adjust timestamp casting

dask-contrib · May 2, 2022 · 3b7cb4d · 3b7cb4d
1 parent 8beb813
commit 3b7cb4d
Show file tree

Hide file tree

Showing 4 changed files with 5 additions and 16 deletions.
diff --git a/dask_planner/src/expression.rs b/dask_planner/src/expression.rs
@@ -96,7 +96,6 @@ impl PyExpr {
                                     }
                                     _ => {
                                         println!("Encountered a non-Aggregate type");
-
                                         name.clone().to_ascii_uppercase()
                                     }
                                 }
@@ -109,7 +108,10 @@ impl PyExpr {
             }
             Expr::Column(column) => column.name.clone(),
             Expr::ScalarVariable(..) => unimplemented!("ScalarVariable!!!"),
-            Expr::Literal(..) => unimplemented!("Literal!!!"),
+            Expr::Literal(scalar_value) => {
+                println!("Scalar Value: {:?}", scalar_value);
+                unimplemented!("Literal!!!")
+            }
             Expr::BinaryExpr { .. } => {
                 // If the BinaryExpr does not have an Alias
                 // Ex: `df.a - Int64(1)` then use the String

diff --git a/dask_sql/physical/rel/logical/project.py b/dask_sql/physical/rel/logical/project.py
@@ -38,10 +38,6 @@ def convert(self, rel: "LogicalPlan", context: "dask_sql.Context") -> DataContai
         new_columns = {}
         new_mappings = {}
 
-        # Debugging only
-        for key, expr in named_projects:
-            print(f"Key: {key} - Expr: {expr.toString()}", str(key), expr)
-
         # Collect all (new) columns this Projection will limit to
         for key, expr in named_projects:
 

diff --git a/dask_sql/physical/rex/core/call.py b/dask_sql/physical/rex/core/call.py
@@ -235,7 +235,7 @@ def cast(self, operand, rex=None) -> SeriesOrScalar:
         # TODO: ideally we don't want to directly access the datetimes,
         # but Pandas can't truncate timezone datetimes and cuDF can't
         # truncate datetimes
-        if output_type == "DATE":
+        if output_type == "DATE" or output_type == "TIMESTAMP":
             return return_column.dt.floor("D").astype(python_type)
 
         return return_column
@@ -887,24 +887,16 @@ def convert(
         context: "dask_sql.Context",
     ) -> SeriesOrScalar:
 
-        print(f"\n\nEntering call.py convert for expr: {expr.toString()}")
-
-        for ex in expr.getOperands():
-            print(f"convert operand expr: {ex.toString()}")
-
         # Prepare the operands by turning the RexNodes into python expressions
         operands = [
             RexConverter.convert(rel, o, dc, context=context)
             for o in expr.getOperands()
         ]
 
-        print(f"\nOperands post conversion: {operands}")
-
         # Now use the operator name in the mapping
         # TODO: obviously this needs to not be hardcoded but not sure of the best place to pull the value from currently???
         schema_name = "root"
         operator_name = expr.getOperatorName().lower()
-        print(f"Operator Name: {operator_name}")
 
         try:
             operation = self.OPERATION_MAPPING[operator_name]

diff --git a/tests/integration/test_groupby.py b/tests/integration/test_groupby.py
@@ -22,7 +22,6 @@ def timeseries_df(c):
     return None
 
 
-@pytest.mark.skip(reason="WIP DataFusion")
 def test_group_by(c):
     return_df = c.sql(
         """