From 1ae3aa5d6b9af3954bc4d8f9bff762295ed4c468 Mon Sep 17 00:00:00 2001
From: Khalim Conn-Kowlessar <kconnkowlessar@gmail.com>
Date: Mon, 15 Dec 2025 11:06:00 +0800
Subject: [PATCH] upgrade storage type of rds to gp3

---
 .../d_restart_failed_subtasks.py                          | 3 +++
 etl/epc/Dataset.py                                        | 8 ++++++--
 infrastructure/terraform/main.tf                          | 4 ++++
 recommendations/Recommendations.py                        | 7 +++++++
 4 files changed, 20 insertions(+), 2 deletions(-)

diff --git a/etl/customers/peabody/Nov 2025 Consulting Project/d_restart_failed_subtasks.py b/etl/customers/peabody/Nov 2025 Consulting Project/d_restart_failed_subtasks.py
index 79a8bdc2..5ef901b2 100644
--- a/etl/customers/peabody/Nov 2025 Consulting Project/d_restart_failed_subtasks.py	
+++ b/etl/customers/peabody/Nov 2025 Consulting Project/d_restart_failed_subtasks.py	
@@ -6,4 +6,7 @@ This script will fetch those failed subtasks, get the associated properties and
 from the database so that the subtasks can be restarted cleanly.
 
 Additionally, we wil find the problematic records and remove them
+
+Given we ran an EPC C scenario, we should check how many properties, below EPC C we have, that have no plan
+or recommendations in case something went wrong
 """
diff --git a/etl/epc/Dataset.py b/etl/epc/Dataset.py
index 01391874..7c27de51 100644
--- a/etl/epc/Dataset.py
+++ b/etl/epc/Dataset.py
@@ -697,8 +697,12 @@ class TrainingDataset(BaseDataset):
                 for x in missing_descriptions:
                     desc_cleaner = cleaner(x)
                     cleaned = desc_cleaner.process()
-                    # IF NODATA, REMAP TO NONE VALUES
-                    if all((pd.DataFrame(cleaned, index=[0]).T)[0] == False):
+                    # IF NODATA, REMAP TO NONE VALUES, apart from walls which we want to keep as is
+                    # If we convert the walls data to None, we end up converting booleans to None which
+                    # causes issues downstream
+                    if all(
+                        (pd.DataFrame(cleaned, index=[0]).T)[0] == False
+                    ) and component != "walls":
                         cleaned = {key: None for key in cleaned.keys()}
                     cleaned_data.append(
                         {
diff --git a/infrastructure/terraform/main.tf b/infrastructure/terraform/main.tf
index f345c985..a26a1af4 100644
--- a/infrastructure/terraform/main.tf
+++ b/infrastructure/terraform/main.tf
@@ -85,6 +85,10 @@ resource "aws_db_instance" "default" {
   ca_cert_identifier = "rds-ca-rsa2048-g1"
   # Temporary to enfore immediate change
   apply_immediately = true
+  # Set up storage type to gp3 for better performance
+  storage_type      = "gp3"
+  iops              = 3000
+  storage_throughput = 125
 }
 
 # Set up the bucket that recieve the csv uploads of epc to be retrofit
diff --git a/recommendations/Recommendations.py b/recommendations/Recommendations.py
index 5d97a554..f984acc3 100644
--- a/recommendations/Recommendations.py
+++ b/recommendations/Recommendations.py
@@ -698,6 +698,13 @@ class Recommendations:
                     "heating_cop": 0.85,
                     "hotwater_cop": 0.85
                 }
+            if main_fuel_description in ['coal (community)']:
+                return {
+                    "heating_fuel_type": "Coal",
+                    "hotwater_fuel_type": "Coal",
+                    "heating_cop": 0.85,
+                    "hotwater_cop": 0.85
+                }
             logger.warning(
                 "Unhandled community fuel."
                 f"Fuel: {main_fuel_description}"