To reproduce:
R --vanilla -s -e 'x <- 42; package.skeleton("use.tinytest")'
cd use.tinytest
R --vanilla -s -e 'tinytest::setup_tinytest(getwd())'
echo "stop()" > inst/tinytest/test_aaa.R
R --vanilla -s -e 'tinytest::run_test_dir()'
I see:
$ R --vanilla -s -e 'tinytest::run_test_dir()'
Error in eval(expr, envir = e) :
Calls: <Anonymous> -> lapply -> FUN -> eval -> eval
Execution halted
Could tinytest instead capture errors seen when executing test files, and report those and allow other tests to continue executing? (Or at least, report what file the error was caught in, or other diagnostic information?)