[plt-scheme] Are new Schemers supposed to be reading SRFIs?

From: Danny Yoo (dyoo at cs.wpi.edu)
Date: Fri May 11 14:38:54 EDT 2007

> Would it be possible to do the same with the SRFIs?
> When referencing them in the code, we can use names like "strings" 
> "lists", etc. and the interpreter will just translate that to the proper 
> number without us needing to know or even see the number.
> Then when SRFI-300 comes along and replaces SRFI-1, all you need to do 
> is to update this translation table. Just my 0.02... At least it 
> *sounds* simple :)

Hi everyone,

Hmmm...  Ok, got it.  It's doable with the existing infrastructure in 
PLaneT.  It'd be very direct to create a PLaneT package that aliases the 
common SRFI modules.

This I have now done, since it seems like doing this from the official 
srfi-collection is the wrong thing to do.  Chongkai, you don't make the 
change to the srfi collection then.  I've submitted a PLaneT package that 
does absolutely nothing but provide the SRFI names:


As a concrete example:

     (require (planet "list.ss" ("dyoo" "srfi-alias.plt")))


     (require (planet "string.ss" ("dyoo" "srfi-alias.plt")))

should now work.

Since it's PLaneT, this isn't officially sanctioned, so that should 
resolve the objections about doing this on the official srfi collection. 
And it still preserves what I wanted before: being able to get at the 
common srfi's without having to remember their numbers.  So it works out.

(And if an SRFI does pop up with a name that conflicts with existing 
names, I could now choose by dictatorial fiat how to resolve the 
conflict... *grin*)

Hope this helps!

Posted on the users mailing list.