[m-rev.] for review: transform X = f to true or fail if X's binding is known

[Zoltan Somogyi]

On Fri, 19 Sep 2025 11:36:43 +1000 (AEST), "Zoltan Somogyi" <zoltan.somogyi at runbox.com> wrote:
> > Does compile-time evaluation of library calls lead to simple code warnings?
> > Or do you want to do this because re-running simplify after library
> > call evaluation
> > has occurred may allow further simplifications to occur?
> 
> The former, though this applies only to situations in which the replacement
> code is simple enough to warrant a --warn-simple-code warning. I think
> there are some, but I haven't checked yet.

There are some, and the attached diff handles them. However, I couldn't
come up with a test case for which the compiler from before this diff
does the wrong thing, so I am attaching the diff only for the record;
I won't commit it.

I will update the XXX in the previous diff.

Zoltan.



-------------- next part --------------
A non-text attachment was scrubbed...
Name: DIFF.rhs2
Type: application/octet-stream
Size: 8040 bytes
Desc: not available
URL: <http://lists.mercurylang.org/archives/reviews/attachments/20250921/e5869131/attachment.obj>