Merge pull request #540 from lorenzwalthert/enfore-blank-at-EOF

- Ensure styled files also have EOF line break if no tokens were changed (#540).

Author: lorenzwalthert

R/io.R

@@ -15,10 +15,12 @@ transform_utf8 <- function(path, fun, write_back = TRUE) {
 
 #' @importFrom rlang with_handlers warn
 transform_utf8_one <- function(path, fun, write_back) {
-  old <- xfun::read_utf8(path)
   with_handlers({
-    new <- fun(old)
-    identical <- identical(unclass(old), unclass(new))
+    file_with_info <- read_utf8(path)
+    # only write back when changed OR when there was a missing newline
+    new <- fun(file_with_info$text)
+    identical_content <- identical(unclass(file_with_info$text), unclass(new))
+    identical <- identical_content && !file_with_info$missing_EOF_line_break
     if (!identical && write_back) {
       xfun::write_utf8(new, path)
     }
@@ -28,3 +30,58 @@ transform_utf8_one <- function(path, fun, write_back) {
     NA
   })
 }
+
+#' Read UTF-8
+#'
+#' Reads an UTF-8 file, returning the content and whether or not the final line
+#' was blank. This information is required higher up in the callstack because
+#' we should write back if contents changed or if there is no blank line at the
+#' EOF. A perfectly styled file with no EOF blank line will gain such a line
+#' with this implementation.
+#' @param path A path to a file to read.
+#' @keywords internal
+read_utf8 <- function(path) {
+  out <- rlang::with_handlers(
+    read_utf8_bare(path),
+    warning = function(w) w,
+    error = function(e) e
+  )
+  if (inherits(out, "character")) {
+    list(
+      text = out,
+      missing_EOF_line_break = FALSE
+    )
+  } else if (inherits(out, "error")) {
+    rlang::abort(out$message)
+  } else if (inherits(out, "warning")) {
+    list(
+      text = read_utf8_bare(path, warn = FALSE),
+      missing_EOF_line_break = grepl("incomplete", out$message)
+    )
+  }
+}
+
+#' Drop-in replacement for [xfun::read_utf8()], with an optional `warn`
+#' argument.
+#' @keywords internal
+read_utf8_bare <- function(con, warn = TRUE) {
+  x <- readLines(con, encoding = "UTF-8", warn = warn)
+  i <- invalid_utf8(x)
+  n <- length(i)
+  if (n > 0) {
+    stop(
+      c(
+        "The file ", con, " is not encoded in UTF-8. ",
+        "These lines contain invalid UTF-8 characters: "
+      ),
+      paste(c(head(i), if (n > 6) "..."), collapse = ", ")
+    )
+  }
+  x
+}
+
+#' Drop-in replacement for [xfun:::invalid_utf8()]
+#' @keywords internal
+invalid_utf8 <- function(x) {
+  which(!is.na(x) & is.na(iconv(x, "UTF-8", "UTF-8")))
+}

man/read_utf8.Rd

@@ -0,0 +1 @@
+x

man/read_utf8_bare.Rd

@@ -0,0 +1 @@
+x

tests/testthat/reference-objects/missing-blank-at-EOF.R

@@ -6,3 +6,15 @@ test_that("non-comment-helpers", {
   expect_equal(previous_non_comment(child, 4), 2)
   expect_equal(next_non_comment(child, 2), 4)
 })
+
+test_that("files with and without blank EOF line are read correctly", {
+  expect_known_value(
+    read_utf8(test_path("reference-objects/missing-blank-at-EOF.R")),
+    test_path("reference-objects/return-read-utf8-missing-EOF")
+  )
+
+  expect_known_value(
+    read_utf8(test_path("reference-objects/non-missing-blank-at-EOF.R")),
+    test_path("reference-objects/return-read-utf8-non-missing-EOF")
+  )
+})