Merge pull request #21 from VedantPatel04/frontEnd

Front-End functionally complete

Author: VedantPatel04

.gitignore

@@ -25,8 +25,10 @@ coverage.xml
 # Environment / secrets
 .env
 backend/.env
+frontend/.env
 .env.*
 !backend/.env.example
+!frontend/.env.example
 
 # OS / editor cruft
 .DS_Store

backend/apps/cards/__pycache__/models.cpython-314.pyc

@@ -0,0 +1,46 @@
+from django.db import migrations, models
+
+
+def backfill_chase_payment_thank_you(apps, schema_editor):
+    """
+    Existing rows never stored Chase Type. Mark the common Chase bill-payment
+    label so summaries stop treating those credits as spend without a re-upload.
+
+    Statement credits / adjustments are NOT guessed here — re-upload those files
+    so the adapter can read Type=Adjustment.
+    """
+    Transactions = apps.get_model("transactions", "Transactions")
+    Transactions.objects.filter(description__icontains="Payment Thank You").update(
+        entry_type="payment"
+    )
+
+
+def noop_reverse(apps, schema_editor):
+    # Irreversible data fix; schema reverse still drops the column.
+    pass
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ("transactions", "0009_globalmerchantalias_transactions_confidence_and_more"),
+    ]
+
+    operations = [
+        migrations.AddField(
+            model_name="transactions",
+            name="entry_type",
+            field=models.CharField(
+                choices=[
+                    ("spend", "Spend"),
+                    ("refund", "Refund"),
+                    ("payment", "Payment"),
+                    ("adjustment", "Adjustment"),
+                ],
+                db_index=True,
+                default="spend",
+                max_length=16,
+            ),
+        ),
+        migrations.RunPython(backfill_chase_payment_thank_you, noop_reverse),
+    ]

backend/apps/transactions/__pycache__/models.cpython-314.pyc

@@ -11,6 +11,19 @@
 SOURCE_BANK = "bank"      # adapter's bank category (tier 4)
 SOURCE_NONE = ""          # genuinely unresolved or categorizedto "other"
 
+ENTRY_SPEND = "spend"
+ENTRY_REFUND = "refund"
+ENTRY_PAYMENT = "payment"
+ENTRY_ADJUSTMENT = "adjustment"
+ENTRY_TYPE_CHOICES = (
+    (ENTRY_SPEND, "Spend"),
+    (ENTRY_REFUND, "Refund"),
+    (ENTRY_PAYMENT, "Payment"),
+    (ENTRY_ADJUSTMENT, "Adjustment"),
+)
+# Rows that count toward by_category spending AKA ottal_spend
+SPEND_SUMMARY_ENTRY_TYPES = (ENTRY_SPEND, ENTRY_REFUND)
+
 
 class Transactions(models.Model):
     class Meta:
@@ -35,6 +48,14 @@ class Meta:
     # 0.0–1.0: user=1.0, global=0.9, bank=0.7, unresolved=0.0
     confidence = models.FloatField(default=0.0)
 
+    #spend, refund, payment, adjustment — see module constants above
+    entry_type = models.CharField(
+        max_length=16,
+        choices=ENTRY_TYPE_CHOICES,
+        default=ENTRY_SPEND,
+        db_index=True,
+    )
+
     # sign convention: positive = spend, negative = refund / credit
     amount = models.DecimalField(max_digits=10, decimal_places=2)
     transaction_date = models.DateField()

backend/apps/transactions/migrations/0010_transactions_entry_type.py

@@ -86,6 +86,7 @@ def transaction_list(request):
             "normalized_description": row.normalized_description,
             "merchant_key": row.merchant_key,
             "category": row.category,
+            "entry_type": row.entry_type,
             "resolution_source": row.resolution_source,
             "confidence": row.confidence,
         }
@@ -236,13 +237,17 @@ def summary_view(request):
 
     All-time spend totals across all wallet cards, by rewards category.
 
-    Returns net spend (refunds reduce category totals). Unresolved rows
-    (category="") are excluded from by_category and reported separately in
+    Returns purchase spend (entry_type spend + refund). Refunds reduce
+    category totals. Bill payments and statement adjustments are excluded
+    from by_category / total_spend but still appear on the transaction list
+    and still count toward statement-cycle months_covered.
+
+    Unresolved rows (category="") among spend/refund are reported in
     unresolved_count / unresolved_amount.
 
-    All 7 category buckets are always present even if zero. Negative totals
-    (e.g. other when card payments outweigh real spend) are not clamped —
-    the Day 6 Confidence Check (Stage 5) handles distortion detection.
+    All 7 category buckets are always present even if zero.
+    months_breakdown is calendar months (display); months_covered is
+    statement-cycle evidence used for annualize / signup.
 
     The Day 6 recommendation engine calls get_spend_summary() directly as a
     service; this view serialises the same output for the HTTP API.