Merge pull request #62 from aminebeh/fix/xslx/wrong_encoding_for_xlsx…

…_files fix(xlsx): Fix error issue for xlsx
ets · Jan 6, 2024 · 1c9a462 · 1c9a462
2 parents f811e40 + a032edf
commit 1c9a462
Showing 1 changed file with 3 additions and 1 deletion.
diff --git a/tap_spreadsheets_anywhere/format_handler.py b/tap_spreadsheets_anywhere/format_handler.py
@@ -157,10 +157,12 @@ def get_row_iterator(table_spec, uri):
             reader = get_streamreader(uri, universal_newlines=universal_newlines, open_mode='r', encoding=encoding)
             iterator = tap_spreadsheets_anywhere.csv_handler.get_row_iterator(table_spec, reader)
         elif format == 'excel':
-            reader = get_streamreader(uri, universal_newlines=universal_newlines,newline=None, open_mode='rb')
             if uri.lower().endswith(".xls"):
+                reader = get_streamreader(uri, universal_newlines=universal_newlines,newline=None, open_mode='rb')
                 iterator = tap_spreadsheets_anywhere.excel_handler.get_legacy_row_iterator(table_spec, reader)
             else:
+                # If encoding is set, smart_open will override binary mode ('b' in open_mode) and it will result in a BadZipFile error
+                reader = get_streamreader(uri, universal_newlines=universal_newlines,newline=None, open_mode='rb', encoding=None)
                 iterator = tap_spreadsheets_anywhere.excel_handler.get_row_iterator(table_spec, reader)
         elif format == 'json':
             reader = get_streamreader(uri, universal_newlines=universal_newlines, open_mode='r', encoding=encoding)