[Turn] Postman return determination result is correct, local data interfaces and data comparison

Disclaimer: This article is a blogger original article, follow the CC 4.0 by-sa copyright agreement, reproduced, please attach the original source link and this statement.
This link: HTTPS: //blog.csdn.net/hqsary/article/details/81940077
----------------

An interface for, when we know the parameters corresponding to the parameters and the results can be judged by Postman, to verify whether the data is equal to the expected return data. This allows us to test the interface more convenient and simple.

1, prepare data.

postman can accept a file format as shown, may be generally we need parameters of the data stored in excel format, as shown, when the postman in the read data, the read column names as shown below according to the first row taken, and excel, for a plurality of parameters may be separated by commas, txt not so troublesome.

FIG, city, and the interface parameters are needed income, Insurance_ability and score is the value returned is expected, of course, may not correspond with the interface returned by the parameter name.

Read data postman in time, it will be considered a default table data. Therefore, when reading data, the read data such as city this row is written as data.city. postman reads the data until the data is empty according to the read mode data.city.

Note that: using the saved excel data when saving, please save as csv format, and use Notepad ++ to convert it to the format utf-8, otherwise identified in the postman is garbled format, judge of the data caused by mistakes.

2, postman parameters

In this interface, we need parameters of the parameter is Income and city, for both parameters, including the use of {}.

3, before the request is sent, the parameter setting
we need the data in the table is parameterized, where before sending a request, we want the postman data file can be read in the postman, Pre-request Script statements postman will be executed before the request is sent, so read the statement of operations, we execute on here.

Code statement read operation.

var city = data.city;
var income =data.income;

4, judgment operation in Tests in
the fourth step should be the most important operations, and now we have to read the data file before sending the request, then how to judge it? Tests carried out in this step we are in, Tests will be performed after the request, Tests also known as assertions.

How to judge whether the returned data is correct thinking is:

1, the document expected results set environment variables

2, the returned data acquired

3, the determination

var score=(data.score);//获取文件中数据
var Insurance_ability=(data.Insurance_ability);//设置成环境变量
pm.environment.set("score",score);
pm.environment.set("Insurance_ability",Insurance_ability);
 
 
if(tests["code is 200"]=responseCode.code===200){ //[postman断言输出，返回200打印code is 200
    var d = JSON.parse(responseBody); //将返回数据解析成json格式
if( d.得分==score&&d.保险能力==Insurance_ability){ //将返回数据与环境变量进行对比，如果同时满足
    var a=1; //因为pm.test语句中含有function（）设置满足条件是a=1,
    pm.test("测试通过",function(){ //if中判断为真，则断言输入测试通过
        if(a==1);
    });
}
else{
    var a=0;
    pm.test("测试不通过",function(){
    if(a===0);
    });
}
 
}else if(tests["code is 500"]=responseCode.code===500){//状态为500时断言输出
    
}else{
    tests["code is 400"]=responseCode.code===400 //状态为400时断言输出
}

5、参数化执行

以上工作准备完成之后，我们就可以开始正式的参数化执行了

1、点击Runner

2、接口选择，环境准备，选择文件

3、点击Run,开始执行

4、检查执行结果

在数据跑完之后，我们可以直接在页面看到结果，其中pass和failed是代码tests中断言通过的数量，其中，我们可以看到测试通过，和code is 200的提示，

我们看到failed数量为1，我们找到该请求

看到coed is 200 状态为FALL，coed is 500状态为pass，可以判断该请求状态码是500

此时就可以查看该请求的请求地址和返回参数，再对比数据，查看具体是哪里的错误。

打开请求地址和返回内容，可以看到是参数我们传递错了，该原因应该是我表格中这行数据为空，所以他直接传递的{{ciry}}和{{income}}，

总结
以上就是如何使用postman判断返回结果是否正确的左右过程，如果有任何写的不好的地方，望指教。

除此之外，也遇到问题，比如在数据量大的时候，再去对比结果，就会比较麻烦，并且tests中的输出只是一个断言输出，无法通过该运行的通过或者不通过去判断有多少用例失败，所以这个是还需要研究的。

大家也可以通过在tests中的各种操作和打印使得自己能通过判断输出就判断出是哪里数据的问题。

更新
小编写了一个小时的更新，按了下谷歌的control+s，页面就崩，小编很无奈，只能重新写。

之前我们已经实现了对预期数据和返回数据做一个基本的判断，但是不能简要的从运行结果查看到是通过了多少测试用例数据，没有通过多少测试用例数据，并且去查找错误数据的时候还是很麻烦。

查阅资料发现，针对tests语句，我们是可以设置他的结果的，就是设置通过或者不通过，所以我们新的思路就是：

1、对预期结果和返回结果进行判断

2、判断相等则设置tests语句为true

3、判断不相等则设置tests语句为false，即失败，并且打印参数、预期结果、实际结果，这样我们简要的从runner运行界面就知道我们那些数据不对，就不用花大量的时间查找那些数据出错。

语句如下：

tests["测试通过"]=true;   //设置为true,运行时表示pass
tests["测试不通过"]=false;  //设置为false,运行时表示fail

实现整个判断的语句如下：

var city = data.city;
var income =data.income;
var score=(data.score);
var Insurance_ability=(data.Insurance_ability);
pm.environment.set("score",score);
pm.environment.set("Insurance_ability",Insurance_ability);
 
var d = JSON.parse(responseBody);
 if(d.得分==score&&d.保险能力==Insurance_ability){
        tests["测试通过"]=true;
    }else{
        tests["测试不通过"+"(输入参数"+city+'、'+income+")预期结果（得分="+score+"、保险能力"+Insurance_ability+"）（实际结果得分="+d.得分+"、保险能力="+d.保险能力+')']=false;

}

如图，数据表中，存在19个用例，通过17个，失败两个，并且针对失败的用例，清晰的可以看到参数、预期结果以及返回结果。

如果数据较多的时候，我们可以直接通过运行结果页面的红色方块（失败区域）展示的用例查找失败的数据。

代码更新：

针对接口出现的问题，简要的分为三种情况，服务器问题、客户端问题、数据校验，所以更新了一下验证代码

var state=responseCode.code;//获取返回状态
var number=(state.toString()).substr(0,1);//将返回的number类型转为string类型，并获取第一位
switch(number){
case '2':
    test();
    break;
case '4':
    clientQue(); //4开头的状态，简单定义为客户端问题
    break;
case '5':
    serverQue(); //5开头的状态，简单定义为服务器问题
    break;
default:
    tests['测试不通过,状态='+state]=false;  //如出现其他情况，则打印状态，并测试不通过
    break;
}
 
function test(){ //状态为200执行的函数
var city = data.city;
var income =data.income;
var score=(data.score);  //可使用request。url获取url，解析参数字段
var Insurance_ability=(data.Insurance_ability);
 
var result = JSON.parse(responseBody);
 if(result.得分==score&&result.保险能力==Insurance_ability){
        tests["测试通过"]=true;
    }else{
        tests["测试失败"+"(输入参数"+city+'、'+income+")预期结果（得分="+score+"、保险能力"+Insurance_ability+"）（实际结果得分="+result.得分+"、保险能力="+result.保险能力+')']=false;
       
        }
}
 
//客户端问题
function clientQue(){
    tests['客户端问题(请求参数或方式错误)---测试失败---状态码为'+state+'   requestURl为'+request.url]=false;
}
 
//服务器或者网关问题
function serverQue(){
    tests['服务器或网关问题---测试失败---状态码为'+state+'   requestURl为'+request.url]=false;

}