Interview with RakAPIt development team leader!

I interviewed the leader of the RakAPIt development team！

Hello everyone.
I'm Shirahama from JPQ.

We are pleased to announce the release of the RakAPIt beta version of our API test automation service.
We are offering this service free of charge until September 2020, so please give it a try.

In this issue, we present an interview with Mai, leader of the RakAPIt development team.

(Shirahama) First of all, What is RakAPIt?

(Mai) RakAPIt is a testing service for APIs.
While there are probably other tools for APIs, RakAPIt is unique in that it automatically generates test cases on the web, automatically executes the created cases, and allows test case management on the web.
By simply entering API specifications, test cases are automatically generated, increasing the coverage of API testing and making it less time-consuming.
For more detailed information on how to use the software, please visit the tutorial on the official website!

(Shirahama) Thank you. By the way, how did you come up with this name RakAPIt?

(Mai) The team members came up with this name together. We named it RakAPIt from "Raku(楽) API Testing" to mean "easy" and "fun" API testing. It reads as "Ra-ka-pit!"
※The kanji 楽 in Japanese is read as "ra-ku" that means easy. This word is also used in the word “楽しい”, which means fun.

(Shirahama) That's easy to remember! How did you decide to develop this service this time?

(Mai) RakAPIt is a tool I created to make API testing easier.
There was a time in the past when I developed APIs. While test analysis and test design are interesting in themselves, the implementation was very time-consuming and difficult.
We started planning this service with the idea of having a computer do the labor-intensive and unskilled part of the work, so that humans could focus on more creative things.
Since our company specializes in testing, we are sometimes asked to perform API testing, which is a time-consuming process that involves combining data patterns, creating item descriptions, and then writing the code.
In addition, API testing is a particularly repetitive process that requires a lot of human resources if you want to test it properly. We wanted to automate this process, and also to be able to share it with others by being able to manage it, so we decided to create a tool that would allow us to do everything in one place.

(Shirahama) If everything can be managed centrally on the Web, it can be unified within the company and easily shared even if members change. Do you have any tips on how to use RakAPIt?

(Mai) Recently, when developing APIs, I think it is common to write API definitions in Swagger.
Since RakAPIt is Swagger-compatible, you can simply have the definitions entered as they are, and it will even auto-execute them.
When automatically executing the generated test cases, you can choose between the all-pair method and the all-combination method, so we recommend using the all-combination method to improve coverage if possible, but depending on the number of cases, you may want to use the all-pair method to focus on the necessary parts for more efficient execution.

*All-pairs method:
Refers to a combination test in which all combinations of all levels are entered at least once in all pairs of factors. The test is not an "All pairs" test, but rather a "pair-only combinations" test.
(Source：https://www.atmarkit.co.jp/ait/articles/1506/11/news013_2.html )

When entering data patterns, it would be good to have the Boundary Value Analysis and Equivalence Partitioning Techniques used. In the future, additional development will be done so that data patterns and assertions are also entered automatically.

(Shirahama) What do you think about the future of test automation?

(Mai) I am interested in AI-based test automation.
RakAPIt is still in the process of additional development, but we hope to include AI to do the automatic entry of data patterns and assertions. At the moment, we are following the Swagger file, so we cannot extract anything other than what is described in the Swagger file. By incorporating AI, it may be possible to predict things that are not mentioned in the Swagger file, such as implicit specifications.

Thanks you very much.
We would like to encourage all development engineers who are in charge of API testing to use it.
We would appreciate it if you could try it once and contact us with any feedback for future reference for additional development.