[ie/matchtv] Fix extractor (#10190)

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

diff --git a/yt_dlp/extractor/eporner.py b/yt_dlp/extractor/eporner.py
index bfecd3a418805761fc3516cf7afb1e14f9d16283..e0b6ca96fb743429cab610db14d985e28d60898e 100644 (file)
--- a/yt_dlp/extractor/eporner.py
+++ b/yt_dlp/extractor/eporner.py
@@ -1,13 +1,10 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
-import re
-
 from .common import InfoExtractor
 from ..utils import (
-    encode_base_n,
     ExtractorError,
+    encode_base_n,
+    get_elements_by_class,
     int_or_none,
+    join_nonempty,
     merge_dicts,
     parse_duration,
     str_to_int,
@@ -33,8 +30,8 @@ class EpornerIE(InfoExtractor):
             'age_limit': 18,
         },
         'params': {
-            'proxy': '127.0.0.1:8118'
-        }
+            'proxy': '127.0.0.1:8118',
+        },
     }, {
         # New (May 2016) URL layout
         'url': 'http://www.eporner.com/hd-porn/3YRUtzMcWn0/Star-Wars-XXX-Parody/',
@@ -51,15 +48,15 @@ class EpornerIE(InfoExtractor):
     }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        mobj = self._match_valid_url(url)
         video_id = mobj.group('id')
         display_id = mobj.group('display_id') or video_id
 
         webpage, urlh = self._download_webpage_handle(url, display_id)
 
-        video_id = self._match_id(urlh.geturl())
+        video_id = self._match_id(urlh.url)
 
-        hash = self._search_regex(
+        vid_hash = self._search_regex(
             r'hash\s*[:=]\s*["\']([\da-f]{32})', webpage, 'hash')
 
         title = self._og_search_title(webpage, default=None) or self._html_search_regex(
@@ -67,13 +64,13 @@ def _real_extract(self, url):
 
         # Reverse engineered from vjs.js
         def calc_hash(s):
-            return ''.join((encode_base_n(int(s[lb:lb + 8], 16), 36) for lb in range(0, 32, 8)))
+            return ''.join(encode_base_n(int(s[lb:lb + 8], 16), 36) for lb in range(0, 32, 8))
 
         video = self._download_json(
-            'http://www.eporner.com/xhr/video/%s' % video_id,
+            f'http://www.eporner.com/xhr/video/{video_id}',
             display_id, note='Downloading video JSON',
             query={
-                'hash': calc_hash(hash),
+                'hash': calc_hash(vid_hash),
                 'device': 'generic',
                 'domain': 'www.eporner.com',
                 'fallback': 'false',
@@ -81,11 +78,12 @@ def calc_hash(s):
 
         if video.get('available') is False:
             raise ExtractorError(
-                '%s said: %s' % (self.IE_NAME, video['message']), expected=True)
+                '{} said: {}'.format(self.IE_NAME, video['message']), expected=True)
 
         sources = video['sources']
 
         formats = []
+        has_av1 = bool(get_elements_by_class('download-av1', webpage))
         for kind, formats_dict in sources.items():
             if not isinstance(formats_dict, dict):
                 continue
@@ -111,7 +109,14 @@ def calc_hash(s):
                         'height': height,
                         'fps': fps,
                     })
-        self._sort_formats(formats)
+                    if has_av1:
+                        formats.append({
+                            'url': src.replace('.mp4', '-av1.mp4'),
+                            'format_id': join_nonempty('av1', format_id),
+                            'height': height,
+                            'fps': fps,
+                            'vcodec': 'av1',
+                        })
 
         json_ld = self._search_json_ld(webpage, display_id, default={})