improve code quality and fix signal handling

tests/performance.bats

@@ -0,0 +1,288 @@
+#!/usr/bin/env bats
+# Performance benchmark tests for Mole optimizations
+# Tests the performance improvements introduced in V1.14.0+
+
+setup_file() {
+    PROJECT_ROOT="$(cd "${BATS_TEST_DIRNAME}/.." && pwd)"
+    export PROJECT_ROOT
+
+    # Create test data directory
+    TEST_DATA_DIR="$(mktemp -d "${BATS_TEST_DIRNAME}/tmp-perf.XXXXXX")"
+    export TEST_DATA_DIR
+}
+
+teardown_file() {
+    rm -rf "$TEST_DATA_DIR"
+}
+
+setup() {
+    source "$PROJECT_ROOT/lib/core/base.sh"
+}
+
+# ============================================================================
+# bytes_to_human Performance Tests
+# ============================================================================
+
+@test "bytes_to_human handles large values efficiently" {
+    local start end elapsed
+
+    # Warm up
+    bytes_to_human 1073741824 > /dev/null
+
+    # Benchmark: 1000 iterations should complete in < 2 seconds (relaxed threshold)
+    start=$(date +%s%N)
+    for i in {1..1000}; do
+        bytes_to_human 1073741824 > /dev/null
+    done
+    end=$(date +%s%N)
+
+    elapsed=$(( (end - start) / 1000000 )) # Convert to milliseconds
+
+    # Should complete in less than 2000ms (2 seconds)
+    [ "$elapsed" -lt 2000 ]
+}
+
+@test "bytes_to_human produces correct output for GB range" {
+    result=$(bytes_to_human 1073741824)
+    [ "$result" = "1.00GB" ]
+
+    result=$(bytes_to_human 5368709120)
+    [ "$result" = "5.00GB" ]
+}
+
+@test "bytes_to_human produces correct output for MB range" {
+    result=$(bytes_to_human 1048576)
+    [ "$result" = "1.0MB" ]
+
+    result=$(bytes_to_human 104857600)
+    [ "$result" = "100.0MB" ]
+}
+
+@test "bytes_to_human produces correct output for KB range" {
+    result=$(bytes_to_human 1024)
+    [ "$result" = "1KB" ]
+
+    result=$(bytes_to_human 10240)
+    [ "$result" = "10KB" ]
+}
+
+@test "bytes_to_human handles edge cases" {
+    # Zero bytes
+    result=$(bytes_to_human 0)
+    [ "$result" = "0B" ]
+
+    # Invalid input returns 0B (with error code 1)
+    run bytes_to_human "invalid"
+    [ "$status" -eq 1 ]
+    [ "$output" = "0B" ]
+
+    # Negative should also fail validation
+    run bytes_to_human "-100"
+    [ "$status" -eq 1 ]
+    [ "$output" = "0B" ]
+}
+
+# ============================================================================
+# BSD Stat Wrapper Performance Tests
+# ============================================================================
+
+@test "get_file_size is faster than multiple stat calls" {
+    # Create test file
+    local test_file="$TEST_DATA_DIR/size_test.txt"
+    dd if=/dev/zero of="$test_file" bs=1024 count=100 2> /dev/null
+
+    # Benchmark: 100 calls should complete quickly
+    local start end elapsed
+    start=$(date +%s%N)
+    for i in {1..100}; do
+        get_file_size "$test_file" > /dev/null
+    done
+    end=$(date +%s%N)
+
+    elapsed=$(( (end - start) / 1000000 ))
+
+    # Should complete in less than 1000ms (relaxed threshold)
+    [ "$elapsed" -lt 1000 ]
+}
+
+@test "get_file_mtime returns valid timestamp" {
+    local test_file="$TEST_DATA_DIR/mtime_test.txt"
+    touch "$test_file"
+
+    result=$(get_file_mtime "$test_file")
+
+    # Should be a valid epoch timestamp (10 digits)
+    [[ "$result" =~ ^[0-9]{10,}$ ]]
+}
+
+@test "get_file_owner returns current user for owned files" {
+    local test_file="$TEST_DATA_DIR/owner_test.txt"
+    touch "$test_file"
+
+    result=$(get_file_owner "$test_file")
+    current_user=$(whoami)
+
+    [ "$result" = "$current_user" ]
+}
+
+# ============================================================================
+# User Context Detection Performance Tests
+# ============================================================================
+
+@test "get_invoking_user executes quickly" {
+    local start end elapsed
+
+    start=$(date +%s%N)
+    for i in {1..100}; do
+        get_invoking_user > /dev/null
+    done
+    end=$(date +%s%N)
+
+    elapsed=$(( (end - start) / 1000000 ))
+
+    # Should complete in less than 200ms
+    [ "$elapsed" -lt 200 ]
+}
+
+@test "get_darwin_major caches correctly" {
+    # Multiple calls should return same result
+    local first second
+    first=$(get_darwin_major)
+    second=$(get_darwin_major)
+
+    [ "$first" = "$second" ]
+    [[ "$first" =~ ^[0-9]+$ ]]
+}
+
+# ============================================================================
+# Temporary File Management Performance Tests
+# ============================================================================
+
+@test "create_temp_file and cleanup_temp_files work efficiently" {
+    local start end elapsed
+
+    # Ensure MOLE_TEMP_DIRS is initialized (base.sh should do this)
+    declare -a MOLE_TEMP_DIRS=()
+
+    # Create 50 temp files (reduced from 100 for faster testing)
+    start=$(date +%s%N)
+    for i in {1..50}; do
+        create_temp_file > /dev/null
+    done
+    end=$(date +%s%N)
+
+    elapsed=$(( (end - start) / 1000000 ))
+
+    # Should complete in less than 1000ms
+    [ "$elapsed" -lt 1000 ]
+
+    # Verify temp files were tracked
+    [ "${#MOLE_TEMP_FILES[@]}" -eq 50 ]
+
+    # Cleanup should also be reasonably fast
+    start=$(date +%s%N)
+    cleanup_temp_files
+    end=$(date +%s%N)
+
+    elapsed=$(( (end - start) / 1000000 ))
+    # Relaxed threshold: should complete within 2 seconds
+    [ "$elapsed" -lt 2000 ]
+
+    # Verify cleanup
+    [ "${#MOLE_TEMP_FILES[@]}" -eq 0 ]
+}
+
+@test "mktemp_file creates files with correct prefix" {
+    local temp_file
+    temp_file=$(mktemp_file "test_prefix")
+
+    # Should contain prefix
+    [[ "$temp_file" =~ test_prefix ]]
+
+    # Should exist
+    [ -f "$temp_file" ]
+
+    # Cleanup
+    rm -f "$temp_file"
+}
+
+# ============================================================================
+# Brand Name Lookup Performance Tests
+# ============================================================================
+
+@test "get_brand_name handles common apps efficiently" {
+    local start end elapsed
+
+    # Warm up (first call includes defaults read which is slow)
+    get_brand_name "wechat" > /dev/null
+
+    # Benchmark: 50 lookups (reduced from 100)
+    start=$(date +%s%N)
+    for i in {1..50}; do
+        get_brand_name "wechat" > /dev/null
+        get_brand_name "QQ" > /dev/null
+        get_brand_name "dingtalk" > /dev/null
+    done
+    end=$(date +%s%N)
+
+    elapsed=$(( (end - start) / 1000000 ))
+
+    # Relaxed threshold: defaults read is called multiple times
+    # Should complete within 5 seconds on most systems
+    [ "$elapsed" -lt 5000 ]
+}
+
+@test "get_brand_name returns correct localized names" {
+    # Test should work regardless of system language
+    local result
+    result=$(get_brand_name "wechat")
+
+    # Should return either "WeChat" or "微信"
+    [[ "$result" == "WeChat" || "$result" == "微信" ]]
+}
+
+# ============================================================================
+# Parallel Job Calculation Tests
+# ============================================================================
+
+@test "get_optimal_parallel_jobs returns sensible values" {
+    local result
+
+    # Default mode
+    result=$(get_optimal_parallel_jobs)
+    [[ "$result" =~ ^[0-9]+$ ]]
+    [ "$result" -gt 0 ]
+    [ "$result" -le 128 ]
+
+    # Scan mode (should be higher)
+    local scan_jobs
+    scan_jobs=$(get_optimal_parallel_jobs "scan")
+    [ "$scan_jobs" -gt "$result" ]
+
+    # Compute mode (should be lower)
+    local compute_jobs
+    compute_jobs=$(get_optimal_parallel_jobs "compute")
+    [ "$compute_jobs" -le "$scan_jobs" ]
+}
+
+# ============================================================================
+# Section Tracking Performance Tests
+# ============================================================================
+
+@test "section tracking has minimal overhead" {
+    local start end elapsed
+
+    # Warm up
+    note_activity
+
+    start=$(date +%s%N)
+    for i in {1..1000}; do
+        note_activity
+    done
+    end=$(date +%s%N)
+
+    elapsed=$(( (end - start) / 1000000 ))
+
+    # Should complete in less than 1000ms (relaxed threshold)
+    [ "$elapsed" -lt 1000 ]
+}