Jimexist commented on pull request #687: URL: https://github.com/apache/arrow-datafusion/pull/687#issuecomment-874740222
> Lag/Lead signatures has to be changed to support 2nd/3nd optional arguments, is this correct or am I misunderstanding the use of Signature::OneOf ? that part was confusing to me as well. i believe a most elegant way is to create support for heterogenous support for function arguments > I'd like to test create_built_in_window_expr with the optional arguments on and off, to match against the lead/lag signatures, i'm okay with integration tests that cover these > Had to copy and change the original arrow::compute::kernels::window::shift to allow filling the start/end nulls with the default value (this new fn is called shift_with_default_value inside lead_lag crate) changing arrow-rs repo is one way to achieve it although it'll needs two more weeks to be released. i'm okay with keeping the copied and modified version here. -- 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. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
