Re: [Matlisp-users] Matlisp subset in pure CL

[Nicolas N. <Nic...@iw...>]

Raymond Toy <to...@rt...> writes:

>     Nicolas> I have thought about this, but I am completely uncertain how to do such a
>     Nicolas> thing in a well-behaved way within ANSI CL.  The best thing would probably
>     Nicolas> be to pay Gerd Moellmann for implementing method inlining in CMUCL.  But
> 
> According to the CMUCL User's manual, Section 2.23.4 says that
> effective methods can be inlined.  Perhaps that is good enough.
> 
> Ray

Thanks for the pointer (I had to update my CMUCL manual to read it,
probably I should switch to a more recent CMUCL version).  I am not sure if
it helps me because of:

   "Please note that this form of inlining has no noticeable effect for
   effective methods that consist of a primary method only, which doesn't
   have keyword arguments. In such cases, PCL uses the primary method
   directly for the effective method."

Since I avoided method combination for BLAS routines anyhow, this might not
change much.  Anyway, I think I will be able to phrase my needs more
precisely when I have some working code.

Yours, Nicolas.