DougLeonard

Thanks for the very detailed reply. No worries about responses. I'm not paying much for your tool. I guess the situation is about what I suspected, as implied by "dealing...

Oh regarding the standard replace refs, that's a nice tip. However, I do need replace refs active during the re-write to repatriate the split commits. I build those in reverse...

I did find this: https://stackoverflow.com/questions/44112593/how-to-get-parent-of-specific-commit-in-git And in the second answer by Nayagam specifically addresses shallow commits with this solution to find parents: ''' git cat-file -p commit_id ''' This does...

Oops, yes, you did write that ;) I was actually thinking to check for shallow commits by looking for the sha in the .git/shallow file (which I wrote ;) )...

It leaves the correct commits (originally 1 and 3). Master just doesn't get placed on the 1st parent (1), which does seem odd.

> Returning non-subclasses from `\_\_new\_\_()` is a hack for sure, but as explained in [#8330 (comment)](https://github.com/python/mypy/issues/8330#issuecomment-876872608) and #14426 there are real-world libraries like Django REST Framework which are using this...

quoting [intgr](https://github.com/intgr) from PR thread #14471: > Please make sure that if the `__new__` return is untyped (returns `Any`) then it uses the class type implicitly. Not sure if that...

@Viicos, Right but pyright is inferring Wolf even when there are no type annotations at all. Should it? You said narrowing is part of checking. I'm not sure if it's...

Thanks. I've seen it. And of course pyright can do whatever pyright wants to do. I believe mypy is meant to follow spec, but then neither really do.

> Please make sure that if the `__new__` return is untyped (returns `Any`) then it uses the class type implicitly. Not sure if that is already covered, but I suspect...