[FIX] account_statement_import_sheet_file_bg: handle OpenPyXL Cell values in sheet header parsing

Prevent crash when importing XLSX files where header rows are iterated as OpenPyXL Cell objects instead of plain strings.

add parser override in miscellaneous module to normalize header values
support both raw strings and Cell instances before calling strip
keep OCA module untouched as required
preserve existing behavior for CSV/XLS flows
Fixes:
AttributeError: 'Cell' object has no attribute 'strip'

Alternative (more concise) title:
fix: normalize XLSX header cells before strip in bg statement import

Author: rov-adhoc

account_statement_import_sheet_file_bg/models/__init__.py

@@ -1 +1,2 @@
 from . import account_statement_import
+from . import account_statement_import_sheet_parser

account_statement_import_sheet_file_bg/models/account_statement_import.py

@@ -171,15 +171,16 @@ def split_base64_excel(self, header_rows_count, rows_per_file_limit):
         mapping = self.sheet_mapping_id
         journal = self.env["account.journal"].browse(self.env.context.get("journal_id"))
         currency_code = (journal.currency_id or journal.company_id.currency_id).name
-
         try:
             file_bytes = base64.b64decode(self.statement_file)
             read_buffer = BytesIO(file_bytes)
 
             # Try openpyxl (xlsx)
             input_workbook = load_workbook(read_buffer)
             input_worksheet = input_workbook.active
-            all_rows = list(input_worksheet.rows)
+            # Normalize rows to plain values to keep parser/output logic
+            # consistent with CSV/xls flows.
+            all_rows = [[cell.value for cell in row] for row in input_worksheet.rows]
             csv_or_xlsx = (input_workbook, input_worksheet)
 
         except Exception:
@@ -277,10 +278,14 @@ def _filter_rows_with_date(self, data_rows, date_column_index):
 
         filtered_rows = []
         for row in data_rows:
+            date_value = None
+            if len(row) > date_column_index:
+                date_cell_or_value = row[date_column_index]
+                date_value = date_cell_or_value.value if hasattr(date_cell_or_value, "value") else date_cell_or_value
             # Check if the row has enough columns and the date column is not empty
-            if len(row) > date_column_index and row[date_column_index].value:
+            if len(row) > date_column_index and date_value:
                 filtered_rows.append(row)
-            elif len(row) > date_column_index and not row[date_column_index].value:
+            elif len(row) > date_column_index and not date_value:
                 # Stop processing when we find the first empty date
                 break
 

account_statement_import_sheet_file_bg/models/account_statement_import_sheet_parser.py

@@ -0,0 +1,31 @@
+# Copyright 2026 ADHOC SA
+# License AGPL-3.0 or later (https://www.gnu.org/licenses/agpl).
+
+from odoo import api, models
+
+
+class AccountStatementImportSheetParser(models.TransientModel):
+    _inherit = "account.statement.import.sheet.parser"
+
+    @api.model
+    def parse_header(self, csv_or_xlsx, mapping):
+        if mapping.no_header:
+            return []
+
+        header_line = mapping.header_lines_skip_count
+        # Prevent negative indexes.
+        if header_line > 0:
+            header_line -= 1
+
+        if isinstance(csv_or_xlsx, tuple):
+            return super().parse_header(csv_or_xlsx, mapping)
+
+        [next(csv_or_xlsx) for _i in range(header_line)]
+        header = []
+        for value in next(csv_or_xlsx):
+            raw_value = value.value if hasattr(value, "value") else value
+            header.append(str(raw_value).strip() if raw_value is not None else "")
+
+        if mapping.offset_column:
+            header = header[mapping.offset_column :]
+        return header