[HUDI-4364]: changes for integrating column stats index into presto-h…

[pratyakshsharma]

…udi connector

Tips

• Thank you very much for contributing to Apache Hudi.
• Please review https://hudi.apache.org/contribute/how-to-contribute before opening a pull request.

What is the purpose of the pull request

(For example: This pull request adds quick-start document.)

Brief change log

(for example:)

• Modify AnnotationLocation checkstyle rule in checkstyle.xml

Verify this pull request

(Please pick either of the following options)

This pull request is a trivial rework / code cleanup without any test coverage.

(or)

This pull request is already covered by existing tests, such as (please describe tests).

(or)

This change added tests and can be verified as follows:

(example:)

• Added integration tests for end-to-end.
• Added HoodieClientWriteTest to verify the change.
• Manually verified the change by running a job locally.

Committer checklist

• Has a corresponding JIRA in PR title & commit

• Commit message is descriptive of the change

• CI is green

• Necessary doc changes done or have another open PR

• For large changes, please consider breaking it into sub-tasks under an umbrella JIRA.

[pratyakshsharma]

@hudi-bot run azure

[xiarixiaoyao]

@pratyakshsharma nice work!
A little question:
Why don't we put those codes into Presto Hudi connector so that we can reuse related classes of Presto directly

[pratyakshsharma]

@xiarixiaoyao this is a good question, something I have been thinking about too. The idea is to build a layer that will help in integrating column stats index with all java based engines like presto, trino and hive. This lays the foundation, since we need something like ranges or column domains so as to be able to filter the files using min and max values. Few classes here are actually inspired from those present in presto, but they are not exactly similar.
We can end up writing this logic in presto, but then a similar work will have to be done for trino as well. With this piece of code in Hudi, we just need an adapter in presto/trino to be able to call the api exposed for filtering files in this PR.

Although since this is just the beginning of this work, I am open to hear others' thoughts on this.

[codope]

cc @alexeykudinkin

[alexeykudinkin]

@pratyakshsharma thanks for taking the time to contribute this!

We definitely want to make sure that the code integrating w/ Presto/Trino/Hive is reusable as much as possible, and i think we should start to think about it upfront to avoid churn of refactoring things back and forth. Given the scope of this integration as well as its impact, i think we'd def go for RFC for it to make sure we solicit the feedback from the community before go too far w/ the implementation.

[pratyakshsharma]

@alexeykudinkin An epic is filed here - https://issues.apache.org/jira/browse/HUDI-4394.

Please note this draft PR is intended as a POC and would work well with Presto. We were actually planning to get this with 0.12 release. If not, we can target this for 1.0.0

[hudi-bot]

CI report:

• cf13b8b Azure: FAILURE

Bot commands

@hudi-bot supports the following commands:

• @hudi-bot run azure re-run the last Azure build

[alexeykudinkin]

Got it. Yeah, i don't think we'll be able to make it into 0.12 given that we're planning to do a code freeze next week.

And again, i don't think we can go with the project of this size, scope and more importantly impact (it'll be affecting all forthcoming execution engines like Flink, Presto, Trino, Hive, etc) w/o an RFC.

[pratyakshsharma]

Agree with you on this. Let me draft an RFC and we can take it up from there.