You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I purportedly implemented a feature allowing the use of Lisp-Binary without CFFI. But I forgot about the fact that nanp and infinityp are just front-ends to their C counterparts. Also, the constants +inf, -inf, and quiet-nan are created by casting bit-patterns to float via CFFI.
In CCL, I avoid some of the problem by using CCL's extended constants. Newer versions of SBCL have constants for infinity, too, but I don't use them because SBCL didn't have them when I wrote the feature. I could begin using them now.
SBCL does not have a constant for NaN, but it does have sb-ext:float-nan-p. So I could avoid using C's isnan(), but I have no way to create one without CFFI.
The text was updated successfully, but these errors were encountered:
I purportedly implemented a feature allowing the use of Lisp-Binary without CFFI. But I forgot about the fact that
nanp
andinfinityp
are just front-ends to their C counterparts. Also, the constants+inf
,-inf
, andquiet-nan
are created by casting bit-patterns to float via CFFI.In CCL, I avoid some of the problem by using CCL's extended constants. Newer versions of SBCL have constants for infinity, too, but I don't use them because SBCL didn't have them when I wrote the feature. I could begin using them now.
SBCL does not have a constant for NaN, but it does have
sb-ext:float-nan-p
. So I could avoid using C'sisnan()
, but I have no way to create one without CFFI.The text was updated successfully, but these errors were encountered: