Terry Kim
Terry Kim
cc @suhsteve @jonsequitur (FYI)
That will be great, thanks @spzSource!
Yea, I think that's one way to do it. But we have to make sure `microsoft-spark` is a fat JAR so that we don't break the existing customers' pipeline. Or...
Cool, that will be great. Please let me know if we can retain most of the existing formatting or not. Thanks!
Can you explain `tpcdsBasic` vs. `tpcdsOther`?
> I quickly checked and there was only one breaking test. Interesting, which test was failing?
Any reason we want to use the latest patch version? Do we need to keep track/update this every time a new patch version is released? Do you want to check...
> Question to @rapoth : Does the customer see listing file during index creation as bottleneck, or if CX really wanted to refresh only couple of partitions where the data...
> To apply the index, a user should set the exact file lists when they are writing the query, without Hybrid Scan. In other words, the relation in their query...
> * `createIndex(df, mode = "include", columns = Seq("C1", "C2"))`: Pick `{"C1", "C2"}` as the included columns. The current `createIndex` API takes in `IndexConfig`. Can you update the examples with...