Utf8Parser.TryParse ignores the provided format for bool

[joegoldman2]

Description

The current implementation of Utf8Parser.TryParse for bool ignores the format:

runtime/src/libraries/System.Private.CoreLib/src/System/Buffers/Text/Utf8Parser/Utf8Parser.Boolean.cs

Lines 29 to 58 in beac274

	public static bool TryParse(ReadOnlySpan<byte> source, out bool value, out int bytesConsumed, char standardFormat = default)
	{
	if (!(standardFormat == default(char) || standardFormat == 'G' || standardFormat == 'l'))
	ThrowHelper.ThrowFormatException_BadFormatSpecifier();
	if (source.Length >= 4)
	{
	int dw = BinaryPrimitives.ReadInt32LittleEndian(source) & ~0x20202020;
	if (dw == 0x45555254 /* 'EURT' */)
	{
	bytesConsumed = 4;
	value = true;
	return true;
	}
	if (source.Length > 4)
	{
	if (dw == 0x534c4146 /* 'SLAF' */ && (source[4] & ~0x20) == 'E')
	{
	bytesConsumed = 5;
	value = false;
	return true;
	}
	}
	}
	bytesConsumed = 0;
	value = default;
	return false;
	}

Reproduction Steps

byte[] bytes = Encoding.UTF8.GetBytes("true");
var result = Utf8Parser.TryParse(bytes, out bool value, out int byteConsumed, standardFormat: 'G');
Console.WriteLine(result);

Expected behavior

result should be false because for the format G, True or False beginning with uppercase are expected.

Actual behavior

result is true.

Regression?

I don't think so. In dotnet/corefx, the implementation already didn't take the format into account.

Known Workarounds

No response

Configuration

9.0.0-preview.2.24128.5

Other information

No response

[dotnet-policy-service]

Tagging subscribers to this area: @dotnet/area-system-memory
See info in area-owners.md if you want to be subscribed.

[jeffhandley]

Note that we decided against making the behavioral change per the comments in #100676, so a docs change is being pursued instead in #103861.

Thanks for submitting this issue and driving the changes, @joegoldman2!