Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

build: Move shim directories #318

Merged
merged 1 commit into from
Apr 25, 2024
Merged

Conversation

kazuyukitanimura
Copy link
Contributor

Which issue does this PR close?

Closes #140

Rationale for this change

To add Spark 3.5 and 4.0 shims later.

What changes are included in this PR?

This PR moves the shims to spark version specific directories

How are these changes tested?

Existing tests

Copy link
Member

@andygrove andygrove left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM. Thanks @kazuyukitanimura

Copy link
Member

@viirya viirya left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think the original idea in the ticket is to create different shim for different Spark version instead of using reflection API for all Spark versions?

@kazuyukitanimura
Copy link
Contributor Author

Thanks @andygrove @viirya

I think the original idea in the ticket is to create different shim for different Spark version instead of using reflection API for all Spark versions?

That's right. Currently it is spark-3.x dir only, but I plan to add spark-4.0 for example. If we want to split spark-3.4 dir out of spark-3.x we can do so as follow up.

@viirya
Copy link
Member

viirya commented Apr 24, 2024

That's right. Currently it is spark-3.x dir only, but I plan to add spark-4.0 for example. If we want to split spark-3.4 dir out of spark-3.x we can do so as follow up.

I think the original idea is to have spark-3.2 and spark-3.3 shims which use reflection API, and spark-3.4 which uses Spark API as it is available. So for the Spark versions which have the API available, its shim layer doesn't need to use reflection.

That's what I got from the ticket.

I'm not sure if you are going to do that? If so, then it is fine to have this as initial move.

@kazuyukitanimura
Copy link
Contributor Author

I'm not sure if you are going to do that? If so, then it is fine to have this as initial move.

My focus would be spark-4.0 without reflection API for now. If anyone is up for taking the task of splitting spark-3.x to 3.2, 3.3, and 3.4, they will be welcomed. Yes this as an initial move.

Copy link
Contributor

@advancedxy advancedxy left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM as an initial move

@viirya viirya merged commit ef94c55 into apache:main Apr 25, 2024
29 checks passed
@viirya
Copy link
Member

viirya commented Apr 25, 2024

Merged. Thanks.

@kazuyukitanimura
Copy link
Contributor Author

Thank you all!

himadripal pushed a commit to himadripal/datafusion-comet that referenced this pull request Sep 7, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Improve Spark shim layer
4 participants