Re: [MIT-Scheme-devel] working directory

[Matt Birkholz]

> From: Taylor R Campbell <address@hidden>
> Date: Sun, 13 Feb 2011 19:47:25 +0000
> 
> [...]  Part of the reason for the klunkiness is that
> WITH-WORKING-DIRECTORY-PATHNAME doesn't really match
> SET-WORKING-DIRECTORY-PATHNAME!

How about renaming with-working-directory-pathname
"with-DEFAULT-directory-pathname", and set-working-directory-pathname!
"set-PROCESS-current-working-directory!"?  Then the two procedures
will not appear to be related, and we might avoid these frustrated
expectations.

> [...]  Nothing uses fchdir -- not even UX-MAKE-SUBPROCESS, as a
> consequence of which Scheme can't report a sensible error message if
> you try to run a subprocess in a bogus working directory.

THAT sounds like a klunk... which can be fixed without craziness.

>    If I pass "/absolute/filename" to a primitive that theoretically
>    depends on the cwd (as I am wont to do), will it now have to frob some
>    OS randomness before doing as I ask?
> 
> It could avoid that.  I wasn't planning to make it avoid that. [...]

I was not trying to pick a performance nit.  I just wish to emphasize
that the runtime makes frequent use of merge-pathnames, such that the
OS never sees a relative pathname.  The frobbing of the cwd is, I
think, appropriately confined to ux- procedures like
ux-make-subprocess.

I AM trying to put the brakes on loading up threads with tons of bells
and whistles.  I want to spawn threads with wild abandon, e.g. in
Edwin every time it needs to load an image, fontify a buffer, or
animate anything.  I want MultiScheme, and all 16 cores in my (future)
CPU doing useful work all at the same time.  (Is that wyrd?)