Skip to main content

Selenium Remote Control - Introduction


Selenium IDE have limitations in terms of language support (Only Selenese - InBuilt Selenium Commands) and browser support (Firefox only) . These limitations can be diminished by using Selenium Remote Control.

Selenium RC can be used for automating web applications for different web browsers on different platforms using your favorite language like JAVA, C#, Perl, Python, Ruby etc. You can use Selenium RC with or without Selenium IDE. Though, You might find it easier to start with Selenium IDE and than move to Selenium RC for customization and making your automation more robust.

Selenium RC is a java server which allows you to interact with the web application using any language. Your script ( If not hosted along with your application on the server) can not talk to your web application directly because of the security restrictions enforced by browser ( Same origin policy ). Using selenium server, you can bypass this restriction and have your automation scripts running against any web application. There are still some limitations, for example crossing domain boundaries, using it for https etc., but you can find solutions for most of them.

Selenium RC server is written in Java and in order to use Selenium RC you should have java runtime environment (newer than 1.5) installed on your machine. Assuming that you have downloaded and unzipped Selenium Remote Control from openqa.org and also have java runtime on your system, you can launch Selenium Remote Control server using following command ( Make sure that you are into directory where selenium-server.jar is located or this directory is in your path variable ) -

java -jar selenium-server.jar

This command should start Selenium Server on your machine and your automation scripts are ready to interact with your web application using this server. You can leave this server up and running while you are developing and executing your automation scripts. If you wish, you can also start Selenium Remote Control in interactive mode and start playing with its various commands on command prompt itself. In order to do that, you need to supply a command line option called interactive.

java -jar selenium-server.jar -interactive

There are many command line options that you can specify while starting selenium RC. Some of the interesting options are multiWindow , forcedBrowserMode, userExtensions, browserSessionReuse and many more.

For now lets move on to client drivers for Selenium, so that you can start using it. In the above paragraphs, we discussed that using Selenium RC you can automate your web application in most of the popular languages. This support is provided by the mechanism of client drivers. These client drivers provide the functionality to write automation script in any language and interaction with the Selenium Remote Control Server. Client drivers for most of the languages are already available and since it is open source you can create your own client driver if it is not available for your language. Details of extending existing client drivers or creating your own client driver can be found on openqa's website.

Once these client drivers are available, you can create a Selenium object in your automation script by specifying host name and port of the Selenium server. Using this Selenium object you can handle all the browser specific tasks for your web application. Some of the well known and popularly known client drivers are distributed along with the selenium RC itself. If you notice the directories where you unzipped the Selenium RC, you will find folders for various languages like Java, Perl, Python etc. Detailed instructions of using these client drivers are available on the respective section in openqa, we will summarize them here very briefly.

  • Java - selenium-java-client--driver.jar should be in the classpath, you can use this with JUnit or TestNG

  • .NET - add the ThoughtWorks.Selenium.Core.dll assembly as a reference to your VS project, use it with NUnit

  • Perl - Install Test::WWW::Selenium . You will need make (on unix) or nmake (on windows) in order to use it. You will also find it useful to use Test::more

  • Ruby - To use the Ruby Client Driver, just "require" or "load" selenium.rb in your Ruby script and create a new Selenium::SeleneseInterpreter object

Similarly, for any client driver steps are more or less same. i.e, install or use driver so that Selenium object can be created in your automation scripts which can interact with the browser.

If you have followed along, at this point you should have Selenium RC server up and running and you have also installed Selenium client drivers. You can now create your automation scripts and run them as you would run any program in the language of your choice. For example -

java YourAutomationClass
perl AutomationScript.pl
ruby AutomationScript.rb
And so on..

Though you can write your automation scripts using Selenium RC alone, we have found it very easy to start with Seleinum IDE and export test cases in specific language. Once you have basic test case, you can use language specific features to customize them and making them more robust.

Hope this document will help you all ! ! !

Comments

Popular posts from this blog

Change IE Browser ZOOM settings

Lot of UI automation testers could have faced this problem as we could change the zoom settings while operating manually for our convenience and forgot to reset to 100%.
But our QTP and some other related tools would operate the browser perfectly if browser zoom is 100%. So wee need to change the zoom before start to run the scripts. Its better to have a code snippet in our framework to change this zoom setting right?
Here we go...

1. We can simply change the Registry values before Invoking IE
Function ChangeRegistry Dim objShell Set objShell = CreateObject("WScript.Shell")  objShell.RegWrite "HKEY_CURRENT_USER\Software\Microsoft\Internet Explorer\Zoom\ZoomFactor", "100000", "REG_DWORD" Set objShell = Nothing EndFunction This option is very useful. But in real time, lot of customers could have restricted write access to windows registry. So we can try other options.

2. Use IE COM object and Change Settings
Function ChangeIEZoom Dim intZoomLevel, objIE  intZoo…

QTP - Object Identification - Prerequesties

While starting a new project or proof of concept or after new QTP installation, we would face the issue in identifying objects as QTP built objects like WinList, WebEdit, etc.; instead everything will be identified as WinObject.

Here are some steps to ensure you did the environment setup RIGHT before start automating.
Did you enabled required Add-ins only while launchingDid you selected "Record >> Record and Run Settings" - This should be based on what type of application you are working with
Windows app - Windows >> Record and run test on any windows applications. Web app - Web >> Record and run test on any open browser.
Did you tried to launch UFT/QTP and then AUT.If its web, ensure browser zoom level set to 100%Try disabling the protected mode in IE for web.Launch UFT and AUT using RUN AS ADMINISTRATORAlso see the Object Identification Methods if you are interested.

Run JavaScript from QTP

Yeah, You can run your pure JavaScript from QTP using RunScript method. Lot of times, we are in need of firing events or simulating actions on web page which is not supported by QTP. At that time, you can use your direct DOM methods and directly execute your script on the web page from QTP like,
Dim MyPage, SearchBox Set MyPage = Browser("title:=Google").Page("title:=Google") Set SearchBox = MyPage.RunScript("document.getElementsByName('q')(0);") SearchBox.Value="testing"
'if objects available in frames, Set SearchBox = MyPage.RunScript("document.frames(4).document.getElementsByName('q')(0);")
'OR
Set objFrame = Browser("title:=Google").Page("title:=Google").Frame("title:=something")
objFrame.RunScript("document.getElementsByName('q')(0);") Also, you can execute by obtaining actual browser window object if you have IE app like, Dim IEApp Set IEApp = CreateObject("Internet…