mirror of
https://github.com/kuhyx/testsAndMisc.git
synced 2026-11-11 21:14:01 +01:00
Krzysztof kuhy Rudnicki
d42d8ff20d
Gives linux_configuration/scripts/lib a real test suite, which takes all nine of its libraries off meta/shell-coverage-allowlist.txt at once — coverage is per-directory, so this is the highest-ratio first batch. 139 assertions across four suites, all offline: the mtk toolkit is driven through its existing MTK_ROOT_FIXTURE seam, and adb is shimmed onto a narrowed PATH. Nothing reads the wall clock, this host's udev rules, or a real device, because the suite runs in shell-tests.yml and again on every pre-push via ci-mirror — a host-dependent assertion would be a flake that blocks unrelated pushes. The gate only checks that tests/run_all.sh EXISTS, so an empty one would mark all nine libraries covered while testing nothing. Each of the four load-bearing rules was therefore mutation-tested and the suite required to go red: classify_device's && -> ||, discover_partitions' init_boot/boot ordering, mtk_select_device's refusal to guess between two phones, and is_hour_in_range's 10#$ octal guard. That last one initially SURVIVED. Dropping 10#$ makes bash abort with "value too great for base", a non-zero exit indistinguishable from an honest out-of-range result, so the status-only assertion passed either way. It now asserts stderr is empty. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
120 lines
4.0 KiB
Bash
Executable File
120 lines
4.0 KiB
Bash
Executable File
#!/usr/bin/env bash
|
|
# lib/tests/mtk_harness.sh — shared fixture for linux_configuration/scripts/lib.
|
|
#
|
|
# Sourced, not executed.
|
|
#
|
|
# WHAT IS EXERCISED
|
|
# The pure decision logic and everything reachable through the
|
|
# MTK_ROOT_FIXTURE seam that mtk_common.sh documents as its only boundary
|
|
# with real hardware: classify_device, discover_partitions,
|
|
# mtk_ramdisk_carrier, mtk_check_layout_expectation, mtk_assert_mediatek,
|
|
# mtk_sanitize, mtk_sane_ramdisk_size, mtk_select_device/require_authorized
|
|
# under a fixture, plus the string helpers in common_datetime.sh.
|
|
#
|
|
# WHAT IS DELIBERATELY NOT EXERCISED, AND WHY
|
|
# The coverage gate only checks that this directory HAS a run_all.sh, so it
|
|
# cannot tell a real suite from an empty one. Naming the gaps is therefore
|
|
# the only honest record of them:
|
|
# - require_hosts_readable / init_android_script — `exec sudo -E bash "$0"`
|
|
# would replace the test process with a sudo re-exec of the suite.
|
|
# - check_adb_device / check_adb_root — call die(), which exits 1 and
|
|
# would take the whole run down; both also need real hardware.
|
|
# - collapse_mounts — calls `umount -l` on a live path.
|
|
# - install_missing_pacman_packages / require_imagemagick / notify —
|
|
# effecting code (pacman, desktop notifications).
|
|
# - mtk_udev_catchall_files — globs /etc/udev/rules.d with no seam, so it
|
|
# reads this host's real rules; the result differs between here and CI.
|
|
# - mtk_sha256 / mtk_file_size — one-line wrappers over sha256sum/stat.
|
|
# This matches Phase 1's measured pattern: decision logic reaches ~90%, the
|
|
# effecting code reaches 0%.
|
|
#
|
|
# TIME AND HOST DEPENDENCE
|
|
# This suite runs in shell-tests.yml and again on every pre-push via
|
|
# ci-mirror, so a wall-clock assertion would pass locally and fail at some
|
|
# other hour on some other machine. get_hour and get_day_of_week are
|
|
# therefore shimmed where a specific value is needed, and the readers of
|
|
# /proc/uptime are asserted on SHAPE, never on value.
|
|
set -euo pipefail
|
|
|
|
PASS=0
|
|
FAIL=0
|
|
|
|
_t_ok() {
|
|
PASS=$((PASS + 1))
|
|
printf ' OK: %s\n' "$1"
|
|
}
|
|
|
|
_t_bad() {
|
|
FAIL=$((FAIL + 1))
|
|
printf ' FAIL: %s\n' "$1"
|
|
}
|
|
|
|
# Assert equality. Every check in this suite funnels through here or _t_rc so
|
|
# there is exactly one place that counts a result.
|
|
_t_is() {
|
|
if [[ $2 == "$1" ]]; then
|
|
_t_ok "$3"
|
|
else
|
|
_t_bad "$3 (want '${1}', got '${2}')"
|
|
fi
|
|
}
|
|
|
|
# Assert a command's exit status. Run under `if` so a non-zero return is data
|
|
# rather than a `set -e` abort.
|
|
_t_rc() {
|
|
local want="$1" what="$2"
|
|
shift 2
|
|
local got=0
|
|
"$@" >/dev/null 2>&1 || got=$?
|
|
_t_is "$want" "$got" "$what"
|
|
}
|
|
|
|
# Assert a substring, for messages whose exact wording is not the contract.
|
|
_t_in() {
|
|
if [[ $2 == *"$1"* ]]; then
|
|
_t_ok "$3"
|
|
else
|
|
_t_bad "$3 (want substring '${1}')"
|
|
fi
|
|
}
|
|
|
|
# ---------------------------------------------------------------------------
|
|
# Isolation
|
|
# ---------------------------------------------------------------------------
|
|
|
|
# android.sh runs `ensure_dir "$ANDROID_WORK_DIR"` at SOURCE time, and that
|
|
# path is under $HOME. Redirect HOME before sourcing anything or merely
|
|
# loading the library writes to the real ~/.cache. MTK_ROOT_CACHE moves the
|
|
# mtk toolkit's artifacts for the same reason.
|
|
_t_sandbox() {
|
|
T_TMP="$(mktemp -d)"
|
|
export T_TMP
|
|
HOME="$T_TMP/home"
|
|
mkdir -p "$HOME"
|
|
export HOME
|
|
export MTK_ROOT_CACHE="$T_TMP/cache"
|
|
}
|
|
|
|
_t_cleanup() {
|
|
[[ -n ${T_TMP:-} && -d ${T_TMP:-} ]] && rm -rf "$T_TMP"
|
|
return 0
|
|
}
|
|
|
|
# Build a device fixture. mtk_common.sh reads props.txt as `key=value` lines
|
|
# and by-name.txt as one partition per line; those two files ARE the seam.
|
|
# Usage: _t_fixture <name> <props-text> <by-name-text>
|
|
_t_fixture() {
|
|
local dir="$T_TMP/fixtures/$1"
|
|
mkdir -p "$dir"
|
|
printf '%s\n' "$2" >"$dir/props.txt"
|
|
printf '%s\n' "$3" >"$dir/by-name.txt"
|
|
printf '%s' "$dir"
|
|
}
|
|
|
|
# Report and set the process exit status. Called at the end of every suite.
|
|
_t_done() {
|
|
printf '\n %s: %d passed, %d failed\n' "$1" "$PASS" "$FAIL"
|
|
_t_cleanup
|
|
[[ $FAIL -eq 0 ]]
|
|
}
|