rishvin commented on code in PR #1971:
URL: https://github.com/apache/datafusion-comet/pull/1971#discussion_r2201439622
##########
native/spark-expr/src/comet_scalar_funcs.rs:
##########
@@ -53,13 +54,23 @@ macro_rules! make_comet_scalar_udf {
);
Ok(Arc::new(ScalarUDF::new_from_impl(scalar_func)))
}};
+ ($name:expr, $func:ident, without $data_type:ident, $fail_on_error:ident)
=> {{
+ let scalar_func = CometScalarFunction::new(
+ $name.to_string(),
+ Signature::variadic_any(Volatility::Immutable),
+ $data_type,
+ Arc::new(move |args| $func(args, $fail_on_error)),
+ );
+ Ok(Arc::new(ScalarUDF::new_from_impl(scalar_func)))
+ }};
}
/// Create a physical scalar function.
pub fn create_comet_physical_fun(
fun_name: &str,
data_type: DataType,
registry: &dyn FunctionRegistry,
+ fail_on_error: Option<bool>,
Review Comment:
Thanks @mbutrovich. I used Option<bool> under the assumption that the
`fail_on_error` flag may not be applicable to all functions. Using
`fail_on_error = None` indicates that this flag will be irrelevant to the
corresponding `fun_name`. Since `fail_on_error = true|false` means whether to
execute operation in ANSI|non-ANSI mode, `Option<bool>` would indicate that
ANSI|non-ANSI mode not applicable to that function.
Let me know if you think a default `false` would be more
appropriate/consistent here instead.
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
