Prior to this commit, building this system in SBCL signaled a
style-warning of type sb-kernel:&optional-and-&key-in-lambda-list
from line 243 of [src/named-readtables.lisp], the definition of
make-readtable. This was despite obvious work in the definition of
define-api to prevent such warnings. This commit makes two
SBCL-specific changes, which result in that warning no longer being
printed:
define-api muffles only
sb-kernel:&optional-and-&key-in-lambda-list, around its defun,
not all of style-warning. This makes more obvious why the
locally muffle-conditions is there, and prevents accidentally
muffling other conditions.
define-api no longer unmuffles any conditions around its
body. This means that sb-kernel:&optional-and-&key-in-lambda-list
is muffled in both the definition and the body of define-api's
defun form. I suspect that unmuffling in the body is what caused
make-readtable to issue that warning.
Prior to this commit, building this system in SBCL signaled a style-warning of type
sb-kernel:&optional-and-&key-in-lambda-list
from line 243 of [src/named-readtables.lisp], the definition ofmake-readtable
. This was despite obvious work in the definition ofdefine-api
to prevent such warnings. This commit makes two SBCL-specific changes, which result in that warning no longer being printed:define-api
muffles onlysb-kernel:&optional-and-&key-in-lambda-list
, around itsdefun
, not all ofstyle-warning
. This makes more obvious why thelocally muffle-conditions
is there, and prevents accidentally muffling other conditions.define-api
no longer unmuffles any conditions around its body. This means thatsb-kernel:&optional-and-&key-in-lambda-list
is muffled in both the definition and the body ofdefine-api
'sdefun
form. I suspect that unmuffling in the body is what causedmake-readtable
to issue that warning.