Daniel Lakeland <dlakelan@...> writes:
> It seems like there is a bug in format under sbcl:
> (format t "~eE" 1.0d0)
> (format t "~,1,,,,,eE" 1.0d0)
> I'm not sure that I understand what the hyperspec thinks the first
> thing should do, but the second one should almost certainly print:
> the character before the E in the format spec is supposed to be used
> as the exponent seperator.
> I think what's going on is that SBCL is interpreting ~e as ~E
Um, but ~e and ~E have the same meaning; from CLHS 22.3, "The case of
the directive character is ignored.". (Consider also how, under your
interpretation, you would get #\E as the exponent separator).
Literal characters in format directives have particular syntax, also
described in CLHS 22.3