youtube-dl

commit 045c48847ad066fc273337c96b25eacc35d146cd
parent 90644a68436e53e3ee74af53841c750549816cd7
Author: Philipp Hagemeister <phihag@phihag.de>
Date:   Sat,  6 Dec 2014 00:42:39 +0100

[tagesschau] Add suppot for sendung (Fixes #4378)

Diffstat:
M
youtube_dl/extractor/tagesschau.py
 | 
101
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++-----------------

1 file changed, 79 insertions(+), 22 deletions(-)
diff --git a/youtube_dl/extractor/tagesschau.py b/youtube_dl/extractor/tagesschau.py
@@ -4,10 +4,11 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..utils import parse_filesize
 
 
 class TagesschauIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?tagesschau\.de/multimedia/video/video(?P<id>-?[0-9]+)\.html'
+    _VALID_URL = r'https?://(?:www\.)?tagesschau\.de/multimedia/(?:sendung/ts|video/video)(?P<id>-?[0-9]+)\.html'
 
     _TESTS = [{
         'url': 'http://www.tagesschau.de/multimedia/video/video1399128.html',
@@ -19,6 +20,16 @@ class TagesschauIE(InfoExtractor):
             'description': 'md5:69da3c61275b426426d711bde96463ab',
             'thumbnail': 're:^http:.*\.jpg$',
         },
+    }, {
+        'url': 'http://www.tagesschau.de/multimedia/sendung/ts-5727.html',
+        'md5': '3c54c1f6243d279b706bde660ceec633',
+        'info_dict': {
+            'id': '5727',
+            'ext': 'mp4',
+            'description': 'md5:695c01bfd98b7e313c501386327aea59',
+            'title': 'Sendung: tagesschau \t04.12.2014 20:00 Uhr',
+            'thumbnail': 're:^http:.*\.jpg$',
+        }
     }]
 
     _FORMATS = {
@@ -32,32 +43,78 @@ class TagesschauIE(InfoExtractor):
         display_id = video_id.lstrip('-')
         webpage = self._download_webpage(url, display_id)
 
-        playerpage = self._download_webpage(
-            'http://www.tagesschau.de/multimedia/video/video%s~player_autoplay-true.html' % video_id,
-            display_id, 'Downloading player page')
-
-        medias = re.findall(
-            r'"(http://media.+?)", type:"video/(.+?)", quality:"(.+?)"',
-            playerpage)
+        player_url = self._html_search_meta(
+            'twitter:player', webpage, 'player URL', default=None)
+        if player_url:
+            playerpage = self._download_webpage(
+                player_url, display_id, 'Downloading player page')
 
-        formats = []
-        for url, ext, res in medias:
-            f = {
-                'format_id': res + '_' + ext,
-                'url': url,
-                'ext': ext,
-            }
-            f.update(self._FORMATS.get(res, {}))
-            formats.append(f)
+            medias = re.findall(
+                r'"(http://media.+?)", type:"video/(.+?)", quality:"(.+?)"',
+                playerpage)
+            formats = []
+            for url, ext, res in medias:
+                f = {
+                    'format_id': res + '_' + ext,
+                    'url': url,
+                    'ext': ext,
+                }
+                f.update(self._FORMATS.get(res, {}))
+                formats.append(f)
+            thumbnail_fn = re.findall(r'"(/multimedia/.+?\.jpg)"', playerpage)[-1]
+            title = self._og_search_title(webpage).strip()
+            description = self._og_search_description(webpage).strip()
+        else:
+            download_text = self._search_regex(
+                r'(?s)<p>Wir bieten dieses Video in folgenden Formaten zum Download an:</p>\s*<div class="controls">(.*?)</div>\s*<p>',
+                webpage, 'download links')
+            links = re.finditer(
+                r'<div class="button" title="(?P<title>[^"]*)"><a href="(?P<url>[^"]+)">(?P<name>.+?)</a></div>',
+                webpage)
+            formats = []
+            for l in links:
+                format_id = self._search_regex(
+                    r'.*/[^/.]+\.([^/]+)\.[^/.]+', l.group('url'), 'format ID')
+                format = {
+                    'format_id': format_id,
+                    'url': l.group('url'),
+                    'format_name': l.group('name'),
+                }
+                m = re.match(
+                    r'''(?x)
+                        Video:\s*(?P<vcodec>[a-zA-Z0-9/._-]+)\s*&\#10;
+                        (?P<width>[0-9]+)x(?P<height>[0-9]+)px&\#10;
+                        (?P<vbr>[0-9]+)kbps&\#10;
+                        Audio:\s*(?P<abr>[0-9]+)kbps,\s*(?P<audio_desc>[A-Za-z\.0-9]+)&\#10;
+                        Gr&ouml;&szlig;e:\s*(?P<filesize_approx>[0-9.,]+\s+[a-zA-Z]*B)''',
+                    l.group('title'))
+                if m:
+                    format.update({
+                        'format_note': m.group('audio_desc'),
+                        'vcodec': m.group('vcodec'),
+                        'width': int(m.group('width')),
+                        'height': int(m.group('height')),
+                        'abr': int(m.group('abr')),
+                        'vbr': int(m.group('vbr')),
+                        'filesize_approx': parse_filesize(m.group('filesize_approx')),
+                    })
+                formats.append(format)
+            thumbnail_fn = self._search_regex(
+                r'(?s)<img alt="Sendungsbild".*?src="([^"]+)"',
+                webpage, 'thumbnail', fatal=False)
+            description = self._html_search_regex(
+                r'(?s)<p class="teasertext">(.*?)</p>',
+                webpage, 'description', fatal=False)
+            title = self._html_search_regex(
+                r'<span class="headline".*?>(.*?)</span>', webpage, 'title')
 
         self._sort_formats(formats)
-
-        thumbnail = re.findall(r'"(/multimedia/.+?\.jpg)"', playerpage)[-1]
+        thumbnail = 'http://www.tagesschau.de' + thumbnail_fn
 
         return {
             'id': display_id,
-            'title': self._og_search_title(webpage).strip(),
-            'thumbnail': 'http://www.tagesschau.de' + thumbnail,
+            'title': title,
+            'thumbnail': thumbnail,
             'formats': formats,
-            'description': self._og_search_description(webpage).strip(),
+            'description': description,
         }