fix(test): Slice rows properly when reading in tests#1632
fix(test): Slice rows properly when reading in tests#1632kodiakhq[bot] merged 3 commits intomainfrom
Conversation
6a25f17 to
7173025
Compare
This reverts commit 7173025.
erezrokah
left a comment
There was a problem hiding this comment.
Can you explain what bug this PR fixes? Maybe via a unit test?
During development of the Databricks destination I found the issue with returned records: in Databricks it's easy to have multirow arrow.Record returned directly from the database. If we pass this as is (with transformation) to the sdk, some of the methods we use in tests will fail:
Basically, it means that our SDK expects to receive a single-row record, but it's never documented. We have a similar issue inDuckDB destination, where we opted to reslice the returned record manually, but I think the slicing should be handled by the SDK itself: |
🤖 I have created a release *beep* *boop* --- ## [4.38.1](v4.38.0...v4.38.1) (2024-04-12) ### Bug Fixes * **deps:** Update golang.org/x/exp digest to c0f41cb ([#1615](#1615)) ([0c21bfb](0c21bfb)) * **deps:** Update module github.com/cloudquery/plugin-pb-go to v1.19.9 ([#1631](#1631)) ([5d45003](5d45003)) * **deps:** Update module google.golang.org/grpc to v1.63.2 ([#1617](#1617)) ([02461b1](02461b1)) * **test:** Slice rows properly when reading in tests ([#1632](#1632)) ([537b64c](537b64c)) --- This PR was generated with [Release Please](https://github.com/googleapis/release-please). See [documentation](https://github.com/googleapis/release-please#release-please).
We never actually required the read records to be a single-row ones, so we may as well slice them ourselves