MLton 20011006

Stephen Weeks MLton@sourcelight.com
Thu, 11 Oct 2001 10:16:59 -0700


> Thanks!  Unfortunately, I can't yet report success.  I tried it, with
> these results:
...
> Again, I'm not sure if this is a bug or something I did wrong.
> The offending source is at http://www.eecs.harvard.edu/~nr/sled.sml

Thanks for the report. This is a bug, and is unrelated to the bugs exposed by
lrtl.  We have fixed the bug in our internal working version.  The bug is in the
type checker, so there is an easy workaround with 20011006.  Change line 100 of
elabsm.nw to read

fun extendSyntaxMap ({opcode=(opcode,_)}, operands) (existingOperands: 'a Constructor.operand' list)

This change is sufficient to unconfuse the type checker.  

The bugfix will appear in any future experimental versions or releases.  If you
can't survive with the workaround and would like an experimental release that
includes the fix, please let us know.