handle missing header

pandas_ods_reader/parser.py

@@ -24,15 +24,15 @@ def load_ods(doc, sheet_id, headers=True, columns=None):
             # columns as lists in a dictionary
             columns = []
             for cell in row:
-                if cell.value:
-                    if cell.value not in columns:
-                        columns.append(cell.value)
-                    else:
-                        # add count to column name
-                        idx = 1
-                        while "{}.{}".format(cell.value, idx) in columns:
-                            idx += 1
-                        columns.append("{}.{}".format(cell.value, idx))
+                if cell.value and cell.value not in columns:
+                    columns.append(cell.value)
+                else:
+                    column_name = cell.value if cell.value else "unnamed"
+                    # add count to column name
+                    idx = 1
+                    while "{}.{}".format(column_name, idx) in columns:
+                        idx += 1
+                    columns.append("{}.{}".format(column_name, idx))
 
             df_dict = OrderedDict((column, []) for column in columns)
             # create index for the column headers

pandas_ods_reader/tests/test_read_ods.py

@@ -13,6 +13,7 @@ header_file = "example_headers.ods"
 no_header_file = "example_no_headers.ods"
 duplicated_column_names_file = "example_duplicated_column_names.ods"
 col_len_file = "example_col_lengths.ods"
+missing_header_file = "example_missing_header.ods"
 
 
 class TestOdsReader(object):
@@ -79,3 +80,11 @@ class TestOdsReader(object):
         with pytest.raises(ValueError) as e_info:
             read_ods(path, sheet_name)
             assert e_info.match(f"There is no sheet named {sheet_name}")
+
+    def test_missing_header(self):
+        path = os.path.join(rsc, missing_header_file)
+        df = read_ods(path, 1)
+        assert isinstance(df, pd.DataFrame)
+        assert len(df) == 10
+        assert (len(df.columns) == 5)
+        assert df.columns[2] == "unnamed.1"