]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/heise.py
Rapsys Git / youtubedl / blobdiff
? search:
summary | shortlog | log | commit | commitdiff | tree
raw | inline | side by side
debian/README.source: Update steps used to create new release.
[youtubedl] / youtube_dl / extractor / heise.py
diff --git a/youtube_dl/extractor/heise.py b/youtube_dl/extractor/heise.py
index d41c0413f015e262237f66161d4021c32f56ff0e..5c03780a3389fa8272663b805855c305bf934689 100644 (file)
--- a/youtube_dl/extractor/heise.py
+++ b/youtube_dl/extractor/heise.py
@@ -2,74 +2,162 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from .kaltura import KalturaIE
+from .youtube import YoutubeIE
 from ..utils import (
 from ..utils import (
-    get_meta_content,
+    determine_ext,
     int_or_none,
     int_or_none,
+    NO_DEFAULT,
     parse_iso8601,
     parse_iso8601,
+    smuggle_url,
+    xpath_text,
 )
 
 
 class HeiseIE(InfoExtractor):
 )
 
 
 class HeiseIE(InfoExtractor):
-    _VALID_URL = r'''(?x)
-        https?://(?:www\.)?heise\.de/video/artikel/
-        .+?(?P<id>[0-9]+)\.html(?:$|[?#])
-    '''
-    _TEST = {
-        'url': (
-            'http://www.heise.de/video/artikel/Podcast-c-t-uplink-3-3-Owncloud-Tastaturen-Peilsender-Smartphone-2404147.html'
-        ),
-        'md5': 'ffed432483e922e88545ad9f2f15d30e',
+    _VALID_URL = r'https?://(?:www\.)?heise\.de/(?:[^/]+/)+[^/]+-(?P<id>[0-9]+)\.html'
+    _TESTS = [{
+        # kaltura embed
+        'url': 'http://www.heise.de/video/artikel/Podcast-c-t-uplink-3-3-Owncloud-Tastaturen-Peilsender-Smartphone-2404147.html',
         'info_dict': {
         'info_dict': {
-            'id': '2404147',
+            'id': '1_kkrq94sm',
             'ext': 'mp4',
             'ext': 'mp4',
-            'title': (
-                "Podcast: c't uplink 3.3 – Owncloud / Tastaturen / Peilsender Smartphone"
-            ),
-            'format_id': 'mp4_720',
-            'timestamp': 1411812600,
-            'upload_date': '20140927',
-            'description': 'In uplink-Episode 3.3 geht es darum, wie man sich von Cloud-Anbietern emanzipieren kann, worauf man beim Kauf einer Tastatur achten sollte und was Smartphones über uns verraten.',
-            'thumbnail': 're:https?://.*\.jpg$',
-        }
-    }
+            'title': "Podcast: c't uplink 3.3 – Owncloud / Tastaturen / Peilsender Smartphone",
+            'timestamp': 1512734959,
+            'upload_date': '20171208',
+            'description': 'md5:c934cbfb326c669c2bcabcbe3d3fcd20',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        # YouTube embed
+        'url': 'http://www.heise.de/newsticker/meldung/Netflix-In-20-Jahren-vom-Videoverleih-zum-TV-Revolutionaer-3814130.html',
+        'md5': 'e403d2b43fea8e405e88e3f8623909f1',
+        'info_dict': {
+            'id': '6kmWbXleKW4',
+            'ext': 'mp4',
+            'title': 'NEU IM SEPTEMBER | Netflix',
+            'description': 'md5:2131f3c7525e540d5fd841de938bd452',
+            'upload_date': '20170830',
+            'uploader': 'Netflix Deutschland, Österreich und Schweiz',
+            'uploader_id': 'netflixdach',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://www.heise.de/video/artikel/nachgehakt-Wie-sichert-das-c-t-Tool-Restric-tor-Windows-10-ab-3700244.html',
+        'info_dict': {
+            'id': '1_ntrmio2s',
+            'ext': 'mp4',
+            'title': "nachgehakt: Wie sichert das c't-Tool Restric'tor Windows 10 ab?",
+            'description': 'md5:47e8ffb6c46d85c92c310a512d6db271',
+            'timestamp': 1512470717,
+            'upload_date': '20171205',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://www.heise.de/ct/artikel/c-t-uplink-20-8-Staubsaugerroboter-Xiaomi-Vacuum-2-AR-Brille-Meta-2-und-Android-rooten-3959893.html',
+        'info_dict': {
+            'id': '1_59mk80sf',
+            'ext': 'mp4',
+            'title': "c't uplink 20.8: Staubsaugerroboter Xiaomi Vacuum 2, AR-Brille Meta 2 und Android rooten",
+            'description': 'md5:f50fe044d3371ec73a8f79fcebd74afc',
+            'timestamp': 1517567237,
+            'upload_date': '20180202',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://www.heise.de/ct/artikel/c-t-uplink-3-3-Owncloud-Tastaturen-Peilsender-Smartphone-2403911.html',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.heise.de/newsticker/meldung/c-t-uplink-Owncloud-Tastaturen-Peilsender-Smartphone-2404251.html?wt_mc=rss.ho.beitrag.atom',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.heise.de/ct/ausgabe/2016-12-Spiele-3214137.html',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
+        def extract_title(default=NO_DEFAULT):
+            title = self._html_search_meta(
+                ('fulltitle', 'title'), webpage, default=None)
+            if not title or title == "c't":
+                title = self._search_regex(
+                    r'<div[^>]+class="videoplayerjw"[^>]+data-title="([^"]+)"',
+                    webpage, 'title', default=None)
+            if not title:
+                title = self._html_search_regex(
+                    r'<h1[^>]+\bclass=["\']article_page_title[^>]+>(.+?)<',
+                    webpage, 'title', default=default)
+            return title
+
+        title = extract_title(default=None)
+        description = self._og_search_description(
+            webpage, default=None) or self._html_search_meta(
+            'description', webpage)
+
+        kaltura_url = KalturaIE._extract_url(webpage)
+        if kaltura_url:
+            return {
+                '_type': 'url_transparent',
+                'url': smuggle_url(kaltura_url, {'source_url': url}),
+                'ie_key': KalturaIE.ie_key(),
+                'title': title,
+                'description': description,
+            }
+
+        yt_urls = YoutubeIE._extract_urls(webpage)
+        if yt_urls:
+            return self.playlist_from_matches(
+                yt_urls, video_id, title, ie=YoutubeIE.ie_key())
+
+        title = extract_title()
+
         container_id = self._search_regex(
         container_id = self._search_regex(
-            r'<div class="videoplayerjw".*?data-container="([0-9]+)"',
+            r'<div class="videoplayerjw"[^>]+data-container="([0-9]+)"',
             webpage, 'container ID')
             webpage, 'container ID')
+
         sequenz_id = self._search_regex(
         sequenz_id = self._search_regex(
-            r'<div class="videoplayerjw".*?data-sequenz="([0-9]+)"',
+            r'<div class="videoplayerjw"[^>]+data-sequenz="([0-9]+)"',
             webpage, 'sequenz ID')
             webpage, 'sequenz ID')
-        data_url = 'http://www.heise.de/videout/feed?container=%s&sequenz=%s' % (container_id, sequenz_id)
-        doc = self._download_xml(data_url, video_id)
 
 
-        info = {
-            'id': video_id,
-            'thumbnail': self._og_search_thumbnail(webpage),
-            'timestamp': parse_iso8601(get_meta_content('date', webpage)),
-            'description': self._og_search_description(webpage),
-        }
-
-        title = get_meta_content('fulltitle', webpage)
-        if title:
-            info['title'] = title
-        else:
-            info['title'] = self._og_search_title(webpage)
+        doc = self._download_xml(
+            'http://www.heise.de/videout/feed', video_id, query={
+                'container': container_id,
+                'sequenz': sequenz_id,
+            })
 
         formats = []
         for source_node in doc.findall('.//{http://rss.jwpcdn.com/}source'):
             label = source_node.attrib['label']
             height = int_or_none(self._search_regex(
                 r'^(.*?_)?([0-9]+)p$', label, 'height', default=None))
 
         formats = []
         for source_node in doc.findall('.//{http://rss.jwpcdn.com/}source'):
             label = source_node.attrib['label']
             height = int_or_none(self._search_regex(
                 r'^(.*?_)?([0-9]+)p$', label, 'height', default=None))
+            video_url = source_node.attrib['file']
+            ext = determine_ext(video_url, '')
             formats.append({
             formats.append({
-                'url': source_node.attrib['file'],
+                'url': video_url,
                 'format_note': label,
                 'format_note': label,
+                'format_id': '%s_%s' % (ext, label),
                 'height': height,
             })
         self._sort_formats(formats)
                 'height': height,
             })
         self._sort_formats(formats)
-        info['formats'] = formats
 
 
-        return info
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'thumbnail': (xpath_text(doc, './/{http://rss.jwpcdn.com/}image') or
+                          self._og_search_thumbnail(webpage)),
+            'timestamp': parse_iso8601(
+                self._html_search_meta('date', webpage)),
+            'formats': formats,
+        }
Backup clone of https://github.com/ytdl-org unavailable since RIAA fucking raid...