Dandandan commented on pull request #288: URL: https://github.com/apache/arrow-datafusion/pull/288#issuecomment-835811345
@msathis Thanks for this PR! I believe we should do it a bit differently and not calculate the timestamp inside the optimization rule. One could disable the optimization rule, or execute an un-optimized plan. The result of the query needs to be correct in those cases too. My suggestion would be to set field to the `ExecutionContextState` (something like `queryExecutionStartTime`) at the start of a query and pass it to the optimizers / evaluation code. This requires maybe some more changes to the code, but is needed to correctly run the queries. For the optimizer you could use `utils::expr_sub_expressions` (see use in filter push down) to recurse into the expressions and replace *all* cases of the `NOW` with the timestamp of the start of the query. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org
