youtube-dl

commit 59b8ab583491410f4f365b1cb268a4360d7be877
parent a5681804415a9b8d011cd238b7d61e6eb4ec7c4b
Author: Philipp Hagemeister <phihag@phihag.de>
Date:   Mon, 16 Feb 2015 15:45:01 +0100

[rtlnl|generic] Add support for rtl.nl embeds (Fixes #4959)

Diffstat:
M
youtube_dl/extractor/__init__.py
 | 
2
+-
M
youtube_dl/extractor/generic.py
 | 
17
++++++++++++++++-
M
youtube_dl/extractor/rtlnl.py
 | 
63
++++++++++++++++++++++++++++++++++++++++++++++++---------------

3 files changed, 65 insertions(+), 17 deletions(-)
diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
@@ -371,7 +371,7 @@ from .rottentomatoes import RottenTomatoesIE
 from .roxwel import RoxwelIE
 from .rtbf import RTBFIE
 from .rte import RteIE
-from .rtlnl import RtlXlIE
+from .rtlnl import RtlNlIE
 from .rtlnow import RTLnowIE
 from .rtl2 import RTL2IE
 from .rtp import RTPIE
diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
@@ -537,6 +537,15 @@ class GenericIE(InfoExtractor):
                 'uploader_id': 'NationalArchives08',
                 'title': 'Webinar: Using Discovery, The National Archives’ online catalogue',
             },
+        },
+        # rtl.nl embed
+        {
+            'url': 'http://www.rtlnieuws.nl/nieuws/buitenland/aanslagen-kopenhagen',
+            'playlist_mincount': 5,
+            'info_dict': {
+                'id': 'aanslagen-kopenhagen',
+                'title': 'Aanslagen Kopenhagen | RTL Nieuws',
+            }
         }
     ]
 
@@ -782,6 +791,13 @@ class GenericIE(InfoExtractor):
                 'entries': entries,
             }
 
+        # Look for embedded rtl.nl player
+        matches = re.findall(
+            r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+video_embed[^"]+)"',
+            webpage)
+        if matches:
+            return _playlist_from_matches(matches, ie='RtlNl')
+
         # Look for embedded (iframe) Vimeo player
         mobj = re.search(
             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
@@ -789,7 +805,6 @@ class GenericIE(InfoExtractor):
             player_url = unescapeHTML(mobj.group('url'))
             surl = smuggle_url(player_url, {'Referer': url})
             return self.url_result(surl)
-
         # Look for embedded (swf embed) Vimeo player
         mobj = re.search(
             r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
diff --git a/youtube_dl/extractor/rtlnl.py b/youtube_dl/extractor/rtlnl.py
@@ -1,16 +1,25 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
-from ..utils import parse_duration
+from ..utils import (
+    int_or_none,
+    parse_duration,
+)
 
 
-class RtlXlIE(InfoExtractor):
-    IE_NAME = 'rtlxl.nl'
-    _VALID_URL = r'https?://(www\.)?rtlxl\.nl/#!/[^/]+/(?P<uuid>[^/?]+)'
+class RtlNlIE(InfoExtractor):
+    IE_NAME = 'rtl.nl'
+    IE_DESC = 'rtl.nl and rtlxl.nl'
+    _VALID_URL = r'''(?x)
+        https?://(www\.)?
+        (?:
+            rtlxl\.nl/\#!/[^/]+/|
+            rtl\.nl/system/videoplayer/[^?#]+?/video_embed\.html\#uuid=
+        )
+        (?P<id>[0-9a-f-]+)'''
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.rtlxl.nl/#!/rtl-nieuws-132237/6e4203a6-0a5e-3596-8424-c599a59e0677',
         'md5': 'cc16baa36a6c169391f0764fa6b16654',
         'info_dict': {
@@ -22,21 +31,30 @@ class RtlXlIE(InfoExtractor):
             'upload_date': '20140814',
             'duration': 576.880,
         },
-    }
+    }, {
+        'url': 'http://www.rtl.nl/system/videoplayer/derden/rtlnieuws/video_embed.html#uuid=84ae5571-ac25-4225-ae0c-ef8d9efb2aed/autoplay=false',
+        'md5': 'dea7474214af1271d91ef332fb8be7ea',
+        'info_dict': {
+            'id': '84ae5571-ac25-4225-ae0c-ef8d9efb2aed',
+            'ext': 'mp4',
+            'timestamp': 1424039400,
+            'title': 'RTL Nieuws - Nieuwe beelden Kopenhagen: chaos direct na aanslag',
+            'thumbnail': 're:^https?://screenshots\.rtl\.nl/system/thumb/sz=[0-9]+x[0-9]+/uuid=84ae5571-ac25-4225-ae0c-ef8d9efb2aed$',
+            'upload_date': '20150215',
+            'description': 'Er zijn nieuwe beelden vrijgegeven die vlak na de aanslag in Kopenhagen zijn gemaakt. Op de video is goed te zien hoe omstanders zich bekommeren om één van de slachtoffers, terwijl de eerste agenten ter plaatse komen.',
+        }
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        uuid = mobj.group('uuid')
-
+        uuid = self._match_id(url)
         info = self._download_json(
             'http://www.rtl.nl/system/s4m/vfd/version=2/uuid=%s/fmt=flash/' % uuid,
             uuid)
 
         material = info['material'][0]
-        episode_info = info['episodes'][0]
-
         progname = info['abstracts'][0]['name']
         subtitle = material['title'] or info['episodes'][0]['name']
+        description = material.get('synopsis') or info['episodes'][0]['synopsis']
 
         # Use unencrypted m3u8 streams (See https://github.com/rg3/youtube-dl/issues/4118)
         videopath = material['videopath'].replace('.f4m', '.m3u8')
@@ -58,14 +76,29 @@ class RtlXlIE(InfoExtractor):
                 'quality': 0,
             }
         ])
-
         self._sort_formats(formats)
 
+        thumbnails = []
+        meta = info.get('meta', {})
+        for p in ('poster_base_url', '"thumb_base_url"'):
+            if not meta.get(p):
+                continue
+
+            thumbnails.append({
+                'url': self._proto_relative_url(meta[p] + uuid),
+                'width': int_or_none(self._search_regex(
+                    r'/sz=([0-9]+)', meta[p], 'thumbnail width', fatal=False)),
+                'height': int_or_none(self._search_regex(
+                    r'/sz=[0-9]+x([0-9]+)',
+                    meta[p], 'thumbnail height', fatal=False))
+            })
+
         return {
             'id': uuid,
             'title': '%s - %s' % (progname, subtitle),
             'formats': formats,
             'timestamp': material['original_date'],
-            'description': episode_info['synopsis'],
+            'description': description,
             'duration': parse_duration(material.get('duration')),
+            'thumbnails': thumbnails,
         }