Jeff Niu

Closing as not reproducible

@soraros with your file I can reproduce the error. Not sure why neither I nor Ewa were getting it in the REPL/Jupyter notebook.

A lot of this is work in progress, which includes some hacks added to make "common cases" work while we spin on general solutions (if they exist). I don't think...

What @nmsmith "discovered" is actually pretty accurate, with the exception that `alias` values are only substituted when it is not declared in a function. This divergence in behaviour is "just"...

> Mojo currently appears to "solve" this problem by disallowing side-effects during specialization, but I assume this constraint will be lifted at some point, because side-effects such as reading from...

Hey @nmsmith thanks for chiming in. Yeah it's the use of tuple literal syntax to represent a tuple type in the return slot that's causing issues, and in fact it...

We ultimately decided on `fn foo() -> String as output:` as the syntax to move this feature forward. This will be included in the next release! (Or tonight's nightly). Thank...

You are correct. We're getting operator precedence wrong here. Thanks for narrowing this down!!

@jon-chuang do you have code that we can run to reproduce this issue?

Thanks so much! @jon-chuang . We'll take a look