Fix GH-21041: Dom\HTMLDocument corrupts closing tags within scripts

lexborisov · ndossche · commit 56e5a80c442b · 2026-01-26T19:20:10.000+01:00
diff --git a/NEWS b/NEWS
@@ -20,6 +20,10 @@ PHP                                                                        NEWS
 - Date:
   . Update timelib to 2022.16. (Derick)
 
+- DOM:
+  . Fixed GH-21041 (Dom\HTMLDocument corrupts closing tags within scripts).
+    (lexborisov)
+
 - MbString:
   . Fixed bug GH-20833 (mb_str_pad() divide by zero if padding string is
     invalid in the encoding). (ndossche)
diff --git a/ext/dom/lexbor/lexbor/html/tokenizer/state_script.c b/ext/dom/lexbor/lexbor/html/tokenizer/state_script.c
@@ -364,6 +364,7 @@ lxb_html_tokenizer_state_script_data_end_tag_name(lxb_html_tokenizer_t *tkz,
                 if (lexbor_str_res_alpha_character[*data]
                     == LEXBOR_STR_RES_SLIP)
                 {
+                    lxb_html_tokenizer_state_append_data_m(tkz, data);
                     goto anything_else;
                 }
 

Original file line number	Diff line number	Diff line change
`@@ -364,6 +364,7 @@ lxb_html_tokenizer_state_script_data_end_tag_name(lxb_html_tokenizer_t *tkz,`
`364`	`364`	`if (lexbor_str_res_alpha_character[*data]`
`365`	`365`	`== LEXBOR_STR_RES_SLIP)`
`366`	`366`	`{`
	`367`	`+ lxb_html_tokenizer_state_append_data_m(tkz, data);`
`367`	`368`	`goto anything_else;`
`368`	`369`	`}`
`369`	`370`