Open cgay opened 4 years ago
SLIME's package-parsing logic has an even bigger problem, but it's related to this one.
Today, I was M-.
-ing my way through the CFFI-GROVEL library. At line 206 of grovel.lisp
there
is a case
form that checks if a variably is the symbol in-package
, and if it is,
it calls a setf
form. But that case branch superficially looks like an in-package
form,
so SLIME assumes it is.
I isolated that particular aspect of SLIME's in-package processing. This will cause a ^J to appear in the package name in the buffer's title:
(in-package
:cl-user)
SLIME should be ignoring the in-package
form in grovel.lisp because it
occurs within a function, and it should be accounting for the fact that in-package
is an ordinary function, so its argument doesn't have to be a literal. Forms that
compute the package at load time ought to work, or at least not trick SLIME
into thinking that a SETF form is literally the name of a package.
SLIME version cd8cc3c95c3391b1f1cffa9e100336250a4509a7
The subject pretty much says it all. If you have a file of the form
SLIME will think the code is in :other-package.
M-.
will get confused.