JavaScriptCore:
Reviewed by Darin.
- fixed <rdar://problem/3800315> encode-URI-test layout test is failing
* kjs/function.cpp:
(KJS::GlobalFuncImp::call): Make sure to escape null
characters. This is a bug in the new code that made part of the
test fail.
WebCore:
Reviewed by Darin.
- fixed <rdar://problem/3800315> encode-URI-test layout test is failing
- updated escape/encodeURI layout test to match our new, more
compatible behavior.
* layout-tests/fast/js/global/encode-URI-test.html:
git-svn-id: http://svn.webkit.org/repository/webkit/trunk@7558 268f45cc-cd09-0410-ab3c-d52691b4dbfc
diff --git a/LayoutTests/fast/js/global/encode-URI-test.html b/LayoutTests/fast/js/global/encode-URI-test.html
index 3ed9682..2cabd33 100644
--- a/LayoutTests/fast/js/global/encode-URI-test.html
+++ b/LayoutTests/fast/js/global/encode-URI-test.html
@@ -54,7 +54,7 @@
return "\"" + p + "\"";
}
-function escapedCharacter(c)
+function encodedCharacter(c)
{
// UTF-8 is what Gecko does, but not what WinIE 6 does.
// It makes much more sense, though, to produce encodings that actually work in URLs.
@@ -71,9 +71,30 @@
case 0xE000: return "%EE%80%80";
case 0xFFFC: return "%EF%BF%BC";
case 0xFFFD: return "%EF%BF%BD";
- case 0xFFFE: return "%EF%BF%BE";
- case 0xFFFF: return "%EF%BF%BF";
- case 0x10000: return "%F0%90%80%80";
+ }
+
+ if (c < 0 || c > 0x7F) {
+ window.alert("encodedCharacter doesn't know how to escape character code " + c);
+ return "?";
+ }
+
+ return "%" + hexDigit(c >> 4) + hexDigit(c - (c >> 4 << 4));
+}
+
+function escapedCharacter(c)
+{
+ // UTF-8 is what Gecko does, but we do what Win IE does instead. We use Latin1 for
+ // unicode characters less than or equal to 255, and %u encoding for everything else.
+ switch (c) {
+ case 0x80: return "%80";
+ case 0x7FF: return "%u07FF";
+ case 0x800: return "%u0800";
+ case 0x2022: return "%u2022";
+ case 0xD7FF: return "%uD7FF";
+ case 0xD800: return "%uD800";
+ case 0xE000: return "%uE000";
+ case 0xFFFC: return "%uFFFC";
+ case 0xFFFD: return "%uFFFD";
}
if (c < 0 || c > 0x7F) {
@@ -133,8 +154,8 @@
function addURINoEscape(c)
{
addExpectedNonEscaped("encodeURI", c);
- expectedResult["decodeURI(" + escapedCharacter(c) + ")"] = escapedCharacter(c);
- expectedResult["decodeURI(" + escapedCharacter(c).toLowerCase() + ")"] = escapedCharacter(c).toLowerCase();
+ expectedResult["decodeURI(" + escapedCharacter(c) + ")"] = encodedCharacter(c);
+ expectedResult["decodeURI(" + escapedCharacter(c).toLowerCase() + ")"] = encodedCharacter(c).toLowerCase();
}
forEachCharacterCode(addNoEscape, "*0123456789-.ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz_");
@@ -164,11 +185,11 @@
}
}
-function testEscape(parameter, expected)
+function testEscapeAndEncode(parameter, escapeExpected, encodeExpected)
{
- test("escape", parameter, expected);
- test("encodeURI", parameter, expected);
- test("encodeURIComponent", parameter, expected);
+ test("escape", parameter, escapeExpected);
+ test("encodeURI", parameter, encodeExpected);
+ test("encodeURIComponent", parameter, encodeExpected);
}
function testUnescape(parameter, expected)
@@ -195,12 +216,13 @@
{
var s = String.fromCharCode(c);
var escaped = escapedCharacter(c);
+ var encoded = encodedCharacter(c);
- testEscape(s, escaped);
+ testEscapeAndEncode(s, escaped, encoded);
testUnescape(escaped, s);
testUnescape(escaped.toLowerCase(), s);
- testDecode(escaped, s);
- testDecode(escaped.toLowerCase(), s);
+ testDecode(encoded, s);
+ testDecode(encoded.toLowerCase(), s);
}
for (var c = 0; c <= 128; c++) {
@@ -240,21 +262,21 @@
testUnescapeAndDecode(String.fromCharCode(0x80), String.fromCharCode(0x80), String.fromCharCode(0x80));
testUnescapeAndDecode(String.fromCharCode(0xD800), String.fromCharCode(0xD800), String.fromCharCode(0xD800));
-testUnescapeAndDecode("%C2%80", String.fromCharCode(0x80), String.fromCharCode(0x80));
-testUnescapeAndDecode("%C2", "%C2", "exception");
-testUnescapeAndDecode("%C2" + String.fromCharCode(0x80), "%C2" + String.fromCharCode(0x80), "exception");
+testUnescapeAndDecode("%C2%80", String.fromCharCode(0xC2) + String.fromCharCode(0x80), String.fromCharCode(0x80));
+testUnescapeAndDecode("%C2", String.fromCharCode(0xC2), "exception");
+testUnescapeAndDecode("%C2" + String.fromCharCode(0x80), String.fromCharCode(0xC2) + String.fromCharCode(0x80), "exception");
// The characters below hav to be literal because String.fromCharCode will make a single character
// and the \u syntax won't allow us to specify characters with Unicode values higher than U+FFFF.
// For most JavaScript engines, this will turn into two characters because they use UTF-16
// instead of Unicode; it's not clear to me at the moment if the standard asks for this UTF-16
// behavior, forbids it, or doesn't say either way.
-testEscape("𐀀", "%F0%90%80%80");
-testUnescapeAndDecode("%F0%90%80%80", "𐀀", "𐀀");
-testEscape("𦏵", "%F0%A6%8F%B5");
-testUnescapeAndDecode("%F0%A6%8F%B5", "𦏵", "𦏵");
-testEscape("", "%F0%AF%BF%BF");
-testUnescapeAndDecode("%F0%AF%BF%BF", "", "");
+testEscapeAndEncode("𐀀", "%uD800%uDC00", "%F0%90%80%80");
+testUnescapeAndDecode("%F0%90%80%80", "\xF0\x90\x80\x80", "𐀀");
+testEscapeAndEncode("𦏵", "%uD858%uDFF5", "%F0%A6%8F%B5");
+testUnescapeAndDecode("%F0%A6%8F%B5", "\xF0\xA6\x8F\xB5", "𦏵");
+testEscapeAndEncode("", "%uD87F%uDFFF", "%F0%AF%BF%BF");
+testUnescapeAndDecode("%F0%AF%BF%BF", "\xF0\xAF\xBF\xBF", "");
// --------
