Postman结合Newman做自动化测试

Postman结合Newman做自动化测试
1. Newman安装
2. Newman使用
  1. 使用cli运行Newman
  2. 编写js运行Newman
附录

本文讲述了如何使用Newman配合Postman进行自动化测试;

Postman结合Newman做自动化测试

在Postman通过Collection进行一组测试的基础上，Postman官方提供了Newman这个命令行工具，可以通过Newman对Collection导出的测试集进行测试

Newman的Github仓库：

https://github.com/postmanlabs/newman

关于Postman的用法以及Postman中的tests脚本编写见：

Postman简明教程

Postman中Tests脚本总结

这里不多做介绍；

Newman安装

Newman依赖于node.js > v10，并且需要npm；

关于Node.js的安装参考Node官网：

https://nodejs.org/en/download/package-manager/

安装Newman：

通过npm即可很方便的安装newman：

npm install -g newman

安装完成后，可以执行下面的命令测试是否成功安装：

newman -version
5.0.0

Newman使用

Newman支持两种使用方式，一种是通过命令行+参数的方式进行测试，另一种是编写js文件，并通过node运行测试，下面分别介绍这两种方式；

使用cli运行Newman

例如我从Postman中导出了api-pl-kylinkzhang-helloworld.proto.postman_collection.json的测试集以及对应的测试环境local-test.postman_environment.json；

则可以简单通过下面的命令进行整体测试：

newman run api-pl-kylinkzhang-helloworld.proto.postman_collection.json --reporters cli --environment local-test.postman_environment.json

上面的命令通过newman run collection.json指定Postman导出的Collection测试集；

通过--reporters指定测试结果输出的格式，如需要输出多种结果可使用逗号分割，类似于： --reporters cli, html

通过--environment environment.json指定Postman导出的测试环境文件；

执行完成后，测试接口和测试结果都会列出在Terminal中，如下：

api/pl-kylinkzhang-helloworld.proto

→ HelloWorld Stub
  POST http://9.134.243.6:8080/HelloWorld [502 Bad Gateway, 531B, 95ms]
  1. status code test
  2⠄ JSONError in test-script

→ HelloWorld Stub2
  POST http://9.134.243.6:8080/HelloWorld [502 Bad Gateway, 531B, 62ms]
  3. status code test
  4⠄ JSONError in test-script

┌─────────────────────────┬───────────────────┬──────────────────┐
│                         │          executed │           failed │
├─────────────────────────┼───────────────────┼──────────────────┤
│              iterations │                 1 │                0 │
├─────────────────────────┼───────────────────┼──────────────────┤
│                requests │                 2 │                0 │
├─────────────────────────┼───────────────────┼──────────────────┤
│            test-scripts │                 2 │                2 │
├─────────────────────────┼───────────────────┼──────────────────┤
│      prerequest-scripts │                 0 │                0 │
├─────────────────────────┼───────────────────┼──────────────────┤
│              assertions │                 2 │                2 │
├─────────────────────────┴───────────────────┴──────────────────┤
│ total run duration: 377ms                                      │
├────────────────────────────────────────────────────────────────┤
│ total data received: 694B (approx)                             │
├────────────────────────────────────────────────────────────────┤
│ average response time: 78ms [min: 62ms, max: 95ms, s.d.: 16ms] │
└────────────────────────────────────────────────────────────────┘

  #  failure                  detail

 1.  AssertionError           status code test
                              expected response to have status code 200 but got 502
                              at assertion:0 in test-script
                              inside "HelloWorld Stub"

 2.  JSONError                Unexpected token '<' at 1:1
                              <h1>无法访问此网站</h1>9.134.243.6:8080
                              拒绝了您的连接请求。<br><br>请试试以下办法：<ul><li>确认目标服务器是否宕机：
                              ^
                              at test-script
                              inside "HelloWorld Stub"

 3.  AssertionError           status code test
                              expected response to have status code 400 but got 502
                              at assertion:0 in test-script
                              inside "HelloWorld Stub2"

 4.  JSONError                Unexpected token '<' at 1:1
                              <h1>无法访问此网站</h1>9.134.243.6:8080
                              拒绝了您的连接请求。<br><br>请试试以下办法：<ul><li>确认目标服务器是否宕机：
                              ^
                              at test-script
                              inside "HelloWorld Stub2"

可以通过--reporters指定多种形式的输出，官方提供的格式有：cli, json, junit, progress and emojitrain；

第三方例如：newman-reporter-html提供了html的测试结果输出格式，可以使用npm install -g newman-reporter-html安装，并通过下面的命令生成测试结果：

newman run api-pl-kylinkzhang-helloworld.proto.postman_collection.json --reporters cli,html --environment local-test.postman_environment.json --reporter-html-export result.html

当然你也可以自己编写相应的reporter，详见：lib/reporters

更多的，官方提供的所有命令行参数如下：

-h, --help output usage information

-V, --version output the version number

-c, --collection [file] Specify a Postman collection as a JSON [file]

-u, --url [url] Specify a Postman collection as a [url]

-f, --folder [folder-name] Run a single folder from a collection. To be used with -c or -u

-e, --environment [file] Specify a Postman environment as a JSON [file]

    --environment-url [url] Specify a Postman environment as a URL

-E, --exportEnvironment [file] Specify an output file to dump the Postman environment before exiting [file]

-d, --data [file] Specify a data file to use either json or csv

-g, --global [file] Specify a Postman globals file [file]

-G, --exportGlobals [file] Specify an output file to dump Globals before exiting [file]

-y, --delay [number] Specify a delay (in ms) between requests

-r, --requestTimeout [number] Specify a request timeout (in ms) for requests (Defaults to 15000 if not set)

-R, --avoidRedirects Prevents Newman from automatically following redirects

-s, --stopOnError Stops the runner with code=1 when a test case fails

-j, --noSummary Doesn't show the summary for each iteration

-n, --number [number] Define the number of iterations to run

-C, --noColor Disable colored output

-S, --noTestSymbols Disable symbols in test output and use PASS|FAIL instead

-k, --insecure Disable strict ssl

-l, --tls Use TLSv1

-N, --encoding [encoding-type] Specify an encoding for the response. Supported values are ascii,utf8,utf16le,ucs2,base64,binary,hex

-x, --exitCode Continue running tests even after a failure, but exit with code=1. Incompatible with --stopOnError

-o, --outputFile [file] Path to file where output should be written [file]

-O, --outputFileVerbose [file] Path to file where full request and responses should be logged [file]

-t, --testReportFile [file] Path to file where results should be written as JUnit XML [file]

-i, --import [file] Import a Postman backup file, and save collections, environments, and globals [file] (Incompatible with any option except pretty)

-p, --pretty Enable pretty-print while saving imported collections, environments, and globals

-H, --html [file] Export a HTML report to a specified file [file]

-W, --whiteScreen Black text for white screen

-L, --recurseLimit [limit] Do not run recursive resolution more than [limit] times. Default = 10. Using 0 will prevent any variable resolution

可通过newman -h和newman run -h查看更多；

编写js运行Newman

如果你想更自由的定制自动化测试方案，可以通过编写js脚本并通过node执行来实现；

Newman提供了相当丰富的API供用户使用，具体可见：

https://github.com/postmanlabs/newman#api-reference

下面以一个简单的例子来展示如何编写js来调用Newman的API进行自动化测试；

const newman = require('newman');
const mail = require('./mail');

// 错误消息
var messages = "";

newman.run({
    collection: require('./data/api-pl-kylinkzhang-helloworld.proto.postman_collection.json'),
    reporters: ['cli', 'html'],
    environment: require('./data/local-test.postman_environment.json'),
    // 禁止SSL检查，并允许自签名的SSL证书
    insecure: true,
}).on('start', function(err) {
    if (err) { throw err; }
    console.log('running a collection..');
}).on('done', function (err, summary) {
    if (err) { throw err; }
    var failures = summary.run.failures;
    if (failures.length) {
        messages += "Test failed count: " + failures.length + "<br/><br/>"
        failures.forEach(e => {
            messages += "Test failed from: " + e.source.name + "<br/>"
                + "error case: " + e.error.test + "<br/>"
                + "error at: " + e.at + "<br/><br/>"
        });

        var mailOptions = {
          // 发送邮件的地址
          from: '"<username>" <sender_email>',
          // 接收邮件的地址
          to: "receiver_email",
          // 邮件主题
          subject: "项目自动化测试未通过",
          // 以HTML的格式显示，这样可以显示图片、链接、字体颜色等信息
          html: messages
        };

        mail.sendMail(mailOptions, (error, info = {}) => {
            if (error) {
              return console.log(error);
            }
            console.log('Test failed messages sent');
        });
    } else {
        console.log("项目测试通过!")
    }
    console.log('collection run complete!');
});

上面的js文件通过collection指定待测试数据集，reporters指定测试结果输出的格式，environment指定测试环境文件；

之后通过on方法指定各个阶段，以及各个阶段的回调函数。在整个测试过程中，在不同阶段会调用对应的回调函数；官方提供的阶段有：

Event	Description
start	The start of a collection run
beforeIteration	Before an iteration commences
beforeItem	Before an item execution begins (the set of prerequest->request->test)
beforePrerequest	Before `prerequest` script is execution starts
prerequest	After `prerequest` script execution completes
beforeRequest	Before an HTTP request is sent
request	After response of the request is received
beforeTest	Before `test` script is execution starts
test	After `test` script execution completes
beforeScript	Before any script (of type `test` or `prerequest`) is executed
script	After any script (of type `test` or `prerequest`) is executed
item	When an item (the whole set of prerequest->request->test) completes
iteration	After an iteration completes
assertion	This event is triggered for every test assertion done within `test` scripts
console	Every time a `console` function is called from within any script, this event is propagated
exception	When any asynchronous error happen in `scripts` this event is triggered
beforeDone	An event that is triggered prior to the completion of the run
done	This event is emitted when a collection run has completed, with or without errors

通过编写在测试不同阶段的回调函数，即可定制你自己的自动化测试方案；

例如上例中就通过调用mail模块封装的发送邮件接口完成了在测试未通过时向指定用户发送测试未通过邮件警告；

附录

如果觉得文章写的不错, 可以关注微信公众号: Coder张小凯

内容和博客同步更新~