Implemented processor for baremetal

Author: QuanMPhm

process_report/invoices/bm_invoice.py

@@ -0,0 +1,29 @@
+from dataclasses import dataclass
+
+
+from process_report.invoices import invoice
+
+
+@dataclass
+class BMInvoice(invoice.Invoice):
+    name: str = "bm_projects"
+    export_columns_list = [
+        invoice.INVOICE_DATE_FIELD,
+        invoice.PROJECT_FIELD,
+        invoice.PROJECT_ID_FIELD,
+        invoice.PI_FIELD,
+        invoice.INVOICE_EMAIL_FIELD,
+        invoice.INVOICE_ADDRESS_FIELD,
+        invoice.INSTITUTION_FIELD,
+        invoice.INSTITUTION_ID_FIELD,
+        invoice.SU_HOURS_FIELD,
+        invoice.SU_TYPE_FIELD,
+        invoice.RATE_FIELD,
+        invoice.COST_FIELD,
+        invoice.CREDIT_FIELD,
+        invoice.CREDIT_CODE_FIELD,
+        invoice.BALANCE_FIELD,
+    ]
+
+    def _prepare_export(self):
+        self.export_data = self.data[self.data[invoice.CLUSTER_NAME_FIELD] == "bm"]

process_report/process_report.py

@@ -10,6 +10,7 @@
 from process_report import util
 from process_report.invoices import (
     invoice,
+    bm_invoice,
     lenovo_invoice,
     nonbillable_invoice,
     billable_invoice,
@@ -98,6 +99,7 @@ def main():
             MOCA_prepaid_invoice.MOCAPrepaidInvoice,
             prepay_credits_snapshot.PrepayCreditsSnapshot,
             ocp_test_invoice.OcpTestInvoice,
+            bm_invoice.BMInvoice,
         ],
         invoice_settings.upload_to_s3,
     )

process_report/processors/bm_usage_processor.py

@@ -1,15 +1,13 @@
 from dataclasses import dataclass
 
-import pandas
-
 from process_report.invoices import invoice
 from process_report.processors import processor
 
 
 @dataclass
 class BMUsageProcessor(processor.Processor):
     def _get_bm_project_mask(self):
-        return pandas.Series(True, index=self.data.index)  # TODO: Remove dummy mask
+        return self.data[invoice.CLUSTER_NAME_FIELD] == "bm"
 
     def _process(self):
         bm_projects_mask = self._get_bm_project_mask()

process_report/processors/bu_subsidy_processor.py

@@ -34,11 +34,9 @@ def _get_subsidy_eligible_projects(data):
         filtered_data = filtered_data[
             filtered_data[invoice.INSTITUTION_FIELD] == "Boston University"
         ]
-        filtered_data = (
-            filtered_data[  # TODO Does it make sense to test this filter in test cases?
-                ~(filtered_data[invoice.PROJECT_ID_FIELD] == "ESI Bare Metal")
-            ]
-        )
+        filtered_data = filtered_data[
+            ~(filtered_data[invoice.CLUSTER_NAME_FIELD] == "bm")
+        ]
 
         return filtered_data
 

process_report/processors/new_pi_credit_processor.py

@@ -139,7 +139,7 @@ def _filter_missing_pis(self, data):
         return data[~data["Missing PI"]]
 
     def _filter_bm_projects(self, data):
-        return data[~(data[invoice.PROJECT_ID_FIELD] == "ESI Bare Metal")]
+        return data[~(data[invoice.CLUSTER_NAME_FIELD] == "bm")]
 
     def _get_credit_eligible_projects(self, data: pandas.DataFrame):
         filtered_data = self._filter_nonbillables(data)

process_report/tests/e2e/test_e2e_pipeline.py

@@ -21,6 +21,7 @@
     "MOCA-A_Prepaid_Groups-2025-06-Invoice.csv",
     "NERC_Prepaid_Group-Credits-2025-06.csv",
     "OCP_TEST 2025-06.csv",
+    "bm_projects 2025-06.csv",
 ]
 
 EXPECTED_DIRECTORIES = ["pi_invoices"]

process_report/tests/unit/processors/test_bm_usage_processor.py

@@ -5,33 +5,31 @@
 from process_report.tests import util as test_utils
 
 
-class TestBUSubsidyProcessor(TestCase):
-    def test_get_bm_project_mask(self):
-        test_invoice = pandas.DataFrame({})
-
-        answer_invoice = test_invoice.iloc[[0, 2]]
-
-        bm_usage_proc = test_utils.new_bm_usage_processor(data=test_invoice)
-        bm_project_mask = bm_usage_proc._get_bm_project_mask()
-        self.assertTrue(test_invoice[bm_project_mask].equals(answer_invoice))
-
+class TestBMUsageProcessor(TestCase):
     def test_process_bm_usage(self):
         test_invoice = pandas.DataFrame(
             {
-                "Project - Allocation": ["test", "test bm-bm"],
-                "Project - Allocation ID": [None] * 2,
-                "Invoice Email": [None] * 2,
+                "Project - Allocation": ["test", "test bm-bm", "not-bm"],
+                "Project - Allocation ID": [None] * 3,
+                "Invoice Email": [None] * 3,
+                "Cluster Name": ["bm", "bm", "ocp"],
             }
         )
 
         answer_invoice = pandas.DataFrame(
             {
-                "Project - Allocation": ["test BM Usage", "test bm-bm BM Usage"],
-                "Project - Allocation ID": ["ESI Bare Metal"] * 2,
-                "Invoice Email": ["nclinton@bu.edu"] * 2,
+                "Project - Allocation": [
+                    "test BM Usage",
+                    "test bm-bm BM Usage",
+                    "not-bm",
+                ],
+                "Project - Allocation ID": ["ESI Bare Metal"] * 2 + [None],
+                "Invoice Email": ["nclinton@bu.edu"] * 2 + [None],
+                "Cluster Name": ["bm", "bm", "ocp"],
             }
         )
 
         bm_usage_proc = test_utils.new_bm_usage_processor(data=test_invoice)
         bm_usage_proc.process()
+        answer_invoice = answer_invoice.astype(bm_usage_proc.data.dtypes)
         self.assertTrue(bm_usage_proc.data.equals(answer_invoice))

process_report/tests/unit/processors/test_bu_subsidy_processor.py

@@ -32,6 +32,7 @@ def _get_test_invoice(
         institution=None,
         is_billable=None,
         missing_pi=None,
+        clusters=None,
     ):
         if not balances:
             balances = pi_balances
@@ -48,6 +49,9 @@ def _get_test_invoice(
         if not missing_pi:
             missing_pi = [False for _ in range(len(pi))]
 
+        if not clusters:
+            clusters = ["" for _ in range(len(pi))]
+
         return pandas.DataFrame(
             {
                 "Manager (PI)": pi,
@@ -57,6 +61,7 @@ def _get_test_invoice(
                 "Institution": institution,
                 "Is Billable": is_billable,
                 "Missing PI": missing_pi,
+                "Cluster Name": clusters,
             }
         )
 
@@ -175,3 +180,21 @@ def test_two_pi(self):
         answer_invoice["PI Balance"] = [0, 60, 0, 0]
 
         self._assert_result_invoice(subsidy_amount, test_invoice, answer_invoice)
+
+    def test_exclude_bm_cluster(self):
+        """Projects in the 'bm' cluster should be excluded from BU subsidy calculation."""
+        subsidy_amount = 100
+        test_invoice = self._get_test_invoice(
+            ["PI"] * 2,  # single PI (will be broadcast to two rows by lengths below)
+            pi_balances=[60, 60],
+            project_names=["P1", "P2"],
+            clusters=["bm", "ocp"],
+        )
+
+        answer_invoice = test_invoice.copy()
+        answer_invoice["Project"] = answer_invoice["Project - Allocation"]
+        # bm allocation gets no subsidy, non-bm allocation gets up to its PI balance (60)
+        answer_invoice["Subsidy"] = [0, 60]
+        answer_invoice["PI Balance"] = [60, 0]
+
+        self._assert_result_invoice(subsidy_amount, test_invoice, answer_invoice)