implemented first version of MainhearAttributes - wip

epc_data/app.py

@@ -40,11 +40,11 @@ def handler():
     cleaner.clean()
 
     # For testing:
-    from epc_data.attributes.FloorAttributes import FloorAttributes
-    descriptions = {x["floor-description"] for x in data}
+    from epc_data.attributes.MainheatAttributes import MainHeatAttributes
+    descriptions = {x["mainheat-description"] for x in data}
     out = []
     for description in descriptions:
-        res = FloorAttributes(description).process()
+        res = MainHeatAttributes(description).process()
         out.append(
             {
                 "original_description": description,
@@ -53,5 +53,29 @@ def handler():
         )
     df = pd.DataFrame(out)
     df = df.sort_values("original_description")
+    df = df.reset_index(drop=True)
+
+    z = df[df["original_description"] == 'Air source heat pump, radiators and underfloor, electric']
+
+    # Up to index: 14
+    # Bugs:
+    # 1)
+    # Description: 'Air source heat pump fan coil units, electric'
+    # Issue: Because "oil" is a fuel type, "oil" is stripped out of the description and the description
+    #        gets converted to "fan c units". It also marks this description as having oil, which it doesn't
+    #        So this code probably won't detect any "fan coil units"
+    # 2)
+    # Description: 'Air source heat pump, Systems with radiators, electric'
+    # Issue: Check detecton of Systems with radiators - it's only searching for "radiators" in DISTRIBUTION_SYSTEMS
+    #        This may actually be fine as we have other descriptions such as
+    #        'Air source heat pump, Underfloor heating and radiators, pipes in insulated timber floor, electric'
+    # 3)
+    # Description: 'Air source heat pump, radiators and underfloor, electric'
+    # Issue: We don't have any logic which identifies this heating system has having underfloor heating.
+    #        Currently, we look for "electric underfloor heating" and "underfloor heating" so we miss
+    #        the underfloor characterisation. There are a few descriptions that just include "underfloor"
+    #        e.g. 'Air source heat pump, radiators, electric' which will get missed
+    # 4)
+    #
 
     df.to_dict("records")

epc_data/attributes/MainheatAttributes.py

@@ -0,0 +1,64 @@
+from typing import Dict, List, Union
+
+
+class MainHeatAttributes:
+    HEAT_SYSTEMS = ["boiler", "air source heat pump", "room heaters", "electric storage heaters", "warm air",
+                    "electric underfloor heating", "electric ceiling heating", "community scheme"]
+    FUEL_TYPES = ["electric", "mains gas", "wood logs", "LPG", "coal", "oil", "wood pellets", "anthracite",
+                  "dual fuel (mineral and wood)", "smokeless fuel"]
+    DISTRIBUTION_SYSTEMS = ["underfloor heating", "radiators", "fan coil units", "pipes in screed above insulation",
+                            "pipes in insulated timber floor", "pipes in concrete slab"]
+    OTHERS = ["assumed", "Electricaire"]
+
+    def __init__(self, description: str):
+        self.description: str = description.lower()
+
+        if not description or not any(
+            rt in self.description for rt in
+            self.HEAT_SYSTEMS + self.FUEL_TYPES + self.DISTRIBUTION_SYSTEMS + self.OTHERS
+        ):
+            raise ValueError('Invalid description')
+
+    def process(self) -> Dict[str, Union[str, bool]]:
+        result: Dict[str, Union[str, bool]] = {}
+        description = self.description.split(',')
+
+        # Process each part separately
+        for part in description:
+            part = part.strip()  # remove leading/trailing white spaces
+
+            # Heating Systems
+            self._process_part(result, part, self.HEAT_SYSTEMS, 'has_')
+
+            # Fuel Types
+            self._process_part(result, part, self.FUEL_TYPES, 'has_')
+
+            # Distribution Systems
+            self._process_part(result, part, self.DISTRIBUTION_SYSTEMS, 'has_')
+
+            # Other attributes
+            self._process_part(result, part, self.OTHERS, 'has_')
+
+        return result
+
+    @staticmethod
+    def _process_part(result: Dict[str, Union[str, bool]], part: str, attr_list: List[str], prefix: str):
+        """
+        Process a part of the description with a given list of attributes
+        and update the result dictionary.
+        """
+        part_words = part.split()
+        for attr in attr_list:
+            attr_words = attr.split()
+            if set(attr_words).issubset(set(part_words)):
+                result[f'{prefix}{attr.replace(" ", "_")}'] = True
+                for word in attr_words:
+                    part_words.remove(word)  # remove the attribute words from part
+
+                part = " ".join(part_words)
+
+                # Check for variations of "underfloor heating"
+                if "underfloor" in part.split():
+                    result[f'{prefix}underfloor_heating'] = True
+
+        return result

epc_data/tests/test_epc_clean.py

@@ -2,8 +2,6 @@ import pytest
 import pickle
 from epc_data.EpcClean import EpcClean
 from pathlib import Path
-from epc_data.tests.test_data.test_roof_attributes_cases import clean_roof_test_cases
-from epc_data.attributes.RoofAttributes import RoofAttributes
 
 # For local testing
 if __file__ == "<input>":