Jukka Lehtosalo

Results 479 comments of Jukka Lehtosalo

The problem with inferring a literal type here is that then code like this could generate a false positive: ``` for x in ('foo', 'bar'): x = x.upper() # str...

Actually, we could maybe infer `str` as the actual type of `x`, and narrow it down to a union of literal types in the body of the for loop. I...

Thanks! I'll update the readme to say that macOS works.

Value types are a great feature, thanks! This seems to address mypyc/mypyc#841 -- what do you think? Since using a value type can change semantics (at least `is` / object...

> Additionally, I think that functionality is already provided by python using NamedTuple I don't think NamedTuple can be a value type, since it supports subclassing and subclasses can add...

Can you merge master? It now includes the dunder changes. This is a big PR so it will take some time to review and test. I'll start looking into this...

These features may not have tests (and some are probably not implemented): * Fixed-length tuple items with value types * Undefined attribute/local variable with value type (e.g. conditionally assign value)...

Here is a potential way to achieve this, if I understood this correctly. I haven't verified that this works though. Let's assume we have package `pkg` with submodule `pkg.sub`: ```...

#10686 has most of the implementation, but tests are failing and the PR is a bit stale. Any help with finishing up the PR would be appreciated!

The behavior changed in #17348. @ilevkivskyi Do you have an idea of what is going on?