fish-shell/tests/test_driver.sh

89 lines
3.4 KiB
Bash

# vim: set ts=4 sw=4 tw=100 et:
# POSIX sh test driver to reduce dependency on fish in tests.
# Executes the specified *fish script* with the provided arguments, after setting up a clean test
# environment (see `test_env.sh`) and then importing the fish-related helper functions and
# performing some state initialization required by the various fish tests. Each payload script is
# executed in its own environment, this script waits for fish to exit then cleans up the target
# environment and bubbles up the fish exit code.
# macOS has really weird default IFS behavior that splits output in random places, and the trailing
# backspace is to prevent \n from being gobbled up by the subshell output substitution.
# Folks, this is why you should use fish!
IFS="$(printf "\n\b")"
# The first argument is the path to the script to launch; all remaining arguments are forwarded to
# the script.
fish_script="$1"
shift 1
script_args="${@}"
# Prevent $@ from persisting to sourced commands
shift $# 2>/dev/null
die() {
if test "$#" -ge 0; then
printf "%s\n" "$@" 1>&2
fi
exit 1
}
# To keep things sane and to make error messages comprehensible, do not use relative paths anywhere
# in this script. Instead, make all paths relative to one of these or the new $HOME ($homedir)."
TESTS_ROOT="$(dirname "$0")"
BUILD_ROOT="${TESTS_ROOT}/.."
# macOS (still!) doesn't have `readlink -f` or `realpath`. That's OK, this is just for aesthetics.
if command -v realpath 1>/dev/null 2>/dev/null; then
TESTS_ROOT="$(realpath --no-symlinks "${TESTS_ROOT}")"
BUILD_ROOT="$(realpath --no-symlinks "${BUILD_ROOT}")"
fi
if ! test -z "$__fish_is_running_tests"; then
echo "Recursive test invocation detected!" 1>&2
exit 10
fi
# Set up the test environment. Does not change the current working directory.
. ${TESTS_ROOT}/test_env.sh
# These are used read-only so it's OK to symlink instead of copy
rm -f "$XDG_CONFIG_HOME/fish/functions"
ln -s "$PWD/test_functions" "$XDG_CONFIG_HOME/fish/functions" || die "Failed to symlink"
# Set the function path at startup, referencing the default fish functions and the test-specific
# functions.
fish_init_cmd="set fish_function_path ${XDG_CONFIG_HOME}/fish/functions ${BUILD_ROOT}/share/functions"
__fish_is_running_tests="$homedir"
export __fish_is_running_tests
# Set a marker to indicate whether colored output should be suppressed (used in `test_util.fish`)
suppress_color=""
if ! tty 0>&1 > /dev/null; then
suppress_color="yes"
fi
export suppress_color
# Source test util functions at startup
fish_init_cmd="${fish_init_cmd} && source ${TESTS_ROOT}/test_util.fish";
# Run the test script, but don't exec so we can clean up after it succeeds/fails. Each test is
# launched directly within its TMPDIR, so that the fish tests themselves do not need to refer to
# TMPDIR (to ensure their output as displayed in case of failure by littlecheck is reproducible).
(cd $TMPDIR; env HOME="$homedir" "${BUILD_ROOT}/test/root/bin/fish" \
--init-command "${fish_init_cmd}" \
"$fish_script" "$script_args")
test_status="$?"
# CMake less than 3.9.0 "fully supports" setting an exit code to denote a skipped test, but then
# it just goes ahead and reports them as failed anyway. Really?
if test -n $CMAKE_SKIPPED_HACK; then
if test $test_status -eq 125; then
echo "Overriding SKIPPED return code from test" 1>&2
test_status=0
fi
fi
rm -rf "$homedir"
exit "$test_status"