소스 검색

LibWeb: Don't skip filtering when CSS contains null or surrogates

Gingeh 8 달 전
부모
커밋
4b1deb6fe1

+ 6 - 6
Libraries/LibWeb/CSS/Parser/Tokenizer.cpp

@@ -204,15 +204,15 @@ Vector<Token> Tokenizer::tokenize(StringView input, StringView encoding)
 
         auto decoded_input = MUST(decoder->to_utf8(input));
 
-        // OPTIMIZATION: If the input doesn't contain any CR or FF, we can skip the filtering
-        bool const contains_cr_or_ff = [&] {
-            for (auto byte : decoded_input.bytes()) {
-                if (byte == '\r' || byte == '\f')
+        // OPTIMIZATION: If the input doesn't contain any filterable characters, we can skip the filtering
+        bool const contains_filterable = [&] {
+            for (auto code_point : decoded_input.code_points()) {
+                if (code_point == '\r' || code_point == '\f' || code_point == 0x00 || is_unicode_surrogate(code_point))
                     return true;
             }
             return false;
         }();
-        if (!contains_cr_or_ff) {
+        if (!contains_filterable) {
             return decoded_input;
         }
 
@@ -242,7 +242,7 @@ Vector<Token> Tokenizer::tokenize(StringView input, StringView encoding)
                 } else if (code_point == '\f') {
                     builder.append('\n');
                     // Replace any U+0000 NULL or surrogate code points in input with U+FFFD REPLACEMENT CHARACTER (�).
-                } else if (code_point == 0x00 || (code_point >= 0xD800 && code_point <= 0xDFFF)) {
+                } else if (code_point == 0x00 || is_unicode_surrogate(code_point)) {
                     builder.append_code_point(REPLACEMENT_CHARACTER);
                 } else {
                     builder.append_code_point(code_point);

BIN
Tests/LibWeb/Text/expected/wpt-import/css/css-syntax/input-preprocessing.txt


+ 46 - 0
Tests/LibWeb/Text/input/wpt-import/css/css-syntax/input-preprocessing.html

@@ -0,0 +1,46 @@
+<!doctype html>
+<title>Input Preprocessing</title>
+<script src="../../resources/testharness.js"></script>
+<script src="../../resources/testharnessreport.js"></script>
+<style>
+
+foo { color: blue; }
+
+</style>
+
+<meta name="author" title="Tab Atkins-Bittner">
+<link rel=help href="https://drafts.csswg.org/css-syntax/#input-preprocessing">
+
+<script>
+
+function roundtripIdent(str) {
+    const rule = document.styleSheets[0].cssRules[0];
+    rule.selectorText = "original-ident";
+    rule.selectorText = str;
+    // Check for parse error.
+    if(rule.selectorText == "original-ident") return "parse error";
+    return rule.selectorText;
+}
+function testParsing(input, expected) {
+    test(()=>{
+        assert_equals(roundtripIdent(input), expected);
+    }, `"${input}" becomes "${expected}"`);
+}
+
+/* Can't figure out how to test the newline normalization... */
+
+/* NULL becomes FFFD */
+testParsing("foo\x00", "foo\ufffd");
+testParsing("f\x00oo", "f\ufffdoo");
+testParsing("\x00foo", "\ufffdfoo");
+testParsing("\x00", "\ufffd");
+testParsing("\x00\x00\x00", "\ufffd\ufffd\ufffd");
+
+/* surrogates become FFFD */
+testParsing("foo\ud800", "foo\ufffd");
+testParsing("f\ud800oo", "f\ufffdoo");
+testParsing("\ud800foo", "\ufffdfoo");
+testParsing("\ud800", "\ufffd");
+testParsing("\ud800\ud800\ud800", "\ufffd\ufffd\ufffd");
+
+</script>