Invalid JSON payload received. Unexpected token. error when bulk uploading with update_cells(cell_list)

Hello,

I started getting the error when I updated to Python 3.7 two days ago. I use the following code to upload a pandas df to a gsheet - the error happens with any df.

Re the error: this seems to pertain to some cells with value ‘180901_millbrae_1’, may or may not be relevant. This exact same data and code used to work fine.

Any ideas?

Error:

Traceback (most recent call last):

  File "<ipython-input-31-85b52784b441>", line 11, in <module>
    ws.update_cells(cell_list)

  File "/Users/../anaconda/lib/python3.7/site-packages/gspread/models.py", line 782, in update_cells
    'values': values_rect

  File "/Users/../anaconda/lib/python3.7/site-packages/gspread/models.py", line 176, in values_update
    r = self.client.request('put', url, params=params, json=body)

  File "/Users/../anaconda/lib/python3.7/site-packages/gspread/client.py", line 79, in request
    raise APIError(response)

APIError: {
  "error": {
    "code": 400,
    "message": "Invalid JSON payload received. Unexpected token.\nae\", 43344, \"None\", NaN], [\"180901_millb\n                    ^",
    "status": "INVALID_ARGUMENT"
  }
}

Code:

n_rows, n_cols = df.shape
k = 1000
for i in range(n_rows//k + 1):    
    row0 = start_row + i*k
    col0 = start_col
    row1 = start_row + min((i+1)*k, n_rows-1)
    col1 = start_col + n_cols - 1
    cell_list = ws.range(inputRange(row0, col0, row1, col1))
    for cell in cell_list:
        val = df.iloc[cell.row-start_row, cell.col-start_col]
        cell.value = val
    ws.update_cells(cell_list)

Operating System: macOS Sierra 10.12.6 Python version: 3.7.0 gspread version: 3.1.0

[eanunez]

Hi, I encountered the same problem. I filled up ‘NaN’ values with empty string, df.fillna('', inplace=True) and the problem was solved.

[Imperssonator]

I found an alternate solution that preserves the numeric data type of numeric columns:

worksht.update(
    [df.columns.values.tolist()] + [[vv if pd.notnull(vv) else '' for vv in ll] for ll in df.values.tolist()]
)

Probably quite slow if your sheet is large but it works 🤷‍♂️