You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 3 Next »



Overview

Postman, more than a utility, is a collaboration platform for developing APIs.

Normally, it is used as a way to quickly interact with existing APIs without having to code HTTP requests by hand.

It provides support for HTTP based APIs, including REST and GraphQL.

Postman also provides the ability to write tests and use Chai assertions, as seen on these Postman test examples.

With Postman, comes also a built-in (test) collection runner; it is also possible to execute tests from the outside, using a CLI tool named Newman.

Concepts

  • request: an API request (e.g. HTTP POST on some URL, with some values)
    • authentication: authentication for the API request (e.g. HTTP basic auth, etc); can be defined at multiple levels and inherited
  • collection: a way of grouping multiple requests
  • folders within the collection: a way to better organize requests within the collection
  • variables: can be defined at multiple levels (e.g. global, collection, environment, local, ...)
  • test: a test; can be defined at request, folder or collection level
  • pre-request script: some code execute before each test; can also be defined at request, folder or collection level
  • environment: an abstraction of some test environment that describes a context for running the requests; it consists of one description plus a set of variables with their corresponding values

Implementing tests

Tests can be implemented using Javascript and making use of Postman APIs/objects supported by Chai assertions. 

One or more tests can be defined at the request level, or even at the whole collection level.

Pre-request scripts may be useful as a means to initialize some data before the test or to implement some test setup code.

Variables

Authentication

Requirements

  • Postman
  • Jenkins plugin (optional)´

Example

Postman Echo API

In this example, we're going to use Postman' sample Echo API as a way to showcase some tests and their integration with Xray.

The Postman Echo API provides a set of endpoints that we'll exercise.


We start by cloning an existing Postman collection from a template and importing it to Postman.

 


The collection contains a request per each endpoint, where each request has one or more tests.

   

In the previous example, we can see two tests: one for validating a successful HTTP request based on the status code and another that checks the response's JSON content. 


The collection (or a subset of its tests) can be run using the Collection Runner.

    

The runner shows the overall count for the number of passed and failed tests. We can also see the assertion error on failed tests; in this case, saving the response (setting the proper flag above) can help us better understand what is happening.


Running the tests can also be done from the command line or from within Jenkins (or any other CI/CD tool). This can be achieved using Newman.

In order to run Newman, we need to provide a path or a URL to our collection.

In this case, we'll obtain a public link to it.

    


Then we need to decide which reporter to use. Newman provides a built-in JUnit reporter; however, better alternatives exist such as junitxray or junitfull.

; this will produce a XML based report (e.g. output.xml).


Which Newman reporter should I use?

The standard Newman junit reporter produces <testcase> entries in the JUnit XML report that can be misleading as tests will be identified on the Postman test description, which can be similar between different tests (e.g. "response is ok").

Therefore, two alternative reporters arise: newman-reporter-junitxray and newman-reporter-junitfull

"newman-reporter-junitxray" (simply known as "junitxray"), will create <testcases> per each request, which in the end will lead to corresponding Test issues in Xray. This means that there won't be explicit visibility for each Postman test on that request, as they will be treated just as one.

"newman-reporter-junitfull" (simply known as "junitfull"), on the other hand, will produce one <testcase> per each Postman test, which will lead to the same number of corresponding Test issues in Xray.

If you aim just to have high-level overview of the request, then "junitxray" reporter will be preferable; otherwise, "junitfull" may be a better option.



junitjunitxrayjunitfull
tests
  • 40 Tests (one per each PM test name/description)
  • 37 Tests (one per request)
  • 90 Tests (one per each PM test)
generic definition field

<collection>.<pm_test_description>


 "PostmanEcho.response is ok"

<collection>.<request_name>
 

"PostmanEcho.Object representation"

<folder_path>/<request_name>.<pm_test_description>
 

"Utilities / Date and Time / Object representation.response is ok"

notes
  • leads to a collision of tests made for different requests
  • ignores folder path, which can lead to the collision of requests having the same name
  • one Test issue per each PM test
  • ignores folder path, which can lead to the collision of requests having the same name
  • doesn't present the multiple PM tests
  • few Tests, one per each request
  • can lead to many Test issues
  • one Test issue per each PM test, identified by the full (folder) path of the request



newman run https://www.getpostman.com/collections/c7334a5cf52a90639a48 -r 'cli,junit,junitfull,junitxray' --reporter-junit-export postman_echo_junit.xml --reporter-junitfull-export postman_echo_junitfull.xml --reporter-junitxray-export postman_echo_junitxray.xml -n 1



Importing results is as easy as submitting them to the REST API with a POST request (e.g. curl), or by using one of the CI plugins available for free (e.g. Xray Jenkins plugin).



An unstructured (i.e. "Generic") Test issue will be auto-provisioned the first time you import the results, based on the name of the test case and of the corresponding test suites.

If you maintain the test case name and the respective test suites, the Test will be reused on subsequent result imports. You may always enforce the results to be reported against an existing Test, if you wish so: just specify its issue key as a tag.

Tags can also be used to cover an existing requirement/user story (e.g. “ROB-11”): when a requirement issue key is given, a link between the test and the requirement is created during the results import process. 

Otherwise, tags are mapped as labels on the corresponding Test issue.






A Test Execution will be created containing results for all test cases executed. In this case, you can see that it is also linked back to an existing Test Plan where you can track the consolidated results from multiple "iterations" (i.e. Test Executions).


Within the execution screen details, accessible from each row, you can look at the Test Run details which include the overall result and also specifics about each keyword, including duration and status.



References

  • No labels