如何在post man Tool中构建框架
请帮助我，我们是否有任何工具可以整合和运行该工具邮递员。
提前感谢!! 阿西
答案 0 :(得分：5)
Newman是邮递员的配套工具，可以从命令行运行邮递员测试。它还支持将newman用作库。我们使用它在我们的生产服务器上进行连续测试。
您可以使用邮递员作为该框架的一部分来开发您想要的任何框架。
以下是我用来测试服务器返回结果的519个已知良好值的测试。如果您在邮递员集合运行器中提供数据文件，或者您将其作为newman的命令行选项提供，则会填充数据。
tests["Status code " + responseCode.code] = responseCode.code === 200;
var response = JSON.parse(responseBody);
tests[ 'response.errorsCount == ' + response.errorsCount ] = response.errorsCount === 0 ;
var outputs = response.outputs ;
var outputsLength = outputs.length ;
if( data.hasOwnProperty("outputs") ) {
tests[ 'data.outputs.length == response.outputs.length ' ] = data.outputs.length == response.outputs.length ;
var dataOutputsLength = data.outputs.length ;
for ( index = 0; index < outputsLength ; index++){
var output = outputs[index] ;
var expectedOutput = data.outputs[index] ;
tests[expectedOutput.cell + ': ' + expectedOutput.value + ' == ' + output.cell + ': ' + output.value ] = expectedOutput.cell == output.cell && expectedOutput.value == output.value ;
}
}
以下是使用newman运行相同测试的示例。
newman --collection score-card-export-all-respondents.postman_collection --environment melbourne.postman_environment --data score-card-export-all-respondents-data.json --requestTimeout 60000
这是我们用来测试其中一台服务器的脚本。由./postman-runner.js调用 - 测试staging.json。存储库网址为postman-runner
#!/usr/bin/env node
var Newman = require('newman') ;
var ResponseExporter = require('newman/src/utilities/ResponseExporter') ;
var SparkPost = require('sparkpost');
var fs = require('fs') ;
var path = require('path') ;
var argv = require('minimist')(process.argv.slice(2)) ;
var errors = [] ;
var testResultsTests = [] ;
function EvaluateNewmanTestResults(options) {
arguments = eval(options) ;
if (!arguments){
arguments = {} ;
}
this.created = new Date() ;
if (arguments.name) {
this.name = name ;
} else {
this.name = this.created.toUTCString() ;
}
if (arguments.failed){
this.failed = failed ;
} else {
this.failed = [] ;
}
if( arguments.resultsPath){
this.resultsPath = arguments.resultsPath ;
} else {
this.resultsPath = null ;
}
this.tests = [] ;
this.urls = [] ;
this.runCount = 0 ;
this.failedCount = 0 ;
this.summary = function(){
return {name: this.name, urls: this.urls, runCount: this.runCount, failedCount: this.failedCount, failed: this.failed} ;
}
this.JSON = function(){
return JSON.stringify( this.summary ) ;
}
this.evaluateResults = function(results_){
this.tests = [] ;
this.testResults = results_ ;
this.overallResults = JSON.parse(fs.readFileSync(this.resultsPath, 'utf8')) ;
if ( this.overallResults.collection.name) {
this.name = this.overallResults.collection.name ;
}
for( testResultsIndex=0; testResultsIndex<this.testResults.length;testResultsIndex++){
testResult = this.testResults[testResultsIndex] ;
url = testResult.url ;
this.urls.push(url) ;
tests = testResult.tests ;
for (var key in tests){
value = tests[key] ;
this.runCount++ ;
if ( value ) {
/* passed */
} else {
/* failed */
this.failed.push(key) ;
this.failedCount++ ;
}
}
}
}
}
function notifyViaSparkPost( key, arguments, from, subject, html, notify, results, summary ){
holdArguments = arguments ;
arguments = JSON.stringify(arguments) ;
/* failed tests is swapped into the html*/
failedTests = JSON.stringify(summary) ;
/* swap out {} to [] to avoid errors when populating html */
failedTests = failedTests.replace(/{{/g, '[[') ;
failedTests = failedTests.replace(/}}/g, ']]') ;
/* collection name is swapped into the html*/
collectionName = summary.name ;
var regularExpression = /(?:{)(.*?)(?:})/g
match = regularExpression.exec(subject);
while (match != null) {
subject = subject.replace( match[0], eval(match[1])) ;
match = regularExpression.exec(subject) ;
}
match = regularExpression.exec(html);
while (match != null) {
console.log(match[0]) ;
console.log(match[1]) ;
html = html.replace( match[0], eval(match[1])) ;
match = regularExpression.exec(html) ;
}
html = html.replace(/\[\[/g,'{{') ;
html = html.replace(/\]\]/g,'}}') ;
var sparkPost = new SparkPost(key);
sparkPost.transmissions.send({
transmissionBody: {
content: {
from: from,
subject: subject,
html: html
},
recipients: notify
}
}, function(err, res) {
if (err) {
console.log('Unexpected error sending email notification');
console.log(err);
} else {
console.dir({sent:'email', 'with': holdArguments, 'and': failedTests });
}
});
}
function nextTest (arguments,sparkpostApiKey,tests,failed,callback) {
var test = tests.shift() ;
if (!test){
callback(failed,arguments) ;
} else {
handleTest(arguments,sparkpostApiKey,test,tests,failed,callback) ;
}
}
function handleTest(arguments,sparkpostApiKey,test,tests,failed,callback){
var description = test.description ;
var resultsJson = null ;
var ran = {} ;
var newmanOptions = {} ;
var holdArguments = arguments ;
if ( description ) {
console.log('') ;
console.log( 'Running ' + description ) ;
}
var collection = test.collection.join(path.sep) ;
var environment = test.environment.join(path.sep) ;
dataFile = test.data
if ( dataFile ) {
dataFile = dataFile.join(path.sep) ;
}
results = test.results.join(path.sep) ;
requestTimeout = test.requestTimeout ;
collectionJson = JSON.parse(fs.readFileSync(collection, 'utf8')) ;
environmentJson = JSON.parse(fs.readFileSync(environment, 'utf8')) ;
newmanOptions = {
envJson: environmentJson ,
iterationCount: 1, // define the number of times the runner should run
outputFile: results, // the file to export to
responseHandler: "TestResponseHandler", // the response handler to use
asLibrary: true, // this makes sure the exit code is returned as an argument to the callback function
stopOnError: false,
dataFile: dataFile,
requestTimeout: requestTimeout
}
arguments = {collection:collection, environment:environment, dataFile:dataFile, results:results} ;
notificationArguments = {collection:collection, environment:environment, dataFile:dataFile, results:results} ;
ouch = new EvaluateNewmanTestResults() ;
ouch.resultsPath = results ;
ouch.runCount = 0 ;
if ( argv.simulateTestFailure ) {
resultsJson = JSON.parse(fs.readFileSync(results, 'utf8')) ;
ran = { name: collection.name, runCount:0, failedCount:0, exitCode: exitCode, failed: ["Some example failed tests","and another"] } ;
notifyViaSparkPost(
sparkpostApiKey,
arguments,
test.notify.sparkpost.from,
test.notify.sparkpost.subject,
test.notify.sparkpost.html,
test.notify.sparkpost.recipients,
resultsJson,
ran
) ;
nextTest(holdArguments, sparkpostApiKey, tests, failed, callback) ;
} else {
/* clear the results from any previous run */
ResponseExporter._results = [] ;
Newman.execute(collectionJson, newmanOptions, function(exitCode){
ouch.evaluateResults(ResponseExporter._results) ;
console.dir(ouch.summary()) ;
if (!holdArguments.totalTestsFailed) {
holdArguments.totalTestsFailed = 0 ;
}
if (!holdArguments.totalTestsRun) {
holdArguments.totalTestsRun = 0 ;
}
holdArguments.totalTestsFailed = holdArguments.totalTestsFailed + ouch.failedCount ;
holdArguments.totalTestsRun = holdArguments.totalTestsRun + ouch.runCount ;
if (ouch.failedCount>0){
notifyViaSparkPost(
sparkpostApiKey,
notificationArguments,
test.notify.sparkpost.from,
test.notify.sparkpost.subject,
test.notify.sparkpost.html,
test.notify.sparkpost.recipients,
resultsJson,
ouch.summary() ) ;
}
nextTest(holdArguments, sparkpostApiKey, tests, failed, callback) ;
}) ;
}
}
if ( !argv.tests ) {
errors.push('--tests parameter is missing') ;
} else {
if( !fs.existsSync(argv.tests)){
errors.push( argv.tests + ' is an invalid path') ;
}
}
if ( errors.length > 0 ) {
console.dir({ errors: errors }) ;
} else {
fs.readFile( argv.tests, 'utf8', function( error, data) {
var tests = JSON.parse(data) ;
var sparkpostApiKey = tests.sparkpostApiKey ;
var run = tests.run ; // array of tests to run
var failed = [] ;
argv.totalTestsRun = 0 ;
argv.totalTestsFailed = 0 ;
nextTest(argv, sparkpostApiKey, run, failed, function(failed,arguments){
console.log('finished test runs') ;
if ( failed.length > 0){
console.dir(failed) ;
}
console.dir({ totalTestsRun: arguments.totalTestsRun, totalTestsFailed: arguments.totalTestsFailed } ) ;
}) ;
}) ;
}
答案 1 :(得分：4)
我不会依赖Postman作为测试框架的基础。
Postman非常适合调试和填充，但我怀疑它是与“持续构建集成”相结合的最佳工具。 ＆安培;提供全面的报告。
我会考虑调查Apache jMeter。
它有点复杂，但它确实有很多非常酷的功能，插件等，它可以轻松地与任何你想要的东西集成。它完成了邮递员所做的一切。
在我的项目中，我们使用Postman在编码完成时快速检查，以及可用性＆amp;负载测试 - jMeter。