Refactored BU Internal Invoice

A new `bu_internal_invoice.py` file has been added.
Note that the BU Internal invoice depends on the billable invoice.
The test case for the BU Internal invoice has been modified appropriately

process_report/invoices/bu_internal_invoice.py

@@ -0,0 +1,68 @@
+from dataclasses import dataclass
+from decimal import Decimal
+
+import process_report.invoices.invoice as invoice
+
+
+@dataclass
+class BUInternalInvoice(invoice.Invoice):
+    subsidy_amount: int
+
+    def _prepare(self):
+        def get_project(row):
+            project_alloc = row[invoice.PROJECT_FIELD]
+            if project_alloc.rfind("-") == -1:
+                return project_alloc
+            else:
+                return project_alloc[: project_alloc.rfind("-")]
+
+        self.data = self.data[
+            self.data[invoice.INSTITUTION_FIELD] == "Boston University"
+        ].copy()
+        self.data["Project"] = self.data.apply(get_project, axis=1)
+        self.data[invoice.SUBSIDY_FIELD] = Decimal(0)
+        self.data = self.data[
+            [
+                invoice.INVOICE_DATE_FIELD,
+                invoice.PI_FIELD,
+                "Project",
+                invoice.COST_FIELD,
+                invoice.CREDIT_FIELD,
+                invoice.SUBSIDY_FIELD,
+                invoice.BALANCE_FIELD,
+            ]
+        ]
+
+    def _process(self):
+        project_list = self.data["Project"].unique()
+        data_no_dup = self.data.drop_duplicates("Project", inplace=False)
+        sum_fields = [invoice.COST_FIELD, invoice.CREDIT_FIELD, invoice.BALANCE_FIELD]
+        for project in project_list:
+            project_mask = self.data["Project"] == project
+            no_dup_project_mask = data_no_dup["Project"] == project
+
+            sum_fields_sums = self.data[project_mask][sum_fields].sum().values
+            data_no_dup.loc[no_dup_project_mask, sum_fields] = sum_fields_sums
+
+        self.data = self._apply_subsidy(data_no_dup, self.subsidy_amount)
+
+    def _apply_subsidy(self, dataframe, subsidy_amount):
+        pi_list = dataframe[invoice.PI_FIELD].unique()
+
+        for pi in pi_list:
+            pi_projects = dataframe[dataframe[invoice.PI_FIELD] == pi]
+            remaining_subsidy = subsidy_amount
+            for i, row in pi_projects.iterrows():
+                project_remaining_cost = row[invoice.BALANCE_FIELD]
+                applied_subsidy = min(project_remaining_cost, remaining_subsidy)
+
+                dataframe.at[i, invoice.SUBSIDY_FIELD] = applied_subsidy
+                dataframe.at[i, invoice.BALANCE_FIELD] = (
+                    row[invoice.BALANCE_FIELD] - applied_subsidy
+                )
+                remaining_subsidy -= applied_subsidy
+
+                if remaining_subsidy == 0:
+                    break
+
+        return dataframe

process_report/process_report.py

@@ -2,7 +2,6 @@
 import os
 import sys
 import datetime
-from decimal import Decimal
 
 import json
 import pandas
@@ -13,6 +12,7 @@
     lenovo_invoice,
     nonbillable_invoice,
     billable_invoice,
+    bu_internal_invoice,
 )
 
 
@@ -166,7 +166,7 @@ def main():
     parser.add_argument(
         "--BU-invoice-file",
         required=False,
-        default="BU_Internal.csv",
+        default="BU_Internal",
         help="Name of output csv for BU invoices",
     )
     parser.add_argument(
@@ -270,8 +270,19 @@ def main():
         bucket = get_invoice_bucket()
         billable_inv.export_s3(bucket)
 
+    bu_internal_inv = bu_internal_invoice.BUInternalInvoice(
+        name=args.BU_invoice_file,
+        invoice_month=invoice_month,
+        data=billable_inv.data,
+        subsidy_amount=args.BU_subsidy_amount,
+    )
+    bu_internal_inv.process()
+    bu_internal_inv.export()
+    if args.upload_to_s3:
+        bucket = get_invoice_bucket()
+        bu_internal_inv.export_s3(bucket)
+
     export_pi_billables(billable_inv.data, args.output_folder, invoice_month)
-    export_BU_only(billable_inv.data, args.BU_invoice_file, args.BU_subsidy_amount)
     export_HU_BU(billable_inv.data, args.HU_BU_invoice_file)
 
     if args.upload_to_s3:
@@ -421,63 +432,6 @@ def export_pi_billables(dataframe: pandas.DataFrame, output_folder, invoice_mont
         )
 
 
-def export_BU_only(dataframe: pandas.DataFrame, output_file, subsidy_amount):
-    def get_project(row):
-        project_alloc = row[PROJECT_FIELD]
-        if project_alloc.rfind("-") == -1:
-            return project_alloc
-        else:
-            return project_alloc[: project_alloc.rfind("-")]
-
-    BU_projects = dataframe[dataframe[INSTITUTION_FIELD] == "Boston University"].copy()
-    BU_projects["Project"] = BU_projects.apply(get_project, axis=1)
-    BU_projects[SUBSIDY_FIELD] = Decimal(0)
-    BU_projects = BU_projects[
-        [
-            INVOICE_DATE_FIELD,
-            PI_FIELD,
-            "Project",
-            COST_FIELD,
-            CREDIT_FIELD,
-            SUBSIDY_FIELD,
-            BALANCE_FIELD,
-        ]
-    ]
-
-    project_list = BU_projects["Project"].unique()
-    BU_projects_no_dup = BU_projects.drop_duplicates("Project", inplace=False)
-    sum_fields = [COST_FIELD, CREDIT_FIELD, BALANCE_FIELD]
-    for project in project_list:
-        project_mask = BU_projects["Project"] == project
-        no_dup_project_mask = BU_projects_no_dup["Project"] == project
-
-        sum_fields_sums = BU_projects[project_mask][sum_fields].sum().values
-        BU_projects_no_dup.loc[no_dup_project_mask, sum_fields] = sum_fields_sums
-
-    BU_projects_no_dup = _apply_subsidy(BU_projects_no_dup, subsidy_amount)
-    BU_projects_no_dup.to_csv(output_file)
-
-
-def _apply_subsidy(dataframe, subsidy_amount):
-    pi_list = dataframe[PI_FIELD].unique()
-
-    for pi in pi_list:
-        pi_projects = dataframe[dataframe[PI_FIELD] == pi]
-        remaining_subsidy = subsidy_amount
-        for i, row in pi_projects.iterrows():
-            project_remaining_cost = row[BALANCE_FIELD]
-            applied_subsidy = min(project_remaining_cost, remaining_subsidy)
-
-            dataframe.at[i, SUBSIDY_FIELD] = applied_subsidy
-            dataframe.at[i, BALANCE_FIELD] = row[BALANCE_FIELD] - applied_subsidy
-            remaining_subsidy -= applied_subsidy
-
-            if remaining_subsidy == 0:
-                break
-
-    return dataframe
-
-
 def export_HU_BU(dataframe, output_file):
     HU_BU_projects = dataframe[
         (dataframe[INSTITUTION_FIELD] == "Harvard University")

process_report/tests/unit_tests.py

@@ -635,13 +635,14 @@ def setUp(self):
             ],  # Test case where subsidy does/doesn't cover fully balance
         }
         self.dataframe = pandas.DataFrame(data)
-        output_file = tempfile.NamedTemporaryFile(delete=False, mode="w", suffix=".csv")
-        self.output_file = output_file.name
         self.subsidy = 100
 
     def test_apply_BU_subsidy(self):
-        process_report.export_BU_only(self.dataframe, self.output_file, self.subsidy)
-        output_df = pandas.read_csv(self.output_file)
+        test_invoice = test_utils.new_bu_internal_invoice(
+            data=self.dataframe, subsidy_amount=self.subsidy
+        )
+        test_invoice.process()
+        output_df = test_invoice.data.reset_index()
 
         self.assertTrue(
             set(

process_report/tests/util.py

@@ -1,6 +1,6 @@
 import pandas
 
-from process_report.invoices import billable_invoice
+from process_report.invoices import billable_invoice, bu_internal_invoice
 
 
 def new_billable_invoice(
@@ -19,3 +19,11 @@ def new_billable_invoice(
         nonbillable_projects,
         old_pi_filepath,
     )
+
+
+def new_bu_internal_invoice(
+    name="", invoice_month="0000-00", data=pandas.DataFrame(), subsidy_amount=0
+):
+    return bu_internal_invoice.BUInternalInvoice(
+        name, invoice_month, data, subsidy_amount
+    )