| FazBrowse GitHub Viewer | Trending | | Home |
| Tools: [Download Repo ZIP] [Original HTTPS Page] |
Sorry, something went wrong.
Python 3.5 always fails in CI these days Linking python#9950
There was a problem hiding this comment.
I look forward to dropping support so we can use TypeVar without needing to guard it.
Sorry, something went wrong.
|
Agreed. Should we keep the number of compiled suites by making 3.6 compiled? |
Sorry, something went wrong.
|
You probably have a better sense of that than I do — how likely is it that 3.6 compiled will catch issues not caught by 3.7 compiled and 3.6? |
Sorry, something went wrong.
|
I do seem to remember we had a CI false negative once because we didn't have a compiled >= 3.8 (due to some typeshed change to ast), so maybe we make Python 3.9 compiled |
Sorry, something went wrong.
There was a problem hiding this comment.
A compiled 3.9 LGTM!
Sorry, something went wrong.
|
I think that we should run compiled tests on the earliest supported/working Python version and the latest, so it would be 3.6 and 3.9 after this change. |
Sorry, something went wrong.
| Back | FazBrowse Home | New Git URL |
Python 3.5 always fails in CI these days. So a) it's annoying and a bad contributor experience, b) we're effectively not testing against Python 3.5 anyway
Linking #9950