Using Python, I'm trying to check each cell in a Google Sheet for a strikethrough and return the rows without a strikethrough. The problem I run into is reaching my quota limit for the current minute since I'm checking 150 rows. Plus I need to check 11 sheets with anywhere from 50-200 rows each.
Is there a way to "batchGet()" formatting, not just values, so I don't hit my quota limit?
I'm not sure what example code is really needed so here's my working "is this cell formatted with a strikethrough" definition:
def row_has_strikethrough(sheet, i):
return 'strikethrough=True' in str(get_user_entered_format(sheet, 'A' str(i)))
This works within my while loop but again, I hit the quota:
last_row = int(sheet.acell('C1').value) # 166
i = 3
while i <= last_row:
if not row_has_strikethrough(sheet, i):
records = get_record(sheet, MAPPING, i)
i = 1
Any help would be greatly appreciated!
CodePudding user response:
I believe your goal is as follows.
- About
I'm trying to check each cell in a Google Sheet for a strikethrough and return the rows without a strikethrough.
, from your script ofreturn 'strikethrough=True' in str(get_user_entered_format(sheet, 'A' str(i)))
, I thought that you might have wanted to check whether the cell value of column "A" has the strikethrough. - And, from your showing script, I thought that you might be using gspread for python.
- You want to reduce the process cost of the script.
If my understanding is correct, how about the following sample script?
Sample script:
spreadsheetId = '###' # Please set your Spreadsheet ID.
client = gspread.authorize(credentials) # Here, please use your credentials and client
sheetNames = ['Sheet1', 'Sheet2'] # Please set the sheet names you want to retrieve the values.
# 1. Retrieve textFormat from each sheet using one API call.
access_token = credentials.access_token
ranges = '&'.join(['ranges=' e for e in sheetNames])
url = 'https://sheets.googleapis.com/v4/spreadsheets/' spreadsheetId '?fields=sheets(data(rowData(values(userEnteredFormat(textFormat(strikethrough))))))&' ranges
res1 = requests.get(url, headers={'Authorization': 'Bearer ' access_token})
obj = res1.json()
rowNumbers = {}
for i, d in enumerate(obj['sheets']):
sheet = []
for j, r in enumerate(d['data'][0].get('rowData')):
if 'values' not in r:
sheet.append(j)
rowNumbers[sheetNames[i]] = sheet
# 2. Retrieve values from each sheet using one API call.
spreadsheet = client.open_by_key(spreadsheetId)
res2 = spreadsheet.values_batch_get(sheetNames)['valueRanges']
# 3. Retrieve the rows from each sheet by checking the rows without the strikethrough at the column "A".
values = {}
for i, s in enumerate(res2):
temp = []
for j, r in enumerate(s['values']):
sheet = sheetNames[i]
if j in rowNumbers[sheet]:
temp.append(r)
values[sheet] = temp
print(values) # Here, you can see the result value.
In this script, 2 API calls of Sheets API are used.
Testing:
When this script is used in a sample Spreadsheet, the following result is obtained.
{
'Sheet1': [['a2', 'b2', 'c2'], ['a5', 'b5', 'c5'], ['a6', 'b6', 'c6']],
'Sheet2': [['a3', 'b3', 'c3'], ['a5', 'b5', 'c5'], ['a6', 'b6', 'c6']]
}
This output value has the rows of each sheet. Those rows are the rows without the strikethrough at column "A".