changed test backend to duckdb

Bayer-Group · Nov 14, 2024 · dd0d7dc · dd0d7dc
1 parent 6c2b507
commit dd0d7dc
Show file tree

Hide file tree

Showing 3 changed files with 29 additions and 6 deletions.
diff --git a/phenex/filters/aggregator.py b/phenex/filters/aggregator.py
@@ -88,7 +88,6 @@ def aggregate(self, input_table: Table):
             _aggregation_column = getattr(input_table, self.aggregation_column)
 
         ibis.options.interactive = True
-        print(input_table)
         # Define the window specification
         window_spec = ibis.window(group_by=_aggregation_index_cols)
 
@@ -98,7 +97,6 @@ def aggregate(self, input_table: Table):
         elif self.aggregation_function == "mean":
             aggregation = _aggregation_column.mean().over(window_spec)
         elif self.aggregation_function == "max":
-            print("DOING MAX")
             aggregation = _aggregation_column.max().over(window_spec)
         elif self.aggregation_function == "min":
             aggregation = _aggregation_column.min().over(window_spec)

diff --git a/phenex/test/phenotype_test_generator.py b/phenex/test/phenotype_test_generator.py
@@ -4,6 +4,7 @@
 import pandas as pd
 import ibis
 import ibis.expr.datatypes as dt
+from pyarrow import int8
 
 from .util.check_equality import check_equality
 
@@ -25,7 +26,7 @@ class PhenotypeTestGenerator:
 
     def run_tests(self, verbose=False):
         self.verbose = verbose
-        self.con = ibis.sqlite.connect()
+        self.con = ibis.duckdb.connect()
         self._create_artifact_directory(self.name_space)
         self._create_input_data()
         self._run_tests()
@@ -69,10 +70,21 @@ def _create_input_data(self):
             # seed data dates in pandas datetime format and let the
             # TestGenerator format them into strings.
             input_info["df"].to_csv(path, index=False, date_format=self.date_format)
+
+            schema = {}
+            for col in input_info["df"].columns:
+                if 'date' in col.lower():
+                    schema[col] = datetime.date
+                elif 'value' in col.lower():
+                    schema[col] = float
+                else:
+                    schema[col] = str
+
             self.domains[input_info["name"]] = self.con.create_table(
-                input_info["name"], input_info["df"]
+                input_info["name"], input_info["df"], schema = schema
             )
 
+
     def _run_tests(self):
         def df_from_test_info(test_info):
             df = pd.DataFrame()
@@ -124,8 +136,21 @@ def df_from_test_info(test_info):
                 path, index=False, date_format=self.date_format
             )
 
+
+            schema = {}
+            for col in df.columns:
+                if 'date' in col.lower():
+                    schema[col] = datetime.date
+                elif 'value' in col.lower():
+                    schema[col] = float
+                elif 'boolean' in col.lower():
+                    schema[col] = bool
+                else:
+                    schema[col] = str
+
             expected_output_table = self.con.create_table(
-                self.name_output_file(test_info), df
+                self.name_output_file(test_info), df, schema=schema
+
             )
 
             join_on = ['PERSON_ID']

diff --git a/phenex/test/phenotypes/test_measurement_phenotype.py b/phenex/test/phenotypes/test_measurement_phenotype.py
@@ -780,7 +780,7 @@ def test_measurement_phenotype_return_values():
 
 def test_measurement_phenotype_value_aggregation():
     spg = MeasurementPhenotypeValueAggregationTestGenerator()
-    spg.run_tests(verbose=True)
+    spg.run_tests(verbose=False)
 
 
 if __name__ == "__main__":