Merge pull request #6790 from remitamine/tele13

[youtube-dl] / youtube_dl / extractor / rai.py

diff --git a/youtube_dl/extractor/rai.py b/youtube_dl/extractor/rai.py
index 115cc64cc6e08ec414de3c0dc671a19c33f1d7bc..7ff1d06c46aacfa67a0134c878d635dce598028f 100644 (file)
--- a/youtube_dl/extractor/rai.py
+++ b/youtube_dl/extractor/rai.py
@@ -5,6 +5,7 @@ import re
 from .common import InfoExtractor
 from ..compat import (
     compat_urllib_parse,
+    compat_urlparse,
 )
 from ..utils import (
     parse_duration,
@@ -62,6 +63,29 @@ class RaiIE(InfoExtractor):
                 'description': 'Edizione delle ore 20:30 ',
             }
         },
+        {
+            'url': 'http://www.ilcandidato.rai.it/dl/ray/media/Il-Candidato---Primo-episodio-Le-Primarie-28e5525a-b495-45e8-a7c3-bc48ba45d2b6.html',
+            'md5': '02b64456f7cc09f96ff14e7dd489017e',
+            'info_dict': {
+                'id': '28e5525a-b495-45e8-a7c3-bc48ba45d2b6',
+                'ext': 'flv',
+                'title': 'Il Candidato - Primo episodio: "Le Primarie"',
+                'description': 'Primo appuntamento con "Il candidato" con Filippo Timi, alias Piero Zucca presidente!',
+                'uploader': 'RaiTre',
+            }
+        },
+        {
+            'url': 'http://www.report.rai.it/dl/Report/puntata/ContentItem-0c7a664b-d0f4-4b2c-8835-3f82e46f433e.html',
+            'md5': '037104d2c14132887e5e4cf114569214',
+            'info_dict': {
+                'id': '0c7a664b-d0f4-4b2c-8835-3f82e46f433e',
+                'ext': 'flv',
+                'title': 'Il pacco',
+                'description': 'md5:4b1afae1364115ce5d78ed83cd2e5b3a',
+                'uploader': 'RaiTre',
+                'upload_date': '20141221',
+            },
+        }
     ]
 
     def _extract_relinker_url(self, webpage):
@@ -79,12 +103,15 @@ class RaiIE(InfoExtractor):
         relinker_url = self._extract_relinker_url(webpage)
 
         if not relinker_url:
-            iframe_path = self._search_regex(
-                r'<iframe[^>]+src="/?(dl/[^"]+\?iframe\b[^"]*)"',
+            iframe_url = self._search_regex(
+                [r'<iframe[^>]+src="([^"]*/dl/[^"]+\?iframe\b[^"]*)"',
+                 r'drawMediaRaiTV\(["\'](.+?)["\']'],
                 webpage, 'iframe')
-            iframe_page = self._download_webpage(
-                '%s/%s' % (host, iframe_path), video_id)
-            relinker_url = self._extract_relinker_url(iframe_page)
+            if not iframe_url.startswith('http'):
+                iframe_url = compat_urlparse.urljoin(url, iframe_url)
+            webpage = self._download_webpage(
+                iframe_url, video_id)
+            relinker_url = self._extract_relinker_url(webpage)
 
         relinker = self._download_json(
             '%s&output=47' % relinker_url, video_id)
@@ -112,15 +139,15 @@ class RaiIE(InfoExtractor):
             uploader = media.get('author')
             upload_date = unified_strdate(media.get('date'))
         else:
-            title = self._search_regex(
-                r'var\s+videoTitolo\s*=\s*"([^"]+)";',
-                webpage, 'title', default=None) or self._og_search_title(webpage)
+            title = (self._search_regex(
+                r'var\s+videoTitolo\s*=\s*"(.+?)";',
+                webpage, 'title', default=None) or self._og_search_title(webpage)).replace('\\"', '"')
             description = self._og_search_description(webpage)
             thumbnail = self._og_search_thumbnail(webpage)
             duration = None
             uploader = self._html_search_meta('Editore', webpage, 'uploader')
             upload_date = unified_strdate(self._html_search_meta(
-                'item-date', webpage, 'upload date'))
+                'item-date', webpage, 'upload date', default=None))
 
         subtitles = self.extract_subtitles(video_id, webpage)