Merge pull request #3 from unit193/master

[youtubedl] / youtube_dl / extractor / laola1tv.py

diff --git a/youtube_dl/extractor/laola1tv.py b/youtube_dl/extractor/laola1tv.py
index 263f687736c3f7fe3e980d79a10971a6dc4df5b2..fa217365a37853f471440d8138d0e20bc4115d52 100644 (file)
--- a/youtube_dl/extractor/laola1tv.py
+++ b/youtube_dl/extractor/laola1tv.py
@@ -1,77 +1,265 @@
+# coding: utf-8

 from __future__ import unicode_literals
 
 from __future__ import unicode_literals
 

-import random
+import json

 import re
 
 from .common import InfoExtractor
 import re
 
 from .common import InfoExtractor

+from ..utils import (
+    ExtractorError,
+    unified_strdate,
+    urlencode_postdata,
+    xpath_element,
+    xpath_text,
+    update_url_query,
+    js_to_json,
+)

 
 
 
 

-class Laola1TvIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?laola1\.tv/(?P<lang>[a-z]+)-(?P<portal>[a-z]+)/.*?/(?P<id>[0-9]+)\.html'
-    _TEST = {
-        'url': 'http://www.laola1.tv/de-de/live/bwf-bitburger-open-grand-prix-gold-court-1/250019.html',
+class Laola1TvEmbedIE(InfoExtractor):
+    IE_NAME = 'laola1tv:embed'
+    _VALID_URL = r'https?://(?:www\.)?laola1\.tv/titanplayer\.php\?.*?\bvideoid=(?P<id>\d+)'
+    _TESTS = [{
+        # flashvars.premium = "false";
+        'url': 'https://www.laola1.tv/titanplayer.php?videoid=708065&type=V&lang=en&portal=int&customer=1024',

         'info_dict': {
         'info_dict': {

-            'id': '250019',
+            'id': '708065',

             'ext': 'mp4',
             'ext': 'mp4',

-            'title': 'Bitburger Open Grand Prix Gold - Court 1',
-            'categories': ['Badminton'],
-            'uploader': 'BWF - Badminton World Federation',
-            'is_live': True,
+            'title': 'MA Long CHN - FAN Zhendong CHN',
+            'uploader': 'ITTF - International Table Tennis Federation',
+            'upload_date': '20161211',

         },
         },

-        'params': {
-            'skip_download': True,
-        }
-    }
+    }]

 
 

-    _BROKEN = True  # Not really - extractor works fine, but f4m downloader does not support live streams yet.
+    def _extract_token_url(self, stream_access_url, video_id, data):
+        return self._download_json(
+            self._proto_relative_url(stream_access_url, 'https:'), video_id,
+            headers={
+                'Content-Type': 'application/json',
+            }, data=json.dumps(data).encode())['data']['stream-access'][0]

 
 

-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        lang = mobj.group('lang')
-        portal = mobj.group('portal')
+    def _extract_formats(self, token_url, video_id):
+        token_doc = self._download_xml(
+            token_url, video_id, 'Downloading token',
+            headers=self.geo_verification_headers())
+
+        token_attrib = xpath_element(token_doc, './/token').attrib

 
 

+        if token_attrib['status'] != '0':
+            raise ExtractorError(
+                'Token error: %s' % token_attrib['comment'], expected=True)
+
+        formats = self._extract_akamai_formats(
+            '%s?hdnea=%s' % (token_attrib['url'], token_attrib['auth']),
+            video_id)
+        self._sort_formats(formats)
+        return formats
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)

         webpage = self._download_webpage(url, video_id)
         webpage = self._download_webpage(url, video_id)

-        iframe_url = self._search_regex(
-            r'<iframe[^>]*?class="main_tv_player"[^>]*?src="([^"]+)"',
-            webpage, 'iframe URL')
-
-        iframe = self._download_webpage(
-            iframe_url, video_id, note='Downloading iframe')
-        flashvars_m = re.findall(
-            r'flashvars\.([_a-zA-Z0-9]+)\s*=\s*"([^"]*)";', iframe)
-        flashvars = dict((m[0], m[1]) for m in flashvars_m)
-
-        xml_url = ('http://www.laola1.tv/server/hd_video.php?' +
-                   'play=%s&partner=1&portal=%s&v5ident=&lang=%s' % (
-                       video_id, portal, lang))
-        hd_doc = self._download_xml(xml_url, video_id)
-
-        title = hd_doc.find('.//video/title').text
-        flash_url = hd_doc.find('.//video/url').text
-        categories = hd_doc.find('.//video/meta_sports').text.split(',')
-        uploader = hd_doc.find('.//video/meta_organistation').text
-
-        ident = random.randint(10000000, 99999999)
-        token_url = '%s&ident=%s&klub=0&unikey=0&timestamp=%s&auth=%s' % (
-            flash_url, ident, flashvars['timestamp'], flashvars['auth'])
+        flash_vars = self._search_regex(
+            r'(?s)flashvars\s*=\s*({.+?});', webpage, 'flash vars')

 
 

-        token_doc = self._download_xml(
-            token_url, video_id, note='Downloading token')
-        token_attrib = token_doc.find('.//token').attrib
-        if token_attrib.get('auth') == 'blocked':
-            raise ExtractorError('Token error: ' % token_attrib.get('comment'))
+        def get_flashvar(x, *args, **kwargs):
+            flash_var = self._search_regex(
+                r'%s\s*:\s*"([^"]+)"' % x,
+                flash_vars, x, default=None)
+            if not flash_var:
+                flash_var = self._search_regex([
+                    r'flashvars\.%s\s*=\s*"([^"]+)"' % x,
+                    r'%s\s*=\s*"([^"]+)"' % x],
+                    webpage, x, *args, **kwargs)
+            return flash_var
+
+        hd_doc = self._download_xml(
+            'http://www.laola1.tv/server/hd_video.php', video_id, query={
+                'play': get_flashvar('streamid'),
+                'partner': get_flashvar('partnerid'),
+                'portal': get_flashvar('portalid'),
+                'lang': get_flashvar('sprache'),
+                'v5ident': '',
+            })

 
 

-        video_url = '%s?hdnea=%s&hdcore=3.2.0' % (
-            token_attrib['url'], token_attrib['auth'])
+        _v = lambda x, **k: xpath_text(hd_doc, './/video/' + x, **k)
+        title = _v('title', fatal=True)
+
+        token_url = None
+        premium = get_flashvar('premium', default=None)
+        if premium:
+            token_url = update_url_query(
+                _v('url', fatal=True), {
+                    'timestamp': get_flashvar('timestamp'),
+                    'auth': get_flashvar('auth'),
+                })
+        else:
+            data_abo = urlencode_postdata(
+                dict((i, v) for i, v in enumerate(_v('req_liga_abos').split(','))))
+            stream_access_url = update_url_query(
+                'https://club.laola1.tv/sp/laola1/api/v3/user/session/premium/player/stream-access', {
+                    'videoId': _v('id'),
+                    'target': self._search_regex(r'vs_target = (\d+);', webpage, 'vs target'),
+                    'label': _v('label'),
+                    'area': _v('area'),
+                })
+            token_url = self._extract_token_url(stream_access_url, video_id, data_abo)
+
+        formats = self._extract_formats(token_url, video_id)
+
+        categories_str = _v('meta_sports')
+        categories = categories_str.split(',') if categories_str else []
+        is_live = _v('islive') == 'true'

 
         return {
             'id': video_id,
 
         return {
             'id': video_id,

-            'is_live': True,
-            'title': title,
-            'url': video_url,
-            'uploader': uploader,
+            'title': self._live_title(title) if is_live else title,
+            'upload_date': unified_strdate(_v('time_date')),
+            'uploader': _v('meta_organisation'),

             'categories': categories,
             'categories': categories,

-            'ext': 'mp4',
+            'is_live': is_live,
+            'formats': formats,

         }
 
         }
 

+
+class Laola1TvBaseIE(Laola1TvEmbedIE):
+    def _extract_video(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+
+        if 'Dieser Livestream ist bereits beendet.' in webpage:
+            raise ExtractorError('This live stream has already finished.', expected=True)
+
+        conf = self._parse_json(self._search_regex(
+            r'(?s)conf\s*=\s*({.+?});', webpage, 'conf'),
+            display_id,
+            transform_source=lambda s: js_to_json(re.sub(r'shareurl:.+,', '', s)))
+        video_id = conf['videoid']
+
+        config = self._download_json(conf['configUrl'], video_id, query={
+            'videoid': video_id,
+            'partnerid': conf['partnerid'],
+            'language': conf.get('language', ''),
+            'portal': conf.get('portalid', ''),
+        })
+        error = config.get('error')
+        if error:
+            raise ExtractorError('%s said: %s' % (self.IE_NAME, error), expected=True)
+
+        video_data = config['video']
+        title = video_data['title']
+        is_live = video_data.get('isLivestream') and video_data.get('isLive')
+        meta = video_data.get('metaInformation')
+        sports = meta.get('sports')
+        categories = sports.split(',') if sports else []
+
+        token_url = self._extract_token_url(
+            video_data['streamAccess'], video_id,
+            video_data['abo']['required'])
+
+        formats = self._extract_formats(token_url, video_id)
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'title': self._live_title(title) if is_live else title,
+            'description': video_data.get('description'),
+            'thumbnail': video_data.get('image'),
+            'categories': categories,
+            'formats': formats,
+            'is_live': is_live,
+        }
+
+
+class Laola1TvIE(Laola1TvBaseIE):
+    IE_NAME = 'laola1tv'
+    _VALID_URL = r'https?://(?:www\.)?laola1\.tv/[a-z]+-[a-z]+/[^/]+/(?P<id>[^/?#&]+)'
+
+    _TESTS = [{
+        'url': 'http://www.laola1.tv/de-de/video/straubing-tigers-koelner-haie/227883.html',
+        'info_dict': {
+            'id': '227883',
+            'display_id': 'straubing-tigers-koelner-haie',
+            'ext': 'flv',
+            'title': 'Straubing Tigers - Kölner Haie',
+            'upload_date': '20140912',
+            'is_live': False,
+            'categories': ['Eishockey'],
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://www.laola1.tv/de-de/video/straubing-tigers-koelner-haie',
+        'info_dict': {
+            'id': '464602',
+            'display_id': 'straubing-tigers-koelner-haie',
+            'ext': 'flv',
+            'title': 'Straubing Tigers - Kölner Haie',
+            'upload_date': '20160129',
+            'is_live': False,
+            'categories': ['Eishockey'],
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://www.laola1.tv/de-de/livestream/2016-03-22-belogorie-belgorod-trentino-diatec-lde',
+        'info_dict': {
+            'id': '487850',
+            'display_id': '2016-03-22-belogorie-belgorod-trentino-diatec-lde',
+            'ext': 'flv',
+            'title': 'Belogorie BELGOROD - TRENTINO Diatec',
+            'upload_date': '20160322',
+            'uploader': 'CEV - Europäischer Volleyball Verband',
+            'is_live': True,
+            'categories': ['Volleyball'],
+        },
+        'params': {
+            'skip_download': True,
+        },
+        'skip': 'This live stream has already finished.',
+    }]
+
+    def _real_extract(self, url):
+        return self._extract_video(url)
+
+
+class EHFTVIE(Laola1TvBaseIE):
+    IE_NAME = 'ehftv'
+    _VALID_URL = r'https?://(?:www\.)?ehftv\.com/[a-z]+(?:-[a-z]+)?/[^/]+/(?P<id>[^/?#&]+)'
+
+    _TESTS = [{
+        'url': 'https://www.ehftv.com/int/video/paris-saint-germain-handball-pge-vive-kielce/1166761',
+        'info_dict': {
+            'id': '1166761',
+            'display_id': 'paris-saint-germain-handball-pge-vive-kielce',
+            'ext': 'mp4',
+            'title': 'Paris Saint-Germain Handball - PGE Vive Kielce',
+            'is_live': False,
+            'categories': ['Handball'],
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }]
+
+    def _real_extract(self, url):
+        return self._extract_video(url)
+
+
+class ITTFIE(InfoExtractor):
+    _VALID_URL = r'https?://tv\.ittf\.com/video/[^/]+/(?P<id>\d+)'
+    _TEST = {
+        'url': 'https://tv.ittf.com/video/peng-wang-wei-matsudaira-kenta/951802',
+        'only_matching': True,
+    }
+
+    def _real_extract(self, url):
+        return self.url_result(
+            update_url_query('https://www.laola1.tv/titanplayer.php', {
+                'videoid': self._match_id(url),
+                'type': 'V',
+                'lang': 'en',
+                'portal': 'int',
+                'customer': 1024,
+            }), Laola1TvEmbedIE.ie_key())