From d43ce28f16392077cb191c462491b9a5af48f79f Mon Sep 17 00:00:00 2001
From: Jon Surrell <sirreal@users.noreply.github.com>
Date: Tue, 19 Dec 2023 15:20:47 +0100
Subject: [PATCH] Better tag finding

---
 tests/phpunit/tests/html-api/wpHtmlProcessorHtml5lib.php | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/tests/phpunit/tests/html-api/wpHtmlProcessorHtml5lib.php b/tests/phpunit/tests/html-api/wpHtmlProcessorHtml5lib.php
index 2eb898e476913..7c5eca8aa154f 100644
--- a/tests/phpunit/tests/html-api/wpHtmlProcessorHtml5lib.php
+++ b/tests/phpunit/tests/html-api/wpHtmlProcessorHtml5lib.php
@@ -185,8 +185,10 @@ public static function parse_html5_dat_testfile( $filename ) {
 					if ( '|' === $line[0] ) {
 						$candidate = substr( $line, 2 );
 						$trimmed   = trim( $candidate );
-						// We ignore `<!` starters to skip `<!--` (comment opener) and `<!DOCTYPE`.
-						if ( '<' === $trimmed[0] && '<!' !== substr( $trimmed, 0, 2 ) ) {
+						// Only take lines that look like tags
+						// At least 3 chars (< + tag + >)
+						// Tag must start with ascii alphabetic
+						if ( strlen( $trimmed > 2 ) && '<' === $trimmed[0] && ctype_alpha( $trimmed[1] ) ) {
 							$test_dom .= $candidate;
 						}
 					}