Mixed Arithmetic in Opcodes

[Solumin]

I've added a sample of mixed arithmetic to BINARY_ADD. It's really messy -- another way to do this would be great.

One problem is that we're treating 32-bit integers as 'numbers', so they look like floating-point numbers, which messes up the widening hierarchy (see the comment on interpreter.ts:widen or Python Numeric Types)

[Solumin]

I'm tempted to unmarshal all 32-bit integers as gLong types (64 bit), just so they follow the hierarchy a little better.

[cibelemf]

Did you try to do that? I saw you changed the widen and other stuff. Is it working now?

[Solumin]

I got it working for BINARY_ADD but it's so clunky and ugly.

On Sun, Oct 19, 2014 at 4:54 PM, Cibele [email protected] wrote:

Did you try to do that? I saw you changed the widen and other stuff. Is it
working now?

—
Reply to this email directly or view it on GitHub
#5 (comment)
.