From 98d8c1115ed3ddcb8ced3806a8bceaa37d8dfeb5 Mon Sep 17 00:00:00 2001
From: zephyrdark <zephyrdark@gmail.com>
Date: Sat, 14 Feb 2026 00:35:31 +0900
Subject: [PATCH] feat: add backfill API endpoint for historical data
 collection

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---
 backend/app/api/admin.py                 | 29 ++++++++++++++++++++++++
 backend/tests/e2e/test_collection_job.py | 18 +++++++++++++++
 2 files changed, 47 insertions(+)

diff --git a/backend/app/api/admin.py b/backend/app/api/admin.py
index 301ac6b..28e2a2c 100644
--- a/backend/app/api/admin.py
+++ b/backend/app/api/admin.py
@@ -25,6 +25,7 @@ from app.services.collectors import (
     ETFCollector,
     ETFPriceCollector,
 )
+from jobs.collection_job import run_backfill
 
 logger = logging.getLogger(__name__)
 
@@ -71,6 +72,24 @@ def _start_background_collection(collector_cls, **kwargs):
     thread.start()
 
 
+def _run_backfill_background(start_year: int):
+    """Run backfill in a background thread."""
+    try:
+        run_backfill(start_year=start_year)
+    except Exception as e:
+        logger.error("Background backfill failed: %s", e)
+
+
+def run_backfill_background(start_year: int):
+    """Start backfill in a daemon thread."""
+    thread = threading.Thread(
+        target=_run_backfill_background,
+        args=(start_year,),
+        daemon=True,
+    )
+    thread.start()
+
+
 @router.post("/collect/stocks", response_model=CollectResponse)
 async def collect_stocks(
     current_user: CurrentUser,
@@ -132,6 +151,16 @@ async def collect_etf_prices(
     return CollectResponse(message="ETF price collection started")
 
 
+@router.post("/collect/backfill", response_model=CollectResponse)
+async def collect_backfill(
+    current_user: CurrentUser,
+    start_year: int = Query(2000, ge=1990, le=2026, description="Start year for backfill"),
+):
+    """Backfill historical price data from start_year to today (runs in background)."""
+    run_backfill_background(start_year)
+    return CollectResponse(message=f"Backfill started from {start_year}")
+
+
 @router.get("/collect/status", response_model=List[JobLogResponse])
 async def get_collection_status(
     current_user: CurrentUser,
diff --git a/backend/tests/e2e/test_collection_job.py b/backend/tests/e2e/test_collection_job.py
index 340448b..24ac269 100644
--- a/backend/tests/e2e/test_collection_job.py
+++ b/backend/tests/e2e/test_collection_job.py
@@ -157,3 +157,21 @@ def test_scheduler_has_daily_collection_job():
     trigger = jobs["daily_collection"].trigger
     trigger_str = str(trigger)
     assert "18" in trigger_str  # hour=18
+
+
+def test_backfill_api_endpoint(client, admin_auth_headers):
+    """POST /api/admin/collect/backfill should trigger backfill."""
+    with patch("app.api.admin.run_backfill_background") as mock_backfill:
+        response = client.post(
+            "/api/admin/collect/backfill?start_year=2020",
+            headers=admin_auth_headers,
+        )
+    assert response.status_code == 200
+    assert "backfill" in response.json()["message"].lower()
+    mock_backfill.assert_called_once()
+
+
+def test_backfill_api_requires_auth(client):
+    """POST /api/admin/collect/backfill should require authentication."""
+    response = client.post("/api/admin/collect/backfill")
+    assert response.status_code == 401