1. Division of equivalence class  

    Divide the input field of the program into several parts , Then select a few representative data from each part for testing .

    Effective equivalence class ： It is reasonable for the tested object 、 meaningful 、 Acceptable input . In a nutshell Correct data .

    Invalid equivalence class ： It is unreasonable for the tested object 、 Meaningless 、 Unacceptable input . In short, that is bad data .

    give an example ： The length of the specified user name is 6-18 position , If the input data is within this length , It's the efficient equivalence class . Not in this range （ Less than 6 Bit or more 18 position ）, Is an invalid equivalence class . therefore , When designing use cases , In response to this request , One forward use case and two reverse use cases can be designed .

    Use case design method ：

        Design Forward use cases when , Make it Cover as many valid equivalence classes as possible . Repeat this step , Until all valid equivalence classes are covered by test cases .

        Design Reverse use case when , Make it Only one invalid equivalence class is overwritten . Repeat this step , Until all invalid equivalence classes are covered by the use case . With the thinking of scientific experiment , Namely Control variable method .

2. Boundary value analysis

    This is usually used as a supplement to the equivalence class division method , Mainly for Boundary value of input or output To test .

    namely ： If the input condition specifies the value range , Then take the boundary value and the value closest to this boundary to test . 

    give an example ： Assume that the value range of age is [18,30], Then its boundary value is 18,30, The numbers closest to these two boundaries are 17,19 and 29,31. therefore , This can be tested as six boundary values . If the test time is tight , Or you can just take 17、18 and 30、31.

    If it's an open interval （18,30） Words , Because age can only be rounded , So essentially , Its value range is [19,29] Between , That is, its boundary value is 19,29. The nearest number to the boundary is 18,20 and 28,30. Thus we can see that , Under this condition, the boundary value of age has 18,19,20 and 28,29,30.

    Some common boundary values ： The second part of the cycle 0 Time 、 for the first time 、 The penultimate and last ; The first and last elements of an ordered set （ Such as the drop-down list ）; Having some development knowledge can also help us dig out the hidden boundary value conditions , Such as java in byte The value range of type is [-128,127] etc. .

3. Decision table （ Decision table ）

    It is mainly used for A combination of multiple conditions And result analysis .

    Design steps ：

    （1） Determine the number of rules

            Suppose there is n Input conditions , The rules are 2^n individual （ Because each condition has only two values , True or false ）, This is the number of columns in the initial decision table .

    （2） List all condition piles and action piles

            Condition stake ： All input conditions

            Action post ： All output conditions

    （3） Fill in the condition item and action item , Get the initial decision table

            Condition item ： The specific value of this condition ,0 or 1

            Action items ： The output result obtained by the combination of all condition items

    （4） simplify 、 Merge similar rules or actions

            The rules ： A column is a rule

    give an example ： see Qianfeng test course P66- Decision table analysis , Or find it online by yourself .

    notes 1： There can only be two conditions , If not , Then the decision table... Does not apply .

    notes 2： Sometimes there are too many conditions , It's impossible to finish the whole list or test , We need to optimize . The optimization strategy is uniform coverage , Make each condition take true and false values . for example ：

4. Cause and effect diagram

    According to the combination of input conditions 、 Constraints and output conditions Causal relationship , Various combinations of input conditions are analyzed , So as to design test cases .

    The advantage of cause and effect diagram is Find deficiencies in the design . For some cases with only input and no output , This is a kind of defects , Can't be designed as a test case . Use case writing should focus on requirements , There are no output results in the requirements , Then you can't write it out . You can talk to the product manager first , Perfect the needs , Then add use cases . exemplars ： Qian Feng Test course P63-- Vending machine case .

    Cause and effect diagram example ：  

      notes ： Cause and effect diagrams and decision tables are Based on positive thinking Use cases for design . Although both of them are multi combination analysis for input conditions , But the cause and effect diagram is more Focus on the relationship between input conditions 、 Causality between input and output .

5. Orthogonal test method

    factor 、 factors ： The amount that affects the experimental results

    level ： The value of the factor

    The orthogonal test method is only applicable to The number of levels is the same The situation of . If the factor A Yes 3 Species value , factor B Yes 4 Species value , Then this situation is not within the scope of application of the orthogonal test method .

    Design steps ：（1） Analyze all factors that affect the results

                      （2） Analyze the number of levels of each factor

                      （3） Choose the right orthogonal table , Find the closest . Generally speaking , Will be greater than the actual number of factors and levels

    Tools ： Orthogonal design assistant . Input factors and levels , Can automatically generate orthogonal results .

6. Scene method

    be based on Business scenario （ The user's operation process , Right or wrong ） To design test cases .

    Basic flow ： All operations are correct , The process that can finally achieve the goal .

    Alternative streams ： Some operations are wrong , This leads to repeated processes , But in the end, the process that can achieve the goal .

    Abnormal flow ： Operation error leads to the process that fails to achieve the goal in the end .

    give an example ： Go to ATM Machine withdrawal .

        ① The card inserted is right , It's a bank card that can be used ; The password is entered correctly , The amount of money entered does not exceed the balance of the card and ATM There's so much money in the plane , Finally got the money . This is the basic flow , There is no problem with the operation of the whole process .

        ② Wrong card , Reinsert the right card ; Wrong password , Lose again , by the way ; The amount of money entered does not exceed the balance, nor does it exceed ATM The money in the plane , Finally, I took out the money . This is an alternative stream , Although there was a mistake on the way , But in the end, the goal was achieved .

        ③ The card is right , The password was entered incorrectly three times , The card is frozen , Can't withdraw money . This is abnormal flow , The business process has not reached the end .

    Use case design steps ：（1） Find out the basic flow of the program （ The right process ） And alternative streams 、 Abnormal flow  （ The wrong link ）

                             （2） Analyze business processes , Draw a flow chart ;

                             （3） A path is a scene .

7. Function diagram method

    It's a little complicated. , Don't know much about , Too lazy to write. , Let's see later .

8. Wrong guess

     Infer errors in the program based on experience and intuition , So as to design test cases .

9. Other design methods

    9.1 Test outline method

            Focus on needs , List various test conditions , Translate requirements into outlines .

            Tree mind map , A test case starts from the root node to the leaf node . therefore , This method does not need to write specific test cases .

    9.2 Exploratory testing

            Based on experience and intuition

    9.3 Monkey test

            Unconscious behavior , Mess up like an ignorant monkey , Found some unexpected mistakes .

            You don't have to write use cases .