]> jfr.im git - yt-dlp.git/blobdiff - yt_dlp/extractor/periscope.py
[extractors] Use new framework for existing embeds (#4307)
[yt-dlp.git] / yt_dlp / extractor / periscope.py
index b15906390d07715494b5653dce5499ca0ad72141..2ff6589d5f03b00f2f3f1b908fcd67370b2ff1ca 100644 (file)
@@ -1,8 +1,3 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
-import re
-
 from .common import InfoExtractor
 from ..utils import (
     int_or_none,
@@ -12,6 +7,10 @@
 
 
 class PeriscopeBaseIE(InfoExtractor):
+    _M3U8_HEADERS = {
+        'Referer': 'https://www.periscope.tv/'
+    }
+
     def _call_api(self, method, query, item_id):
         return self._download_json(
             'https://api.periscope.tv/api/v2/%s' % method,
@@ -29,7 +28,7 @@ def _parse_broadcast_data(self, broadcast, video_id):
 
         return {
             'id': broadcast.get('id') or video_id,
-            'title': self._live_title(title) if is_live else title,
+            'title': title,
             'timestamp': parse_iso8601(broadcast.get('created_at')),
             'uploader': uploader,
             'uploader_id': broadcast.get('user_id') or broadcast.get('username'),
@@ -54,9 +53,11 @@ def _extract_pscp_m3u8_formats(self, m3u8_url, video_id, format_id, state, width
             m3u8_url, video_id, 'mp4',
             entry_protocol='m3u8_native'
             if state in ('ended', 'timed_out') else 'm3u8',
-            m3u8_id=format_id, fatal=fatal)
+            m3u8_id=format_id, fatal=fatal, headers=self._M3U8_HEADERS)
         if len(m3u8_formats) == 1:
             self._add_width_and_height(m3u8_formats[0], width, height)
+        for f in m3u8_formats:
+            f.setdefault('http_headers', {}).update(self._M3U8_HEADERS)
         return m3u8_formats
 
 
@@ -64,6 +65,7 @@ class PeriscopeIE(PeriscopeBaseIE):
     IE_DESC = 'Periscope'
     IE_NAME = 'periscope'
     _VALID_URL = r'https?://(?:www\.)?(?:periscope|pscp)\.tv/[^/]+/(?P<id>[^/?#]+)'
+    _EMBED_REGEX = [r'<iframe[^>]+src=([\'"])(?P<url>(?:https?:)?//(?:www\.)?(?:periscope|pscp)\.tv/(?:(?!\1).)+)\1']
     # Alive example URLs can be found here https://www.periscope.tv/
     _TESTS = [{
         'url': 'https://www.periscope.tv/w/aJUQnjY3MjA3ODF8NTYxMDIyMDl2zCg2pECBgwTqRpQuQD352EMPTKQjT4uqlM3cgWFA-g==',
@@ -89,13 +91,6 @@ class PeriscopeIE(PeriscopeBaseIE):
         'only_matching': True,
     }]
 
-    @staticmethod
-    def _extract_url(webpage):
-        mobj = re.search(
-            r'<iframe[^>]+src=([\'"])(?P<url>(?:https?:)?//(?:www\.)?(?:periscope|pscp)\.tv/(?:(?!\1).)+)\1', webpage)
-        if mobj:
-            return mobj.group('url')
-
     def _real_extract(self, url):
         token = self._match_id(url)