[youtube] Fix extraction.

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

diff --git a/youtube_dl/extractor/viu.py b/youtube_dl/extractor/viu.py
index 1a81b484541a00648cec80b7106886c20b615dd1..3bd37525b6ffcc53d8cb35bed9ab8974847893cc 100644 (file)
--- a/youtube_dl/extractor/viu.py
+++ b/youtube_dl/extractor/viu.py
@@ -4,7 +4,10 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import compat_str
+from ..compat import (
+    compat_kwargs,
+    compat_str,
+)
 from ..utils import (
     ExtractorError,
     int_or_none,
@@ -25,7 +28,7 @@ class ViuBaseIE(InfoExtractor):
                 'userid': 'guest',
                 'useridtype': 'guest',
                 'ver': '1.0'
-            })
+            }, headers=self.geo_verification_headers())
         self._auth_token = viu_auth_res.info()['X-VIU-AUTH']
 
     def _call_api(self, path, *args, **kwargs):
@@ -36,7 +39,8 @@ class ViuBaseIE(InfoExtractor):
         headers.update(kwargs.get('headers', {}))
         kwargs['headers'] = headers
         response = self._download_json(
-            'https://www.viu.com/api/' + path, *args, **kwargs)['response']
+            'https://www.viu.com/api/' + path, *args,
+            **compat_kwargs(kwargs))['response']
         if response.get('status') != 'success':
             raise ExtractorError('%s said: %s' % (
                 self.IE_NAME, response['message']), expected=True)
@@ -44,7 +48,7 @@ class ViuBaseIE(InfoExtractor):
 
 
 class ViuIE(ViuBaseIE):
-    _VALID_URL = r'(?:viu:|https?://www\.viu\.com/[a-z]{2}/media/)(?P<id>\d+)'
+    _VALID_URL = r'(?:viu:|https?://[^/]+\.viu\.com/[a-z]{2}/media/)(?P<id>\d+)'
     _TESTS = [{
         'url': 'https://www.viu.com/en/media/1116705532?containerId=playlist-22168059',
         'info_dict': {
@@ -69,6 +73,9 @@ class ViuIE(ViuBaseIE):
             'skip_download': 'm3u8 download',
         },
         'skip': 'Geo-restricted to Indonesia',
+    }, {
+        'url': 'https://india.viu.com/en/media/1126286865',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -86,13 +93,17 @@ class ViuIE(ViuBaseIE):
         m3u8_url = None
         url_path = video_data.get('urlpathd') or video_data.get('urlpath')
         tdirforwhole = video_data.get('tdirforwhole')
-        hls_file = video_data.get('hlsfile')
+        # #EXT-X-BYTERANGE is not supported by native hls downloader
+        # and ffmpeg (#10955)
+        # hls_file = video_data.get('hlsfile')
+        hls_file = video_data.get('jwhlsfile')
         if url_path and tdirforwhole and hls_file:
             m3u8_url = '%s/%s/%s' % (url_path, tdirforwhole, hls_file)
         else:
-            m3u8_url = re.sub(
-                r'(/hlsc_)[a-z]+(\d+\.m3u8)',
-                r'\1whe\2', video_data['href'])
+            # m3u8_url = re.sub(
+            #     r'(/hlsc_)[a-z]+(\d+\.m3u8)',
+            #     r'\1whe\2', video_data['href'])
+            m3u8_url = video_data['href']
         formats = self._extract_m3u8_formats(m3u8_url, video_id, 'mp4')
         self._sort_formats(formats)
 
@@ -184,16 +195,29 @@ class ViuOTTIE(InfoExtractor):
         'skip': 'Geo-restricted to Hong Kong',
     }]
 
+    _AREA_ID = {
+        'HK': 1,
+        'SG': 2,
+        'TH': 4,
+        'PH': 5,
+    }
+
     def _real_extract(self, url):
         country_code, video_id = re.match(self._VALID_URL, url).groups()
 
+        query = {
+            'r': 'vod/ajax-detail',
+            'platform_flag_label': 'web',
+            'product_id': video_id,
+        }
+
+        area_id = self._AREA_ID.get(country_code.upper())
+        if area_id:
+            query['area_id'] = area_id
+
         product_data = self._download_json(
             'http://www.viu.com/ott/%s/index.php' % country_code, video_id,
-            'Downloading video info', query={
-                'r': 'vod/ajax-detail',
-                'platform_flag_label': 'web',
-                'product_id': video_id,
-            })['data']
+            'Downloading video info', query=query)['data']
 
         video_data = product_data.get('current_product')
         if not video_data:
@@ -203,6 +227,9 @@ class ViuOTTIE(InfoExtractor):
             'https://d1k2us671qcoau.cloudfront.net/distribute_web_%s.php' % country_code,
             video_id, 'Downloading stream info', query={
                 'ccs_product_id': video_data['ccs_product_id'],
+            }, headers={
+                'Referer': url,
+                'Origin': re.search(r'https?://[^/]+', url).group(0),
             })['data']['stream']
 
         stream_sizes = stream_data.get('size', {})