From 39ec1134fadd4749fc15cdc721ce978a637dde32 Mon Sep 17 00:00:00 2001
From: Turiiya <34311583+tobealive@users.noreply.github.com>
Date: Sun, 7 May 2023 02:55:02 +0200
Subject: [PATCH] net.html: fix parsing of nested quoted strings in code tags
 (#18123)

---
 vlib/net/html/parser.v      | 23 ++++++++++-------------
 vlib/net/html/parser_test.v |  2 +-
 2 files changed, 11 insertions(+), 14 deletions(-)

diff --git a/vlib/net/html/parser.v b/vlib/net/html/parser.v
index c54f1590f..6de0f7b77 100644
--- a/vlib/net/html/parser.v
+++ b/vlib/net/html/parser.v
@@ -111,32 +111,29 @@ fn (mut parser Parser) generate_tag() {
 pub fn (mut parser Parser) split_parse(data string) {
 	parser.init()
 	for chr in data {
-		// returns true if byte is a " or '
 		is_quote := chr == `"` || chr == `'`
 		string_code := match chr {
-			`"` { 1 } // "
-			`'` { 2 } // '
+			`"` { 1 }
+			`'` { 2 }
 			else { 0 }
 		}
-		if parser.lexical_attributes.open_code { // here will verify all needed to know if open_code finishes and string in code
+		if parser.lexical_attributes.open_code { // verify if open_code is complete and handle string code
 			parser.lexical_attributes.lexeme_builder.write_u8(chr)
 			if parser.lexical_attributes.open_string > 0
 				&& parser.lexical_attributes.open_string == string_code {
 				parser.lexical_attributes.open_string = 0
-			} else if is_quote {
-				parser.lexical_attributes.open_string = string_code
-			} else if chr == `>` { // only execute verification if is a > // here will verify < to know if code tag is finished
+			} else if chr == `>` { // code tag is finished
 				name_close_tag := '</${parser.lexical_attributes.opened_code_type}>'
 				if parser.builder_str().to_lower().ends_with(name_close_tag) {
 					parser.lexical_attributes.open_code = false
-					// need to modify lexeme_builder to add script text as a content in next loop (not gave error in dom)
+					// modify lexeme_builder to include script text as content in the next loop
 					parser.lexical_attributes.lexeme_builder.go_back(name_close_tag.len)
 					parser.lexical_attributes.current_tag.closed = true
 					parser.lexical_attributes.current_tag.close_type = .new_tag
 				}
 			}
 		} else if parser.lexical_attributes.open_comment {
-			if chr == `>` && parser.verify_end_comment(false) { // close tag '>'
+			if chr == `>` && parser.verify_end_comment(false) {
 				// parser.print_debug(parser.builder_str() + " >> " + parser.lexical_attributes.line_count.str())
 				parser.lexical_attributes.lexeme_builder.go_back_to(0)
 				parser.lexical_attributes.open_comment = false
@@ -174,10 +171,10 @@ pub fn (mut parser Parser) split_parse(data string) {
 			if parser.lexical_attributes.lexeme_builder.len == 0 && is_quote {
 				parser.lexical_attributes.open_string = string_code
 				parser.lexical_attributes.lexeme_builder.write_u8(chr)
-			} else if chr == `>` { // close tag >
+			} else if chr == `>` {
 				complete_lexeme := parser.builder_str().to_lower()
 				parser.lexical_attributes.current_tag.closed = (complete_lexeme.len > 0
-					&& complete_lexeme[complete_lexeme.len - 1] == `/`) // if equals to /
+					&& complete_lexeme[complete_lexeme.len - 1] == `/`)
 				if complete_lexeme.len > 0 && complete_lexeme[0] == `/` {
 					parser.dom.close_tags[complete_lexeme] = true
 				}
@@ -207,7 +204,7 @@ pub fn (mut parser Parser) split_parse(data string) {
 				} else {
 					parser.lexical_attributes.current_tag.attributes[complete_lexeme] = ''
 					parser.lexical_attributes.current_tag.last_attribute = ''
-					if chr == `=` { // if was a =
+					if chr == `=` {
 						parser.lexical_attributes.current_tag.last_attribute = complete_lexeme
 					}
 				}
@@ -216,7 +213,7 @@ pub fn (mut parser Parser) split_parse(data string) {
 			if parser.builder_str() == '!--' {
 				parser.lexical_attributes.open_comment = true
 			}
-		} else if chr == `<` { // open tag '<'
+		} else if chr == `<` {
 			temp_string := parser.builder_str()
 			if parser.lexical_attributes.lexeme_builder.len >= 1 {
 				if parser.lexical_attributes.current_tag.name.len > 1
diff --git a/vlib/net/html/parser_test.v b/vlib/net/html/parser_test.v
index 60ae631cd..e20ca529d 100644
--- a/vlib/net/html/parser_test.v
+++ b/vlib/net/html/parser_test.v
@@ -62,7 +62,7 @@ fn test_giant_string() {
 
 fn test_script_tag() {
 	mut parser := Parser{}
-	script_content := "\nvar googletag = googletag || {};\ngoogletag.cmd = googletag.cmd || [];if(3 > 5) {console.log('Birl');}\n"
+	script_content := '\nvar googletag = googletag || {};\ngoogletag.cmd = googletag.cmd || [];if(3 > 5) {console.log("Quoted \'message\'");}\n'
 	temp_html := '<html><body><script>${script_content}</script></body></html>'
 	parser.parse_html(temp_html)
 	assert parser.tags[2].content.len == script_content.len
-- 
2.39.5