moved array of characters to own constant

src/compiler/parse/state/tag.ts

@@ -11,6 +11,7 @@ import { closing_tag_omitted, decode_character_references } from '../utils/html'
 
 // characters equivalent to using \s in regex
 const whitespace_characters = ['\r', '\n', '\t', '\f', '\v', ' ', '\u00a0', '\u1680', '\u2000', '\u2001', '\u2002', '\u2003', '\u2004', '\u2005', '\u2006', '\u2007', '\u2008', '\u2009', '\u200a', '\u2028', '\u2029', '\u202f', '\u205f', '\u3000', '\ufeff'];
+const invalid_attribute_characters = [...whitespace_characters, '/', '"', '\'', '=', '<', '>', '`'];
 
 // eslint-disable-next-line no-useless-escape
 const valid_tag_name = /^\!?[a-zA-Z]{1,}:?[a-zA-Z0-9\-]*/;
@@ -367,7 +368,6 @@ function read_attribute(parser: Parser, unique_names: Set<string>) {
 		parser.error(parser_errors.unexpected_token('='), parser.index);
 	}
 
-
 	if (type) {
 		const [directive_name, ...modifiers] = name.slice(colon_index + 1).split('|');
 
@@ -473,8 +473,10 @@ function read_attribute_value(parser: Parser) {
 
 	let value;
 	try {
-		const characters = quote_mark ? [quote_mark] : [...whitespace_characters, '/>', '"', '\'', '=', '<', '>', '`'];
+		value = read_sequence(parser, () => {
-		value = read_sequence(parser, () => characters.some(c => parser.match(c)));
+			if (quote_mark) return parser.match(quote_mark);
+			return invalid_attribute_characters.some(c => parser.match(c));
+		});
 	} catch (error) {
 		if (error.code === 'parse-error') {
 			// if the attribute value didn't close + self-closing tag