Migrating from GSpreadManager

GSpreadManager is now part of gsuite-sdk: its engine runs behind gsuite_sheets.Sheets / Spreadsheet / Worksheet. The GSpreadManager package stays on PyPI at 0.1.5 and won't get new releases.

pip uninstall GSpreadManager
pip install gsuite-sdk                 # + "gsuite-sdk[pandas]" or "[polars]" for DataFrames

Authentication

GSpreadManager used a service account file; gsuite-sdk supports it too (and OAuth user login, shared with Gmail/Drive/Calendar):

from gsuite_core import GoogleAuth, Scopes
from gsuite_sheets import Sheets

auth = GoogleAuth.from_service_account("credentials.json", scopes=Scopes.sheets() + Scopes.drive())
sheets = Sheets(auth)

Share the spreadsheet with the service account's email, as before.

From 0.1.5 (GoogleSheetConector)

GSpreadManager 0.1.5	gsuite-sdk
c = GoogleSheetConector(doc, json_file, sheet_name)	ws = Sheets(auth).open(doc).worksheet(sheet_name)
c.read_sheet_data(output_format="list")	ws.get_all_values()
c.read_sheet_data(output_format="dict")	ws.get_all_records()
c.read_sheet_data(output_format="pandas")	ws.read_dataframe()
c.read_sheet_data(skiprows=n)	ws.read_dataframe(skiprows=n) / ws.get_all_values()[n:]
c.spreadsheet_append(data, tab_name)	doc.worksheet(tab_name).append_rows(data)
c.update_cell(sheet, row, col, value)	ws.update_cell(row, col, value)
c.update_row(sheet, row, data, start_column)	ws.update_row(row, data, start_column)
c.spreadsheet_read_range(sheet, tab, r1, r2, "A", "D")	doc.worksheet(tab).get(f"A{r1}:D{r2}")
c.get_rows_where_column_equals(column, value)	ws.rows_where_column_equals(column, value) (0-based column)
c.batch_update(range_data)	ws.batch_update(range_data)
c.get_last_row(tab_name)	ws.last_row()
c.get_row_with_empty_in_column(sheet, "B")	ws.row_with_empty_in_column("B")
c.spreadsheet_insert(sheet, tab, data, fila=n)	ws.insert(data, row=n)

ws.insert(data, row=n) really inserts at row n, shifting rows down. GSpreadManager sent a values.append starting at that row, which Google treats as "append after the table found there", so the data ended up at the end of the table.

Example

# Before
from gspreadmanager import GoogleSheetConector

sheet = GoogleSheetConector(doc_name="Inversiones", json_google_file="creds.json")
sheet.spreadsheet_append([[today, total]])
df = sheet.read_sheet_data(tab_name="Total diario", output_format="pandas")

# After
from gsuite_core import GoogleAuth, Scopes
from gsuite_sheets import Sheets

auth = GoogleAuth.from_service_account("creds.json", scopes=Scopes.sheets() + Scopes.drive())
doc = Sheets(auth).open("Inversiones")
doc.sheet1.append_row([today, total])
df = doc.worksheet("Total diario").read_dataframe()

From 2.x / 3.0 (SheetManager)

Most WorksheetContext methods keep their names on Worksheet (upsert, update_where, iter_rows, read_as, add_dropdown, add_chart, sort_range, insert_rows, ...). The differences:

GSpreadManager 3.0	gsuite-sdk
SheetManager("Doc", json_google_file=...)	Sheets(auth).open("Doc")
SheetManager(key=...) / .open_by_url(url)	sheets.open_by_key(key) / sheets.open_by_url(url)
SheetManager(..., cache=True, rate_limit=...)	Sheets(auth, cache=True); retries and rate-limit handling are built in
SheetManager(..., dataframe_backend="polars")	Sheets(auth, dataframe_backend="polars")
mgr.worksheet("Tab")	doc.worksheet("Tab")
mgr.list_worksheets()	doc.worksheets
mgr.worksheet_by_index(i)	doc.get_worksheet(i)
mgr.create_sheet(t) / mgr.delete_sheet(t)	doc.add_worksheet(t) / doc.del_worksheet(ws)
mgr.create_spreadsheet(t) / copy_spreadsheet / delete_spreadsheet	sheets.create(t) / sheets.copy(id) / sheets.delete(id)
mgr.export(ExportFormat.PDF)	doc.export("pdf")
ws.read(output_format=...)	ws.get_all_values() / get_all_records() / read_dataframe()
ws.read_range(...)	ws.get("A1:D10")
ws.append(data)	ws.append_rows(data)
ws.insert(data, fila=n)	ws.insert(data, row=n) (now really inserts; see above)
ws.format_range(r, CellFormat(...))	ws.format(r, CellFormat(...))
ws.add_protected_range(r)	ws.protect(r, editors=[...])
ws.find_replace(...)	ws.replace(find, replacement, regex=...)
AsyncSheetManager	not yet; async is planned for the whole suite
backend="gspread" / "native"	not needed: requests go through google-suite's client
gspreadmanager.testing.InMemoryBackend	gsuite_sheets.engine.testing.InMemoryBackend + fake_sheets(backend.client)

Errors are subclasses of gsuite_core.exceptions.GSuiteError; the GSpreadManager names (SchemaError, WorksheetNotFoundError, ...) are importable from gsuite_sheets.