📚 IGCSE WJEC Computer Science: Software Engineering Essentials | IGCSE WJEC 计算机：软件工程考点精讲
In the IGCSE WJEC Computer Science syllabus, software engineering covers the systematic approach to developing reliable, efficient, and maintainable software. Understanding the software development life cycle, testing strategies, debugging, and documentation is key to exam success. This article provides a focused revision guide covering every essential topic.
在 IGCSE WJEC 计算机科学大纲中，软件工程涵盖了开发可靠、高效和可维护软件的系统化方法。理解软件开发生命周期、测试策略、调试和文档化是考试成功的关键。本文提供了一份精准复习指南，涵盖各个核心考点。
1. Software Development Life Cycle (SDLC) | 软件开发生命周期
The SDLC is a structured framework that defines the phases required to create software from initial idea to final retirement. It ensures that projects are completed on time, within budget, and meet user expectations.
SDLC 是一个结构化框架，定义了从最初设想到最终退役创建软件所需的各个阶段。它确保项目按时、在预算内完成，并满足用户期望。
A typical SDLC model includes feasibility study, requirements analysis, design, implementation (coding), testing, deployment, and maintenance. Each phase produces deliverables that feed into the next.
典型的 SDLC 模型包括可行性研究、需求分析、设计、实现（编码）、测试、部署和维护。每个阶段都会产生产出物，并输入到下一阶段。
Following the SDLC reduces risks such as scope creep, miscommunication, and technical failures, while improving quality through systematic review.
遵循 SDLC 可以减少范围蔓延、沟通不畅和技术失败等风险，同时通过系统性审查提高质量。
2. System Analysis and Requirements | 系统分析与需求
During analysis, developers collect detailed information about what the system must do. Common fact-finding methods include interviews, questionnaires, observation, and examining existing documentation.
在分析阶段，开发人员收集关于系统必须做什么的详细信息。常见的调研方法包括访谈、问卷调查、观察和检查现有文档。
Requirements are classified as functional (what the system should do, e.g. 'calculate total cost') and non-functional (how the system should behave, e.g. 'response time under 1 second'). Both must be clearly documented.
需求分为功能性（系统应该做什么，例如'计算总成本'）和非功能性（系统应该如何表现，例如'响应时间低于 1 秒'）。两者都必须清楚地记录。
A feasibility study determines whether the project is technically possible, economically viable, legally permissible, and operationally suitable before significant resources are committed.
可行性研究在投入大量资源之前，确定项目在技术上是否可行、经济上是否合理、法律上是否允许以及操作上是否适合。
3. Design Stage: Algorithms and Interface | 设计阶段：算法与界面
In the design phase, programmers plan the solution using algorithms expressed as pseudocode and flowcharts. These tools describe the logic without worrying about a specific programming language syntax.
在设计阶段，程序员使用伪代码和流程图表达算法来规划解决方案。这些工具描述逻辑而不必担心特定编程语言的语法。
The user interface design defines screens, forms, navigation, and inputs. A good UI is intuitive, consistent, and accessible to all intended users.
用户界面设计定义了屏幕、表单、导航和输入。好的 UI 直观、一致，并且对所有目标用户可访问。
Design documentation may also include a data dictionary specifying each variable's name, data type, expected range, and validation rules, ensuring that implementation follows a clear blueprint.
设计文档还可能包括数据字典，指定每个变量的名称、数据类型、预期范围和验证规则，确保实现遵循清晰的蓝图。
4. Implementation and Coding Standards | 实现与编码规范
Implementation turns the design into working code. Writing code that is easy to read and maintain is as important as making it functional, especially for long-term projects and team collaboration.
实现将设计转变为可工作的代码。编写易于阅读和维护的代码与使其功能正常运行同等重要，尤其是在长期项目和团队协作中。
Good coding practice includes using meaningful variable names (e.g. 'studentAge' instead of 'sa'), adding comments to explain complex logic, and following consistent indentation and spacing.
良好的编码实践包括使用有意义的变量名（例如 'studentAge' 而不是 'sa'），添加注释解释复杂逻辑，以及遵循一致的缩进和间距。
Choosing appropriate data types is essential: integer for whole numbers, real/float for decimals, string for text, Boolean for true/false, and arrays/lists for collections. Using the wrong type can cause data loss or runtime errors.
选择合适的数据类型至关重要：整数用于整数，实数/浮点数用于小数，字符串用于文本，布尔型用于真/假，数组/列表用于集合。使用错误的类型可能导致数据丢失或运行时错误。
5. Testing Techniques | 测试技术
Testing verifies that the software functions correctly according to its requirements and is free from critical defects. It is performed at multiple levels throughout development.
测试验证软件是否按照其需求正确运行，并且没有重大缺陷。它贯穿开发的多个层次进行。
The main testing levels are unit testing (testing individual functions or modules), integration testing (testing how modules work together), system testing (testing the entire integrated system), and acceptance testing (carried out by the end user to confirm the software meets their needs).
主要的测试级别有：单元测试（测试单个函数或模块）、集成测试（测试模块如何协同工作）、系统测试（测试整个集成系统）和验收测试（由最终用户执行，确认软件满足他们的需求）。
Test data must be carefully selected to include normal data (valid, expected inputs), boundary data (values at the limits of accepted ranges), and erroneous data (invalid inputs that should trigger error handling).
测试数据必须仔细选择，包括正常数据（有效、预期的输入）、边界数据（接受范围极限的值）和异常数据（应触发错误处理的无效输入）。
The following table summarises typical test data examples:
下表总结了典型的测试数据示例：
| Data Type / 数据类型 | Description / 描述 | Example / 例子 |
|---|---|---|
| Normal 正常 | Valid data likely to be accepted 应被接受的有效数据 | Exam mark = 65 (range 0-100) 考试分数 = 65（范围 0-100） |
| Boundary 边界 | Data at the extreme limits of validity 处于有效性极端界限的数据 | 0, 100 (for a 0-100 range) 0, 100（对于 0-100 范围） |
| Erroneous 异常 | Data that is clearly invalid and must be rejected 明显无效、必须被拒绝的数据 | Exam mark = -5 or 'seventy' 考试分数 = -5 或 'seventy' |
Black box testing focuses on input/output functionality without examining internal code structure, whereas white box testing uses knowledge of the code to test all possible paths and branches.
黑盒测试专注于输入/输出功能，而不检查内部代码结构；而白盒测试利用代码知识来测试所有可能的路径和分支。
A test plan documents the testing strategy, test cases, expected results, and actual outcomes. It ensures thorough verification and provides evidence for evaluation.
测试计划记录了测试策略、测试用例、预期结果和实际结果。它确保了彻底的验证，并为评估提供证据。
6. Debugging and Error Types | 调试与错误类型
Debugging is the systematic process of identifying, locating, and correcting errors within a program. Programmers use various tools and techniques to pinpoint faults.
调试是识别、定位和纠正程序中错误的系统化过程。程序员使用各种工具和技术来查明故障。
There are three main categories of errors: syntax errors (code that breaks language rules, like missing a colon), logic errors (flaws in the algorithm producing incorrect results, e.g. an off-by-one mistake), and runtime errors (problems occurring during execution, such as division by zero or file not found).
错误主要分为三类：语法错误（违反语言规则的代码，如缺少冒号）、逻辑错误（算法中的缺陷导致不正确的结果，例如差一错误），以及运行时错误（执行期间发生的问题，如除以零或文件未找到）。
Integrated development environments (IDEs) provide debugging features like breakpoints (pause execution at a specific line), step into/over (execute line by line), and variable watches (inspect current values). These help trace program flow.
集成开发环境（IDE）提供调试功能，如断点（在特定行暂停执行）、单步执行/跳过（逐行执行）和变量监视（检查当前值）。这些有助于追踪程序流。
Trace tables are manual tools where you record values of variables after each line of code is executed, helping to uncover logic errors by simulating the algorithm step by step.
跟踪表是一种手动工具，其中记录了每行代码执行后变量的值，通过逐步模拟算法来帮助发现逻辑错误。
7. Evaluation and Maintenance | 评估与维护
After deployment, the final software is evaluated against the original success criteria, such as ease of use, speed, reliability, and fitness for purpose. This evaluation may lead to further improvements.
部署后，最终软件将根据最初的成功标准（如易用性、速度、可靠性和适用性）进行评估。此评估可能导致进一步的改进。
Software maintenance is an ongoing activity and falls into three types: corrective maintenance (fixing bugs discovered after release), adaptive maintenance (modifying software to work on new operating systems or hardware), and perfective maintenance (adding new features or improving performance based on user feedback).
软件维护是一项持续活动，分为三种类型：纠正性维护（修复发布后发现的错误）、适应性维护（修改软件以适应新的操作系统或硬件）和完善性维护（根据用户反馈添加新功能或提升性能）。
Without proper maintenance, software can quickly become outdated, insecure, or incompatible with evolving user needs and technology, reducing its long-term value.
如果没有适当的维护，软件会迅速过时、不安全或与不断变化的用户需求和技术不兼容，从而降低其长期价值。
8. Development Methodologies: Waterfall vs Agile | 开发方法论：瀑布与敏捷
The waterfall model is a traditional linear approach where each SDLC phase is completed in sequence and once a phase is finished, you do not return to it. It works well for projects with clear, stable requirements but is inflexible to change.
瀑布模型是一种传统的线性方法，每个 SDLC 阶段按顺序完成，一旦一个阶段完成，就不再返回。它适用于需求清晰、稳定的项目，但对变化不灵活。
Agile methodologies, such as Scrum, break development into small cycles (sprints) that deliver working increments of software. Regular client feedback and stand-up meetings allow the team to adapt priorities and respond to changing requirements quickly.
敏捷方法（如 Scrum）将开发划分为多个小周期（冲刺），交付可工作的软件增量。定期的客户反馈和站会使得团队能够快速调整优先级并响应变化的需求。
In the WJEC exam, you may be asked to compare the two: waterfall emphasises thorough documentation and upfront planning, while agile values individuals, collaboration, and responding to change over rigid processes.
在 WJEC 考试中，可能会要求比较两者：瀑布模型强调详尽的文档和前期规划，而敏捷则重视个体、协作以及响应变化胜过僵化的流程。
9. Documentation and CASE Tools | 文档与 CASE 工具
Comprehensive documentation supports both end users and developers. User documentation includes quick-start guides, tutorials, and FAQs, while technical documentation covers architecture diagrams, API descriptions, and commented code.
全面的文档支持最终用户和开发人员。用户文档包括快速入门指南、教程和常见问题解答，而技术文档涵盖架构图、API 描述和带注释的代码。
Computer-Aided Software Engineering (CASE) tools automate time-consuming activities in design, coding, testing, and project management. They improve consistency, reduce human error, and boost productivity.
计算机辅助软件工程（CASE）工具自动化了设计、编码、测试和项目管理中耗时的活动。它们提高了一致性，减少了人为错误并提升了生产力。
Examples include version control systems (e.g. Git) to track changes, UML diagram generators for design, and integrated testing frameworks that run automated test suites. Understanding their role helps link theory to real-world practice.
示例包括用于跟踪变更的版本控制系统（如 Git）、用于设计 UML 图生成器，以及运行自动化测试套件的集成测试框架。理解它们的作用有助于将理论与实际实践联系起来。
10. Ethical and Legal Considerations | 道德与法律考量
Software engineers have a professional duty to uphold ethical standards:
Published by TutorHao | IGCSE Computer Science Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
或截图后用「扫一扫 → 相册」选择本图