From 87090a257c6874deaeafde62a53d1520848d6025 Mon Sep 17 00:00:00 2001
From: rileyh <rileyh@umn.edu>
Date: Mon, 27 Nov 2023 17:05:08 +0000
Subject: [PATCH] [#118] Add unit tests for override_column_<a/b> for
 column_mappings

---
 hlink/tests/core/column_mapping_test.py | 104 ++++++++++++++++++++++++
 1 file changed, 104 insertions(+)

diff --git a/hlink/tests/core/column_mapping_test.py b/hlink/tests/core/column_mapping_test.py
index 4db1639..eb694e6 100644
--- a/hlink/tests/core/column_mapping_test.py
+++ b/hlink/tests/core/column_mapping_test.py
@@ -194,6 +194,110 @@ def test_select_column_mapping_column_selects_preserved(spark):
     assert set(column_selects) == {"occ_with_underscores", "occupation"}
 
 
+def test_select_column_mapping_override_column_a(spark):
+    """
+    override_column_a lets the user specify a different column name for
+    dataset A. override_transforms are applied only to dataset A in
+    this case.
+    """
+    column_mapping = {
+        "column_name": "occ",
+        "override_column_a": "occupation",
+        "override_transforms": [{"type": "lowercase_strip"}],
+    }
+    df_a = spark.createDataFrame(TEST_DF_1)
+    df_b = spark.createDataFrame(TEST_DF_2)
+
+    df_selected_a, column_selects_a = select_column_mapping(
+        column_mapping,
+        df_a,
+        is_a=True,
+        column_selects=[],
+    )
+
+    df_selected_b, column_selects_b = select_column_mapping(
+        column_mapping,
+        df_b,
+        is_a=False,
+        column_selects=[],
+    )
+
+    assert column_selects_a == column_selects_b == ["occ"]
+
+    occ_a = df_selected_a.select("occ").toPandas()
+    assert occ_a["occ"].to_list() == [
+        "farmer",
+        "computer scientist",
+        "waitress",
+        "retired",
+        "lawyer",
+        "doctor",
+    ]
+
+    occ_b = df_selected_b.select("occ").toPandas()
+    assert occ_b["occ"].to_list() == [
+        "RETIRED",
+        "CHILDCARE",
+        None,
+        None,
+    ]
+
+
+def test_select_column_mapping_override_column_b(spark):
+    """
+    override_column_b lets the user specify a different column name for
+    dataset B. override_transforms are applied only to dataset B in
+    this case, and transforms are applied only to dataset A.
+    """
+    column_mapping = {
+        "column_name": "occupation",
+        "override_column_b": "occ",
+        "override_transforms": [
+            {"type": "concat_two_cols", "column_to_append": "identifier"}
+        ],
+        "transforms": [
+            {"type": "lowercase_strip"},
+            {"type": "concat_two_cols", "column_to_append": "id"},
+        ],
+    }
+    df_a = spark.createDataFrame(TEST_DF_1)
+    df_b = spark.createDataFrame(TEST_DF_2)
+
+    df_selected_a, column_selects_a = select_column_mapping(
+        column_mapping,
+        df_a,
+        is_a=True,
+        column_selects=[],
+    )
+
+    df_selected_b, column_selects_b = select_column_mapping(
+        column_mapping,
+        df_b,
+        is_a=False,
+        column_selects=[],
+    )
+
+    assert column_selects_a == column_selects_b == ["occupation"]
+
+    occ_a = df_selected_a.select("occupation").toPandas()
+    assert occ_a["occupation"].to_list() == [
+        "farmer0",
+        "computer scientist1",
+        "waitress2",
+        "retired3",
+        "lawyer4",
+        "doctor5",
+    ]
+
+    occ_b = df_selected_b.select("occupation").toPandas()
+    assert occ_b["occupation"].to_list() == [
+        "RETIRED1000",
+        "CHILDCARE1002",
+        None,
+        None,
+    ]
+
+
 def test_select_column_mapping_error_missing_column_name(spark):
     """
     Without a column_name key in the column_mapping, the function raises