Sheets Client - is it required that a bounded gridRange be square?

[nraffuse]

I'm using batchUpdateByDataFilter with a gridRange to update a sheet.
I can get the request to work, but I've noticed that I cannot match the shape of my gridRange to the shape of my values, unless the shape of my values is square (equal length for rows and columns).

I don't know if this is expected behavior or not, but the fact that I'm able to pass a gridRange with independent lengths for width and height seems to suggest that its not. Is this a bug or expected?

Example request body

    values = [
        ['X', 'X', 'X'],
    ]
    body = {
        "valueInputOption": 'USER_ENTERED',
        "data": [
            {"dataFilter": {
                'gridRange': {
                    'sheetId': id_sheet,
                    'startRowIndex': 0,
                    'endRowIndex': 1,
                    'startColumnIndex': 0,
                    'endColumnIndex': 3,
                },
            }, "values": values}
        ]
    }

results in a 400 with this response

Invalid dataFilterValueRange[0]: dataFilterValueRange.dataFilter matched range Sheet2!$A$1:$C$1 whose rows cannot fit the data described in data.values[0]

For this example I have to increment the endRowIndex up to 3 for the request to succeed.
Are bounded gridRanges required to be square?

I can see in the documentation, the only examples are for square, half-bounded, or fully unbounded ranges.
https://developers.google.com/sheets/api/reference/rest/v4/spreadsheets/other#gridrange

UPDATE
Testing a little bit more and I've found that I can not issue an update when passing a greater than 26 rows, suggesting that I am bounded to an equivalent column width, even when I issue the command with and unbounded range.
Returned 400

Invalid dataFilterValueRange[0]: dataFilterValueRange.dataFilter matched range Sheet2!$A$1:$Z$1000 whose dimension_unspecified cannot fit the data described in data.values.

v2.62.0 / Python 3.10 on Windows 10

[Jerther]

I've got the same problem with versions 2.128.0 and 2.147.0, and by trial and error, I have found that it is fixed by specifying the majorDimension of the DataFilterValueRange. So in your case that would be:

    values = [
        ['X', 'X', 'X'],
    ]
    body = {
        "valueInputOption": 'USER_ENTERED',
        "data": [{
            "dataFilter": {
                'gridRange': {
                    'sheetId': id_sheet,
                    'startRowIndex': 0,
                    'endRowIndex': 1,
                    'startColumnIndex': 0,
                    'endColumnIndex': 3,
                },
            }, 
            "majorDimension": 'ROWS',  # <-- THIS
            "values": values
        }]
    }