From 80e8493ee7c3083f4e215794e4a67ba5265f24f7 Mon Sep 17 00:00:00 2001
From: pukkandan <pukkandan.ytdlp@gmail.com>
Date: Wed, 18 May 2022 06:42:43 +0530
Subject: [PATCH] [utils] `is_html`: Handle double BOM

Closes #2885
---
 yt_dlp/extractor/generic.py | 15 +++++++++++++++
 yt_dlp/utils.py             | 13 ++++++-------
 2 files changed, 21 insertions(+), 7 deletions(-)

diff --git a/yt_dlp/extractor/generic.py b/yt_dlp/extractor/generic.py
index dda2b1eef..b0fc176ef 100644
--- a/yt_dlp/extractor/generic.py
+++ b/yt_dlp/extractor/generic.py
@@ -2527,6 +2527,21 @@ class GenericIE(InfoExtractor):
                 'upload_date': '20220504',
             },
         },
+        {
+            # Webpage contains double BOM
+            'url': 'https://www.filmarkivet.se/movies/paris-d-moll/',
+            'md5': 'df02cadc719dcc63d43288366f037754',
+            'info_dict': {
+                'id': 'paris-d-moll',
+                'ext': 'mp4',
+                'upload_date': '20220518',
+                'title': 'Paris d-moll',
+                'description': 'md5:319e37ea5542293db37e1e13072fe330',
+                'thumbnail': 'https://www.filmarkivet.se/wp-content/uploads/parisdmoll2.jpg',
+                'timestamp': 1652833414,
+                'age_limit': 0,
+            }
+        }
     ]
 
     def report_following_redirect(self, new_url):
diff --git a/yt_dlp/utils.py b/yt_dlp/utils.py
index 48a94415d..3b0e6750c 100644
--- a/yt_dlp/utils.py
+++ b/yt_dlp/utils.py
@@ -3290,14 +3290,13 @@ def is_html(first_bytes):
         (b'\xff\xfe', 'utf-16-le'),
         (b'\xfe\xff', 'utf-16-be'),
     ]
-    for bom, enc in BOMS:
-        if first_bytes.startswith(bom):
-            s = first_bytes[len(bom):].decode(enc, 'replace')
-            break
-    else:
-        s = first_bytes.decode('utf-8', 'replace')
 
-    return re.match(r'^\s*<', s)
+    encoding = 'utf-8'
+    for bom, enc in BOMS:
+        while first_bytes.startswith(bom):
+            encoding, first_bytes = enc, first_bytes[len(bom):]
+
+    return re.match(r'^\s*<', first_bytes.decode(encoding, 'replace'))
 
 
 def determine_protocol(info_dict):