Thanks for pointing that out. I've fixed it in SVN and added a regression test. Note that picking up the fix requires rebuilding the compiler.
Just a general warning to everyone that the format of exception handlers on X86 has changed and that in order to rebuild the compiler in SVN it's necessary to run "make compiler" twice so that the compiler itself uses the new format. Otherwise it's quite likely to segfault.
David
Phil Clayton wrote:
I've just noticed that singleton tuples don't seem to be pretty printed as valid SML:
Poly/ML 5.4 Experimental
{1 = true};
val it = (true): {1: bool}
Presumably this should say:
val it = {1 = true}: {1: bool}
Phil
polyml mailing list polyml@inf.ed.ac.uk http://lists.inf.ed.ac.uk/mailman/listinfo/polyml