Merge branch 'main' into issue_2237

Author: hadley

NEWS.md

@@ -1,11 +1,12 @@
 # testthat (development version)
 
 * `set_state_inspector()` gains `tolerance` argument and ignores minor FP differences by default (@mcol, #2237).
+* `expect_vector()` fails, instead of erroring, if `object` is not a vector (@plietar, #2224).
 * New `vignette("mocking")` explains mocking in detail (#1265).
 * New `vignette("challenging-functions")` provides an index to other documentation organised by testing challenges (#1265).
 * When running a test interactively, testthat now reports the number of succeses. The results should also be more useful if you are using nested tests.
 * The hints generated by `expect_snapshot()` and `expect_snapshot_file()` now include the path to the package, if its not in the current working directory (#1577).
-* `expect_snapshot_file()` now clearly errors if the `path` doesnt exist (#2191).
+* `expect_snapshot_file()` now clearly errors if the `path` doesn't exist (#2191).
 * `expect_snapshot_file()` now considers `.json` to be a text file (#1593).
 * `expect_snapshot_file()` now shows differences for text files (#1593).
 * The failure messages for all `expect_` functions have been rewritten to first state what was expected and then what was actually received (#2142).
@@ -20,7 +21,7 @@
 * `local_mock()` and `with_mock()` have been deprecated because they are no longer permitted in R 4.5.
 * `snapshot_review()` now passes `...` on to `shiny::runApp()` (#1928).
 * `expect_named()` now gives more informative errors (#2091).
-* `expect_*()` functions consistently and rigorously check their inputs (#1754). 
+* `expect_*()` functions consistently and rigorously check their inputs (#1754).
 * `test_that()` no longer warns about the absence of `{}` since it no longer seems to be necessary.
 * `test_that()`, `describe()`, and `it()` can now be arbitrarily nested. Each component will skip only if it and its subtests don't contain any expectations. The interactive stop reporter has been fixed so it doesn't duplicate failures. (#2063, #2188).
 * Test filtering now works with `it()`, and the `desc` argument can take a character vector in order to recursively filter subtests (i.e. `it()` nested inside of `describe()`) (#2118).
@@ -39,7 +40,7 @@
 * New `expect_r6_class()` (#2030).
 * `expect_*()` functions consistently and rigorously check their inputs (#1754).
 * `JunitReporter()` no longer fails with `"no applicable method for xml_add_child"` for warnings outside of tests (#1913). Additionally, warnings now save their backtraces.
-* `JunitReporter()` strips ANSI escapes in more placese (#1852, #2032).
+* `JunitReporter()` strips ANSI escapes in more places (#1852, #2032).
 * `try_again()` is now publicised. The first argument is now the number of retries, not tries (#2050).
 * `vignette("custom-expectations)` has been overhauled to make it much clearer how to create high-quality expectations (#2113, #2132, #2072).
 * `expect_snapshot()` and friends will now fail when creating a new snapshot on CI. This is usually a signal that you've forgotten to run it locally before committing (#1461).
@@ -988,7 +989,7 @@ This release mostly focusses on an overhaul of how testthat works with condition
 * In `expect_equal_to_reference()`, the default value for `update` is
   now `FALSE` (@BrodieG, #683).
 
-* `expect_error()` now returns the error object as documentated (#724).
+* `expect_error()` now returns the error object as documented (#724).
   It also now warns if you're using a classed expectation and you're
   not using the `class` argument. This is good practice as it decouples the
   error object (which tends to be stable) from its rendering to the user
@@ -1680,7 +1681,7 @@ The reporters system class has been considerably refactored to make existing rep
 * `safe_digest()` uses a better strategy, and returns NA for directories
   (#138, #146).
 
-* Random praise is renabled by default (again!) (#164).
+* Random praise is re-enabled by default (again!) (#164).
 
 * Teamcity reporter now correctly escapes output messages (#150, @windelinckx).
   It also uses nested suites to include test names.

R/expect-self-test.R

@@ -39,7 +39,7 @@ capture_success_failure <- function(expr) {
 #' Use `show_failure()` in examples to print the failure message without
 #' throwing an error.
 #'
-#' @param expr Code to evalute
+#' @param expr Code to evaluate
 #' @param message Check that the failure message matches this regexp.
 #' @param ... Other arguments passed on to [expect_match()].
 #' @export

R/expect-setequal.R

@@ -28,8 +28,8 @@ expect_setequal <- function(object, expected) {
   act <- quasi_label(enquo(object))
   exp <- quasi_label(enquo(expected))
 
-  check_vector(object)
-  check_vector(expected)
+  check_vector(act$val, error_arg = "object")
+  check_vector(exp$val, error_arg = "expected")
   if (!is.null(names(act$val)) && !is.null(names(exp$val))) {
     testthat_warn("expect_setequal() ignores names")
   }

R/expect-that.R

@@ -15,7 +15,7 @@
 #' @param srcref Location of the failure. Should only needed to be explicitly
 #'   supplied when you need to forward a srcref captured elsewhere.
 #' @param trace_env If `trace` is not specified, this is used to generate an
-#'   informative traceack for failures. You should only need to set this if
+#'   informative traceback for failures. You should only need to set this if
 #'   you're calling `fail()` from a helper function; see
 #'   `vignette("custom-expectation")` for details.
 #' @param trace An optional backtrace created by [rlang::trace_back()].

R/expect-vector.R

@@ -17,19 +17,20 @@
 expect_vector <- function(object, ptype = NULL, size = NULL) {
   check_installed("vctrs")
   check_number_whole(size, min = 0, allow_null = TRUE)
-
   act <- quasi_label(enquo(object))
+  # vec_assert() automatically adds backticks so we hack out the ones
+  # added by as_label()
+  act$lab <- gsub("^`|`$", "", act$lab)
 
-  message <- NULL
-  tryCatch(
+  withCallingHandlers(
     vctrs::vec_assert(act$val, ptype = ptype, size = size, arg = act$lab),
+    vctrs_error_scalar_type = function(e) {
+      fail(e$message)
+    },
     vctrs_error_assert = function(e) {
-      message <<- e$message
+      fail(e$message)
     }
   )
 
-  if (!is.null(message)) {
-    return(fail(message))
-  }
   pass(act$val)
 }

R/mock2.R

@@ -100,7 +100,7 @@
 #'
 #' To mock a function that returns different values in sequence,
 #' for instance an API call whose status would be 502 then 200,
-#' or an user intput to `readline()`, you can use [mock_output_sequence()]
+#' or an user input to `readline()`, you can use [mock_output_sequence()]
 #'
 #' ```R
 #' local_mocked_bindings(readline = mock_output_sequence("3", "This is a note", "n"))

R/quasi-label.R

@@ -59,10 +59,11 @@ quasi_label <- function(quo, label = NULL, arg = NULL) {
 }
 
 labelled_value <- function(value, label) {
-  list(
-    val = value,
-    lab = label
-  )
+  if (missing(value)) {
+    list(val = missing_arg(), lab = label)
+  } else {
+    list(val = value, lab = label)
+  }
 }
 
 quasi_capture <- function(.quo, .label, .capture, ...) {

R/reporter-list.R

@@ -11,50 +11,52 @@ ListReporter <- R6::R6Class(
   "ListReporter",
   inherit = Reporter,
   public = list(
-    current_start_time = NA,
-    current_expectations = NULL,
-    current_file = NULL,
-    current_context = NULL,
-    current_test = NULL,
+    running = NULL,
+    current_file = "", # so we can still subset with this
     results = NULL,
 
     initialize = function() {
       super$initialize()
       self$capabilities$parallel_support <- TRUE
+      self$capabilities$parallel_updates <- TRUE
       self$results <- Stack$new()
+      self$running <- new.env(parent = emptyenv())
     },
 
     start_test = function(context, test) {
+      # is this a new test block?
       if (
-        !identical(self$current_context, context) ||
-          !identical(self$current_test, test)
+        !identical(self$running[[self$current_file]]$context, context) ||
+          !identical(self$running[[self$current_file]]$test, test)
       ) {
-        self$current_context <- context
-        self$current_test <- test
-        self$current_expectations <- Stack$new()
-        self$current_start_time <- proc.time()
+        self$running[[self$current_file]]$context <- context
+        self$running[[self$current_file]]$test <- test
+        self$running[[self$current_file]]$expectations <- Stack$new()
+        self$running[[self$current_file]]$start_time <- proc.time()
       }
     },
 
     add_result = function(context, test, result) {
-      if (is.null(self$current_expectations)) {
+      if (is.null(self$running[[self$current_file]]$expectations)) {
         # we received a result outside of a test:
         # could be a bare expectation or an exception/error
         if (!inherits(result, 'error')) {
           return()
         }
-        self$current_expectations <- Stack$new()
+        self$running[[self$current_file]]$expectations <- Stack$new()
       }
 
-      self$current_expectations$push(result)
+      self$running[[self$current_file]]$expectations$push(result)
     },
 
     end_test = function(context, test) {
-      elapsed <- as.double(proc.time() - self$current_start_time)
+      elapsed <- as.double(
+        proc.time() - self$running[[self$current_file]]$start_time
+      )
 
       results <- list()
-      if (!is.null(self$current_expectations)) {
-        results <- self$current_expectations$as_list()
+      if (!is.null(self$running[[self$current_file]]$expectations)) {
+        results <- self$running[[self$current_file]]$expectations$as_list()
       }
 
       self$results$push(list(
@@ -67,28 +69,39 @@ ListReporter <- R6::R6Class(
         results = results
       ))
 
-      self$current_expectations <- NULL
+      self$running[[self$current_file]]$expectations <- NULL
     },
 
     start_file = function(name) {
+      if (!name %in% names(self$running)) {
+        newfile <- list(
+          start_time = NA,
+          expectations = NULL,
+          context = NULL,
+          test = NULL
+        )
+        assign(name, newfile, envir = self$running)
+      }
       self$current_file <- name
     },
 
     end_file = function() {
       # fallback in case we have errors but no expectations
       self$end_context(self$current_file)
+      rm(list = self$current_file, envir = self$running)
     },
 
     end_context = function(context) {
-      results <- self$current_expectations
+      results <- self$running[[self$current_file]]$expectations
       if (is.null(results)) {
         return()
       }
 
-      self$current_expectations <- NULL
+      self$running[[self$current_file]]$expectations <- NULL
 
       # look for exceptions raised outside of tests
-      # they happened just before end_context since they interrupt the test_file execution
+      # they happened just before end_context since they interrupt the test_
+      # file execution
       results <- results$as_list()
       if (length(results) == 0) {
         return()

R/reporter-stop.R

@@ -2,7 +2,7 @@
 #'
 #' @description
 #' The default reporter used when [expect_that()] is run interactively.
-#' It responds by displaying a summary of the number of successes and faiures
+#' It responds by displaying a summary of the number of successes and failures
 #' and [stop()]ping on if there are any failures.
 #'
 #' @export

R/test-compiled-code.R

@@ -230,8 +230,8 @@ run_cpp_tests <- function(package) {
 #' \strong{Function} \tab \strong{Catch} \tab \strong{Description} \cr
 #' `context` \tab `CATCH_TEST_CASE` \tab The context of a set of tests. \cr
 #' `test_that` \tab `CATCH_SECTION` \tab A test section. \cr
-#' `expect_true` \tab `CATCH_CHECK` \tab Test that an expression evaluates to `true`. \cr
-#' `expect_false` \tab `CATCH_CHECK_FALSE` \tab Test that an expression evalutes to `false`. \cr
+#' `expect_true` \tab `CATCH_CHECK` \tab Test that an expression evaluates to `TRUE`. \cr
+#' `expect_false` \tab `CATCH_CHECK_FALSE` \tab Test that an expression evaluates to `FALSE`. \cr
 #' `expect_error` \tab `CATCH_CHECK_THROWS` \tab Test that evaluation of an expression throws an exception. \cr
 #' `expect_error_as` \tab `CATCH_CHECK_THROWS_AS` \tab Test that evaluation of an expression throws an exception of a specific class. \cr
 #' }