From ce2a4665de675476216d1df6aaae9a362ef1d0e1 Mon Sep 17 00:00:00 2001
From: Michael Duong <michael@11s-MacBook.local>
Date: Fri, 17 May 2024 06:13:28 +0100
Subject: [PATCH] add improvement detials

---
 .../find_my_epc_pipeline.py                   | 69 +++++++++++--------
 1 file changed, 40 insertions(+), 29 deletions(-)

diff --git a/etl/epc_recommendations/find_my_epc_pipeline.py b/etl/epc_recommendations/find_my_epc_pipeline.py
index 2b4a762e..0915823d 100644
--- a/etl/epc_recommendations/find_my_epc_pipeline.py
+++ b/etl/epc_recommendations/find_my_epc_pipeline.py
@@ -51,13 +51,13 @@ def retrieve_find_my_epc_data(uprn: int, postcode: str, address: str):
             'address': [address],
             'epc_certificate': [epc_certificate],
             'current_epc_rating': [current_rating.split(' ')[-6]],
-             'current_epc_efficiency': [current_rating.split(' ')[-1]],
+             'current_epc_efficiency': [int(current_rating.split(' ')[-1])],
             'potential_epc_rating': [potential_rating.split(' ')[-6]],
-            "potential_epc_efficiency": [potential_rating.split(' ')[-1]]
+            "potential_epc_efficiency": [int(potential_rating.split(' ')[-1])]
             }
         )
     
-    print("Find assessor")
+    print("Find assessor details")
     assessor_block = address_res.find('div', {'class': 'epc-contact-assessor'})
     assessor_fields = assessor_block.find_all('dd', {"class": 'govuk-summary-list__value govuk-!-width-one-half'})
     assessor_name = assessor_fields[0].text.strip()
@@ -68,43 +68,54 @@ def retrieve_find_my_epc_data(uprn: int, postcode: str, address: str):
     new_property_df['assessor_number'] = assessor_number
     new_property_df['assessor_email'] = assessor_email
 
-    return new_property_df
+    new_property_df['improvements'] = None
+    new_property_df['improvements_sap'] = None
+    new_property_df['improvements_cost'] = None
+
+    # return new_property_df
 
     # print('### Changes that can be made:')
-    # improvements = address_res.find('div', {"class": "govuk-body printable-area epb-recommended-improvements"})
+    improvements = address_res.find('div', {"class": "govuk-body printable-area epb-recommended-improvements"})
 
-    # if improvements is None:
-    #     print("No changes suggested")
-    # else:
-    #     changes = improvements.find_all('h3')
-    #     changes_impact = improvements.find_all('dl', {"class": 'govuk-summary-list'})
+    if improvements is None:
+        print("No changes suggested")
+        return new_property_df
+    else:
+        changes = improvements.find_all('h3')
+        changes_impact = improvements.find_all('dl', {"class": 'govuk-summary-list'})
 
-    #     for element in zip(changes, changes_impact):
-    #         improvement_header = element[0].text
-    #         print("#### " + improvement_header)
+        sap_running_efficiency = new_property_df['current_epc_efficiency'].values[0]
+        changes_list = []
+        changes_impact_list = []
+        changes_cost_list = []
+        for i, element in enumerate(zip(changes, changes_impact)):
 
-    #         improvement_text = element[1].text
-    #         print(improvement_text)
+            improvement_header = element[0].text
+            # print("#### " + improvement_header)
+
+            improvement_text = element[1].text
+            # print(improvement_text)
             
-    #         col_name = improvement_header.split(":")[1]
-    #         cost = element[1].find('dd', {"class": "govuk-summary-list__value"}).text.lstrip().rstrip()
+            col_name = improvement_header.split(":")[1].strip()
+            cost = element[1].find('dd', {"class": "govuk-summary-list__value"}).text.lstrip().rstrip()
         
-    #         impact = element[1].find('text', {"class": "govuk-!-font-weight-bold"}).text.split(" ")
-    #         impact_num = impact[0]
-    #         impact_cat = impact[1]
-    #         print(cost)
-    #         new_property_df[col_name] = True
-    #         # cost_column = col_name + '-cost'
-    #         # new_property_df.assign(cost_column=cost)
-    #         new_property_df[col_name + '-cost'] = cost
-    #         new_property_df[col_name + '-impact_num'] = impact_num
-    #         new_property_df[col_name + '-impact_cat'] = impact_cat
+            impact = element[1].find('text', {"class": "govuk-!-font-weight-bold"}).text.split(" ")
+            impact_num = int(impact[0])
+            impact_cat = impact[1]
 
+            sap_difference = impact_num - sap_running_efficiency
 
-    #     data = pd.concat([data, new_property_df])
-    #     data.to_csv('./portfolio.csv')
+            changes_list.append(col_name)
+            changes_impact_list.append(sap_difference)
+            changes_cost_list.append(cost)
 
+            sap_running_efficiency = impact_num
 
+        new_property_df['improvements'] = [changes_list]
+        new_property_df['improvements_sap'] = [changes_impact_list]
+        new_property_df['improvements_cost'] = [changes_cost_list]
+
+        return new_property_df
 
 def main():
     """