table of Contents
Test methods
Equivalence partitioning method
Effective equivalence class
Description of the standard procedures of meaningful and reasonable input data collection
If the user enters a valid data equivalence class, the program should be calculated correctly performed
Factors to consider summary
- It can not be empty - empty
- Not repeat - repeat
- Data has a range of - out of range
- The number of characters range - out of range
Invalid equivalence class
Specifications for program descriptions meaningless, unreasonable set of input data
If the user enters data invalid equivalence class, the program gives an error message or do not allow users to input
Considerations summary
Data Format (style) - Style illegal
Such as: Integer - non-integer (letters, characters, symbols, decimal)
Digital - Non-digital
Decimal requirements - illegal digits after the decimal point
Boundary value method
- Boundary value data belong to an equivalence class is essentially a data category
Cause and effect diagram method
Cause and effect diagram of the "cause" - input conditions
FIG cause "fruit" - output condition
Cause and effect diagram method to consider
Mutual restraint relations and relationships of all combinations of input conditions
The output of the dependence of input conditions, that is, what kind of input combination will produce what kind of output that "a causal relationship"
The basic symbol
Constraints symbol
Method determination table
Draw cause and effect diagram is only an auxiliary tool, eventually obtained by analyzing the decision table, then write test cases through the decision table. But sometimes draw cause and effect diagram is very troublesome, affect test efficiency, so skilled in the application later, you can write directly to decision tables, and then writing test cases.
The orthogonal collocation method
Method can be orthogonally arranged with a minimum test procedure set maximum test coverage. When the input data or may be a combination of a large number of input data, since it is impossible to create test cases for all combinations of each of the input, this method may be employed.
French scene
Method scenario is when the user operates the software simulation scenario, the main test system for business processes. When a test to get a job, we are not concerned about the first boundary value of a control, whether equivalence classes to meet the requirements, but must first focus on whether its main functions and business processes implemented correctly, which requires the use of law to the scene complete test. When Business Process Testing is no problem, which is the main feature of the software is no problem, we'll focus on testing the boundaries of control in terms of value, equivalence class, etc.
The basic flow
An operation according to the correct path to achieve business process (simulation of the correct process)
Alternative Flow
导致程序出现错误的操作流程（模拟错误的操作流程）
测试大纲法
在一个程序或程序的某个模块中，涉及到多个窗口，每个窗口中能够完成多个动作，这些窗口又相互联系。为了弄清窗口和窗口之间的关系，或者说动作和动作之间的关系，可以使用测试大纲方法。
- 找出所有的窗口以及每个窗口的输入动作
注意窗口之间的先后顺序 - 找到各个窗口之间的联系，并据此编写测试用例
测试方法选择的综合策略
- 为了测试程序的业务逻辑、业务流程、主要功能的正确性，错误处理能力，使用场景法设计测试用例（其中包含等价类）。
- 需要输入数据的地方，进行等价类划分，包括输入条件和输出条件的等价划分，将无限测试变成有限测试，这是减少工作量和提高测试效率最有效的方法。
- 使用边界值分析方法补充用例。
- 如果程序的功能说明中含有输入条件和输出条件的组合情况，则一开始就可选用因果图法或判定表方法。
- 对于参数配置类的软件，要考虑各个参数之间的组合情况，使用正交排列法选择较少的组合方式达到最佳效果。
- 如果程序的模块涉及多个窗口，并有相关联的操作，可以使用测试大纲方法设计测试用例。