diff --git a/src/netsurf.zig b/src/netsurf.zig
index b24c5d26..dca13979 100644
--- a/src/netsurf.zig
+++ b/src/netsurf.zig
@@ -1445,6 +1445,13 @@ pub fn documentHTMLParse(reader: anytype, enc: ?[:0]const u8) !*DocumentHTML {
     while (ln > 0) {
         ln = try reader.read(&buffer);
         err = c.dom_hubbub_parser_parse_chunk(parser, &buffer, ln);
+        // TODO handle encoding change error return.
+        // When the HTML contains a META tag with a different encoding than the
+        // original one, a c.DOM_HUBBUB_HUBBUB_ERR_ENCODINGCHANGE error is
+        // returned.
+        // In this case, we must restart the parsing with the new detected
+        // encoding. The detected encoding is stored in the document and we can
+        // get it with documentGetInputEncoding().
         try parserErr(err);
     }
 
diff --git a/src/run_tests.zig b/src/run_tests.zig
index 8369ec61..1ba0f8b3 100644
--- a/src/run_tests.zig
+++ b/src/run_tests.zig
@@ -119,6 +119,6 @@ test "bug document html parsing #4" {
     const file = try std.fs.cwd().openFile("tests/html/bug-html-parsing-4.html", .{});
     defer file.close();
 
-    doc = try parser.documentHTMLParse(file.reader(), null);
+    doc = try parser.documentHTMLParse(file.reader(), "UTF-8");
     parser.documentHTMLClose(doc) catch {};
 }