@kravets-levko would you mind also reviewing this PR & releasing afterwards?
I've been using a fork at https://github.com/cloudquery/databricks-sql-go (cqmain branch) with #216 & #217 merged into to achieve everything required for the code I'm developing.

The major bump to v15 just updates the library code, the underlying arrow data (wire protocol) stays the same.

@candiduslynx I totally understand your need in this upgrade, however I'm a bit new to this project in particular and Go language in general. So I need @andrefurlan-db or @rcypher-databricks approval, and then I'll merge and ship this change. Sorry, and thank you for understanding

@andrefurlan-db can you help out on this one?

@yunbodeng-db @andrefurlan-db @kravets-levko I've updated the PR as v16 was released.
I'd like to see this merged sooner than later, though.

@kravets-levko thanks for the review!
Is there a way to expedite the merge & release of this code along with #207?

@candiduslynx #207 should be already available in v1.5.4. For this PR - I think I'll add it to the next release, hopefully will be available on Monday-Tuesday

@kravets-levko it's great to see this merged, thanks!
I'm eagerly waiting for the new release 🙂

@candiduslynx v1.5.5 is out - please give it a try and let me know if everything is okay. Thank you for patience

@candiduslynx sorry for a bad news, but turns out that Arrow upgrade caused issues for library users. Library offers a feature that allows to access raw Arrow batches, and with this upgrade behavior of that feature changed. After discussion with other team members, we decided to revert this upgrade for now. v1.5.5 will remain unchanged, though, so you can keep using it

@kravets-levko thanks for notifying!
I think you need to add some tests to outline the expected behaviour then.
However, I wonder what are the details of the changed behaviour that isn't deemed correct by clients? Is it just a requirement to update the library?