'title': video_title,
'ext': 'flv',
'thumbnail': video_thumbnail,
'description': None,
}]
class GooglePlusIE(InfoExtractor):
2012-11-28 02:04:46 +01:00
"""Information extractor for plus.google.com."""
_VALID_URL = r'(?:https://)?plus\.google\.com/(?:[^/]+/)*?posts/(\w+)'
2012-11-28 02:04:46 +01:00
IE_NAME = u'plus.google'
def report_extract_entry(self, url):
"""Report downloading extry"""
self.to_screen(u'Downloading entry: %s' % url)
2012-11-28 02:04:46 +01:00
def report_date(self, upload_date):
"""Report downloading extry"""
self.to_screen(u'Entry date: %s' % upload_date)
2012-11-28 02:04:46 +01:00
def report_uploader(self, uploader):
"""Report downloading extry"""
self.to_screen(u'Uploader: %s' % uploader)
2012-11-28 02:04:46 +01:00
def report_title(self, video_title):
"""Report downloading extry"""
self.to_screen(u'Title: %s' % video_title)
2012-11-28 02:04:46 +01:00
def report_extract_vid_page(self, video_page):
"""Report information extraction."""
self.to_screen(u'Extracting video page: %s' % video_page)
2012-11-28 02:04:46 +01:00
def _real_extract(self, url):
# Extract id from URL
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-11-28 02:04:46 +01:00
post_url = mobj.group(0)
video_id = mobj.group(1)
2012-11-28 02:04:46 +01:00
video_extension = 'flv'
# Step 1, Retrieve post webpage to extract further information
self.report_extract_entry(post_url)
webpage = self._download_webpage(post_url, video_id, u'Downloading entry webpage')
2012-11-28 02:04:46 +01:00
# Extract update date
upload_date = None
pattern = 'title="Timestamp">(.*?)</a>'
mobj = re.search(pattern, webpage)
if mobj:
upload_date = mobj.group(1)
# Convert timestring to a format suitable for filename
upload_date = datetime.datetime.strptime(upload_date, "%Y-%m-%d")
upload_date = upload_date.strftime('%Y%m%d')
self.report_date(upload_date)
# Extract uploader
uploader = None
pattern = r'rel\="author".*?>(.*?)</a>'
mobj = re.search(pattern, webpage)
if mobj:
uploader = mobj.group(1)
self.report_uploader(uploader)
# Extract title
# Get the first line for title
video_title = u'NA'
pattern = r'<meta name\=\"Description\" content\=\"(.*?)[\n<"]'
mobj = re.search(pattern, webpage)
if mobj:
video_title = mobj.group(1)
self.report_title(video_title)
# Step 2, Stimulate clicking the image box to launch video
pattern = '"(https\://plus\.google\.com/photos/.*?)",,"image/jpeg","video"\]'
mobj = re.search(pattern, webpage)
if mobj is None:
self._downloader.report_error(u'unable to extract video page URL')
2012-11-28 02:04:46 +01:00
video_page = mobj.group(1)
webpage = self._download_webpage(video_page, video_id, u'Downloading video page')
2012-11-28 02:04:46 +01:00
self.report_extract_vid_page(video_page)
# Extract video links on video page
"""Extract video links of all sizes"""
pattern = '\d+,\d+,(\d+),"(http\://redirector\.googlevideo\.com.*?)"'
mobj = re.findall(pattern, webpage)
if len(mobj) == 0:
self._downloader.report_error(u'unable to extract video links')
2012-11-28 02:04:46 +01:00
# Sort in resolution
links = sorted(mobj)
# Choose the lowest of the sort, i.e. highest resolution
video_url = links[-1]
# Only get the url. The resolution part in the tuple has no use anymore
video_url = video_url[-1]
# Treat escaped \u0026 style hex
try:
video_url = video_url.decode("unicode_escape")
except AttributeError: # Python 3
video_url = bytes(video_url, 'ascii').decode('unicode-escape')
2012-11-28 02:04:46 +01:00
return [{
'id': video_id,
2012-11-28 02:04:46 +01:00
'url': video_url,
'uploader': uploader,
'upload_date': upload_date,
'title': video_title,
'ext': video_extension,
2012-11-28 02:04:46 +01:00
}]
2012-12-13 21:27:57 +01:00
class NBAIE(InfoExtractor):
_VALID_URL = r'^(?:https?://)?(?:watch\.|www\.)?nba\.com/(?:nba/)?video(/[^?]*)(\?.*)?$'
IE_NAME = u'nba'
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2012-12-13 21:27:57 +01:00
video_id = mobj.group(1)
if video_id.endswith('/index.html'):
video_id = video_id[:-len('/index.html')]
2013-01-01 20:52:59 +01:00
webpage = self._download_webpage(url, video_id)
2012-12-13 21:27:57 +01:00
video_url = u'http://ht-mobile.cdn.turner.com/nba/big' + video_id + '_nba_1280x720.mp4'
def _findProp(rexp, default=None):
m = re.search(rexp, webpage)
if m:
return unescapeHTML(m.group(1))
else:
return default
shortened_video_id = video_id.rpartition('/')[2]
title = _findProp(r'<meta property="og:title" content="(.*?)"', shortened_video_id).replace('NBA.com: ', '')
info = {
'id': shortened_video_id,
'url': video_url,
'ext': 'mp4',
'title': title,
'uploader_date': _findProp(r'<b>Date:</b> (.*?)</div>'),
'description': _findProp(r'<div class="description">(.*?)</h1>'),
}
return [info]
class JustinTVIE(InfoExtractor):
"""Information extractor for justin.tv and twitch.tv"""
# TODO: One broadcast may be split into multiple videos. The key
# 'broadcast_id' is the same for all parts, and 'broadcast_part'
# starts at 1 and increases. Can we treat all parts as one video?
2012-12-16 10:45:46 +01:00
_VALID_URL = r"""(?x)^(?:http://)?(?:www\.)?(?:twitch|justin)\.tv/
2013-05-04 10:33:34 +02:00
(?:
(?P<channelid>[^/]+)|
(?:(?:[^/]+)/b/(?P<videoid>[^/]+))|
(?:(?:[^/]+)/c/(?P<chapterid>[^/]+))
)
/?(?:\#.*)?$
"""
2012-12-16 10:45:46 +01:00
_JUSTIN_PAGE_LIMIT = 100
IE_NAME = u'justin.tv'
2012-12-16 10:45:46 +01:00
def report_download_page(self, channel, offset):
"""Report attempt to download a single page of videos."""
self.to_screen(u'%s: Downloading video information from %d to %d' %
(channel, offset, offset + self._JUSTIN_PAGE_LIMIT))
2012-12-16 10:45:46 +01:00
# Return count of items, list of *valid* items
def _parse_page(self, url, video_id):
webpage = self._download_webpage(url, video_id,
u'Downloading video info JSON',
u'unable to download video info JSON')
2012-12-19 15:19:08 +01:00
response = json.loads(webpage)
if type(response) != list:
error_text = response.get('error', 'unknown error')
raise ExtractorError(u'Justin.tv API: %s' % error_text)
info = []
for clip in response:
video_url = clip['video_file_url']
if video_url:
video_extension = os.path.splitext(video_url)[1][1:]
video_date = re.sub('-', '', clip['start_time'][:10])
video_uploader_id = clip.get('user_id', clip.get('channel_id'))
video_id = clip['id']
video_title = clip.get('title', video_id)
info.append({
'id': video_id,
'url': video_url,
'title': video_title,
'uploader': clip.get('channel_name', video_uploader_id),
'uploader_id': video_uploader_id,
'upload_date': video_date,
'ext': video_extension,
})
return (len(response), info)
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
if mobj is None:
2013-05-04 10:33:34 +02:00
raise ExtractorError(u'invalid URL: %s' % url)
2012-12-19 15:19:08 +01:00
2013-05-04 10:33:34 +02:00
api_base = 'http://api.justin.tv'
paged = False
2013-05-04 10:33:34 +02:00
if mobj.group('channelid'):
paged = True
2013-05-04 10:33:34 +02:00
video_id = mobj.group('channelid')
api = api_base + '/channel/archives/%s.json' % video_id
elif mobj.group('chapterid'):
chapter_id = mobj.group('chapterid')
webpage = self._download_webpage(url, chapter_id)
m = re.search(r'PP\.archive_id = "([0-9]+)";', webpage)
if not m:
raise ExtractorError(u'Cannot find archive of a chapter')
2013-05-04 10:33:34 +02:00
archive_id = m.group(1)
api = api_base + '/broadcast/by_chapter/%s.xml' % chapter_id
chapter_info_xml = self._download_webpage(api, chapter_id,
note=u'Downloading chapter information',
errnote=u'Chapter information download failed')
doc = xml.etree.ElementTree.fromstring(chapter_info_xml)
for a in doc.findall('.//archive'):
if archive_id == a.find('./id').text:
break
else:
raise ExtractorError(u'Could not find chapter in chapter information')
video_url = a.find('./video_file_url').text
video_ext = video_url.rpartition('.')[2] or u'flv'
chapter_api_url = u'https://api.twitch.tv/kraken/videos/c' + chapter_id
chapter_info_json = self._download_webpage(chapter_api_url, u'c' + chapter_id,
note='Downloading chapter metadata',
errnote='Download of chapter metadata failed')
chapter_info = json.loads(chapter_info_json)
bracket_start = int(doc.find('.//bracket_start').text)
bracket_end = int(doc.find('.//bracket_end').text)
# TODO determine start (and probably fix up file)
# youtube-dl -v http://www.twitch.tv/firmbelief/c/1757457
#video_url += u'?start=' + TODO:start_timestamp
# bracket_start is 13290, but we want 51670615
self._downloader.report_warning(u'Chapter detected, but we can just download the whole file. '
u'Chapter starts at %s and ends at %s' % (formatSeconds(bracket_start), formatSeconds(bracket_end)))
2013-05-04 10:33:34 +02:00
info = {
'id': u'c' + chapter_id,
'url': video_url,
'ext': video_ext,
'title': chapter_info['title'],
'thumbnail': chapter_info['preview'],
'description': chapter_info['description'],
'uploader': chapter_info['channel']['display_name'],
'uploader_id': chapter_info['channel']['name'],
}
return [info]
else:
2013-05-04 10:33:34 +02:00
video_id = mobj.group('videoid')
api = api_base + '/broadcast/by_archive/%s.json' % video_id
2012-12-19 15:19:08 +01:00
self.report_extraction(video_id)
2012-12-19 15:19:08 +01:00
info = []
offset = 0
2012-12-16 10:45:46 +01:00
limit = self._JUSTIN_PAGE_LIMIT
while True:
if paged:
self.report_download_page(video_id, offset)
page_url = api + ('?offset=%d&limit=%d' % (offset, limit))
page_count, page_info = self._parse_page(page_url, video_id)
info.extend(page_info)
if not paged or page_count != limit:
break
offset += limit
return info
2012-12-20 21:28:27 +01:00
class FunnyOrDieIE(InfoExtractor):
_VALID_URL = r'^(?:https?://)?(?:www\.)?funnyordie\.com/videos/(?P<id>[0-9a-f]+)/.*$'
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'invalid URL: %s' % url)
2012-12-20 21:28:27 +01:00
video_id = mobj.group('id')
2013-01-01 20:52:59 +01:00
webpage = self._download_webpage(url, video_id)
2012-12-20 21:28:27 +01:00
m = re.search(r'<video[^>]*>\s*<source[^>]*>\s*<source src="(?P<url>[^"]+)"', webpage, re.DOTALL)
if not m:
self._downloader.report_error(u'unable to find video information')
2012-12-20 21:28:27 +01:00
video_url = unescapeHTML(m.group('url'))
2013-03-29 15:59:13 +01:00
m = re.search(r"<h1 class='player_page_h1'.*?>(?P<title>.*?)</h1>", webpage, flags=re.DOTALL)
2012-12-20 21:28:27 +01:00
if not m:
m = re.search(r'<title>(?P<title>[^<]+?)</title>', webpage)
if not m:
self._downloader.report_error(u'Cannot find video title')
2013-03-29 15:59:13 +01:00
title = clean_html(m.group('title'))
2012-12-20 21:28:27 +01:00
m = re.search(r'<meta property="og:description" content="(?P<desc>.*?)"', webpage)
if m:
desc = unescapeHTML(m.group('desc'))
else:
desc = None
info = {
'id': video_id,
'url': video_url,
'ext': 'mp4',
'title': title,
'description': desc,
}
return [info]
2012-12-27 01:38:41 +01:00
2013-01-01 14:12:14 +01:00
class SteamIE(InfoExtractor):
_VALID_URL = r"""http://store\.steampowered\.com/
2013-04-27 11:03:34 +02:00
(agecheck/)?
2013-01-01 14:12:14 +01:00
(?P<urltype>video|app)/ #If the page is only for videos or for a game
(?P<gameID>\d+)/?
(?P<videoID>\d*)(?P<extra>\??) #For urltype == video we sometimes get the videoID
"""
2013-01-01 19:37:07 +01:00
@classmethod
def suitable(cls, url):
2013-01-01 14:12:14 +01:00
"""Receives a URL and returns True if suitable for this IE."""
return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
2013-01-01 20:52:59 +01:00
2013-01-01 14:12:14 +01:00
def _real_extract(self, url):
m = re.match(self._VALID_URL, url, re.VERBOSE)
gameID = m.group('gameID')
videourl = 'http://store.steampowered.com/agecheck/video/%s/?snr=1_agecheck_agecheck__age-gate&ageDay=1&ageMonth=January&ageYear=1970' % gameID
self.report_age_confirmation()
2013-01-01 20:52:59 +01:00
webpage = self._download_webpage(videourl, gameID)
game_title = re.search(r'<h2 class="pageheader">(?P<game_title>.*?)</h2>', webpage).group('game_title')
urlRE = r"'movie_(?P<videoID>\d+)': \{\s*FILENAME: \"(?P<videoURL>[\w:/\.\?=]+)\"(,\s*MOVIE_NAME: \"(?P<videoName>[\w:/\.\?=\+-]+)\")?\s*\},"
2013-01-01 14:12:14 +01:00
mweb = re.finditer(urlRE, webpage)
2013-01-03 23:51:48 +01:00
namesRE = r'<span class="title">(?P<videoName>.+?)</span>'
titles = re.finditer(namesRE, webpage)
2013-02-23 16:48:15 +01:00
thumbsRE = r'<img class="movie_thumb" src="(?P<thumbnail>.+?)">'
thumbs = re.finditer(thumbsRE, webpage)
2013-01-01 14:12:14 +01:00
videos = []
2013-02-23 16:48:15 +01:00
for vid,vtitle,thumb in zip(mweb,titles,thumbs):
2013-01-01 14:12:14 +01:00
video_id = vid.group('videoID')
2013-01-01 20:52:59 +01:00
title = vtitle.group('videoName')
video_url = vid.group('videoURL')
2013-02-23 16:48:15 +01:00
video_thumb = thumb.group('thumbnail')
2013-01-01 14:12:14 +01:00
if not video_url:
self._downloader.report_error(u'Cannot find video url for %s' % video_id)
2013-01-01 14:12:14 +01:00
info = {
'id':video_id,
'url':video_url,
'ext': 'flv',
2013-02-23 16:48:15 +01:00
'title': unescapeHTML(title),
'thumbnail': video_thumb
2013-01-01 14:12:14 +01:00
}
videos.append(info)
return [self.playlist_result(videos, gameID, game_title)]
2013-01-12 13:49:14 +01:00
2013-01-01 17:52:46 +01:00
class UstreamIE(InfoExtractor):
2013-01-12 13:49:14 +01:00
_VALID_URL = r'https?://www\.ustream\.tv/recorded/(?P<videoID>\d+)'
2013-01-01 17:52:46 +01:00
IE_NAME = u'ustream'
2013-01-12 13:49:14 +01:00
2013-01-01 17:52:46 +01:00
def _real_extract(self, url):
m = re.match(self._VALID_URL, url)
video_id = m.group('videoID')
video_url = u'http://tcdn.ustream.tv/video/%s' % video_id
2013-01-01 20:43:43 +01:00
webpage = self._download_webpage(url, video_id)
2013-01-01 17:52:46 +01:00
m = re.search(r'data-title="(?P<title>.+)"',webpage)
title = m.group('title')
m = re.search(r'<a class="state" data-content-type="channel" data-content-id="(?P<uploader>\d+)"',webpage)
uploader = m.group('uploader')
info = {
'id':video_id,
'url':video_url,
'ext': 'flv',
'title': title,
'uploader': uploader
}
return [info]
2013-01-01 19:37:07 +01:00
2013-03-07 06:09:55 +01:00
class WorldStarHipHopIE(InfoExtractor):
2013-05-04 08:06:56 +02:00
_VALID_URL = r'https?://(?:www|m)\.worldstar(?:candy|hiphop)\.com/videos/video\.php\?v=(?P<id>.*)'
2013-03-07 06:09:55 +01:00
IE_NAME = u'WorldStarHipHop'
def _real_extract(self, url):
2013-05-02 22:37:26 +02:00
_src_url = r'so\.addVariable\("file","(.*?)"\)'
2013-03-07 06:09:55 +01:00
2013-03-09 07:48:05 +01:00
m = re.match(self._VALID_URL, url)
video_id = m.group('id')
webpage_src = self._download_webpage(url, video_id)
mobj = re.search(_src_url, webpage_src)
2013-03-07 06:09:55 +01:00
if mobj is not None:
2013-05-02 22:37:26 +02:00
video_url = mobj.group(1)
2013-03-07 06:09:55 +01:00
if 'mp4' in video_url:
2013-03-08 00:39:17 +01:00
ext = 'mp4'
2013-03-07 06:09:55 +01:00
else:
2013-03-08 00:39:17 +01:00
ext = 'flv'
2013-03-07 06:09:55 +01:00
else:
2013-05-04 08:06:56 +02:00
raise ExtractorError(u'Cannot find video url for %s' % video_id)
2013-05-04 08:06:56 +02:00
mobj = re.search(r"<title>(.*)</title>", webpage_src)
2013-03-07 06:09:55 +01:00
2013-05-04 08:06:56 +02:00
if mobj is None:
raise ExtractorError(u'Cannot determine title')
title = mobj.group(1)
2013-03-07 06:09:55 +01:00
2013-05-04 08:06:56 +02:00
mobj = re.search(r'rel="image_src" href="(.*)" />', webpage_src)
2013-03-07 06:09:55 +01:00
# Getting thumbnail and if not thumbnail sets correct title for WSHH candy video.
if mobj is not None:
thumbnail = mobj.group(1)
else:
_title = r"""candytitles.*>(.*)</span>"""
mobj = re.search(_title, webpage_src)
if mobj is not None:
title = mobj.group(1)
thumbnail = None
2013-03-08 00:39:17 +01:00
results = [{
'id': video_id,
2013-03-08 00:39:17 +01:00
'url' : video_url,
'title' : title,
'thumbnail' : thumbnail,
'ext' : ext,
}]
2013-03-07 06:09:55 +01:00
return results
2013-01-12 17:58:39 +01:00
class RBMARadioIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?rbmaradio\.com/shows/(?P<videoID>[^/]+)$'
def _real_extract(self, url):
m = re.match(self._VALID_URL, url)
video_id = m.group('videoID')
webpage = self._download_webpage(url, video_id)
m = re.search(r'<script>window.gon = {.*?};gon\.show=(.+?);</script>', webpage)
if not m:
raise ExtractorError(u'Cannot find metadata')
json_data = m.group(1)
try:
data = json.loads(json_data)
except ValueError as e:
raise ExtractorError(u'Invalid JSON: ' + str(e))
video_url = data['akamai_url'] + '&cbr=256'
url_parts = compat_urllib_parse_urlparse(video_url)
video_ext = url_parts.path.rpartition('.')[2]
info = {
'id': video_id,
'url': video_url,
'ext': video_ext,
'title': data['title'],
'description': data.get('teaser_text'),
'location': data.get('country_of_origin'),
'uploader': data.get('host', {}).get('name'),
'uploader_id': data.get('host', {}).get('slug'),
'thumbnail': data.get('image', {}).get('large_url_2x'),
2013-01-12 17:58:39 +01:00
'duration': data.get('duration'),
}
return [info]
2013-01-01 19:37:07 +01:00
2013-01-05 21:42:35 +01:00
class YouPornIE(InfoExtractor):
"""Information extractor for youporn.com."""
_VALID_URL = r'^(?:https?://)?(?:\w+\.)?youporn\.com/watch/(?P<videoid>[0-9]+)/(?P<title>[^/]+)'
2013-01-05 21:42:35 +01:00
def _print_formats(self, formats):
"""Print all available formats"""
2013-01-12 15:17:04 +01:00
print(u'Available formats:')
print(u'ext\t\tformat')
print(u'---------------------------------')
2013-01-05 21:42:35 +01:00
for format in formats:
print(u'%s\t\t%s' % (format['ext'], format['format']))
2013-01-05 21:42:35 +01:00
def _specific(self, req_format, formats):
for x in formats:
if(x["format"]==req_format):
return x
return None
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2013-01-05 21:42:35 +01:00
video_id = mobj.group('videoid')
2013-01-05 21:42:35 +01:00
req = compat_urllib_request.Request(url)
req.add_header('Cookie', 'age_verified=1')
webpage = self._download_webpage(req, video_id)
2013-01-05 21:42:35 +01:00
# Get the video title
2013-02-18 23:30:33 +01:00
result = re.search(r'<h1.*?>(?P<title>.*)</h1>', webpage)
2013-01-05 21:42:35 +01:00
if result is None:
2013-02-18 23:30:33 +01:00
raise ExtractorError(u'Unable to extract video title')
video_title = result.group('title').strip()
2013-01-05 21:42:35 +01:00
# Get the video date
2013-02-18 23:30:33 +01:00
result = re.search(r'Date:</label>(?P<date>.*) </li>', webpage)
2013-01-05 21:42:35 +01:00
if result is None:
2013-03-02 11:24:07 +01:00
self._downloader.report_warning(u'unable to extract video date')
upload_date = None
else:
upload_date = unified_strdate(result.group('date').strip())
2013-01-05 21:42:35 +01:00
# Get the video uploader
2013-02-18 23:30:33 +01:00
result = re.search(r'Submitted:</label>(?P<uploader>.*)</li>', webpage)
2013-01-05 21:42:35 +01:00
if result is None:
2013-03-02 11:24:07 +01:00
self._downloader.report_warning(u'unable to extract uploader')
video_uploader = None
else:
video_uploader = result.group('uploader').strip()
video_uploader = clean_html( video_uploader )
2013-01-05 21:42:35 +01:00
# Get all of the formats available
DOWNLOAD_LIST_RE = r'(?s)<ul class="downloadList">(?P<download_list>.*?)</ul>'
result = re.search(DOWNLOAD_LIST_RE, webpage)
2013-01-05 21:42:35 +01:00
if result is None:
raise ExtractorError(u'Unable to extract download list')
download_list_html = result.group('download_list').strip()
2013-01-05 21:42:35 +01:00
# Get all of the links from the page
LINK_RE = r'(?s)<a href="(?P<url>[^"]+)">'
links = re.findall(LINK_RE, download_list_html)
2013-01-05 21:42:35 +01:00
if(len(links) == 0):
raise ExtractorError(u'ERROR: no known formats available for video')
self.to_screen(u'Links found: %d' % len(links))
2013-01-05 21:42:35 +01:00
formats = []
for link in links:
# A link looks like this:
# http://cdn1.download.youporn.phncdn.com/201210/31/8004515/480p_370k_8004515/YouPorn%20-%20Nubile%20Films%20The%20Pillow%20Fight.mp4?nvb=20121113051249&nva=20121114051249&ir=1200&sr=1200&hash=014b882080310e95fb6a0
# A path looks like this:
# /201210/31/8004515/480p_370k_8004515/YouPorn%20-%20Nubile%20Films%20The%20Pillow%20Fight.mp4
video_url = unescapeHTML( link )
path = compat_urllib_parse_urlparse( video_url ).path
2013-01-05 21:42:35 +01:00
extension = os.path.splitext( path )[1][1:]
format = path.split('/')[4].split('_')[:2]
size = format[0]
bitrate = format[1]
format = "-".join( format )
title = u'%s-%s-%s' % (video_title, size, bitrate)
formats.append({
'id': video_id,
'url': video_url,
'uploader': video_uploader,
'upload_date': upload_date,
'title': title,
'ext': extension,
'format': format,
'thumbnail': None,
'description': None,
'player_url': None
})
if self._downloader.params.get('listformats', None):
self._print_formats(formats)
return
req_format = self._downloader.params.get('format', None)
self.to_screen(u'Format: %s' % req_format)
2013-01-05 21:42:35 +01:00
if req_format is None or req_format == 'best':
return [formats[0]]
elif req_format == 'worst':
return [formats[-1]]
elif req_format in ('-1', 'all'):
return formats
else:
format = self._specific( req_format, formats )
if result is None:
raise ExtractorError(u'Requested format not available')
2013-01-05 21:42:35 +01:00
return [format]
2013-01-05 21:42:35 +01:00
class PornotubeIE(InfoExtractor):
"""Information extractor for pornotube.com."""
_VALID_URL = r'^(?:https?://)?(?:\w+\.)?pornotube\.com(/c/(?P<channel>[0-9]+))?(/m/(?P<videoid>[0-9]+))(/(?P<title>.+))$'
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2013-01-05 21:42:35 +01:00
video_id = mobj.group('videoid')
video_title = mobj.group('title')
2013-01-05 21:42:35 +01:00
# Get webpage content
webpage = self._download_webpage(url, video_id)
2013-01-05 21:42:35 +01:00
# Get the video URL
VIDEO_URL_RE = r'url: "(?P<url>http://video[0-9].pornotube.com/.+\.flv)",'
result = re.search(VIDEO_URL_RE, webpage)
2013-01-05 21:42:35 +01:00
if result is None:
raise ExtractorError(u'Unable to extract video url')
video_url = compat_urllib_parse.unquote(result.group('url'))
2013-01-05 21:42:35 +01:00
#Get the uploaded date
VIDEO_UPLOADED_RE = r'<div class="video_added_by">Added (?P<date>[0-9\/]+) by'
result = re.search(VIDEO_UPLOADED_RE, webpage)
2013-01-05 21:42:35 +01:00
if result is None:
raise ExtractorError(u'Unable to extract video title')
upload_date = unified_strdate(result.group('date'))
2013-01-05 21:42:35 +01:00
info = {'id': video_id,
'url': video_url,
'uploader': None,
'upload_date': upload_date,
'title': video_title,
'ext': 'flv',
2013-01-12 15:17:04 +01:00
'format': 'flv'}
2013-01-05 21:42:35 +01:00
return [info]
class YouJizzIE(InfoExtractor):
"""Information extractor for youjizz.com."""
_VALID_URL = r'^(?:https?://)?(?:\w+\.)?youjizz\.com/videos/(?P<videoid>[^.]+).html$'
2013-01-05 21:42:35 +01:00
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
video_id = mobj.group('videoid')
# Get webpage content
webpage = self._download_webpage(url, video_id)
2013-01-05 21:42:35 +01:00
# Get the video title
2013-01-12 16:36:51 +01:00
result = re.search(r'<title>(?P<title>.*)</title>', webpage)
2013-01-05 21:42:35 +01:00
if result is None:
2013-01-12 16:36:51 +01:00
raise ExtractorError(u'ERROR: unable to extract video title')
video_title = result.group('title').strip()
2013-01-05 21:42:35 +01:00
# Get the embed page
2013-01-12 16:36:51 +01:00
result = re.search(r'https?://www.youjizz.com/videos/embed/(?P<videoid>[0-9]+)', webpage)
2013-01-05 21:42:35 +01:00
if result is None:
2013-01-12 16:36:51 +01:00
raise ExtractorError(u'ERROR: unable to extract embed page')
2013-01-05 21:42:35 +01:00
embed_page_url = result.group(0).strip()
video_id = result.group('videoid')
webpage = self._download_webpage(embed_page_url, video_id)
2013-01-05 21:42:35 +01:00
# Get the video URL
2013-01-12 16:36:51 +01:00
result = re.search(r'so.addVariable\("file",encodeURIComponent\("(?P<source>[^"]+)"\)\);', webpage)
2013-01-05 21:42:35 +01:00
if result is None:
2013-01-12 16:36:51 +01:00
raise ExtractorError(u'ERROR: unable to extract video url')
video_url = result.group('source')
2013-01-05 21:42:35 +01:00
info = {'id': video_id,
'url': video_url,
'title': video_title,
'ext': 'flv',
'format': 'flv',
'player_url': embed_page_url}
return [info]
2013-01-27 03:01:23 +01:00
class EightTracksIE(InfoExtractor):
IE_NAME = '8tracks'
2013-01-27 04:15:12 +01:00
_VALID_URL = r'https?://8tracks.com/(?P<user>[^/]+)/(?P<id>[^/#]+)(?:#.*)?$'
2013-01-27 03:01:23 +01:00
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
playlist_id = mobj.group('id')
webpage = self._download_webpage(url, playlist_id)
2013-02-18 19:11:32 +01:00
m = re.search(r"PAGE.mix = (.*?);\n", webpage, flags=re.DOTALL)
2013-01-27 03:01:23 +01:00
if not m:
raise ExtractorError(u'Cannot find trax information')
json_like = m.group(1)
data = json.loads(json_like)
session = str(random.randint(0, 1000000000))
mix_id = data['id']
track_count = data['tracks_count']
first_url = 'http://8tracks.com/sets/%s/play?player=sm&mix_id=%s&format=jsonh' % (session, mix_id)
next_url = first_url
res = []
for i in itertools.count():
api_json = self._download_webpage(next_url, playlist_id,
note=u'Downloading song information %s/%s' % (str(i+1), track_count),
errnote=u'Failed to download song information')
api_data = json.loads(api_json)
track_data = api_data[u'set']['track']
info = {
'id': track_data['id'],
'url': track_data['track_file_stream_url'],
2013-01-27 04:05:53 +01:00
'title': track_data['performer'] + u' - ' + track_data['name'],
'raw_title': track_data['name'],
'uploader_id': data['user']['login'],
2013-01-27 03:01:23 +01:00
'ext': 'm4a',
}
res.append(info)
if api_data['set']['at_last_track']:
break
next_url = 'http://8tracks.com/sets/%s/next?player=sm&mix_id=%s&format=jsonh&track_id=%s' % (session, mix_id, track_data['id'])
return res
2013-01-05 21:42:35 +01:00
2013-02-08 08:25:55 +01:00
class KeekIE(InfoExtractor):
_VALID_URL = r'http://(?:www\.)?keek\.com/(?:!|\w+/keeks/)(?P<videoID>\w+)'
IE_NAME = u'keek'
def _real_extract(self, url):
m = re.match(self._VALID_URL, url)
video_id = m.group('videoID')
video_url = u'http://cdn.keek.com/keek/video/%s' % video_id
thumbnail = u'http://cdn.keek.com/keek/thumbnail/%s/w100/h75' % video_id
webpage = self._download_webpage(url, video_id)
2013-04-11 18:39:13 +02:00
m = re.search(r'<meta property="og:title" content="(?P<title>.*?)"', webpage)
2013-02-08 11:00:28 +01:00
title = unescapeHTML(m.group('title'))
2013-03-20 12:13:52 +01:00
m = re.search(r'<div class="user-name-and-bio">[\S\s]+?<h2>(?P<uploader>.+?)</h2>', webpage)
uploader = clean_html(m.group('uploader'))
2013-02-08 08:25:55 +01:00
info = {
2013-03-20 12:13:52 +01:00
'id': video_id,
'url': video_url,
2013-02-08 08:25:55 +01:00
'ext': 'mp4',
'title': title,
'thumbnail': thumbnail,
'uploader': uploader
2013-02-08 11:00:28 +01:00
}
2013-02-08 08:25:55 +01:00
return [info]
2013-02-17 17:13:06 +01:00
class TEDIE(InfoExtractor):
_VALID_URL=r'''http://www\.ted\.com/
2013-02-18 21:42:06 +01:00
(
((?P<type_playlist>playlists)/(?P<playlist_id>\d+)) # We have a playlist
|
((?P<type_talk>talks)) # We have a simple talk
)
(/lang/(.*?))? # The url may contain the language
2013-02-18 21:42:06 +01:00
/(?P<name>\w+) # Here goes the name and then ".html"
'''
@classmethod
def suitable(cls, url):
2013-02-18 21:42:06 +01:00
"""Receives a URL and returns True if suitable for this IE."""
return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
2013-02-18 21:42:06 +01:00
2013-02-17 17:13:06 +01:00
def _real_extract(self, url):
2013-02-18 21:42:06 +01:00
m=re.match(self._VALID_URL, url, re.VERBOSE)
if m.group('type_talk'):
return [self._talk_info(url)]
else :
playlist_id=m.group('playlist_id')
name=m.group('name')
self.to_screen(u'Getting info of playlist %s: "%s"' % (playlist_id,name))
2013-04-20 13:31:21 +02:00
return [self._playlist_videos_info(url,name,playlist_id)]
2013-02-18 21:42:06 +01:00
def _talk_video_link(self,mediaSlug):
'''Returns the video link for that mediaSlug'''
return 'http://download.ted.com/talks/%s.mp4' % mediaSlug
def _playlist_videos_info(self,url,name,playlist_id=0):
'''Returns the videos of the playlist'''
video_RE=r'''
<li\ id="talk_(\d+)"([.\s]*?)data-id="(?P<video_id>\d+)"
([.\s]*?)data-playlist_item_id="(\d+)"
([.\s]*?)data-mediaslug="(?P<mediaSlug>.+?)"
'''
2013-02-23 17:27:49 +01:00
video_name_RE=r'<p\ class="talk-title"><a href="(?P<talk_url>/talks/(.+).html)">(?P<fullname>.+?)</a></p>'
2013-02-18 21:42:06 +01:00
webpage=self._download_webpage(url, playlist_id, 'Downloading playlist webpage')
m_videos=re.finditer(video_RE,webpage,re.VERBOSE)
m_names=re.finditer(video_name_RE,webpage)
2013-04-20 13:31:21 +02:00
playlist_RE = r'div class="headline">(\s*?)<h1>(\s*?)<span>(?P<playlist_title>.*?)</span>'
m_playlist = re.search(playlist_RE, webpage)
playlist_title = m_playlist.group('playlist_title')
playlist_entries = []
2013-02-18 21:42:06 +01:00
for m_video, m_name in zip(m_videos,m_names):
2013-02-23 17:27:49 +01:00
video_id=m_video.group('video_id')
talk_url='http://www.ted.com%s' % m_name.group('talk_url')
2013-04-20 13:31:21 +02:00
playlist_entries.append(self.url_result(talk_url, 'TED'))
return self.playlist_result(playlist_entries, playlist_id = playlist_id, playlist_title = playlist_title)
2013-02-23 17:27:49 +01:00
2013-02-18 21:42:06 +01:00
def _talk_info(self, url, video_id=0):
"""Return the video for the talk in the url"""
m=re.match(self._VALID_URL, url,re.VERBOSE)
videoName=m.group('name')
webpage=self._download_webpage(url, video_id, 'Downloading \"%s\" page' % videoName)
# If the url includes the language we get the title translated
2013-03-29 15:59:13 +01:00
title_RE=r'<span id="altHeadline" >(?P<title>.*)</span>'
2013-02-17 17:13:06 +01:00
title=re.search(title_RE, webpage).group('title')
info_RE=r'''<script\ type="text/javascript">var\ talkDetails\ =(.*?)
"id":(?P<videoID>[\d]+).*?
"mediaSlug":"(?P<mediaSlug>[\w\d]+?)"'''
2013-02-23 17:27:49 +01:00
thumb_RE=r'</span>[\s.]*</div>[\s.]*<img src="(?P<thumbnail>.*?)"'
thumb_match=re.search(thumb_RE,webpage)
2013-02-17 17:13:06 +01:00
info_match=re.search(info_RE,webpage,re.VERBOSE)
video_id=info_match.group('videoID')
mediaSlug=info_match.group('mediaSlug')
2013-02-18 21:42:06 +01:00
video_url=self._talk_video_link(mediaSlug)
2013-02-17 17:13:06 +01:00
info = {
2013-02-18 21:42:06 +01:00
'id': video_id,
'url': video_url,
2013-02-17 17:13:06 +01:00
'ext': 'mp4',
2013-02-23 17:27:49 +01:00
'title': title,
'thumbnail': thumb_match.group('thumbnail')
2013-02-18 21:42:06 +01:00
}
return info
2013-02-08 08:25:55 +01:00
2013-02-18 18:45:09 +01:00
class MySpassIE(InfoExtractor):
2013-02-16 13:46:13 +01:00
_VALID_URL = r'http://www.myspass.de/.*'
2013-02-16 13:46:13 +01:00
def _real_extract(self, url):
META_DATA_URL_TEMPLATE = 'http://www.myspass.de/myspass/includes/apps/video/getvideometadataxml.php?id=%s'
2013-02-18 18:45:09 +01:00
2013-02-16 13:46:13 +01:00
# video id is the last path element of the URL
# usually there is a trailing slash, so also try the second but last
url_path = compat_urllib_parse_urlparse(url).path
url_parent_path, video_id = os.path.split(url_path)
if not video_id:
_, video_id = os.path.split(url_parent_path)
2013-02-16 13:46:13 +01:00
# get metadata
metadata_url = META_DATA_URL_TEMPLATE % video_id
metadata_text = self._download_webpage(metadata_url, video_id)
metadata = xml.etree.ElementTree.fromstring(metadata_text.encode('utf-8'))
2013-02-16 13:46:13 +01:00
# extract values from metadata
url_flv_el = metadata.find('url_flv')
if url_flv_el is None:
raise ExtractorError(u'Unable to extract download url')
2013-02-16 13:46:13 +01:00
video_url = url_flv_el.text
extension = os.path.splitext(video_url)[1][1:]
title_el = metadata.find('title')
if title_el is None:
raise ExtractorError(u'Unable to extract title')
2013-02-16 13:46:13 +01:00
title = title_el.text
format_id_el = metadata.find('format_id')
if format_id_el is None:
format = ext
else:
format = format_id_el.text
description_el = metadata.find('description')
if description_el is not None:
description = description_el.text
else:
description = None
imagePreview_el = metadata.find('imagePreview')
if imagePreview_el is not None:
thumbnail = imagePreview_el.text
else:
thumbnail = None
info = {
'id': video_id,
'url': video_url,
'title': title,
'ext': extension,
'format': format,
'thumbnail': thumbnail,
'description': description
}
return [info]
2013-03-12 01:08:54 +01:00
class SpiegelIE(InfoExtractor):
2013-03-29 15:31:38 +01:00
_VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P<videoID>[0-9]+)(?:\.html)?(?:#.*)?$'
2013-03-12 01:08:54 +01:00
def _real_extract(self, url):
m = re.match(self._VALID_URL, url)
video_id = m.group('videoID')
webpage = self._download_webpage(url, video_id)
m = re.search(r'<div class="spVideoTitle">(.*?)</div>', webpage)
if not m:
raise ExtractorError(u'Cannot find title')
video_title = unescapeHTML(m.group(1))
xml_url = u'http://video2.spiegel.de/flash/' + video_id + u'.xml'
xml_code = self._download_webpage(xml_url, video_id,
note=u'Downloading XML', errnote=u'Failed to download XML')
idoc = xml.etree.ElementTree.fromstring(xml_code)
last_type = idoc[-1]
filename = last_type.findall('./filename')[0].text
duration = float(last_type.findall('./duration')[0].text)
video_url = 'http://video2.spiegel.de/flash/' + filename
video_ext = filename.rpartition('.')[2]
info = {
'id': video_id,
'url': video_url,
'ext': video_ext,
'title': video_title,
'duration': duration,
}
return [info]
class LiveLeakIE(InfoExtractor):
2013-03-26 21:37:08 +01:00
_VALID_URL = r'^(?:http?://)?(?:\w+\.)?liveleak\.com/view\?(?:.*?)i=(?P<video_id>[\w_]+)(?:.*)'
2013-03-26 21:37:08 +01:00
IE_NAME = u'liveleak'
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
if mobj is None:
raise ExtractorError(u'Invalid URL: %s' % url)
2013-03-26 21:37:08 +01:00
video_id = mobj.group('video_id')
2013-03-26 21:37:08 +01:00
webpage = self._download_webpage(url, video_id)
m = re.search(r'file: "(.*?)",', webpage)
if not m:
raise ExtractorError(u'Unable to find video url')
video_url = m.group(1)
2013-03-26 21:37:08 +01:00
m = re.search(r'<meta property="og:title" content="(?P<title>.*?)"', webpage)
if not m:
self._downloader.report_error(u'Cannot find video title')
title = unescapeHTML(m.group('title')).replace('LiveLeak.com -', '').strip()
2013-03-26 21:37:08 +01:00
m = re.search(r'<meta property="og:description" content="(?P<desc>.*?)"', webpage)
if m:
desc = unescapeHTML(m.group('desc'))
else:
desc = None
m = re.search(r'By:.*?(\w+)</a>', webpage)
if m:
uploader = clean_html(m.group(1))
else:
uploader = None
2013-03-26 21:37:08 +01:00
info = {
'id': video_id,
'url': video_url,
'ext': 'mp4',
'title': title,
'description': desc,
'uploader': uploader
2013-03-26 21:37:08 +01:00
}
return [info]
class ARDIE(InfoExtractor):
2013-04-11 10:47:21 +02:00
_VALID_URL = r'^(?:https?://)?(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P<video_id>[^/\?]+)(?:\?.*)?'
_TITLE = r'<h1(?: class="boxTopHeadline")?>(?P<title>.*)</h1>'
_MEDIA_STREAM = r'mediaCollection\.addMediaStream\((?P<media_type>\d+), (?P<quality>\d+), "(?P<rtmp_url>[^"]*)", "(?P<video_url>[^"]*)", "[^"]*"\)'
def _real_extract(self, url):
# determine video id from url
m = re.match(self._VALID_URL, url)
2013-04-11 10:47:21 +02:00
numid = re.search(r'documentId=([0-9]+)', url)
if numid:
video_id = numid.group(1)
else:
video_id = m.group('video_id')
# determine title and media streams from webpage
html = self._download_webpage(url, video_id)
title = re.search(self._TITLE, html).group('title')
streams = [m.groupdict() for m in re.finditer(self._MEDIA_STREAM, html)]
if not streams:
assert '"fsk"' in html
raise ExtractorError(u'This video is only available after 8:00 pm')
# choose default media type and highest quality for now
2013-04-11 10:47:21 +02:00
stream = max([s for s in streams if int(s["media_type"]) == 0],
key=lambda s: int(s["quality"]))
# there's two possibilities: RTMP stream or HTTP download
info = {'id': video_id, 'title': title, 'ext': 'mp4'}
if stream['rtmp_url']:
self.to_screen(u'RTMP download detected')
assert stream['video_url'].startswith('mp4:')
info["url"] = stream["rtmp_url"]
info["play_path"] = stream['video_url']
else:
assert stream["video_url"].endswith('.mp4')
info["url"] = stream["video_url"]
return [info]
class TumblrIE(InfoExtractor):
_VALID_URL = r'http://(?P<blog_name>.*?)\.tumblr\.com/((post)|(video))/(?P<id>\d*)/(.*?)'
def _real_extract(self, url):
m_url = re.match(self._VALID_URL, url)
video_id = m_url.group('id')
blog = m_url.group('blog_name')
url = 'http://%s.tumblr.com/post/%s/' % (blog, video_id)
webpage = self._download_webpage(url, video_id)
re_video = r'src=\\x22(?P<video_url>http://%s\.tumblr\.com/video_file/%s/(.*?))\\x22 type=\\x22video/(?P<ext>.*?)\\x22' % (blog, video_id)
video = re.search(re_video, webpage)
if video is None:
self.to_screen("No video founded")
return []
video_url = video.group('video_url')
ext = video.group('ext')
re_thumb = r'posters(.*?)\[\\x22(?P<thumb>.*?)\\x22' # We pick the first poster
thumb = re.search(re_thumb, webpage).group('thumb').replace('\\', '')
# The only place where you can get a title, it's not complete,
# but searching in other places doesn't work for all videos
2013-05-02 13:12:41 +02:00
re_title = r'<title>(?P<title>.*?)</title>'
title = unescapeHTML(re.search(re_title, webpage, re.DOTALL).group('title'))
return [{'id': video_id,
'url': video_url,
'title': title,
'thumbnail': thumb,
'ext': ext
}]
2013-05-01 15:55:46 +02:00
class BandcampIE(InfoExtractor):
_VALID_URL = r'http://.*?\.bandcamp\.com/track/(?P<title>.*)'
2013-05-01 15:55:46 +02:00
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
title = mobj.group('title')
webpage = self._download_webpage(url, title)
# We get the link to the free download page
m_download = re.search(r'freeDownloadPage: "(.*?)"', webpage)
if m_download is None:
raise ExtractorError(u'No free songs founded')
2013-05-01 15:55:46 +02:00
download_link = m_download.group(1)
id = re.search(r'var TralbumData = {(.*?)id: (?P<id>\d*?)$',
webpage, re.MULTILINE|re.DOTALL).group('id')
download_webpage = self._download_webpage(download_link, id,
'Downloading free downloads page')
# We get the dictionary of the track from some javascrip code
info = re.search(r'items: (.*?),$',
download_webpage, re.MULTILINE).group(1)
info = json.loads(info)[0]
# We pick mp3-320 for now, until format selection can be easily implemented.
mp3_info = info[u'downloads'][u'mp3-320']
# If we try to use this url it says the link has expired
initial_url = mp3_info[u'url']
re_url = r'(?P<server>http://(.*?)\.bandcamp\.com)/download/track\?enc=mp3-320&fsig=(?P<fsig>.*?)&id=(?P<id>.*?)&ts=(?P<ts>.*)$'
2013-05-01 15:55:46 +02:00
m_url = re.match(re_url, initial_url)
#We build the url we will use to get the final track url
# This url is build in Bandcamp in the script download_bunde_*.js
request_url = '%s/statdownload/track?enc=mp3-320&fsig=%s&id=%s&ts=%s&.rand=665028774616&.vrs=1' % (m_url.group('server'), m_url.group('fsig'), id, m_url.group('ts'))
final_url_webpage = self._download_webpage(request_url, id, 'Requesting download url')
# If we could correctly generate the .rand field the url would be
#in the "download_url" key
final_url = re.search(r'"retry_url":"(.*?)"', final_url_webpage).group(1)
track_info = {'id':id,
'title' : info[u'title'],
'ext' : 'mp3',
'url' : final_url,
'thumbnail' : info[u'thumb_url'],
'uploader' : info[u'artist']
}
return [track_info]
2013-05-03 20:07:35 +02:00
class RedTubeIE(InfoExtractor):
"""Information Extractor for redtube"""
_VALID_URL = r'(?:http://)?(?:www\.)?redtube\.com/(?P<id>[0-9]+)'
def _real_extract(self,url):
mobj = re.match(self._VALID_URL, url)
if mobj is None:
2013-05-03 20:07:35 +02:00
raise ExtractorError(u'Invalid URL: %s' % url)
video_id = mobj.group('id')
video_extension = 'mp4'
webpage = self._download_webpage(url, video_id)
self.report_extraction(video_id)
mobj = re.search(r'<source src="'+'(.+)'+'" type="video/mp4">',webpage)
2013-05-03 20:07:35 +02:00
if mobj is None:
raise ExtractorError(u'Unable to extract media URL')
video_url = mobj.group(1)
mobj = re.search('<h1 class="videoTitle slidePanelMovable">(.+)</h1>',webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract title')
video_title = mobj.group(1)
return [{
'id': video_id,
'url': video_url,
'ext': video_extension,
'title': video_title,
}]
2013-03-12 01:08:54 +01:00
2013-01-01 19:37:07 +01:00
def gen_extractors():
""" Return a list of an instance of every supported extractor.
The order does matter; the first extractor matched is the one handling the URL.
"""
return [
YoutubePlaylistIE(),
YoutubeChannelIE(),
YoutubeUserIE(),
YoutubeSearchIE(),
YoutubeIE(),
MetacafeIE(),
DailymotionIE(),
GoogleSearchIE(),
PhotobucketIE(),
YahooIE(),
YahooSearchIE(),
DepositFilesIE(),
FacebookIE(),
BlipTVUserIE(),
BlipTVIE(),
VimeoIE(),
MyVideoIE(),
ComedyCentralIE(),
EscapistIE(),
CollegeHumorIE(),
XVideosIE(),
SoundcloudSetIE(),
2013-01-01 19:37:07 +01:00
SoundcloudIE(),
InfoQIE(),
MixcloudIE(),
StanfordOpenClassroomIE(),
MTVIE(),
YoukuIE(),
XNXXIE(),
2013-01-06 21:52:33 +01:00
YouJizzIE(),
PornotubeIE(),
YouPornIE(),
2013-01-01 19:37:07 +01:00
GooglePlusIE(),
ArteTvIE(),
NBAIE(),
2013-03-07 06:09:55 +01:00
WorldStarHipHopIE(),
2013-01-01 19:37:07 +01:00
JustinTVIE(),
FunnyOrDieIE(),
SteamIE(),
UstreamIE(),
2013-01-12 17:58:39 +01:00
RBMARadioIE(),
2013-01-27 03:01:23 +01:00
EightTracksIE(),
2013-02-08 08:25:55 +01:00
KeekIE(),
2013-02-17 17:13:06 +01:00
TEDIE(),
2013-02-18 18:45:09 +01:00
MySpassIE(),
2013-03-12 01:08:54 +01:00
SpiegelIE(),
LiveLeakIE(),
ARDIE(),
TumblrIE(),
2013-05-01 15:55:46 +02:00
BandcampIE(),
2013-05-03 20:07:35 +02:00
RedTubeIE(),
2013-01-01 19:37:07 +01:00
GenericIE()
]
def get_info_extractor(ie_name):
"""Returns the info extractor class with the given ie_name"""
return globals()[ie_name+'IE']