}
},
window: {},
- general: {}
+ general: {
+ # Currently we always send character count as position offset,
+ # which meanas only utf-32 is supported.
+ # Adding utf-16 simply for good mesure, as I'm scared some servers will
+ # give up if they don't support utf-32 only.
+ positionEncodings: ['utf-32', 'utf-16']
+ },
+ # This is the way clangd expects to be informated about supported encodings:
+ # https://clangd.llvm.org/extensions#utf-8-offsets
+ offsetEncoding: ['utf-32', 'utf-16']
}
# interface 'InitializeParams'