feat(cubestore): Add XIRR aggregate function to Cube Store (#9520)

Author: srh

rust/cubesql/cubesql/src/compile/engine/udf/extension/xirr.rs

@@ -14,6 +14,9 @@ use datafusion::{
     scalar::ScalarValue,
 };
 
+// Note:  A copy/pasted and minimally(?) modified version of this is in Cubestore in udf_xirr.rs, and you might
+// want to update both.
+
 pub const XIRR_UDAF_NAME: &str = "xirr";
 
 /// Creates a XIRR Aggregate UDF.

rust/cubestore/cubestore-sql-tests/src/tests.rs

@@ -134,6 +134,7 @@ pub fn sql_tests() -> Vec<(&'static str, TestFn)> {
         t("hyperloglog_postgres", hyperloglog_postgres),
         t("hyperloglog_snowflake", hyperloglog_snowflake),
         t("hyperloglog_databricks", hyperloglog_databricks),
+        t("xirr", xirr),
         t(
             "aggregate_index_hll_databricks",
             aggregate_index_hll_databricks,
@@ -2802,6 +2803,122 @@ async fn hyperloglog_databricks(service: Box<dyn SqlClient>) {
     assert_eq!(to_rows(&r), rows(&[(1, 4), (2, 4), (3, 20)]));
 }
 
+async fn xirr(service: Box<dyn SqlClient>) {
+    // XIRR result may differ between platforms, so we truncate the results with LEFT(_, 10).
+    let r = service
+        .exec_query(
+            r#"
+        SELECT LEFT(XIRR(payment, date)::varchar, 10) AS xirr
+        FROM (
+            SELECT '2014-01-01'::date AS date, -10000.0 AS payment
+            UNION ALL
+            SELECT '2014-03-01'::date AS date, 2750.0 AS payment
+            UNION ALL
+            SELECT '2014-10-30'::date AS date, 4250.0 AS payment
+            UNION ALL
+            SELECT '2015-02-15'::date AS date, 3250.0 AS payment
+            UNION ALL
+            SELECT '2015-04-01'::date AS date, 2750.0 AS payment
+        ) AS "t"
+        "#,
+        )
+        .await
+        .unwrap();
+
+    assert_eq!(to_rows(&r), rows(&["0.37485859"]));
+
+    let r = service
+        .exec_query(
+            r#"
+        SELECT LEFT(XIRR(payment, date)::varchar, 10) AS xirr
+        FROM (
+            SELECT '2014-01-01'::date AS date, -10000.0 AS payment
+        ) AS "t"
+        WHERE 0 = 1
+        "#,
+        )
+        .await
+        .unwrap_err();
+    assert_eq!(r.elide_backtrace(), CubeError::internal("Arrow error: External error: Execution error: A result for XIRR couldn't be determined because the arguments are empty".to_owned()));
+
+    let r = service
+        .exec_query(
+            r#"
+        SELECT LEFT(XIRR(payment, date)::varchar, 10) AS xirr
+        FROM (
+            SELECT '2014-01-01'::date AS date, 10000.0 AS payment
+        ) AS "t"
+        "#,
+        )
+        .await
+        .unwrap_err();
+    assert_eq!(r.elide_backtrace(), CubeError::internal("Arrow error: External error: Execution error: The XIRR function couldn't find a solution".to_owned()));
+
+    // --- on_error testing ---
+
+    let r = service
+        .exec_query(
+            r#"
+        SELECT LEFT(XIRR(payment, date, 0, NULL::double)::varchar, 10) AS xirr
+        FROM (
+            SELECT '2014-01-01'::date AS date, -10000.0 AS payment
+            UNION ALL
+            SELECT '2014-03-01'::date AS date, 2750.0 AS payment
+            UNION ALL
+            SELECT '2014-10-30'::date AS date, 4250.0 AS payment
+            UNION ALL
+            SELECT '2015-02-15'::date AS date, 3250.0 AS payment
+            UNION ALL
+            SELECT '2015-04-01'::date AS date, 2750.0 AS payment
+        ) AS "t"
+        "#,
+        )
+        .await
+        .unwrap();
+
+    assert_eq!(to_rows(&r), rows(&["0.37485859"]));
+
+    let r = service
+        .exec_query(
+            r#"
+        SELECT LEFT(XIRR(payment, date, 0, NULL::double)::varchar, 10) AS xirr
+        FROM (
+            SELECT '2014-01-01'::date AS date, -10000.0 AS payment
+        ) AS "t"
+        WHERE 0 = 1
+        "#,
+        )
+        .await
+        .unwrap_err();
+    assert_eq!(r.elide_backtrace(), CubeError::internal("Arrow error: External error: Execution error: A result for XIRR couldn't be determined because the arguments are empty".to_owned()));
+
+    let r = service
+        .exec_query(
+            r#"
+        SELECT LEFT(XIRR(payment, date, 0, NULL::double)::varchar, 10) AS xirr
+        FROM (
+            SELECT '2014-01-01'::date AS date, 10000.0 AS payment
+        ) AS "t"
+        "#,
+        )
+        .await
+        .unwrap();
+    assert_eq!(to_rows(&r), rows(&[()]));
+
+    let r = service
+        .exec_query(
+            r#"
+        SELECT LEFT(XIRR(payment, date, 0, 12345)::varchar, 10) AS xirr
+        FROM (
+            SELECT '2014-01-01'::date AS date, 10000.0 AS payment
+        ) AS "t"
+        "#,
+        )
+        .await
+        .unwrap();
+    assert_eq!(to_rows(&r), rows(&["12345.0"]));
+}
+
 async fn aggregate_index_hll_databricks(service: Box<dyn SqlClient>) {
     service.exec_query("CREATE SCHEMA s").await.unwrap();
     service

rust/cubestore/cubestore/src/queryplanner/mod.rs

@@ -23,6 +23,7 @@ pub mod now;
 pub mod providers;
 #[cfg(test)]
 mod test_utils;
+pub mod udf_xirr;
 pub mod udfs;
 
 use crate::cachestore::CacheStore;
@@ -429,6 +430,7 @@ impl ContextProvider for MetaStoreSchemaProvider {
         // TODO: case-insensitive names.
         let kind = match name {
             "merge" | "MERGE" => CubeAggregateUDFKind::MergeHll,
+            "xirr" | "XIRR" => CubeAggregateUDFKind::Xirr,
             _ => return None,
         };
         return Some(Arc::new(aggregate_udf_by_kind(kind).descriptor()));