]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/wdr.py
Start new release.
[youtubedl] / youtube_dl / extractor / wdr.py
index 45466e31b7445f8dd8da742308dcc69f2ff1152f..cf6f7c7ed6ab5bce442a6cb2079baf687c324417 100644 (file)
-# -*- coding: utf-8 -*-
+# coding: utf-8
 from __future__ import unicode_literals
 
 from __future__ import unicode_literals
 
-import itertools
 import re
 
 from .common import InfoExtractor
 from ..compat import (
 import re
 
 from .common import InfoExtractor
 from ..compat import (
-    compat_parse_qs,
+    compat_str,
     compat_urlparse,
 )
 from ..utils import (
     determine_ext,
     compat_urlparse,
 )
 from ..utils import (
     determine_ext,
+    ExtractorError,
+    js_to_json,
+    strip_jsonp,
+    try_get,
     unified_strdate,
     unified_strdate,
+    update_url_query,
+    urlhandle_detect_ext,
 )
 
 
 class WDRIE(InfoExtractor):
 )
 
 
 class WDRIE(InfoExtractor):
-    _PLAYER_REGEX = '-(?:video|audio)player(?:_size-[LMS])?'
-    _VALID_URL = r'(?P<url>https?://www\d?\.(?:wdr\d?|funkhauseuropa)\.de/)(?P<id>.+?)(?P<player>%s)?\.html' % _PLAYER_REGEX
+    _VALID_URL = r'https?://deviceids-medp\.wdr\.de/ondemand/\d+/(?P<id>\d+)\.js'
+    _GEO_COUNTRIES = ['DE']
+    _TEST = {
+        'url': 'http://deviceids-medp.wdr.de/ondemand/155/1557833.js',
+        'info_dict': {
+            'id': 'mdb-1557833',
+            'ext': 'mp4',
+            'title': 'Biathlon-Staffel verpasst Podest bei Olympia-Generalprobe',
+            'upload_date': '20180112',
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        metadata = self._download_json(
+            url, video_id, transform_source=strip_jsonp)
+
+        is_live = metadata.get('mediaType') == 'live'
+
+        tracker_data = metadata['trackerData']
+        media_resource = metadata['mediaResource']
+
+        formats = []
+
+        # check if the metadata contains a direct URL to a file
+        for kind, media_resource in media_resource.items():
+            if kind not in ('dflt', 'alt'):
+                continue
+
+            for tag_name, medium_url in media_resource.items():
+                if tag_name not in ('videoURL', 'audioURL'):
+                    continue
+
+                ext = determine_ext(medium_url)
+                if ext == 'm3u8':
+                    formats.extend(self._extract_m3u8_formats(
+                        medium_url, video_id, 'mp4', 'm3u8_native',
+                        m3u8_id='hls'))
+                elif ext == 'f4m':
+                    manifest_url = update_url_query(
+                        medium_url, {'hdcore': '3.2.0', 'plugin': 'aasp-3.2.0.77.18'})
+                    formats.extend(self._extract_f4m_formats(
+                        manifest_url, video_id, f4m_id='hds', fatal=False))
+                elif ext == 'smil':
+                    formats.extend(self._extract_smil_formats(
+                        medium_url, 'stream', fatal=False))
+                else:
+                    a_format = {
+                        'url': medium_url
+                    }
+                    if ext == 'unknown_video':
+                        urlh = self._request_webpage(
+                            medium_url, video_id, note='Determining extension')
+                        ext = urlhandle_detect_ext(urlh)
+                        a_format['ext'] = ext
+                    formats.append(a_format)
+
+        self._sort_formats(formats)
+
+        subtitles = {}
+        caption_url = media_resource.get('captionURL')
+        if caption_url:
+            subtitles['de'] = [{
+                'url': caption_url,
+                'ext': 'ttml',
+            }]
+
+        title = tracker_data['trackerClipTitle']
+
+        return {
+            'id': tracker_data.get('trackerClipId', video_id),
+            'title': self._live_title(title) if is_live else title,
+            'alt_title': tracker_data.get('trackerClipSubcategory'),
+            'formats': formats,
+            'subtitles': subtitles,
+            'upload_date': unified_strdate(tracker_data.get('trackerClipAirTime')),
+            'is_live': is_live,
+        }
+
+
+class WDRPageIE(InfoExtractor):
+    _CURRENT_MAUS_URL = r'https?://(?:www\.)wdrmaus.de/(?:[^/]+/){1,2}[^/?#]+\.php5'
+    _PAGE_REGEX = r'/(?:mediathek/)?(?:[^/]+/)*(?P<display_id>[^/]+)\.html'
+    _VALID_URL = r'https?://(?:www\d?\.)?(?:wdr\d?|sportschau)\.de' + _PAGE_REGEX + '|' + _CURRENT_MAUS_URL
 
     _TESTS = [
         {
 
     _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',
+            # HDS download, MD5 is unstable
             'info_dict': {
             'info_dict': {
-                'id': 'mdb-362427',
+                'id': 'mdb-1058683',
                 'ext': 'flv',
                 'ext': 'flv',
-                'title': 'Servicezeit',
-                'description': 'md5:c8f43e5e815eeb54d0b96df2fba906cb',
-                'upload_date': '20140310',
+                'display_id': 'doku-am-freitag/video-geheimnis-aachener-dom-100',
+                'title': 'Geheimnis Aachener Dom',
+                'alt_title': 'Doku am Freitag',
+                'upload_date': '20160304',
+                'description': 'md5:87be8ff14d8dfd7a7ee46f0299b52318',
+                'is_live': False,
+                'subtitles': {'de': [{
+                    'url': 'http://ondemand-ww.wdr.de/medp/fsk0/105/1058683/1058683_12220974.xml',
+                    'ext': 'ttml',
+                }]},
             },
             },
-            'params': {
-                'skip_download': True,
+            'skip': 'HTTP Error 404: Not Found',
+        },
+        {
+            'url': 'http://www1.wdr.de/mediathek/audio/wdr3/wdr3-gespraech-am-samstag/audio-schriftstellerin-juli-zeh-100.html',
+            'md5': 'f4c1f96d01cf285240f53ea4309663d8',
+            'info_dict': {
+                'id': 'mdb-1072000',
+                'ext': 'mp3',
+                'display_id': 'wdr3-gespraech-am-samstag/audio-schriftstellerin-juli-zeh-100',
+                'title': 'Schriftstellerin Juli Zeh',
+                'alt_title': 'WDR 3 Gespräch am Samstag',
+                'upload_date': '20160312',
+                'description': 'md5:e127d320bc2b1f149be697ce044a3dd7',
+                'is_live': False,
+                'subtitles': {}
             },
             },
+            'skip': 'HTTP Error 404: Not Found',
         },
         {
         },
         {
-            'url': 'http://www1.wdr.de/themen/av/videomargaspiegelisttot101-videoplayer.html',
+            'url': 'http://www1.wdr.de/mediathek/video/live/index.html',
             'info_dict': {
             'info_dict': {
-                'id': 'mdb-363194',
-                'ext': 'flv',
-                'title': 'Marga Spiegel ist tot',
-                'description': 'md5:2309992a6716c347891c045be50992e4',
-                'upload_date': '20140311',
+                'id': 'mdb-1406149',
+                'ext': 'mp4',
+                'title': r're:^WDR Fernsehen im Livestream \(nur in Deutschland erreichbar\) [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
+                'alt_title': 'WDR Fernsehen Live',
+                'upload_date': '20150101',
+                'is_live': True,
             },
             'params': {
             },
             'params': {
-                'skip_download': True,
+                'skip_download': True,  # m3u8 download
             },
         },
         {
             },
         },
         {
-            'url': 'http://www1.wdr.de/themen/kultur/audioerlebtegeschichtenmargaspiegel100-audioplayer.html',
-            'md5': '83e9e8fefad36f357278759870805898',
+            'url': 'http://www1.wdr.de/mediathek/video/sendungen/aktuelle-stunde/aktuelle-stunde-120.html',
+            'playlist_mincount': 7,
             'info_dict': {
             'info_dict': {
-                'id': 'mdb-194332',
-                'ext': 'mp3',
-                'title': 'Erlebte Geschichten: Marga Spiegel (29.11.2009)',
-                'description': 'md5:2309992a6716c347891c045be50992e4',
-                'upload_date': '20091129',
+                'id': 'aktuelle-stunde-120',
             },
         },
         {
             },
         },
         {
-            'url': 'http://www.funkhauseuropa.de/av/audioflaviacoelhoamaramar100-audioplayer.html',
-            'md5': '99a1443ff29af19f6c52cf6f4dc1f4aa',
+            'url': 'http://www.wdrmaus.de/aktuelle-sendung/index.php5',
             'info_dict': {
             'info_dict': {
-                'id': 'mdb-478135',
-                'ext': 'mp3',
-                'title': 'Flavia Coelho: Amar é Amar',
-                'description': 'md5:7b29e97e10dfb6e265238b32fa35b23a',
-                'upload_date': '20140717',
+                'id': 'mdb-1552552',
+                'ext': 'mp4',
+                'upload_date': 're:^[0-9]{8}$',
+                'title': 're:^Die Sendung mit der Maus vom [0-9.]{10}$',
+            },
+            'skip': 'The id changes from week to week because of the new episode'
+        },
+        {
+            'url': 'http://www.wdrmaus.de/filme/sachgeschichten/achterbahn.php5',
+            'md5': '803138901f6368ee497b4d195bb164f2',
+            'info_dict': {
+                'id': 'mdb-186083',
+                'ext': 'mp4',
+                'upload_date': '20130919',
+                'title': 'Sachgeschichte - Achterbahn ',
+            },
+        },
+        {
+            'url': 'http://www1.wdr.de/radio/player/radioplayer116~_layout-popupVersion.html',
+            # Live stream, MD5 unstable
+            'info_dict': {
+                'id': 'mdb-869971',
+                'ext': 'mp4',
+                'title': r're:^COSMO Livestream [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
+                'upload_date': '20160101',
             },
             },
+            'params': {
+                'skip_download': True,  # m3u8 download
+            }
         },
         {
         },
         {
-            'url': 'http://www1.wdr.de/mediathek/video/sendungen/quarks_und_co/filterseite-quarks-und-co100.html',
-            'playlist_mincount': 146,
+            'url': 'http://www.sportschau.de/handballem2018/handball-nationalmannschaft-em-stolperstein-vorrunde-100.html',
+            'info_dict': {
+                'id': 'mdb-1556012',
+                'ext': 'mp4',
+                'title': 'DHB-Vizepräsident Bob Hanning - "Die Weltspitze ist extrem breit"',
+                'upload_date': '20180111',
+            },
+            'params': {
+                'skip_download': True,
+            },
+        },
+        {
+            'url': 'http://www.sportschau.de/handballem2018/audio-vorschau---die-handball-em-startet-mit-grossem-favoritenfeld-100.html',
+            'only_matching': True,
         }
     ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         }
     ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        page_url = mobj.group('url')
-        page_id = mobj.group('id')
+        display_id = mobj.group('display_id')
+        webpage = self._download_webpage(url, display_id)
+
+        entries = []
+
+        # Article with several videos
 
 
-        webpage = self._download_webpage(url, page_id)
+        # for wdr.de the data-extension is in a tag with the class "mediaLink"
+        # for wdr.de radio players, in a tag with the class "wdrrPlayerPlayBtn"
+        # for wdrmaus, in a tag with the class "videoButton" (previously a link
+        # to the page in a multiline "videoLink"-tag)
+        for mobj in re.finditer(
+            r'''(?sx)class=
+                    (?:
+                        (["\'])(?:mediaLink|wdrrPlayerPlayBtn|videoButton)\b.*?\1[^>]+|
+                        (["\'])videoLink\b.*?\2[\s]*>\n[^\n]*
+                    )data-extension=(["\'])(?P<data>(?:(?!\3).)+)\3
+                    ''', webpage):
+            media_link_obj = self._parse_json(
+                mobj.group('data'), display_id, transform_source=js_to_json,
+                fatal=False)
+            if not media_link_obj:
+                continue
+            jsonp_url = try_get(
+                media_link_obj, lambda x: x['mediaObj']['url'], compat_str)
+            if jsonp_url:
+                entries.append(self.url_result(jsonp_url, ie=WDRIE.ie_key()))
 
 
-        if mobj.group('player') is None:
+        # Playlist (e.g. https://www1.wdr.de/mediathek/video/sendungen/aktuelle-stunde/aktuelle-stunde-120.html)
+        if not entries:
             entries = [
             entries = [
-                self.url_result(page_url + href, 'WDR')
-                for href in re.findall(r'<a href="/?(.+?%s\.html)" rel="nofollow"' % self._PLAYER_REGEX, webpage)
+                self.url_result(
+                    compat_urlparse.urljoin(url, mobj.group('href')),
+                    ie=WDRPageIE.ie_key())
+                for mobj in re.finditer(
+                    r'<a[^>]+\bhref=(["\'])(?P<href>(?:(?!\1).)+)\1[^>]+\bdata-extension=',
+                    webpage) if re.match(self._PAGE_REGEX, mobj.group('href'))
             ]
 
             ]
 
-            if entries:  # Playlist page
-                return self.playlist_result(entries, page_id)
-
-            # Overview page
-            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(r'<param name="flashvars" value="([^"]+)"', webpage, 'flashvars'))
-
-        page_id = flashvars['trackerClipId'][0]
-        video_url = flashvars['dslSrc'][0]
-        title = flashvars['trackerClipTitle'][0]
-        thumbnail = flashvars['startPicture'][0] if 'startPicture' in flashvars else None
-
-        if 'trackerClipAirTime' in flashvars:
-            upload_date = flashvars['trackerClipAirTime'][0]
-        else:
-            upload_date = self._html_search_meta('DC.Date', webpage, 'upload date')
-
-        if upload_date:
-            upload_date = unified_strdate(upload_date)
-
-        if video_url.endswith('.f4m'):
-            video_url += '?hdcore=3.2.0&plugin=aasp-3.2.0.77.18'
-            ext = 'flv'
-        else:
-            ext = determine_ext(video_url)
-
-        description = self._html_search_meta('Description', webpage, 'description')
+        return self.playlist_result(entries, playlist_id=display_id)
 
 
-        return {
-            'id': page_id,
-            'url': video_url,
-            'ext': ext,
-            'title': title,
-            'description': description,
-            'thumbnail': thumbnail,
-            'upload_date': upload_date,
-        }
+
+class WDRElefantIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)wdrmaus\.de/elefantenseite/#(?P<id>.+)'
+    _TEST = {
+        'url': 'http://www.wdrmaus.de/elefantenseite/#folge_ostern_2015',
+        'info_dict': {
+            'title': 'Folge Oster-Spezial 2015',
+            'id': 'mdb-1088195',
+            'ext': 'mp4',
+            'age_limit': None,
+            'upload_date': '20150406'
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+
+        # Table of Contents seems to always be at this address, so fetch it directly.
+        # The website fetches configurationJS.php5, which links to tableOfContentsJS.php5.
+        table_of_contents = self._download_json(
+            'https://www.wdrmaus.de/elefantenseite/data/tableOfContentsJS.php5',
+            display_id)
+        if display_id not in table_of_contents:
+            raise ExtractorError(
+                'No entry in site\'s table of contents for this URL. '
+                'Is the fragment part of the URL (after the #) correct?',
+                expected=True)
+        xml_metadata_path = table_of_contents[display_id]['xmlPath']
+        xml_metadata = self._download_xml(
+            'https://www.wdrmaus.de/elefantenseite/' + xml_metadata_path,
+            display_id)
+        zmdb_url_element = xml_metadata.find('./movie/zmdb_url')
+        if zmdb_url_element is None:
+            raise ExtractorError(
+                '%s is not a video' % display_id, expected=True)
+        return self.url_result(zmdb_url_element.text, ie=WDRIE.ie_key())
 
 
 class WDRMobileIE(InfoExtractor):
 
 
 class WDRMobileIE(InfoExtractor):
@@ -169,83 +324,7 @@ class WDRMobileIE(InfoExtractor):
             'title': mobj.group('title'),
             'age_limit': int(mobj.group('age_limit')),
             'url': url,
             'title': mobj.group('title'),
             'age_limit': int(mobj.group('age_limit')),
             'url': url,
-            'user_agent': 'mobile',
-        }
-
-
-class WDRMausIE(InfoExtractor):
-    _VALID_URL = 'http://(?: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,
+            'http_headers': {
+                'User-Agent': 'mobile',
+            },
         }
         }