Appending rows with empty cells results in increasing row indentation

Environment info

Operating System: Windows 10 Python version: 3.6.3 (Anaconda) gspread version: both 2.0.0 and 2.1.1

Steps to reproduce

ws = sheet.worksheet('Sheet1')
ws.append_row(['Test1', '', 'Test2'])
ws.append_row(['Test1', '', 'Test2'])
ws.append_row(['Test1', '', 'Test2'])

Output

It looks like new rows appended with append_row() start from the last populated column on the previous rows. This is a relatively new behaviour that started a day or so ago.

[burnash]

It looks like this is not a bug. I tried to run append_row() with different table layouts and it appears that the row shift happens when the Sheets API has multiple logical “tables” to chose from and it selects the rightmost. It’s explained in details in the example in the Sheets API guide here: https://developers.google.com/sheets/api/guides/values#appending_values

So In this particular case, when appending the second row the Sheets API sees two logical “tables” A1 and C1 and decides to append to C1:

When appending the third the same happens with E2 and so on.

As the Sheets API guide suggests the solution is to explicitly tell the Sheets API which table it needs to append to. For that, we’ll need to pass an explicit range parameter to the API call. I’ve implemented this in #726 (and named the parameter table_range).

So for this particular case, after #726 is merged you’ll need to call append_row this way:

ws.append_row(['Test1', '', 'Test2'], table_range='A1')

That way the API will know that you’d like to append to the table that starts with A1 cell. Again, this is only need in the case where there’s ambiguity in the selection of logical “tables”.

[JeffJassky]

… Still happening.