Cuttext: No way to detect encoding (utf8 or latin1)

[nob13]

I am using a cut text handler with nvnc_set_cut_text_fn.

Unfortunately, there seems no way to detect, in which encoding the cut text is. Remmina is using legacy cut text and sends latin1 encoding, while NoVnc uses the extended clipboard encoding and sends UTF8.

As a workaround, i check if the string is valid utf8 encoding, then take utf8, otherwise try to decode latin1 to utf8.

[any1]

Yep. Feel free to suggest an API addition for that.

[nob13]

nvnc_send_cut_text seems to have an analogous problem. If the client speaks the extended clipboard extension, we will send out the text as UTF8. If the client doesn't speak it, we send it as latin1. The server application can't know, but would need in order to figure out in which format it should present it.

How should a changed API look like, just new versions, like nvnc_set_cut_text_fn2 and nvnc_send_cut_text2, given that the API should be stable?

[any1]

I think that nvnc_set_cut_text_with_encoding_fn and nvnc_send_cut_text_with_encoding would be okay names.