public static void GetUnicodeCategory_Char_AllInputs()
{
// This tests calls char.GetUnicodeCategory for every possible input, ensuring that
// the runtime agrees with the data in the core Unicode files.
for (uint i = 0; i <= char.MaxValue; i++)
{
UnicodeCategory expected;
// The code points in the switch block below must be special-cased
// because they switched categories between versions of the Unicode
// specification. For compatibility reasons Char keeps its own copy
// of the categories for the first 256 code points, as it's locked
// to an earlier version of the standard. For an example of a code
// point that switched categories, see the discussion on U+00AD
// SOFT HYPHEN at https://www.unicode.org/versions/Unicode4.0.0/.
switch (i)
{
case '\u00a7':
case '\u00b6':
expected = UnicodeCategory.OtherSymbol;
break;
case '\u00aa':
case '\u00ba':
expected = UnicodeCategory.LowercaseLetter;
break;
case '\u00ad':
expected = UnicodeCategory.DashPunctuation;
break;
default:
expected = UnicodeData.GetUnicodeCategory(i);
break;
}
if (expected != char.GetUnicodeCategory((char)i))
{
// We'll build up the exception message ourselves so the dev knows what code point failed.
throw new AssertActualExpectedException(
expected: expected,
actual: char.GetUnicodeCategory((char)i),
userMessage: FormattableString.Invariant($@"char.GetUnicodeCategory('\u{i:X4}') returned wrong value."));
}
}
}
public static void GetUnicodeCategory_AllInputs()
{
// This tests calls Rune.GetUnicodeCategory for every possible input, ensuring that
// the runtime agrees with the data in the core Unicode files.
foreach (Rune rune in AllRunes())
{
if (UnicodeData.GetUnicodeCategory(rune.Value) != Rune.GetUnicodeCategory(rune))
{
// We'll build up the exception message ourselves so the dev knows what code point failed.
throw new AssertActualExpectedException(
expected: UnicodeData.GetUnicodeCategory(rune.Value),
actual: Rune.GetUnicodeCategory(rune),
userMessage: FormattableString.Invariant($@"Rune.GetUnicodeCategory(U+{rune.Value:X4}) returned wrong value."));
}
}
}
public static void IsUpper_Char_AllInputs()
{
// This tests calls char.IsUpper for every possible input, ensuring that
// the runtime agrees with the data in the core Unicode files.
for (uint i = 0; i <= char.MaxValue; i++)
{
bool expected = UnicodeData.GetUnicodeCategory((char)i) == UnicodeCategory.UppercaseLetter;
if (expected != char.IsUpper((char)i))
{
// We'll build up the exception message ourselves so the dev knows what code point failed.
throw new AssertActualExpectedException(
expected: expected,
actual: char.IsUpper((char)i),
userMessage: FormattableString.Invariant($@"char.IsUpper('\u{i:X4}') returned wrong value."));
}
}
}
public static void IsLower_Char_AllInputs()
{
// This tests calls char.IsLower for every possible input, ensuring that
// the runtime agrees with the data in the core Unicode files.
for (uint i = 0; i <= char.MaxValue; i++)
{
bool expected;
switch (i)
{
case '\u00AA': // FEMININE ORDINAL INDICATOR
case '\u00BA': // MASCULINE ORDINAL INDICATOR
// In Unicode 6.1 the code points U+00AA and U+00BA were reassigned
// from category Ll to category Lo. However, for compatibility reasons,
// Char uses the older version of the Unicode standard for code points
// in the range U+0000..U+00FF. So we'll special-case these here.
// More info: https://www.unicode.org/review/pri181/
expected = true;
break;
default:
expected = UnicodeData.GetUnicodeCategory((char)i) == UnicodeCategory.LowercaseLetter;
break;
}
if (expected != char.IsLower((char)i))
{
// We'll build up the exception message ourselves so the dev knows what code point failed.
throw new AssertActualExpectedException(
expected: expected,
actual: char.IsLower((char)i),
userMessage: FormattableString.Invariant($@"char.IsLower('\u{i:X4}') returned wrong value."));
}
}
}
