public void Ctor_WithUnicodeRanges()
{
// Arrange
JavaScriptStringEncoder encoder = new JavaScriptStringEncoder(UnicodeRanges.Latin1Supplement, UnicodeRanges.MiscellaneousSymbols);
// Act & assert
Assert.Equal(@"\u0061", encoder.JavaScriptStringEncode("a"));
Assert.Equal("\u00E9", encoder.JavaScriptStringEncode("\u00E9" /* LATIN SMALL LETTER E WITH ACUTE */));
Assert.Equal("\u2601", encoder.JavaScriptStringEncode("\u2601" /* CLOUD */));
}
public void Ctor_WithUnicodeRanges()
{
// Arrange
JavaScriptStringEncoder encoder = new JavaScriptStringEncoder(UnicodeRanges.Latin1Supplement, UnicodeRanges.MiscellaneousSymbols);
// Act & assert
Assert.Equal(@"\u0061", encoder.JavaScriptStringEncode("a"));
Assert.Equal("\u00E9", encoder.JavaScriptStringEncode("\u00E9" /* LATIN SMALL LETTER E WITH ACUTE */));
Assert.Equal("\u2601", encoder.JavaScriptStringEncode("\u2601" /* CLOUD */));
}
public void Ctor_WithNoParameters_DefaultsToBasicLatin()
{
// Arrange
JavaScriptStringEncoder encoder = new JavaScriptStringEncoder();
// Act & assert
Assert.Equal("a", encoder.JavaScriptStringEncode("a"));
Assert.Equal(@"\u00E9", encoder.JavaScriptStringEncode("\u00E9" /* LATIN SMALL LETTER E WITH ACUTE */));
Assert.Equal(@"\u2601", encoder.JavaScriptStringEncode("\u2601" /* CLOUD */));
}
public void Ctor_WithNoParameters_DefaultsToBasicLatin()
{
// Arrange
JavaScriptStringEncoder encoder = new JavaScriptStringEncoder();
// Act & assert
Assert.Equal("a", encoder.JavaScriptStringEncode("a"));
Assert.Equal(@"\u00E9", encoder.JavaScriptStringEncode("\u00E9" /* LATIN SMALL LETTER E WITH ACUTE */));
Assert.Equal(@"\u2601", encoder.JavaScriptStringEncode("\u2601" /* CLOUD */));
}
public void JavaScriptStringEncode_NoRangesAllowed_EmitsShortFormForCertainCodePoints()
{
// This test ensures that when we're encoding, we always emit the "\uXXXX" form of the
// code point except for very specific code points where we allow a shorter representation.
// Arrange
JavaScriptStringEncoder encoder = new JavaScriptStringEncoder(UnicodeRanges.None); // allow no codepoints
// "[U+0000][U+0001]...[U+007F]"
string input = new string(Enumerable.Range(0, 128).Select(i => (char)i).ToArray());
// @"\u0000\u0001..\u007F", then replace certain specific code points
string expected = string.Concat(Enumerable.Range(0, 128).Select(i => FormattableString.Invariant($@"\u{i:X4}")));
expected = expected.Replace(@"\u0008", @"\b"); // U+0008 BACKSPACE -> "\b"
expected = expected.Replace(@"\u0009", @"\t"); // U+0009 CHARACTER TABULATION -> "\t"
expected = expected.Replace(@"\u000A", @"\n"); // U+000A LINE FEED -> "\n"
expected = expected.Replace(@"\u000C", @"\f"); // U+000C FORM FEED -> "\f"
expected = expected.Replace(@"\u000D", @"\r"); // U+000D CARRIAGE RETURN -> "\n"
expected = expected.Replace(@"\u005C", @"\\"); // U+005C REVERSE SOLIDUS -> "\\"
// Act
string retVal = encoder.JavaScriptStringEncode(input);
// Assert
Assert.Equal(expected, retVal);
}
public void JavaScriptStringEncode_NullInput_Throws()
{
// Arrange
JavaScriptStringEncoder encoder = new JavaScriptStringEncoder();
Assert.Throws <ArgumentNullException>(() => { encoder.JavaScriptStringEncode(null); });
}
public void JavaScriptStringEncode_NullInput_Throws_Relaxed()
{
// Arrange
JavaScriptStringEncoder encoder = JavaScriptStringEncoder.UnsafeRelaxedJsonEscaping;
Assert.Throws <ArgumentNullException>(() => { encoder.JavaScriptStringEncode(null); });
}
public void JavaScriptStringEncode_EmptyStringInput_ReturnsEmptyString_Relaxed()
{
// Arrange
JavaScriptStringEncoder encoder = JavaScriptStringEncoder.UnsafeRelaxedJsonEscaping;
// Act & assert
Assert.Equal("", encoder.JavaScriptStringEncode(""));
}
public void JavaScriptStringEncode_EmptyStringInput_ReturnsEmptyString()
{
// Arrange
JavaScriptStringEncoder encoder = new JavaScriptStringEncoder();
// Act & assert
Assert.Equal("", encoder.JavaScriptStringEncode(""));
}
public void JavaScriptStringEncode_NullInput_ReturnsNull()
{
// Arrange
JavaScriptStringEncoder encoder = new JavaScriptStringEncoder();
// Act & assert
Assert.Null(encoder.JavaScriptStringEncode(null));
}
public void JavaScriptStringEncode_InputDoesNotRequireEncoding_ReturnsOriginalStringInstance_Relaxed()
{
// Arrange
JavaScriptStringEncoder encoder = JavaScriptStringEncoder.UnsafeRelaxedJsonEscaping;
string input = "Hello, there!";
// Act & assert
Assert.Same(input, encoder.JavaScriptStringEncode(input));
}
public void JavaScriptStringEncode_InputDoesNotRequireEncoding_ReturnsOriginalStringInstance()
{
// Arrange
JavaScriptStringEncoder encoder = new JavaScriptStringEncoder();
string input = "Hello, there!";
// Act & assert
Assert.Same(input, encoder.JavaScriptStringEncode(input));
}
public void JavaScriptStringEncode_AllRangesAllowed_StillEncodesForbiddenChars_Simple(string input, string expected)
{
// Arrange
JavaScriptStringEncoder encoder = new JavaScriptStringEncoder(UnicodeRanges.All);
// Act
string retVal = encoder.JavaScriptStringEncode(input);
// Assert
Assert.Equal(expected, retVal);
}
public void JavaScriptStringEncode_Quotes_Relaxed(string input, string expected)
{
// Arrange
JavaScriptStringEncoder encoder = JavaScriptStringEncoder.UnsafeRelaxedJsonEscaping;
// Act
string retVal = encoder.JavaScriptStringEncode(input);
// Assert
Assert.Equal(expected, retVal);
}
public void JavaScriptStringEncode_Relaxed_StillEncodesForbiddenChars_Simple(string input, string expected)
{
// Arrange
JavaScriptStringEncoder encoder = JavaScriptStringEncoder.UnsafeRelaxedJsonEscaping;
// Act
string retVal = encoder.JavaScriptStringEncode(input);
// Assert
Assert.Equal(expected, retVal);
}
public void JavaScriptStringEncode_DoesOutputHtmlSensitiveCharacters_Relaxed(string input, string expected)
{
// Arrange
JavaScriptStringEncoder encoder = JavaScriptStringEncoder.UnsafeRelaxedJsonEscaping;
// Act
string retVal = encoder.JavaScriptStringEncode(input);
// Assert
Assert.Equal(expected, retVal);
}
public void JavaScriptStringEncode_PositiveTestCase()
{
// Arrange
IJavaScriptStringEncoder encoder = new JavaScriptStringEncoder(UnicodeRanges.All);
StringWriter writer = new StringWriter();
// Act
encoder.JavaScriptStringEncode("Hello+there!", writer);
// Assert
Assert.Equal(@"Hello\u002Bthere!", writer.ToString());
}
public void JavaScriptStringEncode_StringSubstring()
{
// Arrange
JavaScriptStringEncoder encoder = new JavaScriptStringEncoder();
var output = new StringWriter();
// Act
encoder.JavaScriptStringEncode("Hello+world!", 3, 5, output);
// Assert
Assert.Equal(@"lo\u002Bwo", output.ToString());
}
public void JavaScriptStringEncode_PositiveTestCase()
{
// Arrange
IJavaScriptStringEncoder encoder = new JavaScriptStringEncoder(UnicodeRanges.All);
StringWriter writer = new StringWriter();
// Act
encoder.JavaScriptStringEncode("Hello+there!", writer);
// Assert
Assert.Equal(@"Hello\u002Bthere!", writer.ToString());
}
public void JavaScriptStringEncode_StringSubstring_Relaxed()
{
// Arrange
JavaScriptStringEncoder encoder = JavaScriptStringEncoder.UnsafeRelaxedJsonEscaping;
using var output = new StringWriter();
// Act
encoder.JavaScriptStringEncode("Hello\\world!", 3, 5, output);
// Assert
Assert.Equal(@"lo\\wo", output.ToString());
}
public void Relaxed_EquivalentToAll_WithExceptions()
{
// Arrange
JavaScriptStringEncoder controlEncoder = new JavaScriptStringEncoder(UnicodeRanges.All);
JavaScriptStringEncoder testEncoder = JavaScriptStringEncoder.UnsafeRelaxedJsonEscaping;
// Act & assert
for (int i = 0; i <= char.MaxValue; i++)
{
if (i == '"' || i == '&' || i == '<' || i == '>' || i == '+' || i == '\'' || i == '`')
{
string input = new string((char)i, 1);
Assert.NotEqual(controlEncoder.JavaScriptStringEncode(input), testEncoder.JavaScriptStringEncode(input));
continue;
}
if (!IsSurrogateCodePoint(i))
{
string input = new string((char)i, 1);
Assert.Equal(controlEncoder.JavaScriptStringEncode(input), testEncoder.JavaScriptStringEncode(input));
}
}
}
public void JavaScriptStringEncode_BadSurrogates_ReturnsUnicodeReplacementChar()
{
// Arrange
JavaScriptStringEncoder encoder = new JavaScriptStringEncoder(UnicodeRanges.All); // allow all codepoints
// "a<unpaired leading>b<unpaired trailing>c<trailing before leading>d<unpaired trailing><valid>e<high at end of string>"
const string input = "a\uD800b\uDFFFc\uDFFF\uD800d\uDFFF\uD800\uDFFFe\uD800";
const string expected = "a\uFFFDb\uFFFDc\uFFFD\uFFFDd\uFFFD\\uD800\\uDFFFe\uFFFD"; // 'D800' 'DFFF' was preserved since it's valid
// Act
string retVal = encoder.JavaScriptStringEncode(input);
// Assert
Assert.Equal(expected, retVal);
}
public void Default_EquivalentToBasicLatin()
{
// Arrange
JavaScriptStringEncoder controlEncoder = new JavaScriptStringEncoder(UnicodeRanges.BasicLatin);
JavaScriptStringEncoder testEncoder = JavaScriptStringEncoder.Default;
// Act & assert
for (int i = 0; i <= Char.MaxValue; i++)
{
if (!IsSurrogateCodePoint(i))
{
string input = new String((char)i, 1);
Assert.Equal(controlEncoder.JavaScriptStringEncode(input), testEncoder.JavaScriptStringEncode(input));
}
}
}
public void JavaScriptStringEncode_Quotes(string input, string expected)
{
// Per the design document, we provide additional defense-in-depth
// against breaking out of HTML attributes by having the encoders
// never emit the ' or " characters. This means that we want to
// \u-escape these characters instead of using \' and \".
// Arrange
JavaScriptStringEncoder encoder = new JavaScriptStringEncoder(UnicodeRanges.All);
// Act
string retVal = encoder.JavaScriptStringEncode(input);
// Assert
Assert.Equal(expected, retVal);
}
public void Default_EquivalentToBasicLatin()
{
// Arrange
JavaScriptStringEncoder controlEncoder = new JavaScriptStringEncoder(UnicodeRanges.BasicLatin);
JavaScriptStringEncoder testEncoder = JavaScriptStringEncoder.Default;
// Act & assert
for (int i = 0; i <= Char.MaxValue; i++)
{
if (!IsSurrogateCodePoint(i))
{
string input = new String((char)i, 1);
Assert.Equal(controlEncoder.JavaScriptStringEncode(input), testEncoder.JavaScriptStringEncode(input));
}
}
}
public void Ctor_WithCodePointFilter()
{
// Arrange
var filter = new CodePointFilter().AllowChars("ab").AllowChars('\0', '&', '\uFFFF', 'd');
JavaScriptStringEncoder encoder = new JavaScriptStringEncoder(filter);
// Act & assert
Assert.Equal("a", encoder.JavaScriptStringEncode("a"));
Assert.Equal("b", encoder.JavaScriptStringEncode("b"));
Assert.Equal(@"\u0063", encoder.JavaScriptStringEncode("c"));
Assert.Equal("d", encoder.JavaScriptStringEncode("d"));
Assert.Equal(@"\u0000", encoder.JavaScriptStringEncode("\0")); // we still always encode control chars
Assert.Equal(@"\u0026", encoder.JavaScriptStringEncode("&")); // we still always encode HTML-special chars
Assert.Equal(@"\uFFFF", encoder.JavaScriptStringEncode("\uFFFF")); // we still always encode non-chars and other forbidden chars
}
public void Ctor_WithCodePointFilter()
{
// Arrange
var filter = new CodePointFilter().AllowChars("ab").AllowChars('\0', '&', '\uFFFF', 'd');
JavaScriptStringEncoder encoder = new JavaScriptStringEncoder(filter);
// Act & assert
Assert.Equal("a", encoder.JavaScriptStringEncode("a"));
Assert.Equal("b", encoder.JavaScriptStringEncode("b"));
Assert.Equal(@"\u0063", encoder.JavaScriptStringEncode("c"));
Assert.Equal("d", encoder.JavaScriptStringEncode("d"));
Assert.Equal(@"\u0000", encoder.JavaScriptStringEncode("\0")); // we still always encode control chars
Assert.Equal(@"\u0026", encoder.JavaScriptStringEncode("&")); // we still always encode HTML-special chars
Assert.Equal(@"\uFFFF", encoder.JavaScriptStringEncode("\uFFFF")); // we still always encode non-chars and other forbidden chars
}
public void JavaScriptStringEncode_ControlCharacters_Relaxed()
{
// Arrange
JavaScriptStringEncoder encoder = JavaScriptStringEncoder.UnsafeRelaxedJsonEscaping;
// Act & assert
for (int i = 0; i <= 0x1F; i++)
{
// Skip characters that are escaped using '\\' since they are covered in other tests.
if (i == '\b' || i == '\f' || i == '\n' || i == '\r' || i == '\t')
{
continue;
}
string javaScriptStringEncoded = encoder.JavaScriptStringEncode(char.ConvertFromUtf32(i));
string expected = string.Format("\\u00{0:X2}", i);
Assert.Equal(expected, javaScriptStringEncoded);
}
}
public void JavaScriptStringEncode_DoesNotOutputHtmlSensitiveCharacters()
{
// Per the design document, we provide additional defense-in-depth
// by never emitting HTML-sensitive characters unescaped.
// Arrange
JavaScriptStringEncoder javaScriptStringEncoder = new JavaScriptStringEncoder(UnicodeRanges.All);
HtmlEncoder htmlEncoder = new HtmlEncoder(UnicodeRanges.All);
// Act & assert
for (int i = 0; i <= 0x10FFFF; i++)
{
if (IsSurrogateCodePoint(i))
{
continue; // surrogates don't matter here
}
string javaScriptStringEncoded = javaScriptStringEncoder.JavaScriptStringEncode(Char.ConvertFromUtf32(i));
string thenHtmlEncoded = htmlEncoder.HtmlEncode(javaScriptStringEncoded);
Assert.Equal(javaScriptStringEncoded, thenHtmlEncoded); // should have contained no HTML-sensitive characters
}
}
public void JavaScriptStringEncode_AboveAscii_Relaxed()
{
// Arrange
JavaScriptStringEncoder encoder = JavaScriptStringEncoder.UnsafeRelaxedJsonEscaping;
// Act & assert
for (int i = 0x128; i <= 0xFFFF; i++)
{
if (IsSurrogateCodePoint(i))
{
continue; // surrogates don't matter here
}
UnicodeCategory category = char.GetUnicodeCategory((char)i);
if (category != UnicodeCategory.NonSpacingMark)
{
continue; // skip undefined characters like U+0378, or spacing characters like U+2028
}
string javaScriptStringEncoded = encoder.JavaScriptStringEncode(char.ConvertFromUtf32(i));
Assert.True(char.ConvertFromUtf32(i) == javaScriptStringEncoded, i.ToString());
}
}
public void JavaScriptStringEncode_NullInput_ReturnsNull()
{
// Arrange
JavaScriptStringEncoder encoder = new JavaScriptStringEncoder();
// Act & assert
Assert.Null(encoder.JavaScriptStringEncode(null));
}
public void JavaScriptStringEncode_AllRangesAllowed_StillEncodesForbiddenChars_Simple(string input, string expected)
{
// Arrange
JavaScriptStringEncoder encoder = new JavaScriptStringEncoder(UnicodeRanges.All);
// Act
string retVal = encoder.JavaScriptStringEncode(input);
// Assert
Assert.Equal(expected, retVal);
}
public void JavaScriptStringEncode_DoesNotOutputHtmlSensitiveCharacters()
{
// Per the design document, we provide additional defense-in-depth
// by never emitting HTML-sensitive characters unescaped.
// Arrange
JavaScriptStringEncoder javaScriptStringEncoder = new JavaScriptStringEncoder(UnicodeRanges.All);
HtmlEncoder htmlEncoder = new HtmlEncoder(UnicodeRanges.All);
// Act & assert
for (int i = 0; i <= 0x10FFFF; i++)
{
if (IsSurrogateCodePoint(i))
{
continue; // surrogates don't matter here
}
string javaScriptStringEncoded = javaScriptStringEncoder.JavaScriptStringEncode(Char.ConvertFromUtf32(i));
string thenHtmlEncoded = htmlEncoder.HtmlEncode(javaScriptStringEncoded);
Assert.Equal(javaScriptStringEncoded, thenHtmlEncoded); // should have contained no HTML-sensitive characters
}
}
public void JavaScriptStringEncode_BadSurrogates_ReturnsUnicodeReplacementChar()
{
// Arrange
JavaScriptStringEncoder encoder = new JavaScriptStringEncoder(UnicodeRanges.All); // allow all codepoints
// "a<unpaired leading>b<unpaired trailing>c<trailing before leading>d<unpaired trailing><valid>e<high at end of string>"
const string input = "a\uD800b\uDFFFc\uDFFF\uD800d\uDFFF\uD800\uDFFFe\uD800";
const string expected = "a\uFFFDb\uFFFDc\uFFFD\uFFFDd\uFFFD\\uD800\\uDFFFe\uFFFD"; // 'D800' 'DFFF' was preserved since it's valid
// Act
string retVal = encoder.JavaScriptStringEncode(input);
// Assert
Assert.Equal(expected, retVal);
}
public void JavaScriptStringEncode_AllRangesAllowed_StillEncodesForbiddenChars_Extended()
{
// Arrange
JavaScriptStringEncoder encoder = new JavaScriptStringEncoder(UnicodeRanges.All);
// Act & assert - BMP chars
for (int i = 0; i <= 0xFFFF; i++)
{
string input = new String((char)i, 1);
string expected;
if (IsSurrogateCodePoint(i))
{
expected = "\uFFFD"; // unpaired surrogate -> Unicode replacement char
}
else
{
if (input == "\b")
{
expected = @"\b";
}
else if (input == "\t")
{
expected = @"\t";
}
else if (input == "\n")
{
expected = @"\n";
}
else if (input == "\f")
{
expected = @"\f";
}
else if (input == "\r")
{
expected = @"\r";
}
else if (input == "\\")
{
expected = @"\\";
}
else if (input == "/")
{
expected = @"\/";
}
else
{
bool mustEncode = false;
switch (i)
{
case '<':
case '>':
case '&':
case '\"':
case '\'':
case '+':
mustEncode = true;
break;
}
if (i <= 0x001F || (0x007F <= i && i <= 0x9F))
{
mustEncode = true; // control char
}
else if (!UnicodeHelpers.IsCharacterDefined((char)i))
{
mustEncode = true; // undefined (or otherwise disallowed) char
}
if (mustEncode)
{
expected = String.Format(CultureInfo.InvariantCulture, @"\u{0:X4}", i);
}
else
{
expected = input; // no encoding
}
}
}
string retVal = encoder.JavaScriptStringEncode(input);
Assert.Equal(expected, retVal);
}
// Act & assert - astral chars
for (int i = 0x10000; i <= 0x10FFFF; i++)
{
string input = Char.ConvertFromUtf32(i);
string expected = String.Format(CultureInfo.InvariantCulture, @"\u{0:X4}\u{1:X4}", (uint)input[0], (uint)input[1]);
string retVal = encoder.JavaScriptStringEncode(input);
Assert.Equal(expected, retVal);
}
}
public void JavaScriptStringEncode_EmptyStringInput_ReturnsEmptyString()
{
// Arrange
JavaScriptStringEncoder encoder = new JavaScriptStringEncoder();
// Act & assert
Assert.Equal("", encoder.JavaScriptStringEncode(""));
}
public void JavaScriptStringEncode_NullInput_Throws()
{
// Arrange
JavaScriptStringEncoder encoder = new JavaScriptStringEncoder();
Assert.Throws<ArgumentNullException>(() => { encoder.JavaScriptStringEncode(null); });
}
public void JavaScriptStringEncode_StringSubstring()
{
// Arrange
JavaScriptStringEncoder encoder = new JavaScriptStringEncoder();
var output = new StringWriter();
// Act
encoder.JavaScriptStringEncode("Hello+world!", 3, 5, output);
// Assert
Assert.Equal(@"lo\u002Bwo", output.ToString());
}
public void JavaScriptStringEncode_AllRangesAllowed_StillEncodesForbiddenChars_Extended()
{
// Arrange
JavaScriptStringEncoder encoder = new JavaScriptStringEncoder(UnicodeRanges.All);
// Act & assert - BMP chars
for (int i = 0; i <= 0xFFFF; i++)
{
string input = new String((char)i, 1);
string expected;
if (IsSurrogateCodePoint(i))
{
expected = "\uFFFD"; // unpaired surrogate -> Unicode replacement char
}
else
{
if (input == "\b") { expected = @"\b"; }
else if (input == "\t") { expected = @"\t"; }
else if (input == "\n") { expected = @"\n"; }
else if (input == "\f") { expected = @"\f"; }
else if (input == "\r") { expected = @"\r"; }
else if (input == "\\") { expected = @"\\"; }
else if (input == "/") { expected = @"\/"; }
else if (input == "`") { expected = @"\u0060"; }
else
{
bool mustEncode = false;
switch (i)
{
case '<':
case '>':
case '&':
case '\"':
case '\'':
case '+':
mustEncode = true;
break;
}
if (i <= 0x001F || (0x007F <= i && i <= 0x9F))
{
mustEncode = true; // control char
}
else if (!UnicodeHelpers.IsCharacterDefined((char)i))
{
mustEncode = true; // undefined (or otherwise disallowed) char
}
if (mustEncode)
{
expected = String.Format(CultureInfo.InvariantCulture, @"\u{0:X4}", i);
}
else
{
expected = input; // no encoding
}
}
}
string retVal = encoder.JavaScriptStringEncode(input);
Assert.Equal(expected, retVal);
}
// Act & assert - astral chars
for (int i = 0x10000; i <= 0x10FFFF; i++)
{
string input = Char.ConvertFromUtf32(i);
string expected = String.Format(CultureInfo.InvariantCulture, @"\u{0:X4}\u{1:X4}", (uint)input[0], (uint)input[1]);
string retVal = encoder.JavaScriptStringEncode(input);
Assert.Equal(expected, retVal);
}
}
public void JavaScriptStringEncode_InputDoesNotRequireEncoding_ReturnsOriginalStringInstance()
{
// Arrange
JavaScriptStringEncoder encoder = new JavaScriptStringEncoder();
string input = "Hello, there!";
// Act & assert
Assert.Same(input, encoder.JavaScriptStringEncode(input));
}
public void JavaScriptStringEncode_Quotes(string input, string expected)
{
// Per the design document, we provide additional defense-in-depth
// against breaking out of HTML attributes by having the encoders
// never emit the ' or " characters. This means that we want to
// \u-escape these characters instead of using \' and \".
// Arrange
JavaScriptStringEncoder encoder = new JavaScriptStringEncoder(UnicodeRanges.All);
// Act
string retVal = encoder.JavaScriptStringEncode(input);
// Assert
Assert.Equal(expected, retVal);
}