test(substrait): update TPCH tests (#12462)

* test(substrait): rename vendored test files to match upstream

* test(substrait): update vendored TPCH plans from substrait-consumer

version: https://github.com/substrait-io/consumer-testing/tree/d9a0453

* test(substrait): collect schemas through from expressions

Substrait plans can contains relations within expressions, for example
in Subqueries. As such, in order to fully collect schemas we must
traverse through relation and expressions.

* feat(substrait): more flexible IntervalDayToSecond handling

When subseconds is not set, allow for missing precision_mode

* test(substrait): update existing TPCH tests

* test(substrait): include new TPCH tests

* test(substrait): remove unused test csv files

* test(substrait): minor TestSchemaCollector updates

* clippy

datafusion/substrait/src/logical_plan/consumer.rs

@@ -2064,14 +2064,19 @@ fn from_substrait_literal(
             // DF only supports millisecond precision, so for any more granular type we lose precision
             let milliseconds = match precision_mode {
                 Some(PrecisionMode::Microseconds(ms)) => ms / 1000,
+                None =>
+                    if *subseconds != 0 {
+                        return substrait_err!("Cannot set subseconds field of IntervalDayToSecond without setting precision");
+                    } else {
+                        0_i32
+                    }
                 Some(PrecisionMode::Precision(0)) => *subseconds as i32 * 1000,
                 Some(PrecisionMode::Precision(3)) => *subseconds as i32,
                 Some(PrecisionMode::Precision(6)) => (subseconds / 1000) as i32,
                 Some(PrecisionMode::Precision(9)) => (subseconds / 1000 / 1000) as i32,
                 _ => {
                     return not_impl_err!(
-                        "Unsupported Substrait interval day to second precision mode"
-                    )
+                    "Unsupported Substrait interval day to second precision mode: {precision_mode:?}")
                 }
             };
 

datafusion/substrait/tests/cases/function_test.rs

@@ -28,7 +28,7 @@ mod tests {
     #[tokio::test]
     async fn contains_function_test() -> Result<()> {
         let proto_plan = read_json("tests/testdata/contains_plan.substrait.json");
-        let ctx = add_plan_schemas_to_ctx(SessionContext::new(), &proto_plan);
+        let ctx = add_plan_schemas_to_ctx(SessionContext::new(), &proto_plan)?;
         let plan = from_substrait_plan(&ctx, &proto_plan).await?;
 
         let plan_str = format!("{}", plan);

datafusion/substrait/tests/cases/logical_plans.rs

@@ -37,7 +37,7 @@ mod tests {
         // ./isthmus-cli/build/graal/isthmus --create "create table data (d boolean)" "select not d from data"
         let proto_plan =
             read_json("tests/testdata/test_plans/select_not_bool.substrait.json");
-        let ctx = add_plan_schemas_to_ctx(SessionContext::new(), &proto_plan);
+        let ctx = add_plan_schemas_to_ctx(SessionContext::new(), &proto_plan)?;
         let plan = from_substrait_plan(&ctx, &proto_plan).await?;
 
         assert_eq!(
@@ -62,7 +62,7 @@ mod tests {
         // ./isthmus-cli/build/graal/isthmus --create "create table data (d int, part int, ord int)" "select sum(d) OVER (PARTITION BY part ORDER BY ord ROWS BETWEEN 1 PRECEDING AND UNBOUNDED FOLLOWING) AS lead_expr from data"
         let proto_plan =
             read_json("tests/testdata/test_plans/select_window.substrait.json");
-        let ctx = add_plan_schemas_to_ctx(SessionContext::new(), &proto_plan);
+        let ctx = add_plan_schemas_to_ctx(SessionContext::new(), &proto_plan)?;
         let plan = from_substrait_plan(&ctx, &proto_plan).await?;
 
         assert_eq!(
@@ -81,7 +81,7 @@ mod tests {
         // This test confirms that reading a plan with non-nullable lists works as expected.
         let proto_plan =
             read_json("tests/testdata/test_plans/non_nullable_lists.substrait.json");
-        let ctx = add_plan_schemas_to_ctx(SessionContext::new(), &proto_plan);
+        let ctx = add_plan_schemas_to_ctx(SessionContext::new(), &proto_plan)?;
         let plan = from_substrait_plan(&ctx, &proto_plan).await?;
 
         assert_eq!(format!("{}", &plan), "Values: (List([1, 2]))");