Drop Python 3.7 support #2169
Drop Python 3.7 support #2169
Conversation
gsheni
commented
Jul 5, 2022
•
gsheni
commented
- Fixes Drop support for Python 3.7 in the future #1893
|
This is blocked until Woodwork does a release. |
setup.cfg
Outdated
| @@ -49,12 +48,12 @@ install_requires = | |||
| distributed >= 2021.10.0 | |||
| holidays >= 0.13 | |||
| numpy >= 1.21.0 | |||
| pandas >= 1.3.0 | |||
| pandas >= 1.4.0, != 1.4.2 | |||
There was a problem hiding this comment.
Do we need to bump the minimum pandas and woodwork versions for the base install?
setup.cfg
Outdated
| pyspark >=3.2.0 | ||
| pandas >=1.3.0, <1.4.0 | ||
| pyspark >= 3.2.0, <3.3.0 | ||
| pandas >= 1.4.3 |
There was a problem hiding this comment.
Do we need to add this? Won't the dependency resolver figure this out, depending on the Woodwork version we have installed?
There was a problem hiding this comment.
@gsheni @ozzieD Does this change have anything to do with dropping support for Python 3.7?
I'm trying to think thorough all the dependency changes, and I'm not sure I'm understanding everything yet, but this feels more like a change for compatibility with the upcoming release of Woodwork. If that is the case, would it be more appropriate to implement the change to these requirements when the new version of Woodwork is released?
There was a problem hiding this comment.
Yes, the latest released version of Woodwork restricts the pandas version, which makes implementing this PR hard.
We get the following errors:
We are going to wait for Woodwork to do another release before proceeding in this PR.
There was a problem hiding this comment.
I'm confused still - why do we need to have pandas >= 1.4.3 here?
There was a problem hiding this comment.
Are you saying that if we leave the spark requirements section as it is on main currently, that we can't proceed with this PR?
