将 pandas 数据框附加到 Google 电子表格

Question

Case: My script returns a data frame that needs has to be appended to an existing google spreadsheet as new rows of data.As of now, I'm appending a data frame as multiple single rows through gspread.

My Code:

import gspread
import pandas as pd
df = pd.DataFrame()

# After some processing a non-empty data frame has been created.

output_conn = gc.open("SheetName").worksheet("xyz")

# Here 'SheetName' is google spreadsheet and 'xyz' is sheet in the workbook

for i, row in df.iterrows():
    output_conn.append_row(row)

Is there a way to append entire data-frame rather than multiple single rows?

Answer 1

I can recommend gspread-dataframe :

import gspread_dataframe as gd

# Connecting with `gspread` here

ws = gc.open("SheetName").worksheet("xyz")
existing = gd.get_as_dataframe(ws)
updated = existing.append(your_new_data)
gd.set_with_dataframe(ws, updated)

Answer 2

Here is the code to write, append(without loading the existing sheet into memory), and read to google sheets.

import gspread_dataframe as gd
import gspread as gs
gc = gs.service_account(filename="your/cred/file.json")

def export_to_sheets(worksheet_name,df,mode='r'):
    ws = gc.open("SHEET_NAME").worksheet("worksheet_name")
    if(mode=='w'):
        ws.clear()
        gd.set_with_dataframe(worksheet=ws,dataframe=df,include_index=False,include_column_header=True,resize=True)
        return True
    elif(mode=='a'):
        ws.add_rows(df.shape[0])
        gd.set_with_dataframe(worksheet=ws,dataframe=df,include_index=False,include_column_header=False,row=ws.row_count+1,resize=False)
        return True
    else:
        return gd.get_as_dataframe(worksheet=ws)
    
df = pd.DataFrame.from_records([{'a': i, 'b': i * 2} for i in range(100)])
export_to_sheets("SHEET_NAME",df,'a')

---

1. Write Mode: First clear existing worksheet => ws.clear() .Second using set_with_dataframe() uploading the dataframe, here note that resize=True , which strictily set the row and col in worksheet to df.shape. This will help later in append method.
2. Append Mode: First, add rows according to the dataframe. Second setting the parameter resize=False as we are adding rows and row=ws.row_count+1 anchoring its row value for append.
3. Read Mode(Default): returns a dataframe

Answer 3

I was facing the same problem, here's what I did converted the dataframe into list and used gspread's append_rows()

    gc = gspread.service_account(filename="credentials.json")
    sh = gc.open_by_key('<your_key>')
    ws = sh.sheet1
    
    ##data is the original data frame
    data_list = data.values.tolist()
    
    ws.append_rows(data_list)

Answer 4

The following approach, using gspread , may help one understand the procedures and solve the problem

1. Install the libraries in your environment.

2. Import the libraries in the script

   import pandas as pd import gspread from gspread_dataframe import set_with_dataframe

3. Create credentials in Google API console .

4. Add the following to the script, to access the Google Sheet

   gc = gspread.service_account(filename='GoogleAPICredentials.json') sh = gc.open_by_key('GoogleSheetID')

Assuming one wants to add to the first sheet, use 0 in get_worksheet (for the second sheet use 1, and so on)

worksheet = sh.get_worksheet(0)

5. Then, in order to export the dataframe, considering that the dataframe name is df , to a Google Sheet

   set_with_dataframe(worksheet, df)

Answer 5

I came up with the following solution. It does not overwrite current data but just appends entire pandas DataFrame df to the end of Sheet with name sheet in the Spreadsheet with the name spread_sheet .

import gspread
from google.auth.transport.requests import AuthorizedSession
from oauth2client.service_account import ServiceAccountCredentials

def append_df_to_gs(df, spread_sheet:str, sheet_name:str):
    scopes = [
        'https://spreadsheets.google.com/feeds',
        'https://www.googleapis.com/auth/drive',
    ]
    credentials = ServiceAccountCredentials.from_json_keyfile_name(
        path_to_credentials,
        scopes=scopes
    )
    gsc = gspread.authorize(credentials)
    sheet = gsc.open(spread_sheet)
    params = {'valueInputOption': 'USER_ENTERED'}
    body = {'values': df.values.tolist()}
    sheet.values_append(f'{sheet_name:str}!A1:G1', params, body)

For params valueInputOption please consult this . I used USER_ENTERED here as I needed some formulas to be valid once I append the data to Google Sheets.

Answer 6

ws = gc.open("sheet title").worksheet("Sheet1")

gd.set_with_dataframe(ws, dataframe)

#simply transform your dataframe to google sheet

Answer 7

I came up with the following solution using try/catch statement, in case the spreadsheet doesn't exsit he will create it for you and set the dataframe otherwise he will append it.

def load_to_sheet(conn_sheet, spreadsheet_name, df):
try:
    worksheet = conn_sheet.worksheet(spreadsheet_name)
    worksheet.add_rows(df.shape[0])
    set_with_dataframe(worksheet=worksheet, row=worksheet.row_count, dataframe=df, include_index=False,
                       include_column_header=False,
                       resize=False)
except Exception:
    worksheet = conn_sheet.add_worksheet(title=spreadsheet_name, rows=100, cols=100)
    set_with_dataframe(worksheet=worksheet, dataframe=df, include_index=False, include_column_header=True,
                       resize=True)

Answer 8

以下不需要 gspread 以外的外部库：

worksheet.update([dataframe.columns.values.tolist()] + dataframe.values.tolist())

Answer 9

如果 Google 电子表格采用 .csv 格式，那么您可以使用 df.to_csv() 将 pandas 数据帧转换为 csv 并以该格式保存