Merge pull request #56739 from frappe/chore/test-exponential-smoothing-forecasting

test: Exponential Smoothing Forecasting report coverage

erpnext/manufacturing/report/exponential_smoothing_forecasting/test_exponential_smoothing_forecasting.py

@@ -0,0 +1,101 @@
+# Copyright (c) 2026, Frappe Technologies Pvt. Ltd. and Contributors
+# See license.txt
+
+import frappe
+from frappe.utils import flt
+
+from erpnext.manufacturing.report.exponential_smoothing_forecasting.exponential_smoothing_forecasting import (
+	execute,
+)
+from erpnext.selling.doctype.sales_order.test_sales_order import make_sales_order
+from erpnext.stock.doctype.item.test_item import make_item
+from erpnext.tests.utils import ERPNextTestSuite
+
+FROM_DATE = "2026-06-01"
+TO_DATE = "2026-08-31"
+SMOOTHING_CONSTANT = 0.5
+
+
+class TestExponentialSmoothingForecasting(ERPNextTestSuite):
+	"""Drive real submitted Sales Orders and assert the report buckets the ordered
+	quantities into the correct historical periods and produces a forecast."""
+
+	def setUp(self):
+		# The forecast query has no lower date bound, so it would pick up any committed
+		# Sales Order for the item. A uniquely-named item keeps the buckets scoped to
+		# just this test's orders.
+		self.item = make_item(properties={"is_stock_item": 1}).name
+
+	def test_monthly_qty_forecast_from_sales_orders(self):
+		# Historical demand: distinct calendar months strictly before FROM_DATE.
+		# Monthly period keys are derived from the period's last day (e.g. "mar_2026").
+		history = {"mar_2026": 7, "apr_2026": 4, "may_2026": 9}
+		self.create_sales_orders(
+			{
+				"2026-03-15": history["mar_2026"],
+				"2026-04-15": history["apr_2026"],
+				"2026-05-15": history["may_2026"],
+			}
+		)
+
+		columns, row = self.run_report()
+		fields = {col["fieldname"] for col in columns}
+
+		# For Monthly periodicity only future periods are exposed as columns, each as a
+		# forecast_ field. Historical demand lives in the row data (keyed by month) but is
+		# not surfaced as its own column.
+		self.assertIn("forecast_jun_2026", fields, "expected future forecast column")
+		self.assertNotIn("jun_2026", fields, "future period must not expose raw demand column")
+		self.assertNotIn("mar_2026", fields, "historical month is not a Monthly report column")
+
+		# Historical buckets must exactly reflect the ordered quantities.
+		for key, qty in history.items():
+			self.assertEqual(flt(row.get(key)), flt(qty), f"bucket {key} mismatch")
+
+		# The forecast seeds at the average of the non-zero historical months and then
+		# smooths through them in order: F = F + a*(actual - F). Asserting the exact
+		# analytical value pins the smoothing formula (Jun 2026 works out to ~7.2083).
+		expected_avg = sum(history.values()) / len(history)
+		self.assertAlmostEqual(flt(row.get("avg")), expected_avg, places=6)
+
+		forecast = expected_avg
+		for month in ("mar_2026", "apr_2026", "may_2026"):
+			forecast = forecast + SMOOTHING_CONSTANT * (history[month] - forecast)
+		self.assertAlmostEqual(flt(row.get("forecast_jun_2026")), forecast, places=6)
+
+	def test_ignores_documents_outside_range_and_other_docstatus(self):
+		self.create_sales_orders({"2026-05-10": 6})
+		# A draft SO and a future-dated SO must not contribute to historical demand.
+		make_sales_order(item_code=self.item, qty=100, transaction_date="2026-05-20", do_not_submit=True)
+		make_sales_order(item_code=self.item, qty=100, transaction_date=FROM_DATE)
+
+		_columns, row = self.run_report()
+		self.assertEqual(flt(row.get("may_2026")), 6.0)
+
+	def create_sales_orders(self, date_to_qty):
+		for transaction_date, qty in date_to_qty.items():
+			make_sales_order(item_code=self.item, qty=qty, transaction_date=transaction_date)
+
+	def run_report(self, **extra):
+		filters = frappe._dict(
+			{
+				"company": "_Test Company",
+				"based_on_document": "Sales Order",
+				"based_on_field": "Qty",
+				"no_of_years": 3,
+				"periodicity": "Monthly",
+				"from_date": FROM_DATE,
+				"to_date": TO_DATE,
+				"smoothing_constant": SMOOTHING_CONSTANT,
+				"item_code": self.item,
+			}
+		)
+		filters.update(extra)
+
+		columns, data = execute(filters)[:2]
+		item_row = next(
+			(r for r in data if r.get("item_code") == self.item),
+			None,
+		)
+		self.assertIsNotNone(item_row, f"{self.item} row missing from report output")
+		return columns, item_row