From 44b2c836ae93ee67d435d34da6839811e565837a Mon Sep 17 00:00:00 2001
From: kayb94 <>
Date: Thu, 31 Aug 2017 16:49:14 +0200
Subject: [PATCH 1/2] [Heise] Add suport for embedded youtube videos and new
 c't uplink episodes.

---
 youtube_dl/extractor/heise.py | 114 ++++++++++++++++++++++++++++------
 1 file changed, 95 insertions(+), 19 deletions(-)

diff --git a/youtube_dl/extractor/heise.py b/youtube_dl/extractor/heise.py
index 382f32771..c5063f25a 100644
--- a/youtube_dl/extractor/heise.py
+++ b/youtube_dl/extractor/heise.py
@@ -9,16 +9,18 @@ from ..utils import (
     xpath_text,
 )
 
+import re
+
 
 class HeiseIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?heise\.de/(?:[^/]+/)+[^/]+-(?P<id>[0-9]+)\.html'
+    _VALID_URL = r'https?://(?:www\.)?heise\.de/(?:[^/]+/)+[^/]+-(?P<id>[0-9]+)\.html'# noqa
     _TESTS = [{
-        'url': 'http://www.heise.de/video/artikel/Podcast-c-t-uplink-3-3-Owncloud-Tastaturen-Peilsender-Smartphone-2404147.html',
+        'url': 'http://www.heise.de/video/artikel/Podcast-c-t-uplink-3-3-Owncloud-Tastaturen-Peilsender-Smartphone-2404147.html',# noqa
         'md5': 'ffed432483e922e88545ad9f2f15d30e',
         'info_dict': {
             'id': '2404147',
             'ext': 'mp4',
-            'title': "Podcast: c't uplink 3.3 – Owncloud / Tastaturen / Peilsender Smartphone",
+            'title': "Podcast: c't uplink 3.3 – Owncloud / Tastaturen / Peilsender Smartphone",# noqa
             'format_id': 'mp4_720p',
             'timestamp': 1411812600,
             'upload_date': '20140927',
@@ -26,10 +28,18 @@ class HeiseIE(InfoExtractor):
             'thumbnail': r're:^https?://.*/gallery/$',
         }
     }, {
-        'url': 'http://www.heise.de/ct/artikel/c-t-uplink-3-3-Owncloud-Tastaturen-Peilsender-Smartphone-2403911.html',
+        'url': 'https://www.heise.de/ct/artikel/c-t-uplink-18-5-Android-Oreo-Nokia-Galaxy-Note-8-AMD-Ryzen-Threadripper-3812972.html',# noqa
+        'info_dict': {
+            'id': '3812972',
+            'ext': 'mp4',
+            'title': "c't uplink 18.5: Android Oreo, Nokia, Galaxy Note 8, AMD Ryzen Threadripper",# noqa
+            'description': 'md5:0601ade34ae5c4f5058d378327928348'
+            }
+    }, {
+        'url': 'http://www.heise.de/ct/artikel/c-t-uplink-3-3-Owncloud-Tastaturen-Peilsender-Smartphone-2403911.html',# noqa
         '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',
+        'url': 'http://www.heise.de/newsticker/meldung/c-t-uplink-Owncloud-Tastaturen-Peilsender-Smartphone-2404251.html?wt_mc=rss.ho.beitrag.atom',# noqa
         'only_matching': True,
     }, {
         'url': 'http://www.heise.de/ct/ausgabe/2016-12-Spiele-3214137.html',
@@ -40,24 +50,40 @@ class HeiseIE(InfoExtractor):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
+        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')
+
+        # Find embedded youtube videos
+        yt_videos = re.findall(
+            r'<iframe class="yt_video" width="[0-9]+" height="[0-9]+" src="//(.+?)"', # noqa
+            webpage)
+        if yt_videos:
+            return self.playlist_from_matches(yt_videos, title, 'Youtube')
+
         container_id = self._search_regex(
             r'<div class="videoplayerjw"[^>]+data-container="([0-9]+)"',
             webpage, 'container ID')
+
         sequenz_id = self._search_regex(
             r'<div class="videoplayerjw"[^>]+data-sequenz="([0-9]+)"',
             webpage, 'sequenz ID')
 
-        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')
-
-        doc = self._download_xml(
-            'http://www.heise.de/videout/feed', video_id, query={
-                'container': container_id,
-                'sequenz': sequenz_id,
-            })
+        try:
+            doc = self._download_xml(
+                'http://www.heise.de/videout/feed', video_id,
+                query={
+                    'container': container_id,
+                    'sequenz': sequenz_id,
+                })
+        except Exception as e:
+            if e.cause.code == 404:
+                if title.rfind('c\'t') != -1:
+                    return self.ctUplinkRss(title, video_id)
+            else:
+                raise e
 
         formats = []
         for source_node in doc.findall('.//{http://rss.jwpcdn.com/}source'):
@@ -82,9 +108,59 @@ class HeiseIE(InfoExtractor):
             'id': video_id,
             'title': title,
             'description': description,
-            'thumbnail': (xpath_text(doc, './/{http://rss.jwpcdn.com/}image') or
-                          self._og_search_thumbnail(webpage)),
+            'thumbnail': (xpath_text(doc, './/{http://rss.jwpcdn.com/}image') or # noqa
+                self._og_search_thumbnail(webpage)), # noqa
             'timestamp': parse_iso8601(
                 self._html_search_meta('date', webpage)),
-            'formats': formats,
+            'formats': formats
+        }
+
+    def ctUplinkRss(self, title, video_id):
+        formats = []
+
+        # e.g. "18.5" from "c't uplink 18.5:"
+        episode_str = re.findall(r'[0-9]{1,2}.[0-9]{1,2}', title)
+
+        # are there other feeds?
+        sd_rss_feed = self._download_xml(
+                'https://blog.ct.de/ctuplink/ctuplinkvideo.rss',
+                video_id, "Downloading alternative XML (SD)")
+        hd_rss_feed = self._download_xml(
+                'https://blog.ct.de/ctuplink/ctuplinkvideohd.rss',
+                video_id, "Downloading alternative XML (HD)")
+
+        # should be same for both sd/hd
+        titles = hd_rss_feed.findall('./channel/item/title')
+        descriptions = hd_rss_feed.findall('./channel/item/description')
+
+        sd_video_urls = sd_rss_feed.findall('./channel/item/guid')
+        hd_video_urls = hd_rss_feed.findall('./channel/item/guid')
+
+        # try to find the real matching title. it might be misformatted or so.
+        # thereby only rely on the episode_str, e.g. "18.5"
+        episode_index = -1
+        for index, item in enumerate(titles):
+            if titles[index].text.rfind(episode_str[0]) != -1:
+                episode_index = index
+                break
+
+        # in case something went wrong
+        if episode_index == -1:
+            return
+
+        formats.append({
+            'url': sd_video_urls[episode_index].text,
+            'height': 360})
+
+        formats.append({
+            'url': hd_video_urls[episode_index].text,
+            'height': 720})
+
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': descriptions[episode_index].text,
+            'formats': formats
         }

From 25a0f8e2997057205c7cdbee66a4e742da3622a4 Mon Sep 17 00:00:00 2001
From: kayb94 <>
Date: Thu, 31 Aug 2017 18:41:55 +0200
Subject: [PATCH 2/2] [Heise] Improved yt regex.

---
 youtube_dl/extractor/heise.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/youtube_dl/extractor/heise.py b/youtube_dl/extractor/heise.py
index c5063f25a..d97d6db91 100644
--- a/youtube_dl/extractor/heise.py
+++ b/youtube_dl/extractor/heise.py
@@ -58,7 +58,7 @@ class HeiseIE(InfoExtractor):
 
         # Find embedded youtube videos
         yt_videos = re.findall(
-            r'<iframe class="yt_video" width="[0-9]+" height="[0-9]+" src="//(.+?)"', # noqa
+            r'<iframe[^>]+class="yt_video"[^>]+src="//([^"]+)', 
             webpage)
         if yt_videos:
             return self.playlist_from_matches(yt_videos, title, 'Youtube')