Last active
September 28, 2025 18:31
-
-
Save krlmlr/2b3a72e1bda91b06bce8238319350b43 to your computer and use it in GitHub Desktop.
Ensure type stability for an R data frame
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| library(conflicted) | |
| library(rlang) | |
| library(vctrs) | |
| library(dplyr) | |
| ensure_type <- function(.data, ..., .default = NULL, .call = caller_env()) { | |
| try_fetch( | |
| try_ensure_type(.data, ..., .default = {{ .default }}), | |
| error = function(e) { | |
| cli::cli_abort( | |
| call = .call, | |
| "Type stability violated", | |
| parent = e | |
| ) | |
| } | |
| ) | |
| } | |
| try_ensure_type <- function(.data, ..., .default = NULL, .call = caller_env()) { | |
| default <- enquo(.default) | |
| types <- vctrs::data_frame(...) | |
| names <- names(types) | |
| if (!quo_is_null(default)) { | |
| missing <- setdiff(colnames(.data), names) | |
| names <- c(names, missing) | |
| types[missing] <- rlang::eval_tidy(.default) | |
| } | |
| if (!all(names %in% names(.data))) { | |
| cli::cli_abort(call = .call, c( | |
| "Columns missing", | |
| i = "Columns {.var {setdiff(names, names(.data))}} not found in data" | |
| )) | |
| } | |
| out <- .data[names] | |
| for (i in seq_along(out)) { | |
| out[[i]] <- vec_cast(out[[i]], types[[i]], x_arg = names[[i]], call = .call) | |
| } | |
| out |> | |
| as_tibble() | |
| } | |
| test1 <- function() { | |
| data.frame(a = 1:3, b = c(4, 4, 5)) |> | |
| ensure_type(b = integer(), c = character(), .default = numeric()) | |
| } | |
| test2 <- function() { | |
| data.frame(a = 1:3, b = c(4, 4, 5)) |> | |
| ensure_type(b = character(), .default = numeric()) | |
| } | |
| test3 <- function() { | |
| data.frame(a = 1:3, b = c(4, 4, 5)) |> | |
| ensure_type(b = integer(), .default = numeric()) | |
| } | |
| test1() | |
| #> Error in `test1()`: | |
| #> ! Type stability violated | |
| #> Caused by error in `ensure_type()`: | |
| #> ! Columns missing | |
| #> ℹ Columns `c` not found in data | |
| test2() | |
| #> Error in `test2()`: | |
| #> ! Type stability violated | |
| #> Caused by error in `ensure_type()`: | |
| #> ! Can't convert `b` <double> to <character>. | |
| test3() | |
| #> # A tibble: 3 × 2 | |
| #> b a | |
| #> <int> <dbl> | |
| #> 1 4 1 | |
| #> 2 4 2 | |
| #> 3 5 3 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment