adsharma

https://github.com/adsharma/py2many/blob/main/py2many/inference.py#L502 is a good place to do so

So the task is about supporting java as a target language? I can see why people would want it. Access to JVM in general is already possible because of Kotlin.

Somewhat related: I just published https://github.com/adsharma/decaf2many Allows: ``` cat code | decaf2many | py2many ```

I think it's a useful module. We should use it and build functionality on top of it. But it's not a full replacement for inference.py or analysis.py.

The code in scope.py and inference.py is doing less of what is the type of "1.2" and more of data flow analysis that these libraries don't seem to be doing....

Sounds like a good idea - especially if we can demonstrate python -> typed python transpiler. It will be some work though and I'm trying to make the best use...

https://github.com/caterinaurban/Typpete Z3 based constraint solvers are interesting too. But I don't know how practical they are. https://link.springer.com/chapter/10.1007/978-3-319-96142-2_2 On Thu, Jul 1, 2021, 11:04 PM John Vandenberg ***@***.***> wrote: > https://github.com/eliphatfs/typhon...

Thank you for sharing your perspective @eliphatfs. I'm thinking that as a practical matter, we can fix usability issues in existing z3 based solvers and when it can't be solved,...

https://github.com/caterinaurban/Typpete/issues/55

Upstream seems to have stalled. But I've had good success hacking on a fork: https://github.com/adsharma/Typpete. https://github.com/adsharma/Typpete/commit/38068caab098616dd5956affe07a491f181180e6 adds support for fixed width integers. As in: ``` def foo(): a: u8 =...