From 4fad0d6d00fd21370c72e8561646bb0370da463d Mon Sep 17 00:00:00 2001
From: "samuel.oranyeli" <samuel.oranyeli@grow.inc>
Date: Wed, 1 Feb 2023 13:50:00 +1100
Subject: [PATCH] add more tests

---
 janitor/functions/summarize.py    |  2 +-
 tests/functions/test_summarize.py | 20 ++++++++++++++++++--
 2 files changed, 19 insertions(+), 3 deletions(-)

diff --git a/janitor/functions/summarize.py b/janitor/functions/summarize.py
index 1a4995e65..77a39c363 100644
--- a/janitor/functions/summarize.py
+++ b/janitor/functions/summarize.py
@@ -118,7 +118,7 @@ def summarize(
         grp = df.groupby(**by)
     elif by_is_true and isinstance(by, col):
         if by.func:
-            raise ValueError("Function assignment is not required within by.")
+            raise ValueError("Function assignment is not required within by")
         cols = get_index_labels([*by.cols], df, axis="columns")
         if by.remove_cols:
             exclude = get_index_labels([*by.remove_cols], df, axis="columns")
diff --git a/tests/functions/test_summarize.py b/tests/functions/test_summarize.py
index f212af6cb..b78cce7bb 100644
--- a/tests/functions/test_summarize.py
+++ b/tests/functions/test_summarize.py
@@ -3,7 +3,7 @@
 import pandas as pd
 
 from pandas.testing import assert_frame_equal
-from pandas.api.types import is_numeric_dtype
+from pandas.api.types import is_numeric_dtype, is_string_dtype
 from janitor import col
 
 
@@ -19,6 +19,20 @@ def test_Column_agg_error(dataframe):
         dataframe.summarize(col("a").compute(func))
 
 
+@pytest.mark.functions
+def test_Column_by_error(dataframe):
+    """
+    Raise if `by` is a col class, and has a function
+    assigned to it
+    """
+    with pytest.raises(
+        ValueError, match="Function assignment is not required within by"
+    ):
+        dataframe.summarize(
+            col("a").compute("sum"), by=col("b").compute("size")
+        )
+
+
 @pytest.mark.functions
 def test_Column_exclude_error(dataframe):
     """
@@ -210,5 +224,7 @@ def test_dataframe():
 
     df = pd.DataFrame(df)
     expected = df.groupby("A").C.describe().add_prefix("C_")
-    actual = df.summarize(col("C").compute("describe"), by="A")
+    actual = df.summarize(
+        col("C").compute("describe"), by=col(is_string_dtype).exclude("B")
+    )
     assert_frame_equal(expected, actual)