Invalid number type: (FLOAT 0)

[paulapatience]

Compiling

(etypecase 0
  (integer)
  ((float 0)))

results in

;   WARNING during compiler-macro-expansion:
;     Invalid number type: (FLOAT 0)
;     at /tmp/foo.lisp 3:3

[Bike]

I think this is not strictly a bug: see CLHS 12.1.6 Interval Designators. The designators for an interval of the given type have to be numbers of that type, and 0 is not a float.

I'm going to leave this open because arguably we should be nice and coerce it for you.