]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/tv2.py
Merge tag 'upstream/2016.12.01'
[youtubedl] / youtube_dl / extractor / tv2.py
index 1457e524e810c8bda02795c1b8dd78e95c47802e..bd28267b0cb6a0154133c98f567c24f054b5459a 100644 (file)
@@ -1,4 +1,4 @@
-# encoding: utf-8
+# coding: utf-8
 from __future__ import unicode_literals
 
 import re
@@ -8,13 +8,14 @@ from ..utils import (
     determine_ext,
     int_or_none,
     float_or_none,
+    js_to_json,
     parse_iso8601,
     remove_end,
 )
 
 
 class TV2IE(InfoExtractor):
-    _VALID_URL = 'http://(?:www\.)?tv2\.no/v/(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?tv2\.no/v/(?P<id>\d+)'
     _TEST = {
         'url': 'http://www.tv2.no/v/916509/',
         'info_dict': {
@@ -54,10 +55,11 @@ class TV2IE(InfoExtractor):
                 ext = determine_ext(video_url)
                 if ext == 'f4m':
                     formats.extend(self._extract_f4m_formats(
-                        video_url, video_id, f4m_id=format_id))
+                        video_url, video_id, f4m_id=format_id, fatal=False))
                 elif ext == 'm3u8':
                     formats.extend(self._extract_m3u8_formats(
-                        video_url, video_id, 'mp4', m3u8_id=format_id))
+                        video_url, video_id, 'mp4', entry_protocol='m3u8_native',
+                        m3u8_id=format_id, fatal=False))
                 elif ext == 'ism' or video_url.endswith('.ism/Manifest'):
                     pass
                 else:
@@ -100,12 +102,12 @@ class TV2IE(InfoExtractor):
 
 
 class TV2ArticleIE(InfoExtractor):
-    _VALID_URL = 'http://(?:www\.)?tv2\.no/(?:a|\d{4}/\d{2}/\d{2}(/[^/]+)+)/(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?tv2\.no/(?:a|\d{4}/\d{2}/\d{2}(/[^/]+)+)/(?P<id>\d+)'
     _TESTS = [{
         'url': 'http://www.tv2.no/2015/05/16/nyheter/alesund/krim/pingvin/6930542',
         'info_dict': {
             'id': '6930542',
-            'title': 'Russen hetses etter pingvintyveri  innrømmer å ha åpnet luken på buret',
+            'title': 'Russen hetses etter pingvintyveri - innrømmer å ha åpnet luken på buret',
             'description': 'md5:339573779d3eea3542ffe12006190954',
         },
         'playlist_count': 2,
@@ -119,9 +121,23 @@ class TV2ArticleIE(InfoExtractor):
 
         webpage = self._download_webpage(url, playlist_id)
 
+        # Old embed pattern (looks unused nowadays)
+        assets = re.findall(r'data-assetid=["\'](\d+)', webpage)
+
+        if not assets:
+            # New embed pattern
+            for v in re.findall('TV2ContentboxVideo\(({.+?})\)', webpage):
+                video = self._parse_json(
+                    v, playlist_id, transform_source=js_to_json, fatal=False)
+                if not video:
+                    continue
+                asset = video.get('assetId')
+                if asset:
+                    assets.append(asset)
+
         entries = [
-            self.url_result('http://www.tv2.no/v/%s' % video_id, 'TV2')
-            for video_id in re.findall(r'data-assetid="(\d+)"', webpage)]
+            self.url_result('http://www.tv2.no/v/%s' % asset_id, 'TV2')
+            for asset_id in assets]
 
         title = remove_end(self._og_search_title(webpage), ' - TV2.no')
         description = remove_end(self._og_search_description(webpage), ' - TV2.no')