From 5008e175357e681584c9d7dcb4b909998645c72b Mon Sep 17 00:00:00 2001
From: j <j@mailb.org>
Date: Sun, 20 Dec 2015 15:35:05 +0100
Subject: [PATCH 1/2] [tele5] Add new extractor (closes #7805)

---
 youtube_dl/extractor/__init__.py |  1 +
 youtube_dl/extractor/tele5.py    | 76 ++++++++++++++++++++++++++++++++
 2 files changed, 77 insertions(+)
 create mode 100644 youtube_dl/extractor/tele5.py

diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
index eac50eda5..59280a2b4 100644
--- a/youtube_dl/extractor/__init__.py
+++ b/youtube_dl/extractor/__init__.py
@@ -649,6 +649,7 @@ from .teamcoco import TeamcocoIE
 from .techtalks import TechTalksIE
 from .ted import TEDIE
 from .telebruxelles import TeleBruxellesIE
+from .tele5 import Tele5IE
 from .telecinco import TelecincoIE
 from .telegraaf import TelegraafIE
 from .telemb import TeleMBIE
diff --git a/youtube_dl/extractor/tele5.py b/youtube_dl/extractor/tele5.py
new file mode 100644
index 000000000..26a477481
--- /dev/null
+++ b/youtube_dl/extractor/tele5.py
@@ -0,0 +1,76 @@
+# encoding: utf-8
+import re
+
+from .common import InfoExtractor
+
+
+class Tele5IE(InfoExtractor):
+    _VALID_URL = r'http://www.tele5.de/(?P<type>re-play/filme|a-z)/(?P<id>.*?).html'
+
+    _TESTS = [{
+        'url': 'http://www.tele5.de/re-play/filme/flying-swords-of-dragon-gate.html',
+        'info_dict': {
+            'ext': 'mp4',
+            'id': 'flying-swords-of-dragon-gate',
+            'title': 'Flying Swords of Dragon Gate',
+        },
+    }]
+
+    def _get_video_url(self, thumbnail, entity_id, video_js_url):
+        video_js = self._download_webpage(video_js_url, None)
+        flavor_id = re.compile(r'''"id\\":\\"(.*?)\\"''').findall(video_js)[-1]
+        return thumbnail.split('thumbnail/')[0] + 'playManifest/entryId/{}/flavorId/{}/format/url/protocol/http/a.mp4'.format(entity_id, flavor_id)
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        series = mobj.group('type') == 'a-z'
+        webpage = self._download_webpage(url, video_id)
+
+        if series:
+            title = self._html_search_regex(r'<title>(.*?)</title>', webpage, 'title').split('-')[0].strip()
+            partner_data =self._html_search_regex(
+                r'<script src="(http://api.medianac.com/.*embedIframeJs.*?)"></script>',
+                webpage, 'video js').split('/')
+            uiconf_id = partner_data[-3]
+            partner_id = partner_data[-1]
+
+            entries = []
+            for player in re.compile(r'''<div id="kaltura_player_.*?</div>''', re.DOTALL).findall(webpage):
+                data = dict(kv for kv in re.compile(r'itemprop="(.*?)" content="(.*?)"').findall(player))
+                thumbnail = data['thumbnail']
+                entity_id = thumbnail.split('entry_id/')[1].split('/')[0]
+                player_id = re.compile(r'id="kaltura_player_(.*?)"').findall(player)[0]
+                video_js_url = 'http://api.medianac.com/html5/html5lib/v2.35/mwEmbedFrame.php?&wid=_{partner_id}&uiconf_id={uiconf_id}&cache_st={player_id}&entry_id={entity_id}&flashvars[streamerType]=rtmp&flashvars[mediaProtocol]=rtmp&playerId=kaltura_player_{player_id}&ServiceUrl=http://api.medianac.com&CdnUrl=http://api.medianac.com&ServiceBase=/api_v3/index.php?service=&UseManifestUrls=true&forceMobileHTML5=true&urid=2.35&callback=mwi_kalturaplayer{player_id}'.format(**{
+                    'player_id': player_id,
+                    'partner_id': partner_id,
+                    'uiconf_id': uiconf_id,
+                    'entity_id': entity_id,
+                })
+                video_url = self._get_video_url(thumbnail, entity_id, video_js_url)
+                entries.append({
+                    'id': entity_id,
+                    'title': data['name'],
+                    'url': video_url,
+                    'thumbnail': data['thumbnail']
+                })
+            return self.playlist_result(entries, video_id, title)
+        else:
+            title = self._html_search_regex(r'<h1>(.*?)</h1>', webpage, 'title')
+            video_js_url =self._html_search_regex(
+                    r'<script src="(http://api.medianac.com/.*embedIframeJs.*?)"></script>',
+                    webpage, 'video js')
+            thumbnail = self._html_search_regex(
+                r'<span itemprop="thumbnail" content="(.*?)">',
+                webpage, "thumbnail url", fatal=False)
+            entity_id = thumbnail.split('entry_id/')[1].split('/')[0]
+            video_url = self._get_video_url(thumbnail, entity_id, video_js_url)
+
+            return {
+                'ext': 'mp4',
+                'id': video_id,
+                'thumbnail': thumbnail,
+                'title': title,
+                'url': video_url,
+            }
+

From b2ae7e2b5b69170e5413d0d1603bea496ffd7fcd Mon Sep 17 00:00:00 2001
From: tetzank <tetzank@users.sourceforge.net>
Date: Fri, 25 Dec 2015 19:22:01 +0100
Subject: [PATCH 2/2] [tele5] added support for different formats

---
 youtube_dl/extractor/tele5.py | 20 ++++++++++++++++----
 1 file changed, 16 insertions(+), 4 deletions(-)

diff --git a/youtube_dl/extractor/tele5.py b/youtube_dl/extractor/tele5.py
index 26a477481..606b61e9d 100644
--- a/youtube_dl/extractor/tele5.py
+++ b/youtube_dl/extractor/tele5.py
@@ -18,8 +18,20 @@ class Tele5IE(InfoExtractor):
 
     def _get_video_url(self, thumbnail, entity_id, video_js_url):
         video_js = self._download_webpage(video_js_url, None)
-        flavor_id = re.compile(r'''"id\\":\\"(.*?)\\"''').findall(video_js)[-1]
-        return thumbnail.split('thumbnail/')[0] + 'playManifest/entryId/{}/flavorId/{}/format/url/protocol/http/a.mp4'.format(entity_id, flavor_id)
+        match = re.compile(r'''\\"flavorParamsId\\":(\d),\\"width\\":(\d+),\\"height\\":(\d+),\\"bitrate\\":(\d+),\\"frameRate\\":(\d+),.+?,\\"id\\":\\"(\w+)\\",.+?,\\"size\\":(\d+),''').findall(video_js)
+        formats = []
+        for m in match:
+            # assumes matches are already in the right order, from worst to best quality
+            formats.append({
+                'url': thumbnail.split('thumbnail/')[0] + 'playManifest/entryId/{}/flavorId/{}/format/url/protocol/http/a.mp4'.format(entity_id, m[5]),
+                'format_id': m[0], # needs to be string otherwise cmp fails
+                'width':     int(m[1]),
+                'height':    int(m[2]),
+                'tbr':       int(m[3]), # needs to be an integer
+                'frameRate': int(m[4]),
+                'filesize':  int(m[6])
+            })
+        return formats
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -47,11 +59,11 @@ class Tele5IE(InfoExtractor):
                     'uiconf_id': uiconf_id,
                     'entity_id': entity_id,
                 })
-                video_url = self._get_video_url(thumbnail, entity_id, video_js_url)
+                formats = self._get_video_url(thumbnail, entity_id, video_js_url)
                 entries.append({
                     'id': entity_id,
                     'title': data['name'],
-                    'url': video_url,
+                    'formats': formats,
                     'thumbnail': data['thumbnail']
                 })
             return self.playlist_result(entries, video_id, title)