Fixes #465 - Allow importing 'ragged' .xlsx files (#466)

HISTORY.md

@@ -12,6 +12,7 @@
 
 - Fixed minimal openpyxl dependency version to 2.6.0 (#457).
 - Dates from xls files are now read as Python datetime objects (#373).
+- Allow import of "ragged" xlsx files (#465).
 
 ### Improvements
 

src/tablib/formats/_xlsx.py

@@ -92,6 +92,8 @@ def import_book(cls, dbook, in_stream, headers=True):
                 if (i == 0) and (headers):
                     data.headers = row_vals
                 else:
+                    if i > 0 and len(row_vals) < data.width:
+                        row_vals += [''] * (data.width - len(row_vals))
                     data.append(row_vals)
 
             dbook.add_sheet(data)

tests/test_tablib.py

@@ -1015,6 +1015,13 @@ def test_xlsx_import_set(self):
         self.assertEqual(data.dict[0]['float'], 21.55)
         self.assertEqual(data.dict[0]['date/time'], date_time)
 
+    def test_xlsx_import_set_ragged(self):
+        """Import XLSX file when not all rows have the same length."""
+        xlsx_source = Path(__file__).parent / 'files' / 'ragged.xlsx'
+        with open(str(xlsx_source), mode='rb') as fh:
+            book = tablib.Databook().load(fh, 'xlsx')
+        self.assertEqual(book.sheets()[0].pop(), (1.0, ''))
+
     def test_xlsx_wrong_char(self):
         """Bad characters are not silently ignored. We let the exception bubble up."""
         from openpyxl.utils.exceptions import IllegalCharacterError