]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/spiegel.py
New upstream version 2018.04.25
[youtubedl] / youtube_dl / extractor / spiegel.py
index 1e55a9ffb5748b70969de11886c13720ff936be7..fc995e8c14da760dc33c706bfadba532bd86b05d 100644 (file)
@@ -1,15 +1,24 @@
-# encoding: utf-8
+# coding: utf-8
 from __future__ import unicode_literals
 
 import re
 
 from .common import InfoExtractor
-from ..compat import compat_urlparse
+from .nexx import (
+    NexxIE,
+    NexxEmbedIE,
+)
 from .spiegeltv import SpiegeltvIE
+from ..compat import compat_urlparse
+from ..utils import (
+    extract_attributes,
+    unified_strdate,
+    get_element_by_attribute,
+)
 
 
 class SpiegelIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P<id>[0-9]+)(?:-embed)?(?:\.html)?(?:#.*)?$'
+    _VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P<id>[0-9]+)(?:-embed|-iframe)?(?:\.html)?(?:#.*)?$'
     _TESTS = [{
         'url': 'http://www.spiegel.de/video/vulkan-tungurahua-in-ecuador-ist-wieder-aktiv-video-1259285.html',
         'md5': '2c2754212136f35fb4b19767d242f66e',
@@ -19,6 +28,7 @@ class SpiegelIE(InfoExtractor):
             'title': 'Vulkanausbruch in Ecuador: Der "Feuerschlund" ist wieder aktiv',
             'description': 'md5:8029d8310232196eb235d27575a8b9f4',
             'duration': 49,
+            'upload_date': '20130311',
         },
     }, {
         'url': 'http://www.spiegel.de/video/schach-wm-videoanalyse-des-fuenften-spiels-video-1309159.html',
@@ -29,6 +39,7 @@ class SpiegelIE(InfoExtractor):
             'title': 'Schach-WM in der Videoanalyse: Carlsen nutzt die Fehlgriffe des Titelverteidigers',
             'description': 'md5:c2322b65e58f385a820c10fa03b2d088',
             'duration': 983,
+            'upload_date': '20131115',
         },
     }, {
         'url': 'http://www.spiegel.de/video/astronaut-alexander-gerst-von-der-iss-station-beantwortet-fragen-video-1519126-embed.html',
@@ -38,7 +49,15 @@ class SpiegelIE(InfoExtractor):
             'ext': 'mp4',
             'description': 'SPIEGEL ONLINE-Nutzer durften den deutschen Astronauten Alexander Gerst über sein Leben auf der ISS-Station befragen. Hier kommen seine Antworten auf die besten sechs Fragen.',
             'title': 'Fragen an Astronaut Alexander Gerst: "Bekommen Sie die Tageszeiten mit?"',
+            'upload_date': '20140904',
         }
+    }, {
+        'url': 'http://www.spiegel.de/video/astronaut-alexander-gerst-von-der-iss-station-beantwortet-fragen-video-1519126-iframe.html',
+        'only_matching': True,
+    }, {
+        # nexx video
+        'url': 'http://www.spiegel.de/video/spiegel-tv-magazin-ueber-guellekrise-in-schleswig-holstein-video-99012776.html',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -49,47 +68,58 @@ class SpiegelIE(InfoExtractor):
         if SpiegeltvIE.suitable(handle.geturl()):
             return self.url_result(handle.geturl(), 'Spiegeltv')
 
-        title = re.sub(r'\s+', ' ', self._html_search_regex(
-            r'(?s)<(?:h1|div) class="module-title"[^>]*>(.*?)</(?:h1|div)>',
-            webpage, 'title'))
-        description = self._html_search_meta('description', webpage, 'description')
+        nexx_id = self._search_regex(
+            r'nexxOmniaId\s*:\s*(\d+)', webpage, 'nexx id', default=None)
+        if nexx_id:
+            domain_id = NexxIE._extract_domain_id(webpage) or '748'
+            return self.url_result(
+                'nexx:%s:%s' % (domain_id, nexx_id), ie=NexxIE.ie_key(),
+                video_id=nexx_id)
+
+        video_data = extract_attributes(self._search_regex(r'(<div[^>]+id="spVideoElements"[^>]+>)', webpage, 'video element', default=''))
+
+        title = video_data.get('data-video-title') or get_element_by_attribute('class', 'module-title', webpage)
+        description = video_data.get('data-video-teaser') or self._html_search_meta('description', webpage, 'description')
 
         base_url = self._search_regex(
-            r'var\s+server\s*=\s*"([^"]+)\"', webpage, 'server URL')
+            [r'server\s*:\s*(["\'])(?P<url>.+?)\1', r'var\s+server\s*=\s*"(?P<url>[^"]+)\"'],
+            webpage, 'server URL', group='url')
 
         xml_url = base_url + video_id + '.xml'
         idoc = self._download_xml(xml_url, video_id)
 
-        formats = [
-            {
-                'format_id': n.tag.rpartition('type')[2],
-                'url': base_url + n.find('./filename').text,
-                'width': int(n.find('./width').text),
-                'height': int(n.find('./height').text),
-                'abr': int(n.find('./audiobitrate').text),
-                'vbr': int(n.find('./videobitrate').text),
-                'vcodec': n.find('./codec').text,
-                'acodec': 'MP4A',
-            }
-            for n in list(idoc)
-            # Blacklist type 6, it's extremely LQ and not available on the same server
-            if n.tag.startswith('type') and n.tag != 'type6'
-        ]
+        formats = []
+        for n in list(idoc):
+            if n.tag.startswith('type') and n.tag != 'type6':
+                format_id = n.tag.rpartition('type')[2]
+                video_url = base_url + n.find('./filename').text
+                formats.append({
+                    'format_id': format_id,
+                    'url': video_url,
+                    'width': int(n.find('./width').text),
+                    'height': int(n.find('./height').text),
+                    'abr': int(n.find('./audiobitrate').text),
+                    'vbr': int(n.find('./videobitrate').text),
+                    'vcodec': n.find('./codec').text,
+                    'acodec': 'MP4A',
+                })
         duration = float(idoc[0].findall('./duration')[0].text)
 
+        self._check_formats(formats, video_id)
         self._sort_formats(formats)
 
         return {
             'id': video_id,
             'title': title,
-            'description': description,
+            'description': description.strip() if description else None,
             'duration': duration,
+            'upload_date': unified_strdate(video_data.get('data-video-date')),
             'formats': formats,
         }
 
 
 class SpiegelArticleIE(InfoExtractor):
-    _VALID_URL = 'https?://www\.spiegel\.de/(?!video/)[^?#]*?-(?P<id>[0-9]+)\.html'
+    _VALID_URL = r'https?://(?:www\.)?spiegel\.de/(?!video/)[^?#]*?-(?P<id>[0-9]+)\.html'
     IE_NAME = 'Spiegel:Article'
     IE_DESC = 'Articles on spiegel.de'
     _TESTS = [{
@@ -99,6 +129,7 @@ class SpiegelArticleIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Faszination Badminton: Nennt es bloß nicht Federball',
             'description': 're:^Patrick Kämnitz gehört.{100,}',
+            'upload_date': '20140825',
         },
     }, {
         'url': 'http://www.spiegel.de/wissenschaft/weltall/astronaut-alexander-gerst-antwortet-spiegel-online-lesern-a-989876.html',
@@ -106,6 +137,26 @@ class SpiegelArticleIE(InfoExtractor):
 
         },
         'playlist_count': 6,
+    }, {
+        # Nexx iFrame embed
+        'url': 'http://www.spiegel.de/sptv/spiegeltv/spiegel-tv-ueber-schnellste-katapult-achterbahn-der-welt-taron-a-1137884.html',
+        'info_dict': {
+            'id': '161464',
+            'ext': 'mp4',
+            'title': 'Nervenkitzel Achterbahn',
+            'alt_title': 'Karussellbauer in Deutschland',
+            'description': 'md5:ffe7b1cc59a01f585e0569949aef73cc',
+            'release_year': 2005,
+            'creator': 'SPIEGEL TV',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 2761,
+            'timestamp': 1394021479,
+            'upload_date': '20140305',
+        },
+        'params': {
+            'format': 'bestvideo',
+            'skip_download': True,
+        },
     }]
 
     def _real_extract(self, url):
@@ -128,6 +179,9 @@ class SpiegelArticleIE(InfoExtractor):
         entries = [
             self.url_result(compat_urlparse.urljoin(
                 self.http_scheme() + '//spiegel.de/', embed_path))
-            for embed_path in embeds
-        ]
-        return self.playlist_result(entries)
+            for embed_path in embeds]
+        if embeds:
+            return self.playlist_result(entries)
+
+        return self.playlist_from_matches(
+            NexxEmbedIE._extract_urls(webpage), ie=NexxEmbedIE.ie_key())