Question

When generating a QR code with QRCodeGenerator.CreateQrCode, what is the purpose of forceUtf8 versus eciMode: EciMode.Utf8?

I would expect that specifying EciMode.Utf8 would force the generated code to use the UTF-8 character set when encoding the data. But in fact, it does so only when either (a) there are non-ISO-8859-1 characters present, or (b) forceUtf8 is true.

Consider the following:

// produces an unreadable code:
var qrData = new QRCodeGenerator().CreateQrCode("https://en.wikipedia.org/wiki/È", ECCLevel.L, eciMode: EciMode.Utf8);
// works fine:
var qrData = new QRCodeGenerator().CreateQrCode("https://en.wikipedia.org/wiki/È", ECCLevel.L, forceUtf8: true, eciMode: EciMode.Utf8);

// whereas this produces identical QR codes:
var qrData = new QRCodeGenerator().CreateQrCode("https://en.wikipedia.org/wiki/🍕", ECCLevel.L, eciMode: EciMode.Utf8);
var qrData = new QRCodeGenerator().CreateQrCode("https://en.wikipedia.org/wiki/🍕", ECCLevel.L, forceUtf8: true, eciMode: EciMode.Utf8);

A similar phenomenon occurs when specifying EciMode.Iso8859_2 with specific strings -- interestingly, forceUtf8 must be true for the code to function correctly.

If we assume that this is a bug, and EciMode.Utf8 should always encode with UTF-8, then what is the purpose for the forceUtf8 argument? My best guess is that it is some compatibility mode to use UTF-8 encoding when encoded with EciMode.Default (which normally encodes as ISO-8859-1).

Suggestion

I suggest that:

1. When specifying a EciMode besides EciMode.Default, the text is always encoded in the specified encoding.
2. For EciMode.Default with forceUtf8 == false, it uses ISO-8859-1 (per spec, EciMode.Iso8859_1 is default)
3. For EciMode.Default with forceUtf8 == true, it uses UTF-8 (against spec)

Spec

From ISO spec page 20:

The default interpretation for QR Code is ECI 000003 representing the ISO/IEC 8859-1 character set.