There are (at least) two cases where unicode characters in f"..." expressions currently fail due to index arithmetics:
- directly preceeding format expressions
- at the end of the f"..." expression
Replacing index arithmetics by prevind() and nextind(), e.g. j-1 => prevind(s, j), would solve the problem.
StrLiterals has the same problem, and I have already submitted a PR. I also have version for StringLiterals at hand if you are interested ;-)
But you indicated that you are going to no longer support StringLiterals ...
Two expressions which currently fail:
@test f"π(22)" == "π4"
@test f"π = (290)°" == "π = 180°"
There are (at least) two cases where unicode characters in f"..." expressions currently fail due to index arithmetics:
Replacing index arithmetics by
prevind()andnextind(), e.g.j-1=>prevind(s, j), would solve the problem.StrLiteralshas the same problem, and I have already submitted a PR. I also have version for StringLiterals at hand if you are interested ;-)But you indicated that you are going to no longer support
StringLiterals...Two expressions which currently fail:
@test f"π(22)" == "π4"
@test f"π = (290)°" == "π = 180°"