[MLton] mlnlffigen is awesome (was: bootstrap problem)

[Vesa Karvonen]

Quoting Matthew Fluet <fluet@cs.cornell.edu>:
> > I would probably argue (although weakly) for having MLB path variables in
> > a file instead of the command line (with the command line arg just pointing
> > to the file).  The argument is that often different projects can share
> > the variables so the extra indirection is useful.
> 
> Other reasons for only putting them in a file:
>  * paths to libraries tend to be long (albeit, from the shell you might 
>    have path completion)
>  * defining a path in term of other path variables may require extra 
>    quoting to protect an MLB path variable from shell/environment variable
>    expansion

This may be a silly question, but is there a reason why the ML Basis system
does not support binding path variables in the MLB files?

I'm just wondering about the need for three levels of source files:
1. SML files
2. MLB files
3. MLB Path Map files

Wouldn't it be simpler if there were just two levels:
1. SML files
2. MLB files (with the ability to bind path variables; importing
              a basis would bring the path variables to scope)

Just a thought. Has this been discussed earlier?

-Vesa Karvonen