[extractors] Use new framework for existing embeds (#4307)

[yt-dlp.git] / yt_dlp / extractor / anvato.py

diff --git a/yt_dlp/extractor/anvato.py b/yt_dlp/extractor/anvato.py
index 7386ae2e6bd35ea305972b60abd2856e7fde25cc..cb94835693a56b9f101cf1ba071089770dbeefeb 100644 (file)
--- a/yt_dlp/extractor/anvato.py
+++ b/yt_dlp/extractor/anvato.py
@@ -1,6 +1,3 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
 import base64
 import hashlib
 import json
@@ -8,6 +5,7 @@
 import re
 import time
 
+from .anvato_token_generator import NFLTokenGenerator
 from .common import InfoExtractor
 from ..aes import aes_encrypt
 from ..compat import compat_str
@@ -16,21 +14,12 @@
     determine_ext,
     intlist_to_bytes,
     int_or_none,
+    join_nonempty,
     strip_jsonp,
     unescapeHTML,
     unsmuggle_url,
 )
 
-# This import causes a ModuleNotFoundError on some systems for unknown reason.
-# See issues:
-# https://github.com/yt-dlp/yt-dlp/issues/35
-# https://github.com/ytdl-org/youtube-dl/issues/27449
-# https://github.com/animelover1984/youtube-dl/issues/17
-try:
-    from .anvato_token_generator import NFLTokenGenerator
-except ImportError:
-    NFLTokenGenerator = None
-
 
 def md5_text(s):
     if not isinstance(s, compat_str):
@@ -303,13 +292,13 @@ def _get_anvato_videos(self, access_key, video_id):
             tbr = int_or_none(published_url.get('kbps'))
             a_format = {
                 'url': video_url,
-                'format_id': ('-'.join(filter(None, ['http', published_url.get('cdn_name')]))).lower(),
-                'tbr': tbr if tbr != 0 else None,
+                'format_id': join_nonempty('http', published_url.get('cdn_name')).lower(),
+                'tbr': tbr or None,
             }
 
             if media_format == 'm3u8' and tbr is not None:
                 a_format.update({
-                    'format_id': '-'.join(filter(None, ['hls', compat_str(tbr)])),
+                    'format_id': join_nonempty('hls', tbr),
                     'ext': 'mp4',
                 })
             elif media_format == 'm3u8-variant' or ext == 'm3u8':
@@ -351,30 +340,16 @@ def _get_anvato_videos(self, access_key, video_id):
             'subtitles': subtitles,
         }
 
-    @staticmethod
-    def _extract_urls(ie, webpage, video_id):
-        entries = []
-        for mobj in re.finditer(AnvatoIE._ANVP_RE, webpage):
-            anvplayer_data = ie._parse_json(
-                mobj.group('anvp'), video_id, transform_source=unescapeHTML,
-                fatal=False)
-            if not anvplayer_data:
-                continue
-            video = anvplayer_data.get('video')
-            if not isinstance(video, compat_str) or not video.isdigit():
-                continue
-            access_key = anvplayer_data.get('accessKey')
-            if not access_key:
-                mcp = anvplayer_data.get('mcp')
-                if mcp:
-                    access_key = AnvatoIE._MCP_TO_ACCESS_KEY_TABLE.get(
-                        mcp.lower())
+    @classmethod
+    def _extract_from_webpage(cls, url, webpage):
+        for mobj in re.finditer(cls._ANVP_RE, webpage):
+            anvplayer_data = unescapeHTML(json.loads(mobj.group('anvp'))) or {}
+            video_id, access_key = anvplayer_data.get('video'), anvplayer_data.get('accessKey')
             if not access_key:
+                access_key = cls._MCP_TO_ACCESS_KEY_TABLE.get((anvplayer_data.get('mcp') or '').lower())
+            if not (video_id or '').isdigit() or not access_key:
                 continue
-            entries.append(ie.url_result(
-                'anvato:%s:%s' % (access_key, video), ie=AnvatoIE.ie_key(),
-                video_id=video))
-        return entries
+            yield cls.url_result(f'anvato:{access_key}:{video_id}', AnvatoIE, video_id)
 
     def _extract_anvato_videos(self, webpage, video_id):
         anvplayer_data = self._parse_json(
@@ -390,7 +365,7 @@ def _real_extract(self, url):
             'countries': smuggled_data.get('geo_countries'),
         })
 
-        mobj = re.match(self._VALID_URL, url)
+        mobj = self._match_valid_url(url)
         access_key, video_id = mobj.group('access_key_or_mcp', 'id')
         if access_key not in self._ANVACK_TABLE:
             access_key = self._MCP_TO_ACCESS_KEY_TABLE.get(