A-Level CCEA Computer Science: Last-Minute Revision Notes | A-Level CCEA 计算机：考前冲刺笔记

📚 A-Level CCEA Computer Science: Last-Minute Revision Notes | A-Level CCEA 计算机：考前冲刺笔记

This revision guide condenses the entire A-Level CCEA Computer Science specification into key concepts, quick definitions, and exam-ready facts. Use it to reinforce your understanding, spot knowledge gaps, and walk into the exam hall with confidence. Each section pairs an English explanation with a Chinese translation to support bilingual learners and international candidates aiming for top grades.

这份冲刺笔记浓缩了A-Level CCEA计算机科学课程的全部核心概念、快速定义和应试要点。用它来巩固理解、发现知识漏洞，自信地走进考场。每个部分都提供英文解释与中文翻译，帮助双语学习者和国际考生冲刺高分。

---

1. Data Representation and Number Systems | 数据表示与数制

All data inside a computer is stored in binary – sequences of 1s and 0s. Denary (base‑10) numbers are converted to binary using successive division by 2, and hexadecimal (base‑16) is used as a shorter, more readable representation. One hex digit represents four bits (a nibble).

计算机内所有数据都以二进制（0 和 1）存储。十进制数通过反复除以 2 转换为二进制，而十六进制（基数为 16）用于更简洁的表示。一个十六进制数字代表四个比特（一个半字节）。

Signed integers are represented with two’s complement. To obtain the negative of a binary number, invert all bits and add 1. In an 8‑bit register, the range of two’s complement values is -128 to +127. Overflow occurs when a calculation produces a result outside this range.

带符号整数使用二进制补码表示。要得到一个二进制数的负数，将所有位取反后加 1。在 8 位寄存器中，补码的范围是 -128 到 +127。当计算结果超出此范围时会发生溢出。

Real numbers are stored in floating-point format: mantissa × 2^exponent. Increasing the number of bits for the mantissa improves precision, while increasing the exponent bits expands the range. Normalisation ensures a single representation by adjusting the mantissa so that its first bit is 1.

实数以浮点格式存储：尾数 × 2^指数。增加尾数位数可提高精度，增加指数位数可扩大范围。规范化通过调整尾数使得第一位是 1 来确保唯一表示。

Denary	Binary (8-bit)	Hexadecimal
42	0010 1010	2A
-42	1101 0110	D6
15.625	01000011 11101000 (example mantissa/exponent)	–

Character encoding uses standards such as ASCII (7‑bit) and Unicode. Unicode can represent every character in all major languages and uses schemes like UTF‑8 to remain backwards compatible with ASCII.

字符编码使用标准的 ASCII（7 位）和 Unicode。Unicode 能表示所有主要语言的每个字符，并使用诸如 UTF‑8 的编码方案保持与 ASCII 的向后兼容。

Bitwise operations – AND, OR, XOR, and NOT – manipulate individual bits. Masking with AND can clear selected bits, while OR can set them. XOR toggles bits and is useful in parity checks and simple encryption.

按位运算 – AND、OR、XOR 和 NOT – 直接操作个别位。用 AND 掩码可以清除特定位，用 OR 则可置位。XOR 能翻转位，常用于奇偶校验和简单加密。

---

2. CPU Architecture and the FDE Cycle | CPU 架构与取指-解码-执行周期

The central processing unit (CPU) contains the control unit (CU), arithmetic logic unit (ALU), and registers such as the program counter (PC), memory address register (MAR), memory data register (MDR), current instruction register (CIR), and accumulator (ACC). The CU orchestrates the fetch‑decode‑execute cycle.

中央处理器 (CPU) 包含控制单元 (CU)、算术逻辑单元 (ALU)，以及程序计数器 (PC)、内存地址寄存器 (MAR)、内存数据寄存器 (MDR)、当前指令寄存器 (CIR) 和累加器 (ACC) 等寄存器。控制单元负责协调取指-解码-执行周期。

During the fetch stage, the address in the PC is copied to the MAR, the instruction is fetched from RAM into the MDR, and then moved to the CIR. The PC is incremented to point to the next instruction. In the decode stage, the CU interprets the opcode. During execute, the ALU performs the required operation, using the accumulator for temporary results.

在取指阶段，PC 中的地址被复制到 MAR，指令从 RAM 中取出放入 MDR，然后转移到 CIR。PC 递增以指向下一条指令。在解码阶段，CU 解释操作码。在执行阶段，ALU 执行所需的操作，使用累加器暂存结果。

Factors affecting CPU performance include clock speed (cycles per second), number of cores, and cache size. Pipelining improves throughput by overlapping FDE stages for successive instructions. RISC (Reduced Instruction Set Computer) processors use simpler, fixed‑length instructions, while CISC (Complex Instruction Set Computer) can handle multi‑step operations with single complex instructions.

影响 CPU 性能的因素包括时钟频率（每秒周期数）、核心数量和缓存大小。流水线通过重叠续指令的取指-解码-执行阶段来提高吞吐量。RISC（精简指令集计算机）处理器使用简单、定长的指令，而 CISC（复杂指令集计算机）能用一条复杂指令处理多步操作。

Moore’s Law observes that transistor density doubles roughly every two years. As transistors approach atomic scales, heat dissipation and quantum effects present limits, steering development towards multi‑core and specialised architectures.

摩尔定律指出晶体管密度大约每两年翻一番。随着晶体管尺寸接近原子级别，散热和量子效应带来限制，推动了多核和专用架构的发展。

---

3. Memory and Storage | 内存与存储设备

Primary memory is directly accessed by the CPU. RAM (Random Access Memory) is volatile and holds the operating system, running programs, and data in current use. ROM (Read‑Only Memory) is non‑volatile and typically stores the BIOS or firmware needed to boot the computer.

主存由 CPU 直接访问。RAM（随机存取存储器）是易失性的，用于存放操作系统、正在运行的程序和当前数据。ROM（只读存储器）是非易失性的，通常存储引导计算机所需的 BIOS 或固件。

Cache memory uses high‑speed SRAM located on or near the CPU to store frequently accessed instructions and data. Levels L1, L2, and L3 offer a hierarchy of size and speed, reducing the average time to access memory. Virtual memory uses part of the hard drive as an extension of RAM when physical RAM is full, but disk access is much slower.

高速缓存使用位于 CPU 内部或附近的高速 SRAM，存储频繁访问的指令和数据。L1、L2 和 L3 级缓存构成了大小和速度的层次，减少了平均内存访问时间。当物理 RAM 满时，虚拟内存使用部分硬盘作为 RAM 的扩展，但磁盘访问速度要慢得多。

Secondary storage is non‑volatile. Optical discs (CD, DVD, Blu‑ray) use lasers to read pits and lands. Magnetic hard disk drives (HDD) store data on spinning platters, with access time affected by seek time and latency. Solid‑state drives (SSD) use NAND flash memory, offering faster access, lower power consumption, and no moving parts, but have a finite number of write cycles.

二级存储器是非易失性的。光盘（CD、DVD、Blu‑ray）使用激光读取凹坑和平面。磁性硬盘驱动器 (HDD) 将数据存储在旋转的盘片上，访问时间受寻道时间和延迟影响。固态硬盘 (SSD) 使用 NAND 闪存，提供更快的访问速度、更低的功耗且无移动部件，但写入次数有限。

Cloud storage stores data on remote servers accessed via the internet. Advantages include accessibility from any device, automatic backup, and scalability. Disadvantages involve dependence on internet connectivity, ongoing subscription costs, and security concerns.

云存储将数据存放在通过互联网访问的远程服务器上。其优点包括可从任何设备访问、自动备份和可扩展性。缺点则依赖于互联网连接、持续订阅费用和安全问题。

---

4. Input, Output and Sensors | 输入、输出与传感器

Input devices feed data into the computer system. A flatbed scanner captures images using a CCD array, while a barcode reader reflects laser light off printed bars to identify products via a check digit. RFID (Radio Frequency Identification) uses tags and readers for contactless tracking.

输入设备将数据送入计算机系统。平板扫描仪通过 CCD 阵列捕捉图像，而条形码阅读器通过激光在印刷条码上反射，利用校验位识别产品。RFID（射频识别）使用标签和读写器实现非接触式跟踪。

Output devices present processed information. LCD and LED screens use millions of pixels, each composed of red, green, and blue sub‑pixels. Laser printers use static electricity, toner, and a heated fuser to produce high‑quality text. 3D printers build objects layer by layer from materials like PLA or resin.

输出设备展示处理后的信息。LCD 和 LED 屏幕使用数百万个像素，每个像素由红、绿、蓝子像素构成。激光打印机利用静电、碳粉和加热定影器产生高质量文本。3D 打印机通过 PLA 或树脂等材料逐层构建物体。

Sensors continuously monitor the environment. Examples include temperature (thermistor), pressure, light (LDR), and motion (PIR). An ADC (analogue‑to‑digital converter) converts continuous sensor signals into discrete digital values that the processor can handle.

传感器持续监测环境。例如温度（热敏电阻）、压力、光（光敏电阻）和运动（PIR）。模数转换器 (ADC) 将连续的传感器信号转换为处理器可处理的离散数字值。

Assistive technology makes systems accessible. A puff‑sip switch allows users with limited motor skills to input commands via air pressure. Screen readers convert on‑screen text to synthesised speech or Braille displays, supporting visually impaired users.

辅助技术使系统更易访问。吹吸式开关允许运动能力受限的用户通过气压输入指令。屏幕阅读器将屏幕上的文本转换为合成语音或盲文显示，支持视障用户。

---

5. System Software and Operating Systems | 系统软件与操作系统

An operating system (OS) manages hardware resources, provides a user interface, handles memory management, processor scheduling, file management, and security. Common OS types include multi‑user, multi‑tasking, real‑time, and distributed systems.

操作系统 (OS) 管理硬件资源，提供用户界面，处理内存管理、处理器调度、文件管理和安全。常见 OS 类型包括多用户、多任务、实时和分布式系统。

Memory management uses paging and segmentation to allocate RAM to processes efficiently. Pages are fixed‑size blocks moved between RAM and secondary storage, preventing external fragmentation. Virtual memory enables execution of programs larger than physical memory by swapping idle pages to disk.

内存管理使用分页和分段来高效地为进程分配 RAM。页面是固定大小的块，可在 RAM 和二级存储之间移动，防止外部碎片。虚拟内存通过将空闲页面交换到磁盘，使得大于物理内存的程序得以运行。

Scheduling algorithms determine which process runs next. Round‑robin gives each process a fixed time slice. Shortest job first minimises average waiting time. Priority‑based scheduling assigns a priority to each process, but can lead to starvation of low‑priority tasks. The OS kernel remains resident in memory at all times.

调度算法决定下一个运行的进程。轮转法给每个进程固定时间片。最短作业优先可最小化平均等待时间。基于优先级的调度为每个进程分配优先级，但可能导致低优先级任务饥饿。OS 内核始终驻留在内存中。

Utility software performs maintenance tasks. Disk defragmentation reorganises fragmented files so that each file occupies contiguous sectors, improving read times (not needed for SSDs). Encryption utilities like BitLocker protect data by converting plaintext into ciphertext using algorithms such as AES. Backup software automates copying of data to external media or the cloud.

实用工具软件执行维护任务。磁盘碎片整理重新组织碎片文件，使每个文件占用连续的扇区，从而提高读取速度（对 SSD 不需要）。BitLocker 等加密工具使用 AES 等算法将明文转换为密文来保护数据。备份软件自动将数据复制到外部媒介或云端。

---

6. Programming Fundamentals | 编程基础

Programming constructs are sequence, selection, and iteration. Selection uses if‑else and switch‑case statements. Iteration uses definite loops (for) and indefinite loops (while, do‑while). Careful use of Boolean operators (AND, OR, NOT) builds complex conditions.

编程构造包括顺序、选择和迭代。选择使用 if‑else 和 switch‑case 语句。迭代使用定数循环 (for) 和不定数循环 (while、do‑while)。谨慎使用布尔运算符 (AND、OR、NOT) 可构建复杂条件。

Data types include integer, real (float), character, string, and Boolean. Arrays store multiple elements of the same type under one identifier. One‑dimensional arrays are indexed from 0, and two‑dimensional arrays organise data in rows and columns. Sub‑programs (functions and procedures) promote modularity and code reuse; functions return a value, procedures do not.

数据类型包括整型、实型（浮点型）、字符、字符串和布尔型。数组在一个标识符下存储多个相同类型的元素。一维数组从 0 开始索引，二维数组以行列组织数据。子程序（函数和过程）促进模块化和代码重用；函数返回值，过程不返回。

Parameter passing by value copies the arguments into the procedure’s local variables, leaving the original unchanged. Passing by reference allows the procedure to modify the original variable by working with its memory address. Recursion is a technique where a sub‑program calls itself; it must have a base case to terminate.

按值传递参数将实参复制到过程的局部变量中，原变量不变。按引用传递允许过程通过操作内存地址来修改原变量。递归是一种子程序调用自身的技术；必须有一个基本情况来终止。

Testing is crucial. Syntax errors are detected by the compiler or interpreter. Logic errors produce unexpected results but no crash. Run‑time errors occur during execution (e.g., division by zero). Normal, boundary, and erroneous test data should be used. Trace tables help dry‑run algorithms by tracking variable states step by step.

测试至关重要。语法错误由编译器或解释器检测。逻辑错误产生意外结果但不崩溃。运行时错误在执行期间发生（例如除以零）。应使用正常、边界和异常测试数据。跟踪表通过逐步跟踪变量状态来帮助预演算法。

---

7. Databases and SQL | 数据库与 SQL

A relational database stores data in tables (relations) with rows (records/tuples) and columns (fields/attributes). Each table has a primary key that uniquely identifies each record. Foreign keys link tables by referencing the primary key of another table, establishing relationships (one‑to‑one, one‑to‑many, many‑to‑many).

关系型数据库将数据存储在表（关系）中，由行（记录/元组）和列（字段/属性）组成。每个表有一个主键用于唯一标识每条记录。外键通过引用另一个表的主键来连接表，建立关系（一对一、一对多、多对多）。

Normalisation reduces data redundancy and anomalies. First normal form (1NF) requires atomic values and no repeating groups. Second normal form (2NF) requires 1NF and that non‑key fields are fully functionally dependent on the whole primary key (no partial dependencies). Third normal form (3NF) removes transitive dependencies where a non‑key field depends on another non‑key field.

规范化减少数据冗余和异常。第一范式 (1NF) 要求原子值且无重复组。第二范式 (2NF) 在满足 1NF 的基础上，要求非键字段完全函数依赖于整个主键（无部分依赖）。第三范式 (3NF) 消除了非键字段依赖于另一个非键字段的传递依赖。

SQL (Structured Query Language) is used to define, manipulate, and query databases. DDL commands include CREATE TABLE, ALTER TABLE, and DROP TABLE. DML commands include SELECT, INSERT, UPDATE, and DELETE. Typical SELECT with conditions: SELECT StudentName, Grade FROM Results WHERE Subject = 'Maths' ORDER BY Grade DESC;

SQL（结构化查询语言）用于定义、操纵和查询数据库。DDL 命令包括 CREATE TABLE、ALTER TABLE 和 DROP TABLE。DML 命令包括 SELECT、INSERT、UPDATE 和 DELETE。带条件的典型查询：SELECT StudentName, Grade FROM Results WHERE Subject = 'Maths' ORDER BY Grade DESC;

Referential integrity ensures that any foreign key value must point to an existing valid record in the related table, enforced by constraints. ACID properties (Atomicity, Consistency, Isolation, Durability) guarantee reliable database transactions.

引用完整性确保任何外键值必须指向关联表中存在的有效记录，通过约束来实施。ACID 特性（原子性、一致性、隔离性、持久性）保证可靠的数据库事务。

---

8. Computer Networks and Protocols | 计算机网络与协议

Networks are classified by geographical scale: PAN (Personal Area Network), LAN (Local Area Network), WAN (Wide Area Network). Common topologies include star (central switch, easy fault isolation), bus (single backbone, low cost), and mesh (full or partial interconnection, high redundancy).

网络按地理范围分类：PAN（个人区域网）、LAN（局域网）、WAN（广域网）。常见拓扑包括星型（中央交换机，易于故障隔离）、总线型（单主干，低成本）和网状（全互连或部分互连，高冗余）。

The TCP/IP suite has four layers: Application (HTTP, FTP, SMTP), Transport (TCP, UDP), Internet (IP), and Network Access (Ethernet, Wi‑Fi). Packet switching breaks data into packets that travel independently and are reassembled at the destination. Circuit switching establishes a dedicated path before transmission.

TCP/IP 协议族有四层：应用层（HTTP、FTP、SMTP）、传输层（TCP、UDP）、互联网层（IP）和网络接入层（以太网、Wi‑Fi）。分组交换将数据拆分成独立传输并在目的地重新组装的数据包。电路交换在传输前建立专用路径。

IP addresses (IPv4: 32‑bit dotted decimal; IPv6: 128‑bit hexadecimal) uniquely identify devices. MAC addresses are 48‑bit physical addresses burned into NICs. DNS translates domain names to IP addresses, and DHCP automatically assigns IP configuration to hosts.

IP 地址（IPv4：32 位点分十进制；IPv6：128 位十六进制）唯一标识设备。MAC 地址是嵌入网卡的 48 位物理地址。DNS 将域名转换为 IP 地址，DHCP 自动为主机分配 IP 配置。

Network security measures include firewalls (packet filtering, stateful inspection), encryption (WPA3 for Wi‑Fi, SSL/TLS for web), and authentication. Malware types: virus (attaches to files), worm (self‑replicates across networks), Trojan horse (disguises as legitimate software), and ransomware (encrypts files for ransom). Phishing uses fraudulent emails to steal credentials.

网络安全措施包括防火墙（包过滤、状态检测）、加密（Wi‑Fi 用 WPA3，网页用 SSL/TLS）和认证。恶意软件类型：病毒（附着于文件）、蠕虫（通过网络自我复制）、特洛伊木马（伪装为合法软件）和勒索软件（加密文件索要赎金）。网络钓鱼利用欺诈邮件窃取凭据。

---

9. Web Technologies | 网页技术

HTML (HyperText Markup Language) structures web content using elements like <html>, <head>, <body>, <h1>, <p>, <a>, and <img>. CSS (Cascading Style Sheets) controls presentation (colours, fonts, layout) and supports responsive design through media queries and frameworks like Bootstrap.

HTML（超文本标记语言）使用 <html>、<head>、<body>、<h1>、<p>、<a> 和 <img> 等元素构建网页内容。CSS（层叠样式表）控制表现（颜色、字体、布局），并通过媒体查询和 Bootstrap 等框架支持响应式设计。

Client‑side scripting (JavaScript) runs in the browser to validate forms, animate elements, and update content without reloading the page. Server‑side scripting (PHP, Python) processes requests on the web server, interacts with databases, and generates dynamic HTML before sending it to the client.

客户端脚本（JavaScript）在浏览器中运行，用于验证表单、动画元素和无须重新加载页面更新内容。服务器端脚本（PHP、Python）在 Web 服务器上处理请求，与数据库交互，并在发送给客户端之前生成动态 HTML。

HTTP is a request‑response protocol. GET requests retrieve data, while POST submits data to be processed. HTTPS encrypts HTTP traffic using SSL/TLS. Cookies are small text files stored by the browser to maintain state across multiple requests (e.g., shopping carts, user preferences).

HTTP 是一种请求‑响应协议。GET 请求检索数据，POST 提交待处理的数据。HTTPS 使用 SSL/TLS 加密 HTTP 流量。Cookies 是由浏览器存储的小型文本文件，用于在多次请求之间保持状态（例如购物车、用户偏好）。

---

10. Ethical, Legal and Environmental Issues | 伦理、法律与环境问题

The Data Protection Act 2018 (UK GDPR) governs how personal data is collected, processed, and stored. Principles include data minimisation, purpose limitation, accuracy, and the rights of data subjects to access and erase their data. Organisations must appoint a data protection officer where appropriate.

《2018 年数据保护法》（英国 GDPR）规定了个人数据的收集、处理和存储方式。原则包括数据最小化、目的限制、准确性，以及数据主体访问和删除其数据的权利。组织须在合适时任命数据保护官。

The Computer Misuse Act 1990 makes it an offence to gain unauthorised access to computer material, commit further offences such as data theft or fraud, and impair the operation of a computer (e.g., launching a denial‑of‑service attack). The Copyright, Designs and Patents Act protects intellectual property in software and digital content.

《1990 年计算机滥用法》规定未经授权访问计算机材料、实施数据窃取或欺诈等进一步犯罪、以及损害计算机运行（如发起拒绝服务攻击）均为犯罪。《版权、设计和专利法》保护软件和数字内容的知识产权。

Technology impacts the environment. Manufacturing devices consumes energy and rare metals, while e‑waste generates toxic pollutants. Data centres require vast amounts of electricity for computing and cooling. Positive impacts include smart energy grids, telecommuting reducing travel, and efficient logistics.

技术影响环境。制造设备消耗能源和稀有金属，而电子垃圾产生有毒污染物。数据中心需要大量电力用于计算和冷却。积极影响包括智能电网、远程办公减少出行以及高效的物流。

Artificial intelligence raises ethical concerns: bias in training data leads to discriminatory outputs, autonomous weapons make lethal decisions without human oversight, and mass surveillance can erode civil liberties. Algorithmic transparency and accountability are essential.

人工智能引发伦理问题：训练数据中的偏见导致歧视性输出，自主武器在没有人类监督的情况下做出致命决定，以及大规模监控可能侵蚀公民自由。算法透明度和问责制至关重要。

---

11. Software Development Life Cycle | 软件开发生命周期

The waterfall model proceeds linearly through stages: feasibility study, requirements analysis, design, implementation, testing, deployment, and maintenance. It is suitable for projects with stable, well‑understood requirements but lacks flexibility for changes late in the cycle.

瀑布模型按阶段线性推进：可行性研究、需求分析、设计、实现、测试、部署和维护。它适用于需求稳定且充分理解的项目，但缺乏在后期变更的灵活性。

Agile methodologies (e.g., Scrum, Extreme Programming) use iterative development. Features are delivered in short sprints, with constant customer feedback and adaptation. Scrum defines roles such as Product Owner, Scrum Master, and Development Team, and uses daily stand‑ups and sprint reviews.

敏捷方法（如 Scrum、极限编程）采用迭代开发。功能以短冲刺周期交付，伴随持续的客户反馈和调整。Scrum 定义了产品负责人、Scrum Master 和开发团队等角色，并使用每日站会和冲刺评审。

Analysis tools include data flow diagrams (DFDs) showing how data moves through a system, and entity‑relationship diagrams (ERDs) representing database structure. Prototyping creates an early model of the system to clarify requirements and gather user input before full development begins.

分析工具包括数据流图 (DFD)，用于展示数据在系统中的流动方式，以及实体关系图 (ERD)，用于表示数据库结构。原型设计在全面开发之前创建系统的早期模型，以澄清需求并收集用户反馈。

Testing strategies: alpha testing is done by developers in‑house; beta testing involves real users in a live environment. Black‑box testing examines inputs and outputs without knowledge of internal code, while white‑box testing verifies internal logic and all paths. Maintenance types are corrective, adaptive, perfective, and preventive.

测试策略：alpha 测试由内部开发人员完成；beta 测试让真实用户在实际环境中进行。黑盒测试在不了解内部代码的情况下检查输入与输出，而白盒测试验证内部逻辑和所有路径。维护类型包括纠正性、适应性、完善性和预防性维护。

---

12. Exam Technique and Common Pitfalls | 考试技巧与常见错误

Read the question carefully and identify the command word: ‘describe’, ‘explain’, ‘evaluate’, ‘state’, ‘calculate’. Use the mark allocation as a guide to depth – a 4‑mark question expects four distinct points or two well‑developed points with explanation.

仔细阅读题目并识别命令词：“描述”、“解释”、“评估”、“陈述”、“计算”。以分值作为深度的参考——4 分的题目通常需要四个明确的要点或两个充分展开并带解释的要点。

For binary arithmetic questions, always show your working. Convert numbers clearly, indicate where you add 1 for two’s complement, and double‑check for overflow by comparing the result’s range. In floating‑point normalisation, always express the answer in the specified mantissa/exponent bit lengths.

对于二进制算术题，一定要展示步骤。清楚地转换数字，标出补码加 1 的位置，并通过比较结果范围来复核溢出。在浮点规范化中，始终用指定位数表示尾数和指数。

When designing databases, ensure every attribute is atomic, no repeating groups, and clearly underline primary keys. For normalisation, methodically check partial and transitive dependencies. In SQL queries, remember that string literals go in single quotes and order of clauses matters: SELECT, FROM, WHERE, ORDER BY.

设计数据库时，确保每个属性都是原子的，没有重复组，并明确标出主键。在进行规范化时，系统性地检查部分依赖和传递依赖。在 SQL 查询中，记得字符串字面量用单引号，子句的顺序很重要：SELECT、FROM、WHERE、ORDER BY。

Manage your time: allocate roughly 1.2 minutes per mark. Start with questions you find easiest to build confidence. If stuck, mark the question and return later. For essay‑style questions, plan a quick bullet list before writing to ensure a coherent structure.

管理好时间：大约每分值分配 1.2 分钟。从你觉得最简单的题目开始以建立信心。如果卡住，标记题目稍后返回。对于论述型问题，在正式作答前快速列出要点提纲，以确保条理清晰。

Published by TutorHao | Computer Science Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）