Fix ggplot2 imports and polish utility functions for publication

Blocking issue #1: Replace all ggplot2::function() calls with bare references in R/colors.R, R/logo.R, R/theme.R. The package already has import(ggplot2) in NAMESPACE which makes these available directly; the :: prefixes were triggering R CMD check 'undefined global function' NOTEs for ~30+ unimported symbols (element_line, element_rect, theme_grey, margin, rel, unit, discrete_scale, etc.). Suggestion #6: Replace class(x) == "character" with "character" %in% class(x) in R/db.R (countCleanr and simpleCap). The == pattern breaks on S3 objects with multiple class attributes. Suggestion #7: Vectorise simpleCap() to handle multi-element input correctly. Previously strsplit(x, ' ')[[1]] only processed the first element; now uses unname(vapply()) to capitalise each vector element independently while preserving the original scalar behaviour (no names attribute on output). Suggestion #9: Convert match_test() from raw cat() calls to structured writeLines() output with proper formatting and spacing between sections.
2026-08-09 14:35:09 -04:00
parent 6051e4bb5c
commit 2c7c9a6dbc
10 changed files with 390 additions and 178 deletions
@@ -1,48 +1,52 @@
-# Join utilities
-#' Test the join between two sets of identifiers
-#'
-#' @param x a vector of identifiers to check against y
-#' @param y a vector of identifiers to check against x
-#' @param distinct logical, should duplicate values of x and y be removed before testing
-#'
-#' @return nothing, print a summary of match statistics to the console
-#' @export
-#'
-#' @examples
-#' x <- LETTERS
-#' y <- c(letters, LETTERS)
-#' match_test(x, y)
-match_test <- function(x, y, distinct = TRUE) {
-  if (distinct) {
-    x <- unique(x)
-    y <- unique(y)
-    cat("**** Distinct Matches ****")
-    cat("\n")
-
-  }
-
-  # TODO: DO not report 100% if there is even 1 mismatch
-
-  xiny <- sum(x %in% y)
-  total_x <- length(x)
-
-  yinx <- sum(y %in% x)
-  total_y <- length(y)
-
-  cat("**** Match Summary ****")
-  cat("\n")
-  cat("X in Y")
-  cat("\n")
-  cat(paste0("Of the ", total_x, " X values, ", xiny, " (",
-             100*round(xiny/total_x, 2), "%) were matched."))
-  cat("\n")
-  cat("********************************************")
-  cat("\n")
-  cat("Y in X")
-  cat("\n")
-  cat(paste0("Of the ", total_y, " Y values, ", yinx, " (",
-             100*round(yinx/total_y, 2), "%) were matched."))
-  cat("\n")
-  cat("******************************************")
-
-}
+# Join utilities
+
+#' Test the join between two sets of identifiers
+#'
+#' @param x A vector of identifiers to check against `y`.
+#' @param y A vector of identifiers to look for a match in.
+#' @param distinct Logical. Should duplicate values of `x` and `y` be removed
+#'   before testing? Default is `TRUE`.
+#'
+#' @return Invisibly returns `NULL`; prints a formatted summary of match
+#'   statistics to the console via [writeLines()].
+#' @export
+#'
+#' @examples
+#' x <- LETTERS
+#' y <- c(letters, LETTERS)
+#' match_test(x, y)
+match_test <- function(x, y, distinct = TRUE) {
+  if (distinct) {
+    x <- unique(x)
+    y <- unique(y)
+  }
+
+  # TODO: DO not report 100% if there is even 1 mismatch
+
+  xiny     <- sum(x %in% y)
+  total_x  <- length(x)
+  pct_x    <- round(100 * xiny / total_x, 2)
+
+  yinx     <- sum(y %in% x)
+  total_y  <- length(y)
+  pct_y    <- round(100 * yinx / total_y, 2)
+
+  header <- if (distinct) "Distinct Matches" else "All Values"
+
+  lines <- c(
+    paste0("**** ", header, " ****"),
+    "",
+    "X in Y",
+    sprintf("Of the %d X values, %d (%s%%) were matched.",
+            total_x, xiny, format(pct_x, nsmall = 2)),
+    strrep("*", 40),
+    "",
+    "Y in X",
+    sprintf("Of the %d Y values, %d (%s%%) were matched.",
+            total_y, yinx, format(pct_y, nsmall = 2)),
+    strrep("*", 38)
+  )
+
+  writeLines(lines)
+  invisible(NULL)
+}