Merge remote-tracking branch 'Boris-de/wdrmaus_fix#8562'

youtube_dl/extractor/extractors.py

@@ -967,7 +967,6 @@ from .watchindianporn import WatchIndianPornIE
 from .wdr import (
     WDRIE,
     WDRMobileIE,
-    WDRMausIE,
 )
 from .webofstories import (
     WebOfStoriesIE,

youtube_dl/extractor/wdr.py

@@ -1,214 +1,205 @@
 # -*- coding: utf-8 -*-
 from __future__ import unicode_literals
 
-import itertools
 import re
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_parse_qs,
-    compat_urlparse,
-)
 from ..utils import (
+    determine_ext,
+    js_to_json,
+    strip_jsonp,
     unified_strdate,
-    qualities,
+    ExtractorError,
 )
 
 
 class WDRIE(InfoExtractor):
-    _PLAYER_REGEX = '-(?:video|audio)player(?:_size-[LMS])?'
+    _CURRENT_MAUS_URL = r'https?://(?:www\.)wdrmaus.de/(?:[^/]+/){1,2}[^/?#]+\.php5'
-    _VALID_URL = r'(?P<url>https?://www\d?\.(?:wdr\d?|funkhauseuropa)\.de/)(?P<id>.+?)(?P<player>%s)?\.html' % _PLAYER_REGEX
+    _PAGE_REGEX = r'/mediathek/(?P<media_type>[^/]+)/(?P<type>[^/]+)/(?P<display_id>.+)\.html'
+    _VALID_URL = r'(?P<page_url>https?://(?:www\d\.)?wdr\d?\.de)' + _PAGE_REGEX + '|' + _CURRENT_MAUS_URL
 
     _TESTS = [
         {
-            'url': 'http://www1.wdr.de/mediathek/video/sendungen/servicezeit/videoservicezeit560-videoplayer_size-L.html',
+            'url': 'http://www1.wdr.de/mediathek/video/sendungen/doku-am-freitag/video-geheimnis-aachener-dom-100.html',
+            'md5': 'e58c39c3e30077141d258bf588700a7b',
             'info_dict': {
-                'id': 'mdb-362427',
+                'id': 'mdb-1058683',
                 'ext': 'flv',
-                'title': 'Servicezeit',
+                'display_id': 'doku-am-freitag/video-geheimnis-aachener-dom-100',
-                'description': 'md5:c8f43e5e815eeb54d0b96df2fba906cb',
+                'title': 'Geheimnis Aachener Dom',
-                'upload_date': '20140310',
+                'alt_title': 'Doku am Freitag',
-                'is_live': False
+                'upload_date': '20160304',
-            },
+                'description': 'md5:87be8ff14d8dfd7a7ee46f0299b52318',
-            'params': {
+                'is_live': False,
-                'skip_download': True,
+                'subtitles': {'de': [{
+                    'url': 'http://ondemand-ww.wdr.de/medp/fsk0/105/1058683/1058683_12220974.xml'
+                }]},
             },
             'skip': 'Page Not Found',
         },
         {
-            'url': 'http://www1.wdr.de/themen/av/videomargaspiegelisttot101-videoplayer.html',
+            'url': 'http://www1.wdr.de/mediathek/audio/wdr3/wdr3-gespraech-am-samstag/audio-schriftstellerin-juli-zeh-100.html',
+            'md5': 'f4c1f96d01cf285240f53ea4309663d8',
             'info_dict': {
-                'id': 'mdb-363194',
+                'id': 'mdb-1072000',
-                'ext': 'flv',
-                'title': 'Marga Spiegel ist tot',
-                'description': 'md5:2309992a6716c347891c045be50992e4',
-                'upload_date': '20140311',
-                'is_live': False
-            },
-            'params': {
-                'skip_download': True,
-            },
-            'skip': 'Page Not Found',
-        },
-        {
-            'url': 'http://www1.wdr.de/themen/kultur/audioerlebtegeschichtenmargaspiegel100-audioplayer.html',
-            'md5': '83e9e8fefad36f357278759870805898',
-            'info_dict': {
-                'id': 'mdb-194332',
                 'ext': 'mp3',
-                'title': 'Erlebte Geschichten: Marga Spiegel (29.11.2009)',
+                'display_id': 'wdr3-gespraech-am-samstag/audio-schriftstellerin-juli-zeh-100',
-                'description': 'md5:2309992a6716c347891c045be50992e4',
+                'title': 'Schriftstellerin Juli Zeh',
-                'upload_date': '20091129',
+                'alt_title': 'WDR 3 Gespräch am Samstag',
-                'is_live': False
+                'upload_date': '20160312',
-            },
+                'description': 'md5:e127d320bc2b1f149be697ce044a3dd7',
-        },
+                'is_live': False,
-        {
+                'subtitles': {}
-            'url': 'http://www.funkhauseuropa.de/av/audioflaviacoelhoamaramar100-audioplayer.html',
-            'md5': '99a1443ff29af19f6c52cf6f4dc1f4aa',
-            'info_dict': {
-                'id': 'mdb-478135',
-                'ext': 'mp3',
-                'title': 'Flavia Coelho: Amar é Amar',
-                'description': 'md5:7b29e97e10dfb6e265238b32fa35b23a',
-                'upload_date': '20140717',
-                'is_live': False
             },
             'skip': 'Page Not Found',
         },
         {
-            'url': 'http://www1.wdr.de/mediathek/video/sendungen/quarks_und_co/filterseite-quarks-und-co100.html',
+            'url': 'http://www1.wdr.de/mediathek/video/live/index.html',
-            'playlist_mincount': 146,
-            'info_dict': {
-                'id': 'mediathek/video/sendungen/quarks_und_co/filterseite-quarks-und-co100',
-            }
-        },
-        {
-            'url': 'http://www1.wdr.de/mediathek/video/livestream/index.html',
             'info_dict': {
                 'id': 'mdb-103364',
-                'title': 're:^WDR Fernsehen Live [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
+                'ext': 'mp4',
+                'display_id': 'index',
+                'title': r're:^WDR Fernsehen im Livestream [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
+                'alt_title': 'WDR Fernsehen Live',
+                'upload_date': None,
                 'description': 'md5:ae2ff888510623bf8d4b115f95a9b7c9',
-                'ext': 'flv',
+                'is_live': True,
-                'upload_date': '20150101',
+                'subtitles': {}
-                'is_live': True
             },
             'params': {
-                'skip_download': True,
+                'skip_download': True,  # m3u8 download
+            },
+        },
+        {
+            'url': 'http://www1.wdr.de/mediathek/video/sendungen/aktuelle-stunde/aktuelle-stunde-120.html',
+            'playlist_mincount': 8,
+            'info_dict': {
+                'id': 'aktuelle-stunde/aktuelle-stunde-120',
+            },
+        },
+        {
+            'url': 'http://www.wdrmaus.de/aktuelle-sendung/index.php5',
+            'info_dict': {
+                'id': 'mdb-1096487',
+                'ext': 'flv',
+                'upload_date': 're:^[0-9]{8}$',
+                'title': 're:^Die Sendung mit der Maus vom [0-9.]{10}$',
+                'description': '- Die Sendung mit der Maus -',
+            },
+            'skip': 'The id changes from week to week because of the new episode'
+        },
+        {
+            'url': 'http://www.wdrmaus.de/sachgeschichten/sachgeschichten/achterbahn.php5',
+            'md5': 'ca365705551e4bd5217490f3b0591290',
+            'info_dict': {
+                'id': 'mdb-186083',
+                'ext': 'flv',
+                'upload_date': '20130919',
+                'title': 'Sachgeschichte - Achterbahn ',
+                'description': '- Die Sendung mit der Maus -',
+            },
+            'params': {
+                'skip_download': True,  # the file has different versions :(
+            },
         },
-        }
     ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        page_url = mobj.group('url')
+        url_type = mobj.group('type')
-        page_id = mobj.group('id')
+        page_url = mobj.group('page_url')
+        display_id = mobj.group('display_id')
+        webpage = self._download_webpage(url, display_id)
 
-        webpage = self._download_webpage(url, page_id)
+        # for wdr.de the data-extension is in a tag with the class "mediaLink"
+        # for wdrmaus its in a link to the page in a multiline "videoLink"-tag
+        json_metadata = self._html_search_regex(
+            r'class=(?:"mediaLink\b[^"]*"[^>]+|"videoLink\b[^"]*"[\s]*>\n[^\n]*)data-extension="([^"]+)"',
+            webpage, 'media link', default=None, flags=re.MULTILINE)
 
-        if mobj.group('player') is None:
+        if not json_metadata:
             entries = [
-                self.url_result(page_url + href, 'WDR')
+                self.url_result(page_url + href[0], 'WDR')
                 for href in re.findall(
-                    r'<a href="/?(.+?%s\.html)" rel="nofollow"' % self._PLAYER_REGEX,
+                    r'<a href="(%s)"[^>]+data-extension=' % self._PAGE_REGEX,
                     webpage)
             ]
 
             if entries:  # Playlist page
-                return self.playlist_result(entries, page_id)
+                return self.playlist_result(entries, playlist_id=display_id)
 
-            # Overview page
+            raise ExtractorError('No downloadable streams found', expected=True)
-            entries = []
-            for page_num in itertools.count(2):
-                hrefs = re.findall(
-                    r'<li class="mediathekvideo"\s*>\s*<img[^>]*>\s*<a href="(/mediathek/video/[^"]+)"',
-                    webpage)
-                entries.extend(
-                    self.url_result(page_url + href, 'WDR')
-                    for href in hrefs)
-                next_url_m = re.search(
-                    r'<li class="nextToLast">\s*<a href="([^"]+)"', webpage)
-                if not next_url_m:
-                    break
-                next_url = page_url + next_url_m.group(1)
-                webpage = self._download_webpage(
-                    next_url, page_id,
-                    note='Downloading playlist page %d' % page_num)
-            return self.playlist_result(entries, page_id)
 
-        flashvars = compat_parse_qs(self._html_search_regex(
+        media_link_obj = self._parse_json(json_metadata, display_id,
-            r'<param name="flashvars" value="([^"]+)"', webpage, 'flashvars'))
+                                          transform_source=js_to_json)
+        jsonp_url = media_link_obj['mediaObj']['url']
 
-        page_id = flashvars['trackerClipId'][0]
+        metadata = self._download_json(
-        video_url = flashvars['dslSrc'][0]
+            jsonp_url, 'metadata', transform_source=strip_jsonp)
-        title = flashvars['trackerClipTitle'][0]
+
-        thumbnail = flashvars['startPicture'][0] if 'startPicture' in flashvars else None
+        metadata_tracker_data = metadata['trackerData']
-        is_live = flashvars.get('isLive', ['0'])[0] == '1'
+        metadata_media_resource = metadata['mediaResource']
+
+        formats = []
+
+        # check if the metadata contains a direct URL to a file
+        metadata_media_alt = metadata_media_resource.get('alt')
+        if metadata_media_alt:
+            for tag_name in ['videoURL', 'audioURL']:
+                if tag_name in metadata_media_alt:
+                    alt_url = metadata_media_alt[tag_name]
+                    if determine_ext(alt_url) == 'm3u8':
+                        m3u_fmt = self._extract_m3u8_formats(
+                            alt_url, display_id, 'mp4', 'm3u8_native',
+                            m3u8_id='hls')
+                        formats.extend(m3u_fmt)
+                    else:
+                        formats.append({
+                            'url': alt_url
+                        })
+
+        # check if there are flash-streams for this video
+        if 'dflt' in metadata_media_resource and 'videoURL' in metadata_media_resource['dflt']:
+            video_url = metadata_media_resource['dflt']['videoURL']
+            if video_url.endswith('.f4m'):
+                full_video_url = video_url + '?hdcore=3.2.0&plugin=aasp-3.2.0.77.18'
+                formats.extend(self._extract_f4m_formats(full_video_url, display_id, f4m_id='hds', fatal=False))
+            elif video_url.endswith('.smil'):
+                formats.extend(self._extract_smil_formats(video_url, 'stream', fatal=False))
+
+        subtitles = {}
+        caption_url = metadata_media_resource.get('captionURL')
+        if caption_url:
+            subtitles['de'] = [{
+                'url': caption_url
+            }]
+
+        title = metadata_tracker_data.get('trackerClipTitle')
+        is_live = url_type == 'live'
 
         if is_live:
             title = self._live_title(title)
-
+            upload_date = None
-        if 'trackerClipAirTime' in flashvars:
+        elif 'trackerClipAirTime' in metadata_tracker_data:
-            upload_date = flashvars['trackerClipAirTime'][0]
+            upload_date = metadata_tracker_data['trackerClipAirTime']
         else:
-            upload_date = self._html_search_meta(
+            upload_date = self._html_search_meta('DC.Date', webpage, 'upload date')
-                'DC.Date', webpage, 'upload date')
 
         if upload_date:
             upload_date = unified_strdate(upload_date)
 
-        formats = []
-        preference = qualities(['S', 'M', 'L', 'XL'])
-
-        if video_url.endswith('.f4m'):
-            formats.extend(self._extract_f4m_formats(
-                video_url + '?hdcore=3.2.0&plugin=aasp-3.2.0.77.18', page_id,
-                f4m_id='hds', fatal=False))
-        elif video_url.endswith('.smil'):
-            formats.extend(self._extract_smil_formats(
-                video_url, page_id, False, {
-                    'hdcore': '3.3.0',
-                    'plugin': 'aasp-3.3.0.99.43',
-                }))
-        else:
-            formats.append({
-                'url': video_url,
-                'http_headers': {
-                    'User-Agent': 'mobile',
-                },
-            })
-
-        m3u8_url = self._search_regex(
-            r'rel="adaptiv"[^>]+href="([^"]+)"',
-            webpage, 'm3u8 url', default=None)
-        if m3u8_url:
-            formats.extend(self._extract_m3u8_formats(
-                m3u8_url, page_id, 'mp4', 'm3u8_native',
-                m3u8_id='hls', fatal=False))
-
-        direct_urls = re.findall(
-            r'rel="web(S|M|L|XL)"[^>]+href="([^"]+)"', webpage)
-        if direct_urls:
-            for quality, video_url in direct_urls:
-                formats.append({
-                    'url': video_url,
-                    'preference': preference(quality),
-                    'http_headers': {
-                        'User-Agent': 'mobile',
-                    },
-                })
-
         self._sort_formats(formats)
 
-        description = self._html_search_meta('Description', webpage, 'description')
-
         return {
-            'id': page_id,
+            'id': metadata_tracker_data.get('trackerClipId', display_id),
-            'formats': formats,
+            'display_id': display_id,
             'title': title,
-            'description': description,
+            'alt_title': metadata_tracker_data.get('trackerClipSubcategory'),
-            'thumbnail': thumbnail,
+            'formats': formats,
             'upload_date': upload_date,
-            'is_live': is_live
+            'description': self._html_search_meta('Description', webpage),
+            'is_live': is_live,
+            'subtitles': subtitles,
         }
 
 
@@ -241,81 +232,3 @@ class WDRMobileIE(InfoExtractor):
                 'User-Agent': 'mobile',
             },
         }
-
-
-class WDRMausIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?wdrmaus\.de/(?:[^/]+/){,2}(?P<id>[^/?#]+)(?:/index\.php5|(?<!index)\.php5|/(?:$|[?#]))'
-    IE_DESC = 'Sendung mit der Maus'
-    _TESTS = [{
-        'url': 'http://www.wdrmaus.de/aktuelle-sendung/index.php5',
-        'info_dict': {
-            'id': 'aktuelle-sendung',
-            'ext': 'mp4',
-            'thumbnail': 're:^http://.+\.jpg',
-            'upload_date': 're:^[0-9]{8}$',
-            'title': 're:^[0-9.]{10} - Aktuelle Sendung$',
-        }
-    }, {
-        'url': 'http://www.wdrmaus.de/sachgeschichten/sachgeschichten/40_jahre_maus.php5',
-        'md5': '3b1227ca3ed28d73ec5737c65743b2a3',
-        'info_dict': {
-            'id': '40_jahre_maus',
-            'ext': 'mp4',
-            'thumbnail': 're:^http://.+\.jpg',
-            'upload_date': '20131007',
-            'title': '12.03.2011 - 40 Jahre Maus',
-        }
-    }]
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-
-        webpage = self._download_webpage(url, video_id)
-        param_code = self._html_search_regex(
-            r'<a href="\?startVideo=1&amp;([^"]+)"', webpage, 'parameters')
-
-        title_date = self._search_regex(
-            r'<div class="sendedatum"><p>Sendedatum:\s*([0-9\.]+)</p>',
-            webpage, 'air date')
-        title_str = self._html_search_regex(
-            r'<h1>(.*?)</h1>', webpage, 'title')
-        title = '%s - %s' % (title_date, title_str)
-        upload_date = unified_strdate(
-            self._html_search_meta('dc.date', webpage))
-
-        fields = compat_parse_qs(param_code)
-        video_url = fields['firstVideo'][0]
-        thumbnail = compat_urlparse.urljoin(url, fields['startPicture'][0])
-
-        formats = [{
-            'format_id': 'rtmp',
-            'url': video_url,
-        }]
-
-        jscode = self._download_webpage(
-            'http://www.wdrmaus.de/codebase/js/extended-medien.min.js',
-            video_id, fatal=False,
-            note='Downloading URL translation table',
-            errnote='Could not download URL translation table')
-        if jscode:
-            for m in re.finditer(
-                    r"stream:\s*'dslSrc=(?P<stream>[^']+)',\s*download:\s*'(?P<dl>[^']+)'\s*\}",
-                    jscode):
-                if video_url.startswith(m.group('stream')):
-                    http_url = video_url.replace(
-                        m.group('stream'), m.group('dl'))
-                    formats.append({
-                        'format_id': 'http',
-                        'url': http_url,
-                    })
-                    break
-
-        self._sort_formats(formats)
-
-        return {
-            'id': video_id,
-            'title': title,
-            'formats': formats,
-            'thumbnail': thumbnail,
-            'upload_date': upload_date,
-        }