From d4b3954e17aebad14f0e76774e1b6945a1893146 Mon Sep 17 00:00:00 2001
From: Jake Adams <jdadams@utah.gov>
Date: Mon, 4 Nov 2024 17:15:30 -0700
Subject: [PATCH] chore: replace infs with nans

---
 src/wmrc/summarize.py   |  2 ++
 tests/test_summarize.py | 25 +++++++++++++++++++++++++
 2 files changed, 27 insertions(+)

diff --git a/src/wmrc/summarize.py b/src/wmrc/summarize.py
index ff85ded..34e3193 100644
--- a/src/wmrc/summarize.py
+++ b/src/wmrc/summarize.py
@@ -2,6 +2,7 @@
 Calender_Year__c to create dataframes of the reports that will be used to update the AGOL feature services.
 """
 
+import numpy as np
 import pandas as pd
 
 try:
@@ -233,6 +234,7 @@ def recovery_rates_by_tonnage(records: helpers.SalesForceRecords) -> pd.Series:
         )
     )
 
+    clean_rates.replace([np.inf, -np.inf], np.nan, inplace=True)  #: Can arise from division by np.nan
     clean_rates.name = "annual_recycling_uncontaminated_rate"
     clean_rates.index.name = "data_year"
     clean_rates.index = clean_rates.index.map(helpers.convert_to_int)
diff --git a/tests/test_summarize.py b/tests/test_summarize.py
index 47598d7..9977915 100644
--- a/tests/test_summarize.py
+++ b/tests/test_summarize.py
@@ -101,6 +101,31 @@ def test_recovery_rates_by_tonnage_happy_path(self, mocker):
 
         pd.testing.assert_series_equal(output_series, test_df)
 
+    def test_recovery_rates_by_tonnage_replaces_inf_with_zero(self, mocker):
+        records = mocker.Mock()
+        records.df = pd.DataFrame(
+            {
+                "Calendar_Year__c": [2022, 2022, 2023, 2023],
+                "Out_of_State__c": [0, 0, 0, 0],
+                "Municipal_Solid_Waste__c": [100, 100, 100, 100],
+                "Annual_Recycling_Contamination_Rate__c": [np.nan, np.nan, 50, 50],
+                "Combined_Total_of_Material_Recycled__c": [50, 100, 50, 40],
+            }
+        )
+
+        output_series = summarize.recovery_rates_by_tonnage(records)
+
+        test_df = pd.Series(
+            {
+                2022: np.nan,
+                2023: 50.0,
+            },
+            name="annual_recycling_uncontaminated_rate",
+        )
+        test_df.index.name = "data_year"
+
+        pd.testing.assert_series_equal(output_series, test_df)
+
     def test_recovery_rates_by_tonnage_uses_out_of_state_modifier(self, mocker):
         records = mocker.Mock()
         records.df = pd.DataFrame(