Merge pull request #153 from tidymodels/use-cli

use cli functions

DESCRIPTION

@@ -21,7 +21,8 @@ BugReports: https://github.com/tidymodels/themis/issues
 Depends: 
     R (>= 3.6),
     recipes (>= 1.1.0.9000)
-Imports: 
+Imports:
+    cli,
     gower,
     lifecycle (>= 1.0.3),
     dplyr,
@@ -48,4 +49,4 @@ Config/testthat/edition: 3
 Encoding: UTF-8
 LazyData: true
 Roxygen: list(markdown = TRUE)
-RoxygenNote: 7.3.1
+RoxygenNote: 7.3.2

R/adasyn.R

@@ -37,7 +37,7 @@
 #'
 #' When you [`tidy()`][tidy.recipe()] this step, a tibble is retruned with
 #'  columns `terms` and `id`:
-#' 
+#'
 #' \describe{
 #'   \item{terms}{character, the selectors or variables selected}
 #'   \item{id}{character, id of this step}
@@ -151,7 +151,7 @@ step_adasyn_new <-
 prep.step_adasyn <- function(x, training, info = NULL, ...) {
   col_name <- recipes_eval_select(x$terms, training, info)
   if (length(col_name) > 1) {
-    rlang::abort("The selector should select at most a single variable")
+    cli::cli_abort("The selector should select at most a single variable.")
   }
 
   if (length(col_name) == 1) {

R/adasyn_impl.R

@@ -32,21 +32,21 @@
 #' res <- adasyn(circle_numeric, var = "class", over_ratio = 0.8)
 adasyn <- function(df, var, k = 5, over_ratio = 1) {
   if (length(var) != 1) {
-    rlang::abort("Please select a single factor variable for `var`.")
+    cli::cli_abort("Please select a single factor variable for {.arg var}.")
   }
 
   var <- rlang::arg_match(var, colnames(df))
 
   if (!(is.factor(df[[var]]) | is.character(df[[var]]))) {
-    rlang::abort(glue("`{var}` should be a factor or character variable."))
+    cli::cli_abort("{.var {var}} should be a factor or character variable.")
   }
 
   if (length(k) != 1) {
-    rlang::abort("`k` must be length 1.")
+    cli::cli_abort("The {.arg k} argument must be length 1.")
   }
 
   if (k < 1) {
-    rlang::abort("`k` must be non-negative.")
+    cli::cli_abort("The {.arg k} argument must be non-negative.")
   }
 
   predictors <- setdiff(colnames(df), var)
@@ -83,12 +83,7 @@ adasyn_impl <- function(df, var, k = 5, over_ratio = 1, call = caller_env()) {
     minority <- data_mat[!min_class_in, , drop = FALSE]
 
     if (nrow(minority) <= k) {
-      rlang::abort(
-        glue(
-        "Not enough observations of '{min_names[i]}' to perform ADASYN."
-        ),
-        call = call
-      )
+      cli::cli_abort("Not enough observations of {.val {min_names[i]}} to perform ADASYN.", call = call)
     }
 
     tmp_df <- as.data.frame(

R/bsmote.R

@@ -59,7 +59,7 @@
 #'
 #' When you [`tidy()`][tidy.recipe()] this step, a tibble is retruned with
 #'  columns `terms` and `id`:
-#' 
+#'
 #' \describe{
 #'   \item{terms}{character, the selectors or variables selected}
 #'   \item{id}{character, id of this step}
@@ -183,7 +183,7 @@ step_bsmote_new <-
 prep.step_bsmote <- function(x, training, info = NULL, ...) {
   col_name <- recipes_eval_select(x$terms, training, info)
   if (length(col_name) > 1) {
-    rlang::abort("The selector should select at most a single variable")
+    cli::cli_abort("The selector should select at most a single variable.")
   }
 
   if (length(col_name) == 1) {

R/bsmote_impl.R

@@ -57,21 +57,21 @@
 #' res <- bsmote(circle_numeric, var = "class", all_neighbors = TRUE)
 bsmote <- function(df, var, k = 5, over_ratio = 1, all_neighbors = FALSE) {
   if (length(var) != 1) {
-    rlang::abort("Please select a single factor variable for `var`.")
+    cli::cli_abort("Please select a single factor variable for {.arg var}.")
   }
 
   var <- rlang::arg_match(var, colnames(df))
 
   if (!(is.factor(df[[var]]) | is.character(df[[var]]))) {
-    rlang::abort(glue("{var} should be a factor or character variable."))
+    cli::cli_abort("{.var {var}} should be a factor or character variable.")
   }
 
   if (length(k) != 1) {
-    rlang::abort("`k` must be length 1.")
+    cli::cli_abort("{.arg k} must be length 1.")
   }
 
   if (k < 1) {
-    rlang::abort("`k` must be non-negative.")
+    cli::cli_abort("The {.arg k} argument must be non-negative.")
   }
 
   predictors <- setdiff(colnames(df), var)
@@ -100,9 +100,10 @@ bsmote_impl <- function(df, var, k = 5, over_ratio = 1, all_neighbors = FALSE) {
     )
 
     if (sum(danger_ids) <= k) {
-      rlang::abort(glue(
-        "Not enough danger observations of '{min_names[i]}' to perform BSMOTE."
-      ))
+      cli::cli_abort(
+        "Not enough danger observations of {.val {min_names[i]}} to perform
+        BSMOTE."
+      )
     }
 
     if (all_neighbors == FALSE) {

R/downsample.R

@@ -53,7 +53,7 @@
 #'
 #' When you [`tidy()`][tidy.recipe()] this step, a tibble is retruned with
 #'  columns `terms` and `id`:
-#' 
+#'
 #' \describe{
 #'   \item{terms}{character, the selectors or variables selected}
 #'   \item{id}{character, id of this step}
@@ -179,7 +179,7 @@ prep.step_downsample <- function(x, training, info = NULL, ...) {
   }
 
   if (length(col_name) > 1) {
-    rlang::abort("The selector should select at most a single variable")
+    cli::cli_abort("The selector should select at most a single variable.")
   }
 
   if (length(col_name) == 0) {

R/misc.R

@@ -8,18 +8,16 @@ check_na <- function(data, step, call = caller_env()) {
   na_cols <- vapply(data, function(x) any(is.na(x)), FUN.VALUE = logical(1))
   if (any(na_cols)) {
     cols <- paste(names(na_cols)[na_cols], collapse = ", ")
-    rlang::abort(
-      glue(
-        "Cannot have any missing values. NAs found ind: {cols}."
-      ),
+    cli::cli_abort(
+      "Cannot have any missing values. NAs found in {cols}.",
       call = call
     )
   }
 }
 
 check_2_levels_only <- function(data, col_name, call = caller_env()) {
   if (length(levels(data[[col_name]])) != 2) {
-    rlang::abort(glue("`{col_name}` must only have 2 levels."), call = call)
+    cli::cli_abort("The {.code {col_name}} must only have 2 levels.", call = call)
   }
 }
 
@@ -28,14 +26,14 @@ check_numeric <- function(dat) {
   label <- "numeric"
 
   if (!all(all_good)) {
-    rlang::abort("All columns for this function should be numeric.")
+    cli::cli_abort("All columns for this function should be numeric.")
   }
   invisible(all_good)
 }
 
 check_column_factor <- function(data, column, call = caller_env()) {
   if (!is.factor(data[[column]])) {
-    rlang::abort(glue("`{column}` should be a factor variable."), call = call)
+    cli::cli_abort("{.code {column}} should be a factor variable.", call = call)
   }
 }
 

R/nearmiss.R

@@ -40,7 +40,7 @@
 #'
 #' When you [`tidy()`][tidy.recipe()] this step, a tibble is retruned with
 #'  columns `terms` and `id`:
-#' 
+#'
 #' \describe{
 #'   \item{terms}{character, the selectors or variables selected}
 #'   \item{id}{character, id of this step}
@@ -159,7 +159,7 @@ prep.step_nearmiss <- function(x, training, info = NULL, ...) {
   col_name <- recipes_eval_select(x$terms, training, info)
 
   if (length(col_name) > 1) {
-    rlang::abort("The selector should select at most a single variable")
+    cli::cli_abort("The selector should select at most a single variable.")
   }
 
   if (length(col_name) == 1) {

R/nearmiss_impl.R

@@ -32,21 +32,21 @@
 #' res <- nearmiss(circle_numeric, var = "class", under_ratio = 1.5)
 nearmiss <- function(df, var, k = 5, under_ratio = 1) {
   if (length(var) != 1) {
-    rlang::abort("Please select a single factor variable for `var`.")
+    cli::cli_abort("Please select a single factor variable for {.arg var}.")
   }
 
   var <- rlang::arg_match(var, colnames(df))
 
   if (!(is.factor(df[[var]]) | is.character(df[[var]]))) {
-    rlang::abort(glue("`{var}` should be a factor or character variable."))
+    cli::cli_abort("{.var {var}} should be a factor or character variable.")
   }
 
   if (length(k) != 1) {
-    rlang::abort("`k` must be length 1.")
+    cli::cli_abort("The {.arg k} argument must have length 1.")
   }
 
   if (k < 1) {
-    rlang::abort("`k` must be non-negative.")
+    cli::cli_abort("The {.arg k} argument must be non-negative.")
   }
 
   predictors <- setdiff(colnames(df), var)
@@ -69,11 +69,7 @@ nearmiss_impl <- function(df, var, ignore_vars, k = 5, under_ratio = 1) {
     not_class <- subset_to_matrix(df_only, var, names(classes)[i], FALSE)
 
     if (nrow(not_class) <= k) {
-      rlang::abort(
-        glue(
-        "Not enough danger observations of '{names(classes)[i]}' to perform NEARMISS."
-        )
-      )
+      cli::cli_abort("Not enough danger observations of {.val {names(classes)[i]}} to perform NEARMISS.")
     }
 
     dists <- RANN::nn2(

R/rose.R

@@ -52,7 +52,7 @@
 #'
 #' When you [`tidy()`][tidy.recipe()] this step, a tibble is retruned with
 #'  columns `terms` and `id`:
-#' 
+#'
 #' \describe{
 #'   \item{terms}{character, the selectors or variables selected}
 #'   \item{id}{character, id of this step}
@@ -168,7 +168,7 @@ step_rose_new <-
 prep.step_rose <- function(x, training, info = NULL, ...) {
   col_name <- recipes_eval_select(x$terms, training, info)
   if (length(col_name) > 1) {
-    rlang::abort("The selector should select at most a single variable")
+    cli::cli_abort("The selector should select at most a single variable.")
   }
   if (length(col_name) == 1) {
     check_column_factor(training, col_name)

R/smote.R

@@ -44,7 +44,7 @@
 #'
 #' When you [`tidy()`][tidy.recipe()] this step, a tibble is retruned with
 #'  columns `terms` and `id`:
-#' 
+#'
 #' \describe{
 #'   \item{terms}{character, the selectors or variables selected}
 #'   \item{id}{character, id of this step}
@@ -157,7 +157,7 @@ step_smote_new <-
 prep.step_smote <- function(x, training, info = NULL, ...) {
   col_name <- recipes_eval_select(x$terms, training, info)
   if (length(col_name) > 1) {
-    rlang::abort("The selector should select at most a single variable")
+    cli::cli_abort("The selector should select at most a single variable.")
   }
 
   if (length(col_name) == 1) {

R/smote_impl.R

@@ -40,21 +40,21 @@
 #' res <- smote(circle_numeric, var = "class", over_ratio = 0.8)
 smote <- function(df, var, k = 5, over_ratio = 1) {
   if (length(var) != 1) {
-    rlang::abort("Please select a single factor variable for `var`.")
+    cli::cli_abort("Please select a single factor variable for {.arg var}.")
   }
 
   var <- rlang::arg_match(var, colnames(df))
 
   if (!(is.factor(df[[var]]) | is.character(df[[var]]))) {
-    rlang::abort(glue("`{var}` should be a factor or character variable."))
+    cli::cli_abort("{.var {var}} should be a factor or character variable.")
   }
 
   if (length(k) != 1) {
-    rlang::abort("`k` must be length 1.")
+    cli::cli_abort("The {.arg k} must be length 1.")
   }
 
   if (k < 1) {
-    rlang::abort("`k` must be non-negative.")
+    cli::cli_abort("{.arg k} must be non-negative.")
   }
 
   predictors <- setdiff(colnames(df), var)
@@ -80,12 +80,7 @@ smote_impl <- function(df, var, k, over_ratio, call = caller_env()) {
     minority <- as.matrix(minority_df[names(minority_df) != var])
 
     if (nrow(minority) <= k) {
-      rlang::abort(
-        glue(
-          "Not enough observations of '{min_names[i]}' to perform SMOTE."
-        ),
-        call = call
-      )
+      cli::cli_abort("Not enough observations of {.val {min_names[i]}} to perform SMOTE.", call = call)
     }
 
     synthetic <- smote_data(minority, k = k, n_samples = samples_needed[i])

R/smotenc.R

@@ -46,7 +46,7 @@
 #'
 #' When you [`tidy()`][tidy.recipe()] this step, a tibble is retruned with
 #'  columns `terms` and `id`:
-#' 
+#'
 #' \describe{
 #'   \item{terms}{character, the selectors or variables selected}
 #'   \item{id}{character, id of this step}
@@ -143,7 +143,7 @@ step_smotenc_new <-
 prep.step_smotenc <- function(x, training, info = NULL, ...) {
   col_name <- recipes_eval_select(x$terms, training, info)
   if (length(col_name) > 1) {
-    rlang::abort("The selector should select at most a single variable")
+    cli::cli_abort("The selector should select at most a single variable.")
   }
 
   if (length(col_name) == 1) {

R/smotenc_impl.R

@@ -47,21 +47,23 @@ smotenc <- function(df, var, k = 5, over_ratio = 1) {
   # the input variables need to be numeric and contain no NA values
 
   if (length(var) != 1) {
-    rlang::abort("Please select a single factor variable for `var`.")
+    cli::cli_abort("Please select a single factor variable for {.arg var}.")
   }
 
   var <- rlang::arg_match(var, colnames(df))
 
   if (!(is.factor(df[[var]]) | is.character(df[[var]]))) {
-    rlang::abort(paste0(var, " should be a factor or character variable."))
+    cli::cli_abort(
+      "{var} should be {.cls factor} or {.cls character} variable."
+    )
   }
 
   if (length(k) != 1) {
-    rlang::abort("`k` must be length 1.")
+    cli::cli_abort("The {.arg k} must have length 1.")
   }
 
   if (k < 1) {
-    rlang::abort("`k` must be non-negative.")
+    cli::cli_abort("The {.arg k} argument must be non-negative.")
   }
 
   check_na(select(df, -all_of(var)))
@@ -99,10 +101,9 @@ smotenc_impl <- function(df, var, k, over_ratio) {
 
     # Ensure that we have more minority isntances than desired neighbors
     if (nrow(minority) <= k) {
-      rlang::abort(paste0(
-        "Not enough observations of '", min_names[i],
-        "' to perform SMOTE."
-      ))
+      cli::cli_abort(
+        "Not enough observations of {.var {min_names[i]}} to perform SMOTE."
+      )
     }
 
     # Run the smote algorithm (minority data, # of neighbors, # of sampeles needed)

R/tomek.R

@@ -38,7 +38,7 @@
 #'
 #' When you [`tidy()`][tidy.recipe()] this step, a tibble is retruned with
 #'  columns `terms` and `id`:
-#' 
+#'
 #' \describe{
 #'   \item{terms}{character, the selectors or variables selected}
 #'   \item{id}{character, id of this step}
@@ -139,7 +139,7 @@ step_tomek_new <-
 prep.step_tomek <- function(x, training, info = NULL, ...) {
   col_name <- recipes_eval_select(x$terms, training, info)
   if (length(col_name) > 1) {
-    rlang::abort("The selector should select at most a single variable")
+    cli::cli_abort("The selector should select at most a single variable.")
   }
 
   if (length(col_name) == 1) {