Python (scalar UDF)

import krishiv as ks

session = ks.Session.embedded()

@ks.udf(return_type="utf8")
def greet(name: str) -> str:
    return f"hello, {name}"

session.register_udf("greet", greet, ["utf8"], "utf8")

session.sql("SELECT greet(name) FROM users").show()

If you are using DataFusion-style registration, the supported call is session.register_udf(name, fn, input_types, return_type). See Python Session.

Rust (scalar UDF)

use krishiv_api::{col, Session};
use datafusion::logical_expr::Volatility;
use datafusion::functions_nested;

#[tokio::main]
async fn main() -> krishiv_api::Result<()> {
    let session = Session::embedded().await?;
    // Register a closure-based scalar UDF
    session.register_table_udf_fn(
        "shout",
        std::sync::Arc::new(arrow::datatypes::Schema::empty()),
        |_args| Ok(arrow::array::RecordBatch::new_empty(std::sync::Arc::new(arrow::datatypes::Schema::empty()))),
    )?;
    Ok(())
}

Most production UDFs are written as DataFusion ScalarUDF implementations and registered via session.register_udf(udf). See the Rust Expressions page for the patterns.

SQL-body table-valued function

For pure-SQL TVFs you do not need to register anything in code — see the SQL UDFs reference.