[la7.it] Fix the extractor
authorYen Chi Hsuan <yan12125@gmail.com>
Sat, 2 Jul 2016 15:49:03 +0000 (23:49 +0800)
committerYen Chi Hsuan <yan12125@gmail.com>
Sat, 2 Jul 2016 15:49:03 +0000 (23:49 +0800)
youtube_dl/extractor/la7.py

index b08f6e3c9548de02217e43bebbf20b5f2ab871e8..826dc94765067af1ed69d2fb365252578e6da2e6 100644 (file)
@@ -1,60 +1,74 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
 from ..utils import (
-    parse_duration,
+    determine_ext,
+    js_to_json,
 )
 
 
 class LA7IE(InfoExtractor):
-    IE_NAME = 'la7.tv'
-    _VALID_URL = r'''(?x)
-        https?://(?:www\.)?la7\.tv/
-        (?:
-            richplayer/\?assetid=|
-            \?contentId=
-        )
-        (?P<id>[0-9]+)'''
-
-    _TEST = {
-        'url': 'http://www.la7.tv/richplayer/?assetid=50355319',
-        'md5': 'ec7d1f0224d20ba293ab56cf2259651f',
+    IE_NAME = 'la7.it'
+    _VALID_URL = r'''(?x)(https?://)?(?:
+        (?:www\.)?la7\.it/([^/]+)/(?:rivedila7|video)/|
+        tg\.la7\.it/repliche-tgla7\?id=
+    )(?P<id>.+)'''
+
+    _TESTS = [{
+        # 'src' is a plain URL
+        'url': 'http://www.la7.it/crozza/video/inccool8-02-10-2015-163722',
+        'md5': '6054674766e7988d3e02f2148ff92180',
+        'info_dict': {
+            'id': 'inccool8-02-10-2015-163722',
+            'ext': 'mp4',
+            'title': 'Inc.Cool8',
+            'description': 'Benvenuti nell\'incredibile mondo della INC. COOL. 8. dove “INC.” sta per “Incorporated” “COOL” sta per “fashion” ed Eight sta per il gesto  atletico',
+            'thumbnail': 're:^https?://.*',
+        },
+    }, {
+        # 'src' is a dictionary
+        'url': 'http://tg.la7.it/repliche-tgla7?id=189080',
+        'md5': '6b0d8888d286e39870208dfeceaf456b',
         'info_dict': {
-            'id': '50355319',
+            'id': '189080',
             'ext': 'mp4',
-            'title': 'IL DIVO',
-            'description': 'Un film di Paolo Sorrentino con Toni Servillo, Anna Bonaiuto, Giulio Bosetti  e Flavio Bucci',
-            'duration': 6254,
+            'title': 'TG LA7',
         },
-        'skip': 'Blocked in the US',
-    }
+    }, {
+        'url': 'http://www.la7.it/omnibus/rivedila7/omnibus-news-02-07-2016-189077',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        xml_url = 'http://www.la7.tv/repliche/content/index.php?contentId=%s' % video_id
-        doc = self._download_xml(xml_url, video_id)
-
-        video_title = doc.find('title').text
-        description = doc.find('description').text
-        duration = parse_duration(doc.find('duration').text)
-        thumbnail = doc.find('img').text
-        view_count = int(doc.find('views').text)
-
-        prefix = doc.find('.//fqdn').text.strip().replace('auto:', 'http:')
-
-        formats = [{
-            'format': vnode.find('quality').text,
-            'tbr': int(vnode.find('quality').text),
-            'url': vnode.find('fms').text.strip().replace('mp4:', prefix),
-        } for vnode in doc.findall('.//videos/video')]
+
+        webpage = self._download_webpage(url, video_id)
+
+        player_data = self._parse_json(
+            self._search_regex(r'videoLa7\(({[^;]+})\);', webpage, 'player data'),
+            video_id, transform_source=js_to_json)
+
+        source = player_data['src']
+        source_urls = source.values() if isinstance(source, dict) else [source]
+
+        formats = []
+        for source_url in source_urls:
+            ext = determine_ext(source_url)
+            if ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    source_url, video_id, ext='mp4',
+                    entry_protocol='m3u8_native', m3u8_id='hls'))
+            else:
+                formats.append({
+                    'url': source_url,
+                })
         self._sort_formats(formats)
 
         return {
             'id': video_id,
-            'title': video_title,
-            'description': description,
-            'thumbnail': thumbnail,
-            'duration': duration,
+            'title': player_data['title'],
+            'description': self._og_search_description(webpage, default=None),
+            'thumbnail': player_data.get('poster'),
             'formats': formats,
-            'view_count': view_count,
         }