[racket-dev] "Too many" struct definitions leads to literal <local-code> error

From: J. Ian Johnson (ianj at ccs.neu.edu)
Date: Mon Aug 5 14:55:01 EDT 2013

I'm working more on my analysis framework, and added many different kinds of new continuation frames. I get to a point where I comment out all definitions past a certain point (and their uses) and the program 
compiles. I add one more (any one) and I get the following error:

kcfa-instantiations.rkt:136:6: ?: literal data is not allowed;
 no #%datum syntax transformer is bound
  in: #<local-code>
  context...:
   /home/ianj/racket/racket/collects/racket/splicing.rkt:234:0
   /home/ianj/racket/racket/collects/racket/splicing.rkt:181:0
   /home/ianj/racket/racket/collects/racket/splicing.rkt:234:0
   /home/ianj/racket/racket/collects/racket/splicing.rkt:181:0
   /home/ianj/racket/racket/collects/racket/splicing.rkt:234:0
   /home/ianj/racket/racket/collects/racket/splicing.rkt:181:0
   /home/ianj/racket/racket/collects/racket/splicing.rkt:234:0
   /home/ianj/racket/racket/collects/racket/splicing.rkt:181:0
   /home/ianj/racket/racket/collects/racket/splicing.rkt:234:0
   /home/ianj/racket/racket/collects/racket/splicing.rkt:181:0
   /home/ianj/racket/racket/collects/racket/splicing.rkt:234:0
   /home/ianj/racket/racket/collects/racket/splicing.rkt:181:0
   /home/ianj/racket/racket/collects/racket/private/modbeg.rkt:46:4
   /home/ianj/racket/racket/collects/compiler/cm.rkt:345:0: compile-zo*
   /home/ianj/racket/racket/collects/compiler/cm.rkt:552:26
   /home/ianj/racket/racket/collects/compiler/cm.rkt:545:42...

This is blocking progress. I'm about to pull and hope for the best. Any ideas what might be causing this kind of behavior in the expander?
-Ian

Posted on the dev mailing list.