Added function to update rows

.github/workflows/R-CMD-check.yaml

@@ -2,9 +2,9 @@
 # Need help debugging build failures? Start at https://github.com/r-lib/actions#where-to-find-help
 on:
   push:
-    branches: [main, master]
+    branches: [main, master, develop]
   pull_request:
-    branches: [main, master]
+    branches: [main, master, develop]
 
 name: R-CMD-check
 

.gitignore

@@ -4,3 +4,4 @@
 .Ruserdata
 .Rprofile
 CRAN-SUBMISSION
+.Renviron

NAMESPACE

@@ -20,6 +20,7 @@ export(ss_replace_sheet)
 export(ss_row_ids)
 export(ss_sheet_share)
 export(ss_sheetid)
+export(ss_update_rows)
 export(ss_write_sheet)
 export(ss_write_sheet_columns)
 importFrom(memoise,memoise)

R/ss_update_rows.R

@@ -0,0 +1,131 @@
+#' Update rows in a sheet.
+#'
+#' @param ss_id The sheetId, permalink, or name of the Smartsheet sheet to update
+#' @param data A data frame of rows to be updated. Column names should match Smartsheet column names if use_column_names is TRUE.
+#' @param row_ids Optional; a vector of the rowIds corresponding to the rows in the data frame.
+#'                Defaults to `NULL`. If `NULL`, the row IDs will be obtained from the sheet.
+#' @param row_numbers Optional; a vector of the row numbers corresponding to the rows in the data frame.
+#'                    This argument can be used to set a row_offset (e.g. `row_numbers = 1:nrow(data)+row_offset`).
+#'                    Defaults to `NULL`. If `NULL`, `row_ids` will be used. If not `NULL`, requires that `row_ids` are `NULL`.
+#' @param column_ids Optional; a vector of the columnIds of the smartsheets sheetId.
+#'                   If `NULL` and use_column_names is TRUE, the column IDs will be matched based on the column names in `data`.
+#' @param column_offset Optional; The number of columns to skip from the beginning for updating.
+#'                      Defaults to 0 (start at the first column).
+#'                      If non-zero, requires that `column_ids` are `NULL` and `use_column_names` is `FALSE`.
+#' @param use_column_names Optional; If `TRUE`, column names in `data` are used to match column IDs in the sheet.
+#'                         Defaults to `TRUE`.
+#'
+#' @examples
+#' \dontrun{
+#' ss_id = ss_sheetid(ss_write_sheet(paste0("smartsheetr-example-",random_sheet_name())))
+#' # Assuming existing rows and columns
+#' ss_add_rows(ss_id, data.frame("PK"="Value"))
+#' ss_update_rows(ss_id, data.frame("PK"="Updated Value"))
+#' ss_read_sheet(ss_id)
+#' # clean up
+#' ss_delete_sheet(ss_id)
+#' }
+#'
+#' @return A `ss_updaterows_resp` object
+#'
+#' @export
+ss_update_rows <- function(ss_id, data, row_ids = NULL, row_numbers = NULL, column_ids = NULL, column_offset = 0, use_column_names = TRUE) {
+  ss_id = validate_ss_id(ss_id)
+  if (!is.data.frame(data) || nrow(data) == 0) {
+    rlang::abort("data must be a non-empty data frame.")
+  }
+  if (!is.null(row_numbers) && (!is.numeric(row_numbers) || any(row_numbers < 1))) {
+    rlang::abort("row_numbers must be a vector of positive numbers.")
+  }
+  if (!is.null(row_numbers) && (length(row_numbers) != nrow(data))) {
+    rlang::abort("row_numbers must be the same length as the data.")
+  }
+  if (!is.null(row_numbers) && !is.null(row_ids)) {
+    rlang::abort("row_numbers cannot be used with row_ids.")
+  }
+  if (!is.null(column_offset) && (!is.numeric(column_offset) || column_offset < 0)) {
+    rlang::abort("column_offset must be a non-negative number.")
+  }
+  if (column_offset != 0 && !is.null(column_ids)) {
+    rlang::abort("column_offset cannot be used with column_ids.")
+  }
+  if (column_offset != 0 && use_column_names) {
+    rlang::abort("column_offset cannot be used with use_column_names.")
+  }
+  # Fetching Column IDs
+  if(is.null(column_ids)) {
+    col_resp = ss_get(path = paste0('sheets/', ss_id, '/columns'))
+    # Match column names to column IDs
+    if (use_column_names) {
+      # Create dataframe of column IDs and titles
+      sheet_columns = purrr::map(col_resp$content$data, ~data.frame(id=.x$id, title=.x$title)) |>
+        purrr::list_rbind()
+      # Check if all column names are matched and filter out unmatched columns
+      data_column_names = names(data)
+      unmatched_columns = setdiff(data_column_names, sheet_columns$title)
+      if (length(unmatched_columns) > 0) {
+        rlang::abort("No match found for column names: ", paste(unmatched_columns, collapse = ", "))
+      }
+      column_ids = sheet_columns |>
+        dplyr::filter(.data$title %in% data_column_names) |>
+        dplyr::pull(.data$id)
+    } else {
+      # Use all column IDs from the sheet
+      column_ids = purrr::map(col_resp$content$data, ~.x$id) |> purrr::list_c()
+    }
+  }
+
+  # Fetching Row IDs
+  if(is.null(row_ids)) {
+    # Fetching the number of rows in the data
+    if (is.null(row_numbers)) {
+      row_seq = 1:nrow(data)
+      row_numbers = row_seq
+    }
+    # Convert row numbers to comma separated string
+    row_numbers = paste(row_numbers, collapse = ",")
+    if (nchar(row_numbers) >= 5000) {
+      rlang::warn("The number of rows to update is too large. Please use row_ids instead.")
+      row_query = list()
+    } else {
+      row_query = list(rowNumbers=row_numbers)
+    }
+    row_resp = ss_get(path = paste0('sheets/', ss_id), query = row_query)
+    row_ids = purrr::map(row_resp$content$rows, ~.x$id) |> purrr::list_c()
+    # Check if enough row IDs are available
+    if (length(row_ids) == 0) {
+      rlang::abort("No row IDs available to update. Please add rows to the sheet.")
+    }
+    if (length(row_ids) < nrow(data)) {
+      rlang::abort("Not enough row IDs available to update all rows.")
+    }
+  }
+
+  if(length(row_ids) != nrow(data)) {
+    rlang::abort("The number of row_ids must match the number of rows in the data frame.")
+  }
+
+  if(column_offset >= ncol(data)) {
+    rlang::abort("The column_offset must be less than the number of columns in the data frame.")
+  }
+
+  # Format row data for API payload
+  row_data = list()
+  for(i in 1:nrow(data)) {
+    cell_data = list()
+    for(j in (1 + column_offset):ncol(data)) {
+      cell_value = unlist(data[i, j])
+      if (is.na(cell_value)) {
+        cell_value = NA # Convert NaNs to NA (otherwise this is converted to text)
+      }
+      cell_data[[j - column_offset]] = list(columnId = unlist(column_ids[j]), value = cell_value)
+    }
+    row_data[[i]] = list(id = row_ids[i], cells = cell_data)
+  }
+
+  # Update rows with JSON payload
+  path = paste0('sheets/', ss_id, '/rows')
+  resp = ss_put(path = path, body = to_json(row_data))
+  class(resp) = c("ss_updaterows_resp", class(resp))
+  return(resp)
+}

R/util.R

@@ -25,3 +25,7 @@ to_json <- function(...) {
 random_sheet_name <- function(n=10) {
   paste0(sample(c(LETTERS,letters), size=n, replace=TRUE), collapse="")
 }
+
+#' @noRd
+#' @keywords internal
+utils::globalVariables(c(".data"))