when coding in a fresh system

  I have designed a component of a web system with my workmate. In detail, I am just a coder instead of a designer, as all of it are designed by my manager. During the programme, we encounter much of problems,  most of which can be avoided. As a result, we delay the finished time. We spent much of time revising program for the design is being changed all the time. I am baffling with the dynamic requirement. It is common that I just finish a function with spending lots of time when the requirement is changed. What should I do? For example, I have alter the construct of the database table much times, which leads to a result that I have to resvise the routine. It is not the most disgusted thing that I‘m confronted with. I code in a bad system which has a terrible framework. I have to change my code each time a external system modifies their interface or convention. It makes me upset. I think a robost system should be able handle kinds of change without modifying itself. That‘s the goal of our programmer.

  In my opinion, when we design a system, we should design its framework, instead of coding first. If it depends on a external system, we prefer to depend on our inernal interface which will depend on the external system. So when a external system changes its interface or requirement, we have to modify the inernal interface instead of much code. That is to abstract the function .

时间: 2024-10-12 13:23:41

when coding in a fresh system的相关文章

Kickstart Options

The following options can be placed in a kickstart file. If you prefer to use a graphical interface for creating your kickstart file, use the Kickstart Configurator application. Refer to Chapter 29, Kickstart Configurator for details. Note If the opt

HDFS HA架构以及源码引导

HA体系架构 相关知识介绍 HDFS master/slave架构,HDFS节点分为NameNode节点和DataNode节点.NameNode存有HDFS的元数据:主要由FSImage和EditLog组成.FSImage保存有文件的目录.分块ID.文件权限等,EditLog保存有对HDFS的操作记录.DataNode存放分块的数据,并采用CRC循环校验方式对本地的数据进行校验,DataNode周期性向NameNode汇报本机的信息. NameNode单点故障:HDFS只有一个NameNode节

java工具类–自动将数据库表生成javabean

最近和数据库的表打交道挺多的,因为暂时做的是接口活. 在这过程中发现要把表转换成对应的javabean类型,字段少的表还行,如果不小心碰到几十个字段的他妈的写起来就有点麻烦了,万一碰到几百个的呢,那不是要崩溃. 于是想写个工具类,自动生成javabean. 先说下思路: 1.读取数据库表,获取里面的字段名. 准备连接数据库的驱动包,这里只是针对了oracle数据库和mysql数据库 2.构建一个stringBuffer字符串,用来生成我们需要的类. 3.写入文件 要求具备相应的文件流知识. 好了

信息安全管理(2):什么叫作信息安全?信息安全的原则和要求

这个章节将简单介绍一些信息安全的基本概念和原则.包括安全缺陷(Vulnerabilities)的检测,安全威胁(threats)的类别,数据安全的要求,和数据安全的防御措施. 1 信息安全的基本原则(碎碎念的概念) 作为一个计算机安全的专家,在学习技术领域的知识之前,他必须也要先理解那些重要的信息安全的原则!!!Computer security specialists must not only know the technical side of their jobs but also mu

《大道至简》第一章读后感(伪代码)

1.愚公移山: package 大道至简; import java.愚公移山.*; import java.util.*; public Class 愚公移山问题 { public static void main(String[]  args) { int 愚公生命值=愚公生命值; float 山=山; int 子孙=子孙后代生命值; while(山!=0) { while(愚公生命值!=0) { 山=山-1; 愚公生命值=愚公生命值-1; if(山==0) { System.out.prin

HDFS HA架构以及源代码引导

HA体系架构 相关知识介绍 HDFS master/slave架构,HDFS节点分为NameNode节点和DataNode节点. NameNode存有HDFS的元数据:主要由FSImage和EditLog组成. FSImage保存有文件的文件夹.分块ID.文件权限等,EditLog保存有对HDFS的操作记录. DataNode存放分块的数据,并採用CRC循环校验方式对本地的数据进行校验,DataNode周期性向NameNode汇报本机的信息. NameNode单点故障:HDFS仅仅有一个Name

设计模式(五)学习----装饰模式

装饰设计模式:又名包装模式(Wrapper),以对客户端透明的方式扩展对象的功能,是继承关系的一种替代关系.装饰模式以对客户端透明的方式动态的给一个对象附加上更多的责任.换言之,客户端不会觉得在装饰前和装饰后有什么不同. 下面看装饰设计模式的UML类图: 在装饰模式中的各个角色: 抽象构件Component:给出一个抽象接口,以规范准备接收附加责任的对象. 具体构件角色:ConcreteComponent:定义一个要接收附加责任的类. 装饰角色:Decorator,实现抽象构件的接口,拥有一个抽

一个基本的自定义类加载器

实现自定义类加载器的三步: 1.继承ClassLoader 2.重写findClass()方法 3.调用defineClass()方法 一个基本的自定义类加载器代码如下: package cn.xpleaf.coding.c4; import java.io.*; /** * @author xpleaf * @date 2019/3/10 11:10 AM */ public class CustomClassLoader extends ClassLoader { @Override pro

mysql的text字段长度 mysql数据库中text字段长度不够的问题

分享一下我老师大神的人工智能教程吧.零基础!通俗易懂!风趣幽默!还带黄段子!希望你也加入到我们人工智能的队伍中来!http://www.captainbed.net 类型是可变长度的字符串,最多65535个字符: 可以把字段类型改成MEDIUMTEXT(最多存放16777215个字符)或者LONGTEXT(最多存放4294967295个字符). MySQL supports 4 TEXT field types (TINYTEXT, TEXT, MEDIUMTEXT and LONGTEXT)