bats-testing-patterns

Bats 测试模式

使用 Bats（Bash Automated Testing System）为 shell 脚本编写全面单元测试的综合指南，包括测试模式、fixture 以及生产级 shell 测试的最佳实践。

何时使用此技能

• 为 shell 脚本编写单元测试
• 为脚本实施测试驱动开发（TDD）
• 在 CI/CD 流水线中设置自动化测试
• 测试边界情况和错误条件
• 验证不同 shell 环境中的行为
• 为脚本构建可维护的测试套件
• 为复杂测试场景创建 fixture
• 测试多种 shell 方言（bash、sh、dash）

详细模式与完整示例

详细的模式文档位于 references/details.md。当上面的导航层级不够用时，请阅读该文件。

测试错误条件

#!/usr/bin/env bats

@test "Function fails with missing file" {
    run my_function "/nonexistent/file.txt"
    [ "$status" -ne 0 ]
    [[ "$output" == *"not found"* ]]
}

@test "Function fails with invalid input" {
    run my_function ""
    [ "$status" -ne 0 ]
}

@test "Function fails with permission denied" {
    touch "$TMPDIR/readonly.txt"
    chmod 000 "$TMPDIR/readonly.txt"
    run my_function "$TMPDIR/readonly.txt"
    [ "$status" -ne 0 ]
    chmod 644 "$TMPDIR/readonly.txt"  # Cleanup
}

@test "Function provides helpful error message" {
    run my_function --invalid-option
    [ "$status" -ne 0 ]
    [[ "$output" == *"Usage:"* ]]
}

使用依赖项进行测试

#!/usr/bin/env bats

setup() {
    # Check for required tools
    if ! command -v jq &>/dev/null; then
        skip "jq is not installed"
    fi

    export SCRIPT="${BATS_TEST_DIRNAME}/../bin/script.sh"
}

@test "JSON parsing works" {
    skip_if ! command -v jq &>/dev/null
    run my_json_parser '{"key": "value"}'
    [ "$status" -eq 0 ]
}

测试 Shell 兼容性

#!/usr/bin/env bats

@test "Script works in bash" {
    bash "${BATS_TEST_DIRNAME}/../bin/script.sh" arg1
}

@test "Script works in sh (POSIX)" {
    sh "${BATS_TEST_DIRNAME}/../bin/script.sh" arg1
}

@test "Script works in dash" {
    if command -v dash &>/dev/null; then
        dash "${BATS_TEST_DIRNAME}/../bin/script.sh" arg1
    else
        skip "dash not installed"
    fi
}

并行执行

#!/usr/bin/env bats

@test "Multiple independent operations" {
    run bash -c 'for i in {1..10}; do
        my_operation "$i" &
    done
    wait'
    [ "$status" -eq 0 ]
}

@test "Concurrent file operations" {
    for i in {1..5}; do
        my_function "$TMPDIR/file$i" &
    done
    wait
    [ -f "$TMPDIR/file1" ]
    [ -f "$TMPDIR/file5" ]
}

测试助手模式

test_helper.sh

#!/usr/bin/env bash

# Source script under test
export SCRIPT_DIR="${BATS_TEST_DIRNAME%/*}/bin"

# Common test utilities
assert_file_exists() {
    if [ ! -f "$1" ]; then
        echo "Expected file to exist: $1"
        return 1
    fi
}

assert_file_equals() {
    local file="$1"
    local expected="$2"

    if [ ! -f "$file" ]; then
        echo "File does not exist: $file"
        return 1
    fi

    local actual=$(cat "$file")
    if [ "$actual" != "$expected" ]; then
        echo "File contents do not match"
        echo "Expected: $expected"
        echo "Actual: $actual"
        return 1
    fi
}

# Create temporary test directory
setup_test_dir() {
    export TEST_DIR=$(mktemp -d)
}

cleanup_test_dir() {
    rm -rf "$TEST_DIR"
}

与 CI/CD 集成

GitHub Actions 工作流

name: Tests

on: [push, pull_request]

jobs:
  test:
    runs-on: ubuntu-latest

    steps:
      - uses: actions/checkout@v3

      - name: Install Bats
        run: |
          npm install --global bats

      - name: Run Tests
        run: |
          bats tests/*.bats

      - name: Run Tests with Tap Reporter
        run: |
          bats tests/*.bats --tap | tee test_output.tap

Makefile 集成

.PHONY: test test-verbose test-tap

test:
	bats tests/*.bats

test-verbose:
	bats tests/*.bats --verbose

test-tap:
	bats tests/*.bats --tap

test-parallel:
	bats tests/*.bats --parallel 4

coverage: test
	# Optional: Generate coverage reports

最佳实践

1. 每个测试只测试一件事 - 单一职责原则
2. 使用描述性的测试名称 - 清楚说明正在测试什么
3. 测试后进行清理 - 始终在 teardown 中移除临时文件
4. 同时测试成功和失败路径 - 不要只测试顺利路径
5. 模拟外部依赖 - 隔离被测单元
6. 对复杂数据使用 fixture - 让测试更具可读性
7. 在 CI/CD 中运行测试 - 尽早发现回归
8. 跨 shell 方言测试 - 确保可移植性
9. 保持测试快速 - 尽可能并行运行
10. 记录复杂的测试设置 - 解释不寻常的模式