Update upstream source from tag 'upstream/2019.09.28'

[youtubedl] / youtube_dl / extractor / mdr.py

diff --git a/youtube_dl/extractor/mdr.py b/youtube_dl/extractor/mdr.py
index 5fdd19027db3ccad0265601b8d88452a0eaac525..322e5b45a141bacd327288ac58ed658a46e628f7 100644 (file)
--- a/youtube_dl/extractor/mdr.py
+++ b/youtube_dl/extractor/mdr.py
@@ -1,64 +1,184 @@
+# coding: utf-8

 from __future__ import unicode_literals
 
 from __future__ import unicode_literals
 

-import re
-

 from .common import InfoExtractor
 from .common import InfoExtractor

+from ..compat import compat_urlparse
+from ..utils import (
+    determine_ext,
+    int_or_none,
+    parse_duration,
+    parse_iso8601,
+    xpath_text,
+)

 
 
 class MDRIE(InfoExtractor):
 
 
 class MDRIE(InfoExtractor):

-    _VALID_URL = r'^(?P<domain>https?://(?:www\.)?mdr\.de)/(?:.*)/(?P<type>video|audio)(?P<video_id>[^/_]+)(?:_|\.html)'
+    IE_DESC = 'MDR.DE and KiKA'
+    _VALID_URL = r'https?://(?:www\.)?(?:mdr|kika)\.de/(?:.*)/[a-z-]+-?(?P<id>\d+)(?:_.+?)?\.html'

 
 

-    # No tests, MDR regularily deletes its videos
-    _TEST = {
+    _TESTS = [{
+        # MDR regularly deletes its videos

         'url': 'http://www.mdr.de/fakt/video189002.html',
         'only_matching': True,
         'url': 'http://www.mdr.de/fakt/video189002.html',
         'only_matching': True,

-    }
+    }, {
+        # audio
+        'url': 'http://www.mdr.de/kultur/audio1312272_zc-15948bad_zs-86171fdd.html',
+        'md5': '64c4ee50f0a791deb9479cd7bbe9d2fa',
+        'info_dict': {
+            'id': '1312272',
+            'ext': 'mp3',
+            'title': 'Feuilleton vom 30. Oktober 2015',
+            'duration': 250,
+            'uploader': 'MITTELDEUTSCHER RUNDFUNK',
+        },
+        'skip': '404 not found',
+    }, {
+        'url': 'http://www.kika.de/baumhaus/videos/video19636.html',
+        'md5': '4930515e36b06c111213e80d1e4aad0e',
+        'info_dict': {
+            'id': '19636',
+            'ext': 'mp4',
+            'title': 'Baumhaus vom 30. Oktober 2015',
+            'duration': 134,
+            'uploader': 'KIKA',
+        },
+        'skip': '404 not found',
+    }, {
+        'url': 'http://www.kika.de/sendungen/einzelsendungen/weihnachtsprogramm/videos/video8182.html',
+        'md5': '5fe9c4dd7d71e3b238f04b8fdd588357',
+        'info_dict': {
+            'id': '8182',
+            'ext': 'mp4',
+            'title': 'Beutolomäus und der geheime Weihnachtswunsch',
+            'description': 'md5:b69d32d7b2c55cbe86945ab309d39bbd',
+            'timestamp': 1482541200,
+            'upload_date': '20161224',
+            'duration': 4628,
+            'uploader': 'KIKA',
+        },
+    }, {
+        # audio with alternative playerURL pattern
+        'url': 'http://www.mdr.de/kultur/videos-und-audios/audio-radio/operation-mindfuck-robert-wilson100.html',
+        'info_dict': {
+            'id': '100',
+            'ext': 'mp4',
+            'title': 'Feature: Operation Mindfuck - Robert Anton Wilson',
+            'duration': 3239,
+            'uploader': 'MITTELDEUTSCHER RUNDFUNK',
+        },
+    }, {
+        'url': 'http://www.kika.de/baumhaus/sendungen/video19636_zc-fea7f8a0_zs-4bf89c60.html',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.kika.de/sendungen/einzelsendungen/weihnachtsprogramm/einzelsendung2534.html',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.mdr.de/mediathek/mdr-videos/a/video-1334.html',
+        'only_matching': True,
+    }]

 
     def _real_extract(self, url):
 
     def _real_extract(self, url):

-        m = re.match(self._VALID_URL, url)
-        video_id = m.group('video_id')
-        domain = m.group('domain')
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)

 
 

-        # determine title and media streams from webpage
-        html = self._download_webpage(url, video_id)
+        data_url = self._search_regex(
+            r'(?:dataURL|playerXml(?:["\'])?)\s*:\s*(["\'])(?P<url>.+?-avCustom\.xml)\1',
+            webpage, 'data url', group='url').replace(r'\/', '/')

 
 

-        title = self._html_search_regex(r'<h[12]>(.*?)</h[12]>', html, 'title')
-        xmlurl = self._search_regex(
-            r'dataURL:\'(/(?:.+)/(?:video|audio)[0-9]+-avCustom.xml)', html, 'XML URL')
+        doc = self._download_xml(
+            compat_urlparse.urljoin(url, data_url), video_id)
+
+        title = xpath_text(doc, ['./title', './broadcast/broadcastName'], 'title', fatal=True)

 
 

-        doc = self._download_xml(domain + xmlurl, video_id)

         formats = []
         formats = []

-        for a in doc.findall('./assets/asset'):
-            url_el = a.find('.//progressiveDownloadUrl')
-            if url_el is None:
-                continue
-            abr = int(a.find('bitrateAudio').text) // 1000
-            media_type = a.find('mediaType').text
-            format = {
-                'abr': abr,
-                'filesize': int(a.find('fileSize').text),
-                'url': url_el.text,
-            }
-
-            vbr_el = a.find('bitrateVideo')
-            if vbr_el is None:
-                format.update({
-                    'vcodec': 'none',
-                    'format_id': '%s-%d' % (media_type, abr),
-                })
-            else:
-                vbr = int(vbr_el.text) // 1000
-                format.update({
-                    'vbr': vbr,
-                    'width': int(a.find('frameWidth').text),
-                    'height': int(a.find('frameHeight').text),
-                    'format_id': '%s-%d' % (media_type, vbr),
-                })
-            formats.append(format)
+        processed_urls = []
+        for asset in doc.findall('./assets/asset'):
+            for source in (
+                    'progressiveDownload',
+                    'dynamicHttpStreamingRedirector',
+                    'adaptiveHttpStreamingRedirector'):
+                url_el = asset.find('./%sUrl' % source)
+                if url_el is None:
+                    continue
+
+                video_url = url_el.text
+                if video_url in processed_urls:
+                    continue
+
+                processed_urls.append(video_url)
+
+                vbr = int_or_none(xpath_text(asset, './bitrateVideo', 'vbr'), 1000)
+                abr = int_or_none(xpath_text(asset, './bitrateAudio', 'abr'), 1000)
+
+                ext = determine_ext(url_el.text)
+                if ext == 'm3u8':
+                    url_formats = self._extract_m3u8_formats(
+                        video_url, video_id, 'mp4', entry_protocol='m3u8_native',
+                        preference=0, m3u8_id='HLS', fatal=False)
+                elif ext == 'f4m':
+                    url_formats = self._extract_f4m_formats(
+                        video_url + '?hdcore=3.7.0&plugin=aasp-3.7.0.39.44', video_id,
+                        preference=0, f4m_id='HDS', fatal=False)
+                else:
+                    media_type = xpath_text(asset, './mediaType', 'media type', default='MP4')
+                    vbr = int_or_none(xpath_text(asset, './bitrateVideo', 'vbr'), 1000)
+                    abr = int_or_none(xpath_text(asset, './bitrateAudio', 'abr'), 1000)
+                    filesize = int_or_none(xpath_text(asset, './fileSize', 'file size'))
+
+                    f = {
+                        'url': video_url,
+                        'format_id': '%s-%d' % (media_type, vbr or abr),
+                        'filesize': filesize,
+                        'abr': abr,
+                        'preference': 1,
+                    }
+
+                    if vbr:
+                        width = int_or_none(xpath_text(asset, './frameWidth', 'width'))
+                        height = int_or_none(xpath_text(asset, './frameHeight', 'height'))
+                        f.update({
+                            'vbr': vbr,
+                            'width': width,
+                            'height': height,
+                        })
+
+                    url_formats = [f]
+
+                if not url_formats:
+                    continue
+
+                if not vbr:
+                    for f in url_formats:
+                        abr = f.get('tbr') or abr
+                        if 'tbr' in f:
+                            del f['tbr']
+                        f.update({
+                            'abr': abr,
+                            'vcodec': 'none',
+                        })
+
+                formats.extend(url_formats)
+

         self._sort_formats(formats)
 
         self._sort_formats(formats)
 

+        description = xpath_text(doc, './broadcast/broadcastDescription', 'description')
+        timestamp = parse_iso8601(
+            xpath_text(
+                doc, [
+                    './broadcast/broadcastDate',
+                    './broadcast/broadcastStartDate',
+                    './broadcast/broadcastEndDate'],
+                'timestamp', default=None))
+        duration = parse_duration(xpath_text(doc, './duration', 'duration'))
+        uploader = xpath_text(doc, './rights', 'uploader')
+

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

+            'description': description,
+            'timestamp': timestamp,
+            'duration': duration,
+            'uploader': uploader,

             'formats': formats,
         }
             'formats': formats,
         }