refreshable_tibble

mutate2 <- function(.data, ...) {
  dots <- rlang::enquos(...)
  has_tilde_lgl <- sapply(dots, function(x) {
    expr <- rlang::quo_get_expr(x)
    is.call(expr) && identical(expr[[1]], quote(`~`))
    })
  inds <- which(has_tilde_lgl)
  nms <- names(dots)[inds]
  exprs <- vector("list", length(inds))
  for(i in seq_along(inds)) {
    env <- attr(dots[[inds[i]]], ".Environment")
    exprs[[i]] <- rlang::quo_get_expr(dots[[inds[i]]])[[2]]
    dots[[inds[i]]] <- rlang::as_quosure(exprs[[i]], env = env)
  }
  res <- dplyr::mutate(.data, !!!dots)
  for(i in seq_along(inds)) {
    attr(res[[nms[i]]],"expr") <- exprs[[i]]
    class(res[[nms[i]]]) <- 
      c(paste0("~",pillar::type_sum(res[[nms[i]]])), 
        "refreshable_column", class(res[[nms[i]]]))
  }
  res
}

refresh <- function(x) {
  pf <- parent.frame()
  for(i in seq_along(x)) {
    if(inherits(x[[i]], "refreshable_column")) {
      cl <- class(x[[i]])
      expr <- attr(x[[i]],"expr")
      x[[i]] <- eval(expr, x, pf)
      attr(x[[i]],"expr") <- expr
      class(x[[i]]) <- 
        c(paste0("~", pillar::type_sum(col)), 
          "refreshable_column", class(col))
    }
  }
  x
}

# convert so headers are displayed
cars <- tibble::as_tibble(cars)

# mutate2 is like mutate but we use a `~` prefix for refreshable columns
df1 <- mutate2(cars, time1 = dist/speed, time2 = ~dist/speed)

# we see these cols have a header prefixed with "~"
print(df1, n = 2)
#> # A tibble: 50 x 4
#>   speed  dist time1 time2 
#>   <dbl> <dbl> <dbl> <~dbl>
#> 1     4     2   0.5 0.5   
#> 2     4    10   2.5 2.5   
#> # ... with 48 more rows

# let's change the dist column, the time2 column won't update but... 
df2 <- dplyr::mutate(df1, dist = dist*2)
print(df2, n = 2)
#> # A tibble: 50 x 4
#>   speed  dist time1 time2 
#>   <dbl> <dbl> <dbl> <~dbl>
#> 1     4     4   0.5 0.5   
#> 2     4    20   2.5 2.5   
#> # ... with 48 more rows

# ... if I refresh it will
df3 <- refresh(df2)
print(df3, n = 2)
#> # A tibble: 50 x 4
#>   speed  dist time1 time2 
#>   <dbl> <dbl> <dbl> <~dbl>
#> 1     4     4   0.5 1     
#> 2     4    20   2.5 5     
#> # ... with 48 more rows

# str() can be used to see the expr, note that it is a standard tibble
str(df3)
#> tibble [50 x 4] (S3: tbl_df/tbl/data.frame)
#>  $ speed: num [1:50] 4 4 7 7 8 9 10 10 10 11 ...
#>  $ dist : num [1:50] 4 20 8 44 32 20 36 52 68 34 ...
#>  $ time1: num [1:50] 0.5 2.5 0.571 3.143 2 ...
#>  $ time2: '~dbl' num [1:50] 1 5 1.14 6.29 4 ...
#>   ..- attr(*, "expr")= language dist/speed