Faré wrote:
Do you have a backtrace for asdf-pathname-test on CCL and/or SBCL?
This one only failed for me on clisp. Here's the error:
TEST ABORTED: These two expressions yield paths that are not pathname-equal the first expression (MAKE-PATHNAME :HOST "ASDFTEST" :DIRECTORY '(:ABSOLUTE "system2" "module4") :NAME NIL :TYPE NIL) yields this: #P"ASDFTEST:system2;module4;" (:HOST "ASDFTEST" :DEVICE :UNSPECIFIC :DIRECTORY (:ABSOLUTE "system2" "module4") :NAME NIL :TYPE NIL :VERSION NIL)
the other expression (PARSE-NAMESTRING "ASDFTEST:system2;module4;") yields that: #P"ASDFTEST:SYSTEM2;MODULE4;" (:HOST "ASDFTEST" :DEVICE :UNSPECIFIC :DIRECTORY (:ABSOLUTE "SYSTEM2" "MODULE4") :NAME NIL :TYPE NIL :VERSION NIL)
This seems like it might be wrong -- if these are logical pathnames, aren't these two pathnames pathname-equal (at least semantically, if not according to your code), because logical pathnames are not case-sensitive?
I'm actually finding this hard to replicate interactively. I.e.,. I'm finding it hard to make clisp's PARSE-NAMESTRING with that string give the same up-cased pathname as listed above.
Cheers, r