Open jansorg opened 3 years ago
The scanner seems to be quite forgiving:
Even clearly invalid Unicode is parsed and survives the round-trip through cue eval
:
echo -ne 'foo: "\xf0\x28\x8c\x28\x7F\xfe\xff"' >> imports.cue
Not sure if this is a bug or a feature.
I have had issues where UTF-16 json files on windows fail to read.
https://github.com/cuelang/cue/blob/master/doc/ref/spec.md#source-code-representation
UTF-8 seems to be the only valid encoding. We could: