Skip to contents

List rows from a SeaTable table

Source: R/read.R
seatable_list_rows.Rd

Reads rows through SeaTable's list_rows API (rather than SQL), optionally restricted to a view and ordered by a column. Returns an R data.frame, or the raw pandas DataFrame when python = TRUE.

Usage

seatable_list_rows(
  table,
  base = NULL,
  con = default_connection(),
  view_name = NULL,
  order_by = NULL,
  desc = FALSE,
  start = 0L,
  limit = Inf,
  python = FALSE,
  chunksize = NULL,
  collapse_lists = TRUE
)

Arguments

table

Name of the table.

base

Optional base name or Base object; discovered from table when NULL.

con

A seatable_connection.

view_name

Optional view limiting the rows/columns returned.

order_by

Optional column name to order by.

desc

Whether to sort descending (default ascending).

start

Row offset to start from (0-based).

limit

Maximum rows to return; the default Inf returns all rows.

python

Whether to return the pandas DataFrame rather than an R data.frame.

chunksize

Advanced: rows to request per call. The default NULL chooses a size from the column count (SeaTable allows roughly one million cells per request).

collapse_lists

Whether to collapse multiple-select (and other list) columns into simple character vectors. The default TRUE comma-joins multi-valued cells; FALSE keeps them as list-columns. A string is used as the separator instead of ",".

Value

An R data.frame, or a pandas DataFrame when python = TRUE.

Details

The list_rows endpoint caps the number of rows per request, and that cap is server-configurable with no API to discover it, so a page coming back shorter than requested does not signal the end of the table – only an empty page does. Rows are therefore fetched in chunks until an empty page (or limit) is reached, then concatenated once and coerced together, so multiple-select and other list columns flatten uniformly across chunks.

See also

seatable_query(), seatable_list_selected()