feat: add schema to ast builder #26
Conversation
sardination
changed the title
|
Hi @sardination this makes sense indeed. We recently started upstreaming the SQL serialization logic to the main DataFusion repo in apache/datafusion#9596. I've yet to come up with a strategy around how to deal with the transition phase in this repo. In the meantime, let me know if you're interested in upstreaming this change as well or if you prefer I take care of that. |
Awesome! I will go ahead and open an upstream PR for this. |
|
I believe this was completed upstream, so closing the PR here! |
* optimize_recursively should check the provider for subqueries * Preserve the vec<provider>,use apply_subqueries * Federate subtree in subquery * clean up * Rewrite federation as optimizer rule * Revert "Rewrite federation as optimizer rule" This reverts commit e0ff999. * Keep federaion rule as analyzer rule, handle InSubquery federation * Wrap a Projection plan when the entire Insubquery can be federated * not run scalar_subquery_to_join and update scalar subquery handling * Resolve outref federation provider * clean up * Include table reference created by subquery aliases * Remove redundant clone * Do not panic on ScanResult unwrap
SQL languages will typically have schemas, so add support for schema names to the table scan AST builder.