From: Sam Steingold
Subject: (pathname (merge-pathnames ...))
Date:
Message-ID: <ur7g1py02.fsf@gnu.org>
<http://www.lisp.org/HyperSpec/Body/var_stcompile_e-truenamest.html>:
During a call to compile-file, *compile-file-pathname* is bound
to the pathname denoted by the first argument to compile-file,
merged against the defaults; that is, it is bound to (pathname
(merge-pathnames input-file)).
since MERGE-PATHNAMES always returns a pathname,
(pathname (merge-pathnames input-file))
== (merge-pathnames input-file)
right?
why then the added function call?
--
Sam Steingold (http://www.podval.org/~sds) running w2k
<http://www.jihadwatch.org/> <http://www.memri.org/> <http://ffii.org/>
<http://www.camera.org> <http://www.honestreporting.com>
I want Tamagochi! -- What for? Your pet hamster is still alive!
From: Kent M Pitman
Subject: Re: (pathname (merge-pathnames ...))
Date:
Message-ID: <uacmp91q2.fsf@nhplace.com>
Sam Steingold <···@gnu.org> writes:
> <http://www.lisp.org/HyperSpec/Body/var_stcompile_e-truenamest.html>:
> During a call to compile-file, *compile-file-pathname* is bound
> to the pathname denoted by the first argument to compile-file,
> merged against the defaults; that is, it is bound to (pathname
> (merge-pathnames input-file)).
>
> since MERGE-PATHNAMES always returns a pathname,
>
> (pathname (merge-pathnames input-file))
> == (merge-pathnames input-file)
>
> right?
Right.
> why then the added function call?
My best guesses...
Sometimes stuff like this is show-through from CLTL that never got
fixed. My various copies of CLTL weren't handy as I write this (and
CMU's webbed CLTL2 seems to be getting page-not-found errors) so it
was hard to check. You might look up merge-pathnames there and see
if that phrase comes from there.
But also sometimes things like this got added by me or my predecessor
out of some fear that MERGE-PATHNAMES might change definition and we
didn't want the example broken. Think of it as security against a
definition of MERGE-PATHNAMES that allowed the pathname designator
(which as you know is not always a pathname) to be returned as-is in
situations where no merging was required.
Now that the standard is done, of course, such changes don't happen
easily, but while the standard was being worked on, things were
changing, so sometimes things like that were just protection.