Mobile Web Testing with QTP – Take a Peek at the Apexon Solution

Before the introduction of iPhone App Store, mobile phone applications were limited to native applications offered by handset manufacturers.  Users had a very limited choice and had no control on customization. The evolution of Android and iOS ecosystems since 2007 has made mobile applications very popular. It has allowed developers to write and sell their applications through the app store infrastructure.

Mobile browsers have also evolved from being static to advance with technologies such as HTML5 and faster networks.   This advancement has facilitated rendering high definition content on mobile and this has spiked the need for mobile browser testing. The challenge however is to test it on multiple devices, each having its own OS, model and browser version. Testing on each of these variations is not just expensive but also extremely time consuming. So, the need of the hour then is tools which would simulate a ‘mobile browser type’ environment on a regular browser.

In this blog, we will focus on HP’s QC, QTP tool to demonstrate how it can be used to resolve some the challenges that we stated, keeping the existing processes and tools intact.

Let’s understand how much time and material it would take if we were to embark upon the manual mobile browser testing journey:

Manual Mobile Browser Testing

Mobile Web

Functional Regression Approach

Manual

Mobile Web

Functional Regression Coverage

Less than 20%

Mobile Web

Functional Regression TC count

Less than 200 TCs

Mobile Web

Functional Regression Device Profiles

2 Device Profiles (1 Smart Phone and 1 Non-Smart Phone)

Mobile Web

Functional Regression Efforts

2 resources, 3 weeks

Mobile Web

Functional Regression Cycles

1 cycle (80%)

The HP-QTP Solution

And then, for a mobile web simulation using QTP, the actual pages are rendered in a web browser with request made on a mobile device and response sent as if sent to a particular device browser. Quality center was used as a main test execution tool and all reporting repository was stored in QC. Existing QTP implementation was used by modifying components or BPTs. Existing object repositories were used to identify the object with little modification to add new object for different phones. Database was used to store profile information and browser simulation was done to test different devices.

Browser simulation is an effective approach to test mobile web application while using HP QC/BPT framework.

One way to test mobile web application is using browser simulation. Web browser would camouflage itself to appear as a mobile browser by using user-agent headers/profile. Each mobile device contains certain information which identifies one phone behavior from another. Certain tools are available which change user-agent header/profile and simulate a similar environment on a different browser. Such tools help a lot in multiple environment mobile testing. Bayden UA Pick, Fiddler, etc are examples of such tools.

First time automated tests run from HP QC, it will connect to QTP. QTP gets information from QC test lab about what device it has to run. QTP will perform following action

  • Make database call to fetch profile and other necessary information from database for device
  • Open proxy (e.g fiddler), change settings for proxy with device information
  • It will open browser and proxy will update browser with device behavior and it goes back to QTP.
  • QTP will execute rest of test

The outcome

With above implementation at client side, here are the results of simulated functional testing with QTP.

Simulated Functional Automation
Before After Apexon Solution

Mobile Web

Functional Regression Approach

Manual

Automation (using QTP)

Mobile Web

Functional Regression Coverage

Less than 20%

More than 90%

Mobile Web

Functional Regression TC count

200 TCs

2000 TCs

Mobile Web

Functional Regression Device Profiles

1 (iPhone)

40 device profiles

Mobile Web

Functional Regression Efforts

2 resources, 3 weeks

2 resources, 1 week

Mobile Web

Functional Regression Cycles

1 cycle (80%)

4 cycles

In conclusion, Apexon developed a mobile web testing solution using the above approach with existing HP QC / QTP tools and with very little need to change processes or a need to learn a new language or tool.

Interested in our Testing Services?

Please enable JavaScript in your browser to complete this form.
Checkboxes
By submitting this form, you agree that you have read and understand Apexon’s Terms and Conditions. You can opt-out of communications at any time. We respect your privacy.

Other stories you may enjoy...

Healthcare Apps and the Need for Security

One of the most exciting areas in tech right now promises to be “the most personal” ever. A key aspect of making wearable devices like the Apple Watch personal is through...

Developing an App for the 2020 General Election?

Here is a thought: With the UK General Election having just finished, could the next one in 2020 be the first to use a mobile app to allow people to vote? The polling...

Be honest. Describe the state of your test cases.

“There’s some dead wood in there.” “Hmmm…. Someone really needs to clean them up.” “A little outdated.” For those reading this in the northern hemisphere,...