[racket] My PLT->Racket porting experience

From: Robby Findler (robby at eecs.northwestern.edu)
Date: Thu Mar 3 11:30:35 EST 2011

On Thu, Mar 3, 2011 at 10:26 AM, Drew Hess <drew.hess at gmail.com> wrote:
> On Thu, Mar 3, 2011 at 8:15 AM, Robby Findler
> <robby at eecs.northwestern.edu> wrote:
>> If you were to replace the body of 'play' with something like
>>  (car 'not-a-pair)
>> then you see the error, right?
>> If so, then it sounds like the problem is that play-sound is not
>> signaling an error (ie this is a problem specific to play-sound, not
>> to teachpacks).
> Hmm, there's something more insidious going on. I just tried the test
> program I proposed that you try, and I *do* get an error message from
> play-sound (called from inside the teachpack) when the file is not
> found. For some reason, this isn't the behavior I am seeing with the
> Bootstrap teachpack.
> OK, I'll have to package something up to demonstrate the problem.

The bug that I fixed had to do with the initial REPL, before you run
any programs.

Specifically, if you set up some teachpacks, and then restart
drracket, drracket will remember the teachpacks you've set and will
try to run them (without running the program) for the initial REPL
(interactions window). If this caused in error in 5.1, that error was
swallowed (similarly in 4.2.5). In the current pre-release you'll see
the error.

I don't know if that's relevant or not, but that's the change that I
believe Matthew was alluding to.


Posted on the users mailing list.