fix missing case of secondary system in hot water

model_data/app.py

@@ -97,8 +97,7 @@ def handler():
                 )
             )
 
-    test = [x for x in data if "Conservatory" in x["floor-description"]]
-    test = pd.DataFrame(test)
+    z = pd.DataFrame([x for x in data if x["hotwater-description"] == "From second main heating system"])
 
     # Incorporate input data into cleaning
     cleaner = EpcClean(data + [p.data for p in input_properties])

model_data/epc_attributes/HotWaterAttributes.py

@@ -22,6 +22,7 @@ class HotWaterAttributes(BaseUtility):
         'from main system',  # The hot water is provided by the main heating system of the building
         'from secondary system',
         # The hot water is provided by a secondary (or supplementary) heating system in the building
+        'from second main heating system',  # Same as 'from secondary system'
         'community scheme',  # The hot water is provided by a community heating system
     ]
 
@@ -83,6 +84,11 @@ class HotWaterAttributes(BaseUtility):
         'gas range cooker',  # A gas-powered range cooker
     ]
 
+    # Descriptions which represent the same thing
+    SYNONYMS = {
+        'from second main heating system': 'from secondary system',
+    }
+
     def __init__(self, description: str):
         self.description: str = clean_description(description.lower())
 
@@ -128,7 +134,7 @@ class HotWaterAttributes(BaseUtility):
 
         result: Dict[str, Union[str, bool]] = {
             "heater_type": find_keyword(self.description, self.HEATER_TYPES),
-            "system_type": find_keyword(self.description, self.SYSTEM_TYPES),
+            "system_type": find_keyword(self.description, self.SYSTEM_TYPES, self.SYNONYMS),
             "thermostat_characteristics": find_keyword(self.description, self.THERMOSTAT_CHARACTERISTICS),
             "heating_scope": find_keyword(self.description, self.HEATING_SCOPE),
             "energy_recovery": find_keyword(self.description, self.ENERGY_RECOVERY),

model_data/epc_attributes/attribute_utils.py

@@ -118,7 +118,10 @@ def remove_double_spaces(text):
     return cleaned_text
 
 
-def find_keyword(description, keywords):
+def find_keyword(description, keywords, synonyms=None):
+    if synonyms is None:
+        synonyms = {}
+
     # Sort keywords by length, longest first.
     # This ensures that 'time and temperature zone control'
     # will be checked before 'temperature zone control' if both are present in the keywords list
@@ -126,13 +129,13 @@ def find_keyword(description, keywords):
 
     for keyword in keywords:
         if keyword in description:
-            return keyword
+            return synonyms.get(keyword, keyword)
 
     # If no keyword is found, try again after removing punctuation
     description_without_punct = remove_punctuation(description)
 
     for keyword in keywords:
         if keyword in description_without_punct:
-            return keyword
+            return synonyms.get(keyword, keyword)
 
     return None

model_data/tests/test_data/test_hot_water_attributes_cases.py

@@ -120,5 +120,10 @@ hotwater_cases = [
     {'original_description': 'Gas range cooker', 'heater_type': None, 'system_type': None,
      'thermostat_characteristics': None, 'heating_scope': None, 'energy_recovery': None, 'tariff_type': None,
      'extra_features': None, 'chp_systems': None, 'distribution_system': None, 'no_system_present': None,
-     'assumed': False, "appliance": "gas range cooker"}
+     'assumed': False, "appliance": "gas range cooker"},
+    {'original_description': 'From second main heating system', 'heater_type': None,
+     'system_type': 'from secondary system',
+     'thermostat_characteristics': None, 'heating_scope': None, 'energy_recovery': None, 'tariff_type': None,
+     'extra_features': None, 'chp_systems': None, 'distribution_system': None, 'no_system_present': None,
+     'assumed': False, "appliance": None}
 ]