Handle interrupts in tests

I don't see any way to test this programmatically, so I've just been adding `Sys.sleep(10)` to a random test.

Fixes #1464

Author: hadley

NEWS.md

@@ -1,5 +1,6 @@
 # testthat (development version)
 
+* Interrupting a test now prints the test name. This makes it easier to tell where a very slow test might be hanging (#1464)
 * `expect_snapshot_value()` can now handle expressions that generate `-` (#1678) or zero length atomic vectors (#2042).
 * `expect_matches()` failures should be a little easier to read (#2135).
 * New `local_on_cran(TRUE)` allows you to simulate how your tests will run on CRAN (#2112).

R/test-that.R

@@ -178,6 +178,13 @@ test_code <- function(test, code, env, reporter, skip_on_empty = TRUE) {
     signalCondition(e)
   }
 
+  handle_interrupt <- function(e) {
+    if (!is.null(test)) {
+      cat("\n")
+      cli::cli_inform(c("!" = "Interrupting test: {test}"))
+    }
+  }
+
   test_env <- new.env(parent = env)
   old <- options(rlang_trace_top_env = test_env)[[1]]
   on.exit(options(rlang_trace_top_env = old), add = TRUE)
@@ -197,7 +204,8 @@ test_code <- function(test, code, env, reporter, skip_on_empty = TRUE) {
       skip = handle_skip,
       warning = handle_warning,
       message = handle_message,
-      error = handle_error
+      error = handle_error,
+      interrupt = handle_interrupt
     ),
     # some errors may need handling here, e.g., stack overflow
     error = handle_fatal,