#crestech

Soon after the worldwide release of OpKey Personal edition 1.0 we had a Sprint Review Meeting discussing how hectic the sprint was, what all assumptions went wrong, and what good decisions made the release possible. We decided that for the next few months, we will only be focusing on Client Issues and no further Feature Development will be entertained until widely required.

We were happy for the release, and thought that handling the client issues will not be much of a problem, and we will get enough time to rejuvenate from the recent hectic deadlines & work timings. A pleasure much awaited by all the IT Professionals.

Just before the Happy Ending of the review meeting, a young lady from the pre-sales team (probably a new joinee and a fresher) raised a rather rare concern, “OpKey was build for manual testers to do automation testing with quick authoring and speedy executions. But, I feel, that manual testers must be well aware of the keywords to author the TestCases, and authoring takes much more time than the current Selenium IDE. So Can we make TestCase authoring a little bit faster and seamless?”

We ignored the concern (as we had no idea how to do it then), but our Business Analyst was so moved by the concern, that he asked me, “Anshuman, what it takes to build a web recorder for OpKey?” And I was like, “Mmmm, a couple of years, may be”.

“I want the OpKey Recorder in the coming release, which is planned next month…”

So, here was a little concern, raised by a fresher that became a nightmare for the Development Team. But, the release was planned, and the requirement was crystal clear, “OpKey Recorder in one month”

I started with the .Net based WebBrowserControl (Embedded IE) but soon realized, that many-a-sites show up differently on older versions of IE and many-a-users still use IE 6. Then I started making an Extension for Google Chrome which will tell me about the page events, but was stuck handling issues with Google’s Security measures and container object handlings.

Then there came the life savior, Selenium IDE itself. The thought of extending Selenium IDE came from our Business Analyst himself with these backing up his point:

Selenium IDE is OpenSource, so you can hook-in your code-piece in it.

Six years of rich development, bug fixes and new version support.

Huge User base, who are comfortable with the UI &Usability offered by Selenium IDE.

 Support of ancient as well as newest HTML standards & controls.

Open Forums, multiple object identifiers etc etc.

So, twenty days of hardwork, and now we are proud to announce that OpKey supports Recording via Selenium IDE. In addition to Selenium IDE, the Selenium IDE for OpKey supports the following:

It fetches more object properties and element attributes.

Create separate entities for TestScripts & ObjectRepositories, to manage them better.

Intelligently adds keywords as per scenario, like SyncBrowser, VerifyPopup and then HandlePopup.

Promotes usage of more than one property to uniquely identify objects on a webpage.

Redirects runs to the default browser(By using the GlobalVariable feature of OpKey)

Creates more user readable scripts, like Click on a button and on a link is actually recorded as ClickOnButton & ClickOnLink respectively.

OpKey has now become much more usable and user friendly by having the Recorder feature in the automation tool itself. And now I too feel that record-n-Play is one of the coolest features of OpKey.

  Not only this, we charted out an experimental study with a closed group of users which showed us that we have reduced the test case creation time by more than 60% and that was a great achievement in itself. We also figured out generic queries which we believe our end user community will have, we know testers are inquisitive. So here’s a small faq list-

Q1- Selenium IDE for OpKey, where can I find it?

Download your copy from <link>SourgeForge</link> and drag-drop the xpi file onto your Mozilla Firefox. It will prompt you to install and then restart it.

Please Note, that the SeleniumIDE for OpKey is OpenSource and the source code is available on request

Q2- Which version of Selenium IDE is being used?

Currently we are supporting SeleniumIDE version 2.3. We are planning to release the versions 2.2 & 2.4 soon.

Q3- I have a different version of selenium IDE, so will the xpi file work?

Yes, the SeleniumIDE for OpKey must work as it is all together a different Plugin. Although, I would suggest that you create different Firefox Profiles and install different plugins.

Launchfirefox.exe –pfrom Run Window

Or follow this link: https://support.mozilla.org/en-US/kb/profile-manager-create-and-remove-firefox-profiles

Q4- I am not sure I understand the port number which I have to fill, what is it for?

A local machine port is needed for the communication of OpKey & SeleniumIDE for OpKey. You can give any random number ranging from 1 to 65535. The default port(8282) must just work fine.

Just make sure your firewall or antivirus is not blocking it.

Q5- I added the xpi file, and have followed all steps. But my actions are not getting recorded by OpKey what to do?

Can you check the following?

You have entered the same port number in OpKey Recorder & SeleniumIDE-> Tools-> OpKey Settings

You have clicked on Start Recording in OpKey Recorder

You have opened SeleniumIDE for OpKey in firefox and not the SeleniumIDE

  Q6- Will this work only for Firefox?

SeleniumIDE only works with Firefox, so, Yes. Although, we plan to integrate our solution with other recorders as well to support multiple browsers.

You can also ask firefox to behave as InternetExplorer or GoogleChrome or an AndroidBrowser. Check out this link: http://www.howtogeek.com/113439/

Q7- So you are saying for each selenium action; there is an OpKey keyword, where can I find that mapping?

Document is in progress. We shall update it on the website soon.

Q8- Can I add the objects of different test case recording in one Object Repository?

Select the same ObjectRepository for different TestCases while recording.

Q9- Where is the recorder located, I can’t find it?

Soon after the worldwide release of OpKey Personal edition 1.0 we had a Sprint Review Meetingdiscussing how hectic the sprint was, what all assumptions went wrong, and what good decisions made the release possible. We decided that for the next few months,…

As we see the current trend we understand that business is accelerating, and it is accelerating at the Speed of Thought! Well exciting times for business has always meant a huge load on its supporting functions Testing being the major chunk of it. Challenges most commonly faced by our sector are-

We as always have a Limited Test Budget

To add to it a strictly Time bound Schedule

A Large number of test cases to be executed on many environments with different tools

All this Leading to high costs incurred per tests

Geographical distribution of users, add to the challenges.

  Let us take a real time scenario – Automating an E Commerce Application Using Selenium WebDriver

Now Regression Testing on one Browser OS combination takes close to 5 hours. This has to be tested on 16 different combinations of browser and OS and Two parallel test environments with each release. This sounds like one mammoth effort! One way to achieve this is – “Create your Own Test Lab”.

      You can get 30 odd machines or VM’s having different operating systems and browsers. Ask your IT team to ensure that its health is always up to date. Timely updates, handling crashes and providing 24/7 support for the ever changing business requirements. So much effort to be put in a solution which is going to be occasionally used and mostly to report the Failures! Well seems like we should look for an alternative solution.

Another alternate solution is to move your functional test automation to cloud, where we basically execute our created test scripts to identify defects in application on cloud instead of our local infrastructure set up.

This approach provides many benefits, in terms of -

On Demand Infrastructure-

The infrastructure is always available On Demand and is highly scalable and flexible to our business test requirements. We can use as many browsers and OS combinations as we like. Moreover their maintenance overhead is not incurred by us.

Test Run Time

The Test run time is greatly reduced as the Cloud allows Parallel Test Runs on different browser and Os combinations.  And once you have scheduled your runs, your machine is free to work upon for other activities. Thus helping saving test time in total.

Cost

We are no longer bound by any expensive seat licensing solutions as we Pay Per Use only. In cloud test automation, you pay only for those minutes which you need to run your test suite. This causes a significant reduction in the cost incurred for test automation.

There are many big players who are trying to make foray into this area. CresTech is one of the first to launch a cloud enabled solution. CresTech most sought after Tool Agnostic Test Automation Platform Solution OpKey is now – Cloud Enabled! So, create your Automation Test without any scripting with any test tool and run it on cloud.

  With the OpKey Cloud, you author your test scripts on your local machine. And for execution you can chose from the 100+ configurations available of operating systems and browsers to run your test suite on. These tests will be executed in parallel on the cloud environment.

With OpKey, you do not have to worry about knowing any programming language. With its more than 400 keyword available, you can author test scenarios with ease. It also provides in built object and test data management interface.  And with OpKey recorder, an extension of Selenium IDE, create your test case with ease as you navigate steps while recording. Your first test case to run on cloud is ready within moments of installing and setting OpKey Personal edition. Moreover you are provided 100 FREE Automation minutes on Cloud.

Functional Test Automation – Let’s do it the Cloud Way

As we see the current trend we understand that business is accelerating, and it is accelerating at…

OpKey Integration with ALM Tool QC

Application Lifecycle Management

Abbreviated as ALM, Application Lifecycle Management refers to the…

OpKey Integration with CI Tool Hudson

OpKeycan be easily integrated with Continuous Integration tools to automate the execution process.…

Multi-Tool Test Automation Platform

World is moving towards Open Source adoption. Currently 1.5 million Users are registered on www.op…

What is Data Driven Testing ?

Using variables and input data that are stored in external sources like Excel,database to test an AUT (Application under Test)  is a common practice that let the tester to see how the application respond with various inputs. It increases test coverage and also helps to reduce the necessitaties of creating more test cases with different variables.We  generally term it as Parameterization.

Why do we need to test same Test Case using different Datasets?

Parameterization allows user to check same operation with multiple set of data.In some scenario user has to provide multiple set of data for a single test case for example if you want to check the Login credential so user has to provide multiple values for the Login. To perform this operation user has 2 options, one is user can provide these value one by one manually after completing the each flow or second one is user can use multiple set of data value  in the test case so they can easily execute the flow without changing value manually. This is also called the Parameterization when you can test AUT (Application under test) with multiple set of data values.

Advantages of data driven testing (Parameterization):-

You can test your application with multiple set of data values during Regression testing.

You can test your application with positive and negative scenarios.

Different ways to parameterize your test case in OpKey:-

You can do the parameterization in OpKey with the help of Data Repository. We can parameterize test data by two ways:-

Entering Data directly into the Data Repository.

Importing Test Data from External files.

Entering Data Directly into Data Repository:-

Add Data Repository -> Add columns and rows according to your need -> save this Data Repository

Importing Test Data from External files:-

Add Data Repository -> Import any Excel and CSV file -> save this Data Repository

(You can see in detail how to import test data from External files http://opkey.crestechglobal.com/help/pages.php?subid=57&accordid=3#Data-Repository-File-Features)

Consider the simple test case of application login. You want to test the application response to right and wrong pairs of username-password. You might also be interested in block-after-three-attempts scenario. We start with a simple test script as OpenBrowser(“internet Explorer”, “http://www.yoursite.com/loginpage.php”) TypeTextOnEditBox(“txt-box-1″, “your-username”) TypeTextOnEditBox(“txt-box-2″, “password”) ClickButtonAndWait(“bttn1″) VerifyObjectExists(“logout-link”)

The idea is, if the logout-link exists then we say that the page accepted the given username and password. Depending on the validity of username/password pair we provided, the test either passes or fails. A problem with this approach is that if we need to test with a different username/password pair, we need to alter the script and then rerun it. That will not be feasible if there are say, tens or hundreds of data pairs. We need to parameterize the username and password fields (variables) and feed the data from a table. This is where data repositories come in handy.

A data repository is a simple table which holds the data for one test. There could be many-to-many relationships between test cases and data repositories but for now let there be one data repository per test case. A data repository is a table in row/column fashion. Each column represents a variable/parameterized-data from the test case. In our case we would have two columns USERNAME and PASSWORD. Each row represents one iteration of test case with values taken from the cells in that row. Let us create a data repository as such.

Go to test case module. Add step and provide keyword in this step. Select the provide data field under Provide Data value for Data Input Argument panel. Instead of providing static value, go to Data-Repository section. Expand the newly created item. Select the username column. Do same with text keyword. Select the password field of second TypeTextOnEditBox step. Change the static value to Data-Repository. Expand the data-repository from the drop down. Select the password column. Test case looks like below image.

Now execute the test and watch. The whole test will run three times, one with each pair of username password. We say that the test has iterated three times, which was the maximum depth of any column in all the repository columns used in current test case. You can increase iterations by increasing the number of rows. For much complex tests which are heavily parameterized, the number of columns will increase.

So we have seen how we can easily parameterize OpKey test scripts and iterate a test over a large set of test data. In next post we will discuss scenarios when data in one column repeats for data in other columns.

Advanced Test Parametrization

In part 1 of this article we discussed on how we can parameterize OpKey test scripts and how we can iterate the test over a large set of data. Now we will be discussing some advanced aspects of it. In our previous discussion we tested the login and credential validation process of an application. Now we want to test the product-search feature of the same application. We have a text box where we enter search keywords. We hit enter and we are presented with a page full of relevant products. But before we do it we need to login first. So we create a test script as:

//this part does the login

OpenBrowser(“internet Explorer”, “http://www.yoursite.com/loginpage.php”) TypeTextOnEditBox(“txt-box-1″, “your-username”) TypeTextOnEditBox(“txt-box-2″, “password”) ClickAndWait(“bttn1″) //here we search TypeTextOnEditBox (“txt-box-3″, “search-keyword”) VerifyObjectExists(“item-count-link”)

When executed, this test script opens internet explorer, logs in with provided credentials and searches for a product. We now want to parameterize this. There are a total of three variables – username, password and keyword. We could go with a data repository of three columns but then we will have a slight problem: we have just one (or limited) set of login credentials but fairly large set of keywords. A basic data repository will look like:

To try this feature, go to the data repository and delete the redundant cells. You may select the cells and clear them. If you execute the test right now you will see that in fourth iteration OpKey tries to login with blank username and password. This is the default behavior. In the fourth row the username and passwords and empty hence the TypeTextOnEditBox is provided empty. This behavior can be overridden.

Go back to the test case. Locate the username parameter it would be bound to data repository and the icon next to it will tell you so. We need to redo the binding process. Expand the data repository but do not select the column now, not yet. Rather look to the top of the popup. There you will find two radio buttons CYCLIC and CONTINOUS with continuous being selected by default. Select the cyclic button and then select the USERNAME column. What we did is that we bound the username column in cyclic fashion. So once the data in the column finishes (after three) iterations, OpKey will automatically cycle to the first element in the set. Do same with password field. Save and run.