[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Issue: PATHNAME-SYSTEM-TYPE (version 1)
- To: Dan L. Pierson <email@example.com>
- Subject: Re: Issue: PATHNAME-SYSTEM-TYPE (version 1)
- From: firstname.lastname@example.org (Sandra J Loosemore)
- Date: Thu, 25 May 89 16:24:36 MDT
- Cc: email@example.com (Sandra J Loosemore), CL-Cleanup@sail.stanford.edu
- In-reply-to: Dan L. Pierson <firstname.lastname@example.org>, Thu, 25 May 89 18:04:46 EDT
> Date: Thu, 25 May 89 18:04:46 EDT
> From: Dan L. Pierson <email@example.com>
> But if we restricted what Common Lisp supports to what is possible in
> the most restrictive existing environment we wouldn't have modern Lisp
> at all.
Hey, that's my argument! Issue PATHNAME-COMPONENT-CASE even says that
this is the end goal -- to come up with the least common denominator
for pathname component values. I think it's the wrong goal, precisely
because the least common denominator (at least as far as MAKE-PATHNAME
is concerned) isn't very useful.
I think we'd be better off sidestepping the whole issue and simply
putting some discussion in the standard about what kinds of pathname
operations are portable, and which aren't. In the absence of such a
discussion in CLtL, I've seen users (even experienced Lisp
implementors, not just students) make assumptions in their programs
that inhibit portability without realizing they've done so.