Merge pull request #199 from American-Institutes-for-Research/HEA-811/SO14-error-wild-foods-instances

Add guardrails to prevent type TypeError: can't multiply sequence by …

Author: rhunwicks

apps/baseline/migrations/0023_alter_wealthgroup_average_household_size.py

@@ -0,0 +1,26 @@
+# Generated by Django 5.2.7 on 2025-12-29 07:04
+
+from django.db import migrations, models
+
+import common.models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ("baseline", "0022_alter_wealthgroup_options"),
+    ]
+
+    operations = [
+        migrations.AlterField(
+            model_name="wealthgroup",
+            name="average_household_size",
+            field=models.FloatField(
+                blank=True,
+                help_text="Average number of people per household in this Wealth Group",
+                null=True,
+                validators=[common.models.validate_positive],
+                verbose_name="Average household size",
+            ),
+        ),
+    ]

apps/baseline/models.py

@@ -517,8 +517,12 @@ class WealthGroup(common_models.Model):
         verbose_name=_("Percentage of households"),
         help_text=_("Percentage of households in the Community or Livelihood Zone that are in this Wealth Group"),
     )
-    average_household_size = models.PositiveSmallIntegerField(
-        blank=True, null=True, verbose_name=_("Average household size")
+    average_household_size = models.FloatField(
+        blank=True,
+        null=True,
+        validators=[common_models.validate_positive],
+        verbose_name=_("Average household size"),
+        help_text=_("Average number of people per household in this Wealth Group"),
     )
 
     objects = WealthGroupManager()

apps/baseline/serializers.py

@@ -582,7 +582,7 @@ def get_livelihood_zone_baseline_label(self, obj):
     wealth_group_percentage_of_households = serializers.IntegerField(
         source="wealth_group.percentage_of_households", read_only=True
     )
-    wealth_group_average_household_size = serializers.IntegerField(
+    wealth_group_average_household_size = serializers.FloatField(
         source="wealth_group.average_household_size", read_only=True
     )
     wealth_group_category = serializers.CharField(source="wealth_group.wealth_group_category.pk", read_only=True)
@@ -1337,7 +1337,7 @@ def get_livelihood_zone_baseline_label(self, obj):
     wealth_group_percentage_of_households = serializers.IntegerField(
         source="wealth_group.percentage_of_households", read_only=True
     )
-    wealth_group_average_household_size = serializers.IntegerField(
+    wealth_group_average_household_size = serializers.FloatField(
         source="wealth_group.average_household_size", read_only=True
     )
     wealth_group_category = serializers.CharField(source="wealth_group.wealth_group_category.pk", read_only=True)
@@ -1453,7 +1453,7 @@ def get_livelihood_zone_baseline_label(self, obj):
     wealth_group_percentage_of_households = serializers.IntegerField(
         source="wealth_group.percentage_of_households", read_only=True
     )
-    wealth_group_average_household_size = serializers.IntegerField(
+    wealth_group_average_household_size = serializers.FloatField(
         source="wealth_group.average_household_size", read_only=True
     )
     wealth_group_category = serializers.CharField(source="wealth_group.wealth_group_category.pk", read_only=True)

apps/baseline/tests/factories.py

@@ -155,7 +155,7 @@ class Meta:
     )
     wealth_group_category = factory.SubFactory(WealthGroupCategoryFactory)
     percentage_of_households = fuzzy.FuzzyInteger(10, 91)
-    average_household_size = fuzzy.FuzzyInteger(2, 31)
+    average_household_size = fuzzy.FuzzyDecimal(2.0, 31.0, precision=2)
 
 
 class BaselineWealthGroupFactory(WealthGroupFactory):

pipelines/assets/livelihood_activity.py

@@ -847,9 +847,27 @@ def get_instances_from_dataframe(
                             # worksheets
                             column = df.columns[i + 1]
                             household_size = df.iloc[3, i + 1]
+
+                            # Convert to numeric to handle string values (e.g., "6" instead of 6)
+                            try:
+                                percentage_kcals = (
+                                    float(livelihood_activity["percentage_kcals"])
+                                    if livelihood_activity.get("percentage_kcals")
+                                    else None
+                                )
+                            except (ValueError, TypeError):
+                                percentage_kcals = None
+
+                            try:
+                                household_size_numeric = (
+                                    float(household_size) if household_size not in ("", None) else None
+                                )
+                            except (ValueError, TypeError):
+                                household_size_numeric = None
+
                             livelihood_activity["kcals_consumed"] = (
-                                livelihood_activity["percentage_kcals"] * 2100 * 365 * household_size
-                                if livelihood_activity["percentage_kcals"] and household_size
+                                percentage_kcals * 2100 * 365 * household_size_numeric
+                                if percentage_kcals and household_size_numeric
                                 else None
                             )
 

pipelines/assets/wealth_characteristic.py

@@ -490,10 +490,14 @@ def wealth_characteristic_instances(
     # Calculate the kcals_consumed
     # Derive it by multiplying percentage_kcals by:
     #   2100 (kcals per person per day) * 365 (days per year) * average_household_size
+    # Convert columns to numeric to handle string values from the dataframe
+    wealth_group_df["percentage_kcals"] = pd.to_numeric(wealth_group_df["percentage_kcals"], errors="coerce")
+    wealth_group_df["average_household_size"] = pd.to_numeric(
+        wealth_group_df["average_household_size"], errors="coerce"
+    )
     wealth_group_df["kcals_consumed"] = (
         wealth_group_df["percentage_kcals"] * 2100 * 365 * wealth_group_df["average_household_size"]
     )
-
     result = {
         "WealthGroup": wealth_group_df.to_dict(orient="records"),
         "WealthGroupCharacteristicValue": wealth_group_characteristic_values,