[racket-dev] Should BSL signal an error on (require ...) in an unsaved buffer?

From: Matthias Felleisen (matthias at ccs.neu.edu)
Date: Mon Jun 27 19:39:14 EDT 2011

On Jun 27, 2011, at 7:37 PM, John Clements wrote:

> I'm fixing require in the stepper, and I want to make sure that the existing behavior is desirable before I try to simulate it.  In particular, my experiments suggest that "require" in an unsaved buffer implicitly requires things from (current-directory). Is this the right behavior? It would seem to me that we'd be better off signaling an error in this case.  No?

Yes, please signal an error. 

Posted on the dev mailing list.