Re: introspection in SML

In article <fj46l1$rs7$1@[EMAIL PROTECTED]
>,
 Vesa Karvonen <vesa.karvonen@[EMAIL PROTECTED]
> wrote:

> Pascal Costanza <pc@[EMAIL PROTECTED]
> wrote:
> [...]
> > You can also fine-tune things, including across compilation units,
with 
> > inline and notinline declarations, and compiler macros.
> 
> That is precisely what I don't want to do.  For example, in some case
> you might like to inline a particular definition from a library, in
> some other case you might not.

You can do that in CL.

>  How do you know when it is really safe
> to add an inline declaration for a function (CL makes certain
> behaviors undefined when you add such a declaration, so you'd have to
> analyze the program to know whether it is safe)?  What if that
> function wasn't written by you?  What if you are using a library
> written by someone else?

There is no such concept as a 'library' in Common Lisp.
It does not matter who writes the code.

>  Do you go and modify your copy of the
> library by adding / removing / chanding inline and notinline
> declarations?

You don't need to modify the library. You can add these
declarations in your own code referencing other functions
defined elsewhere. They will affect the code that you
are compiling. Already compiled code is not affected.

>  Do you bend over and copy a definition from a library
> to another file (duplicating code) just to please the compiler?  These
> kind of optimizations can be done by a machine.
> 
> -Vesa Karvonen

-- 
http://lispm.dyndns.org/