[MLton-devel] Re: Twelf benchmark

[Frank Pfenning]

> Yeah, MLton doesn't have a proper type checker yet, and so accepts
> many nonstandard programs.  We assume you've run your program through
> a proper typechecker first.
> 
> I was thinking about patching your code so I could compile it with
> Poly/ML.  It's usually not too difficult to replace type definitions
> with datatypes, which can be shared.

The right thing here is actually to expand the type definitions.
I have done this already several times, but I could never convince
myself to check these changes into the repository, basically
because I think in this case SML'97 is broken, and not SML/NJ.

The alternative is to de-functorize the Twelf implementation,
to use functors only where they are necessary, rather than
everywhere.  This would also solve the problem, but it a much
more pervasive change.  I think this is what I will do.

In any case, the version of Twelf in the repository will not check
the realistic example suites.  I will release a new version, and
try to make sure it is portable at least to MLton and Poly/ML.
This should happen over the next few days (since classes are
over :-)

  - Frank

P.S.: It would help if you could send me the diffs, or MLton specific
files so I do not replicate the work that you already did.


-------------------------------------------------------
This sf.net email is sponsored by:
With Great Power, Comes Great Responsibility 
Learn to use your power at OSDN's High Performance Computing Channel
http://hpc.devchannel.org/
_______________________________________________
MLton-devel mailing list
MLton-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mlton-devel