Check column names of dataframes?

DESCRIPTION

@@ -1,7 +1,7 @@
 Type: Package
 Package: datawizard
 Title: Easy Data Wrangling and Statistical Transformations
-Version: 0.13.0.15
+Version: 0.13.0.16
 Authors@R: c(
     person("Indrajeet", "Patil", , "[email protected]", role = "aut",
            comment = c(ORCID = "0000-0003-1995-6531")),

R/adjust.R

@@ -30,7 +30,7 @@
 #'
 #' @return A data frame comparable to `data`, with adjusted variables.
 #'
 #' @examplesIf require("bayestestR", quietly = TRUE) && require("rstanarm", quietly = TRUE) && require("gamm4", quietly = TRUE)
 #' adjusted_all <- adjust(attitude)
 #' head(adjusted_all)
 #' adjusted_one <- adjust(attitude, effect = "complaints", select = "rating")
@@ -75,12 +75,8 @@
                    ignore_case = FALSE,
                    regex = FALSE,
                    verbose = FALSE) {
-  if (!all(colnames(data) == make.names(colnames(data), unique = TRUE))) {
-    insight::format_warning(
-      "Bad column names (e.g., with spaces) have been detected which might create issues in many functions.",
-      "Please fix it (you can run `names(mydata) <- make.names(names(mydata))` for a quick fix)."
-    )
-  }
+  # make sure column names are syntactically valid
+  .check_dataframe_names(data, action = "error")
 
   # check for formula notation, convert to character vector
   if (inherits(effect, "formula")) {

R/utils.R

@@ -47,6 +47,22 @@
 }
 
 
+#' Checks data framesfor syntactically valid column names
+#' Argument "action" can be "warning", "error", or "message".
+#'
+#' @keywords internal
+#' @noRd
+.check_dataframe_names <- function(x, action = "warning", verbose = TRUE) {
+  if (verbose && !all(colnames(x) == make.names(colnames(x), unique = TRUE))) {
+    insight::format_alert(
+      "Bad column names (e.g., with spaces) have been detected which might create issues in many functions.",
+      "Please fix it (you can run `names(mydata) <- make.names(names(mydata))` or use `janitor::clean_names()` for a quick fix).", # nolint
+      type = action
+    )
+  }
+}
+
+
 #' Fuzzy grep, matches pattern that are close, but not identical
 #' @examples
 #' colnames(iris)
@@ -125,7 +141,7 @@
   for (nm in setdiff(names(custom_attr), names(attributes(data.frame())))) {
     attr(data, which = nm) <- custom_attr[[nm]]
   }
   return(data)
 }
 
 
@@ -246,11 +262,11 @@
 
   # Evaluate
   if (eval) {
     args <- lapply(exprs, function(expr) {
       eval(parse(text = expr), envir = src)
     })
   } else {
     args <- unname(mget(exprs, envir = as.environment(src)))
   }
 
   # Create the string(s)

tests/testthat/test-adjust.R

@@ -28,3 +28,13 @@ test_that("adjust regex", {
     adjust(mtcars, select = "mpg")
   )
 })
+
+# select helpers ------------------------------
+test_that("adjust, invalid column names", {
+  data(iris)
+  colnames(iris)[1] <- "I am"
+  expect_error(
+    adjust(iris[c("I am", "Species")], multilevel = FALSE, bayesian = FALSE),
+    regex = "Bad column names"
+  )
+})