Prepare for upload.

[youtubedl] / youtube_dl / extractor / heise.py

diff --git a/youtube_dl/extractor/heise.py b/youtube_dl/extractor/heise.py
index d41c0413f015e262237f66161d4021c32f56ff0e..382f32771db513fc1c41e81f14ce42308ad50276 100644 (file)
--- a/youtube_dl/extractor/heise.py
+++ b/youtube_dl/extractor/heise.py
@@ -3,73 +3,88 @@ from __future__ import unicode_literals
 
 from .common import InfoExtractor
 from ..utils import (
 
 from .common import InfoExtractor
 from ..utils import (

-    get_meta_content,
+    determine_ext,

     int_or_none,
     parse_iso8601,
     int_or_none,
     parse_iso8601,

+    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'
-        ),
+    _VALID_URL = r'https?://(?:www\.)?heise\.de/(?:[^/]+/)+[^/]+-(?P<id>[0-9]+)\.html'
+    _TESTS = [{
+        'url': 'http://www.heise.de/video/artikel/Podcast-c-t-uplink-3-3-Owncloud-Tastaturen-Peilsender-Smartphone-2404147.html',

         'md5': 'ffed432483e922e88545ad9f2f15d30e',
         'info_dict': {
             'id': '2404147',
             'ext': 'mp4',
         'md5': 'ffed432483e922e88545ad9f2f15d30e',
         'info_dict': {
             'id': '2404147',
             'ext': 'mp4',

-            'title': (
-                "Podcast: c't uplink 3.3 – Owncloud / Tastaturen / Peilsender Smartphone"
-            ),
-            'format_id': 'mp4_720',
+            'title': "Podcast: c't uplink 3.3 – Owncloud / Tastaturen / Peilsender Smartphone",
+            'format_id': 'mp4_720p',

             'timestamp': 1411812600,
             'upload_date': '20140927',
             '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$',
+            'description': 'md5:c934cbfb326c669c2bcabcbe3d3fcd20',
+            'thumbnail': r're:^https?://.*/gallery/$',

         }
         }

-    }
+    }, {
+        '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)
 
         container_id = self._search_regex(
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
         container_id = self._search_regex(

-            r'<div class="videoplayerjw".*?data-container="([0-9]+)"',
+            r'<div class="videoplayerjw"[^>]+data-container="([0-9]+)"',

             webpage, 'container ID')
         sequenz_id = self._search_regex(
             webpage, 'container ID')
         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 = self._html_search_meta('fulltitle', webpage, default=None)
+        if not title or title == "c't":
+            title = self._search_regex(
+                r'<div[^>]+class="videoplayerjw"[^>]+data-title="([^"]+)"',
+                webpage, 'title')

 
 

-        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
+        description = self._og_search_description(
+            webpage, default=None) or self._html_search_meta(
+            'description', webpage)
+
+        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,
+        }