WIP

src/data/csv_loader.py

@@ -25,6 +25,12 @@ class InvoiceRow:
     Bankgiro: str | None = None
     Plusgiro: str | None = None
     Amount: Decimal | None = None
+    # New fields
+    split: str | None = None  # train/test split indicator
+    customer_number: str | None = None  # Customer number (no matching needed)
+    supplier_name: str | None = None  # Supplier name (no matching)
+    supplier_organisation_number: str | None = None  # Swedish org number (needs matching)
+    supplier_accounts: str | None = None  # Supplier accounts (needs matching)
 
     # Raw values for reference
     raw_data: dict = field(default_factory=dict)
@@ -40,6 +46,8 @@ class InvoiceRow:
             'Bankgiro': self.Bankgiro,
             'Plusgiro': self.Plusgiro,
             'Amount': str(self.Amount) if self.Amount else None,
+            'supplier_organisation_number': self.supplier_organisation_number,
+            'supplier_accounts': self.supplier_accounts,
         }
 
     def get_field_value(self, field_name: str) -> str | None:
@@ -68,6 +76,12 @@ class CSVLoader:
         'Bankgiro': 'Bankgiro',
         'Plusgiro': 'Plusgiro',
         'Amount': 'Amount',
+        # New fields
+        'split': 'split',
+        'customer_number': 'customer_number',
+        'supplier_name': 'supplier_name',
+        'supplier_organisation_number': 'supplier_organisation_number',
+        'supplier_accounts': 'supplier_accounts',
     }
 
     def __init__(
@@ -200,6 +214,12 @@ class CSVLoader:
             Bankgiro=self._parse_string(row.get('Bankgiro')),
             Plusgiro=self._parse_string(row.get('Plusgiro')),
             Amount=self._parse_amount(row.get('Amount')),
+            # New fields
+            split=self._parse_string(row.get('split')),
+            customer_number=self._parse_string(row.get('customer_number')),
+            supplier_name=self._parse_string(row.get('supplier_name')),
+            supplier_organisation_number=self._parse_string(row.get('supplier_organisation_number')),
+            supplier_accounts=self._parse_string(row.get('supplier_accounts')),
             raw_data=dict(row)
         )
 
@@ -318,14 +338,16 @@ class CSVLoader:
                 row.OCR,
                 row.Bankgiro,
                 row.Plusgiro,
-                row.Amount
+                row.Amount,
+                row.supplier_organisation_number,
+                row.supplier_accounts,
             ]
             if not any(matchable_fields):
                 issues.append({
                     'row': i,
                     'doc_id': row.DocumentId,
                     'field': 'All',
-                    'issue': 'No matchable fields (InvoiceNumber/OCR/Bankgiro/Plusgiro/Amount)'
+                    'issue': 'No matchable fields (InvoiceNumber/OCR/Bankgiro/Plusgiro/Amount/supplier_organisation_number/supplier_accounts)'
                 })
 
         return issues