[Fluxus] racket 5.2 fix

[Kassen]

On Wed, May 02, 2012 at 11:12:21PM +0200, gabor papp wrote:
> >Ooh, that was the big one! Congratulations!
> thanks, but Matthew Flatt from racket deserves this. he suggested
> the fix, which i sent to the list some months ago. it was not
> committed, because that time resetting the interpreter with f6
> didn't work, but they fixed it in racket since then.

Works here, as far as I can tell.

More good news; I suspect that building without places and futures led
to a regression of a unknown bug causing console errors about unknown
incomming MIDI messages when a in port was open. Checking out the
source those seemed to be about length 0 or shorter messages (!!!). I
was still investigating the exact nature of that regression and with
this fix those are gone again.

No news on how the Scheme parts were related to those messages, I'd
say those would be between the C++ part and Alsa, but they were
harmless (if ugly) and are gone again.

Yours,
Kas.