Tracking change and innovation in the enterprise software development community
QCon San Francisco
Nov 19-21:
+ 80 sessions & tracks
Register now
and Save $400!
Create REST
-enabled Web Apps
with Project Zero
from IBM.
Terracotta:
Scale your
Apps w/ease
Offload the DB:
Learn More!
IBM WebSphere
Virtualize apps
to lower costs,
Increase agility
Learn more!
Posted by Scott Delap on Aug 03, 2007 05:42 PM
Is Selenium worth the pain? Atlassian developer Nick Menere has asked that very question on the Atlassian Developer Blog. Selenium is a test tool for web applications that runs directly in a browser. In his blog post Menere looks at the roadblocks found while trying to use Selenium to test two new Ajax features of JIRA 3.10. The roadblocks included:Key Events
It turns out selenium.type(...) doesn't actually simulate a user typing into an input box. Go figure. So we typed in the text and triggered everykey stroke seperately - selenium.keyDown(...); selenium.keyPress(...); selenium.keyUp(...);. It worked!! in Firefox... In other browsers, it would now would print every character twice.
Timings
..Due to our non-selectable items (section labels etc.), we wanted to test that some elements attributes do not change. We fire a mouse event on the element, and then use an xpath locator to check that the element's attribute did not change ... it needed a slight pause between the event fire and test ... This was a very common fix for a lot of the things that would break - "Add/increase the pause".
Mouse Positioning Issues
To make things a little nicer for our users we decided to put in an autoscroll feature.... After putting in/increasing numerous pauses, I still couldn't get them to pass. Why would scrolling the screen cause the tests to fail? We fire events on selected elements - not co-ordinates. The build kept on breaking......I saw the mouse positioned directly in the middle of the screen, when the browser window scrolled, the mouse was positioned over one of the non-selectable items ... How to fix? We installed xwarppointer, to enable us to move the mouse pointer through bash, and tucked it away in a little corner.
Overall Menere learned a number of lessons. The most important of which was that the Selenium client is not a user.
WebSphere Virtual Enterprise 3 minute demo
Scaling a Massively Multi-player Server Casestudy: Terracotta on SmartFoxServer
Spring App Platform, Java Concurrency/Multicore, Eclipse Mylyn and more @ QCon SF Nov 19-21
We have been going through similar pains with Selenium. Sometimes Selenium fails without any deterministic reason. We could figure some issues within our test usage. However, there are other failures which are totally random and do not give any clue. Running from command prompt with maven resulted in test failures at different places than the ones that appear when tested from within eclipse. Arggh.
While I agree Selenium can be a pain to work with against ajax enabled pages, I personally think it is a great product - it's free!!! - it is not even on version 1 yet (0.9.0), can only get better - API's exist for multiple programming languages by the way the 'Atlassian Developer Blog' link in the article is busted - prefixed with 'link:'
Selenium is great! But not so great to put it into production. Developers can use it randomly to assist them. It fits with agile practices, but can not run reliably in production yet. I am not sure why the development is so slow (no updates since dec 2006). With Web2.0 applications, so many languages and browsers, (functional) testing on a regular basis is necessary. I would suggest to develop Selenium without going into conclusions (like pain).
But not so great to put it into production
What do you mean by "production" ?
Selenium is a testing tool, by its very nature it doesn't get used in production.
Are you suggesting that developers should write Selenium tests for the purposes of development, but not run them as a regular test suite?
If so, then that seems like a very strange position to take.
Are you suggesting that developers should write Selenium tests for the purposes of development, but not run them as a regular test suite?
If so, then that seems like a very strange position to take.
I think that what Kumal means is that atm you cannot rely on these tests. You can definitely write and run them periodically, but you shouldn't bet your release on just the fact that your Selenium tests passed. At least this is the way I've read his message.
./alex
--
.w( the_mindstorm )p.
________________________
Alexandru Popescu
Senior Software Eng.
InfoQ TechLead&CoFounder
It is worth it.
I have used HttpUnit for a few years and that had some advantages. The best was that once you have written your test, you could start it on a great many threads and knock seven bells out of your server. This is not so easy with Selenium as it requires one instance of FireFox, IE or another to simulate one user.
However, writting the test in the first place is much easier with Selenium IDE, and keeping it upto date as the UI changes is quite easy too. With HttpUnit I quite often had to revert to debugging my unit test to see what was in the page that had been returned by the server.
OK using Selenium with AJAX isn't a smooth ride yet. I think selenium.type() will set the value of the node in the DOM instead of simulating key presses. However, one of the really nice things about Selenium is where it doesn't quite do what you need, it is very easy to extend. Create your own user-extensions.js, put the following in it, and set up the extensions file in the options:
Selenium.prototype.doTypeKeys = function(locator, value) {
for(var i=0; i
On thinking about it - you probably shouldn't do both a keyPress and a keyUp. You should only need keyDown and a keyUp, any onKeyPress event handler would still be fired. May be this why you get double key strokes in other browsers to FireFox.
http://blogs.atlassian.com/developer/2007/08/selenium_is_the_pain_worth_it.html#more I had to poke around a bit to find it!
http://sahi.co.in/ -- from what I have seen, it's pretty good at behaving like a user
I must confess that at first glance Sahi looks impressive. Thanks for sharing. ./alex -- .w( the_mindstorm )p. ________________________ Alexandru Popescu Senior Software Eng. InfoQ TechLead&CoFounder
First link in this item now fixed, thanks for pointing it out.
I've been working with selenium since last year and it´s great for web test. It´s true that ajax have been a problem for selenium but the extensibility allows anything. The posibility of implemet new commands have allowed me, and my company, to solve the problems implementing personalized comands. Selenium: It works... and if it doesn´t you could make it work!
Another contender may be: Chickenfoot. Funny name, isn't it? :-). ./alex -- .w( the_mindstorm )p. ________________________ Alexandru Popescu Senior Software Eng. InfoQ TechLead&CoFounder
I've been working with selenium since last year and it´s great for web test. It´s true that ajax have been a problem for selenium but the extensibility allows anything. The posibility of implemet new commands have allowed me, and my company, to solve the problems implementing personalized comands. Selenium: It works... and if it doesn´t you could make it work! Thanks from: Las Vegas Entertainment Stress Management Best Web Hosting
Thin client paradigm characterized by web applications is a kludge that needs to be repudiated. Old compromises are no longer needed and it's time to move the presentation tier to where it belongs.
In this presentation filmed during QCon 2007, Jeff Sutherland, the creator of Scrum, talks about his visit at Google to do an analysis of Google's first implementation of Scrum.
In this article, Bryon Jacob and Chris Berry introduce AtomServer, their implementation of a full-fledged Atom Store based on Apache Abdera, which is now available as open source.
It is easy to think that virtualization applies only to servers. In reality the recent resurgence of the concept is also being applied to networking, storage, and application infrastructure.
In this article, Stefan Tilkov explains some of the most common anti-patterns found in applications that claim to follow a "RESTful" design and suggests ways to avoid them.
In this article, Adrien Louis and Marc Dutoo discuss the differences and relative merits of using orchestration vs. routing in a typical ESB setup, and discuss various implementation options.
Wayne Lund discusses batch processing, Spring Batch objectives and features, scenarios for usage, Spring Batch architecture, scaling, example code, failures and retrying, and the future roadmap.
Developer Jay Fields draws on his experiences as a ThoughtWorks consultant to describe effective user story estimation techniques.
14 comments
Reply