Tag Info

New answers tagged

0

It's unclear which part will be "constantly changing". A lot of this depends on which parts of the source page will not change If the row number will change and the row id is dynamic but has part that doesn't change, you could use the following xpath as your locator: //tr[contains(@id,'rgvTransactions')]/td[5] Or an alternative css locator that does the ...


0

Sikuli with Robot Framework might be a good fit for you. Sikuli uses image recognition to identify and control GUI components. Sikuli automates anything you see on the screen.


0

Here is what we did at my company: write a webapp that talks to Gmail over IMAP. The webapp displays your inbox, or individual messages, in a form that is easy for Webdriver to process. It can also display individual messages.


0

First of all, if you get an exception in your test script, add a wait function (see the how-to via this link). That way, you'll only enter values in a control once that control is available. Secondly, if your page is loading very slowly even when going through it manually, it has nothing to do with Selenium. It seems like a performance and/or network issue, ...


0

is it any way to get object by combining UI elements in such way: hp.get(Widget.class, "Page name").get(Link.class, "Link name").click


0

Please check this if it helps.


0

We are human beings (developers, architects, managers and everyone involved in an IT project) we make mistakes. As PaulDonny mentioned earlier sometimes these mistakes can be costly and catastrophic. By finding these problems (and hopefully fixed by devs) the risk is reduced and the product is made safer to use. We don't add value to the product directly ...


1

Answering the original question , Nothing in Java as far as I know , you probably already using test ng/spring-test so not much I can recommend . Concerning peter answer , if you have a running automation with WD / selenium - great for you! Most chances you are still spending too much time on maintenance and can't run it on mobile web/ some browser properly ...


0

As per your question I think you are looking for a tool to automate windows application or desktop applications. I suggest you take a look a Autoit. Autoit works for all the windows or the desktop based application. You can even integrate it with selenium webdrivers to perform combined web and desktop based automation. Visit : ...


1

Since more or less by definition you don't have unlimited time, you can't test everything. As Peter says, you need to work out which subset of "everything" gets highest priority. I've got a few suggestions on that front: If you have data on what operating systems your customers use, great. If not, get it. You want to know which OSs are most popular with ...


3

By definition testing EVERYTHING will take INFINITE TIME. Which I bet is less than optimal (unless your boss is OK with that - then you just found safe job for life :-) So you need to agree with your boss about some metrics to limit and prioritize "everything". Make a survey among your customers which OTHER applications they use, on which platforms. Or ...


1

WebDriver is an upcoming W3C standard, so if CATJS dismisses it as "we don't do WebDriver" it raises a red flag to me immediately. Is there any good reason to NOT use industry standard solutions? Cool is fine for fun, if you can abandon it and move elsewhere with ease. If we invest man-years in developing production code I prefer stable platform. ...


0

Here is how I solved this. Would love to hear better solutions: Set WshShell = CreateObject("WScript.Shell") WshShell.SendKeys("+{TAB}") 'shift + tab wait(1) WshShell.SendKeys("{RIGHT}")'right arrow wait(1) WshShell.SendKeys("{RIGHT}")'right arrow wait(1) WshShell.SendKeys(Chr(13)) 'enter wait(1) WshShell.SendKeys("o") 'o


0

If this has to be the same copy you ship and you cannot modify it, you cannot test beyond the captcha. That is the reason of having captchas - to avoid automation. So the only option is to build in something into your app that makes testing possible. This of course will pose the risk of getting out into the field and mess things up. I only can imagine some ...


1

The thing that helped me sometimes in such cases is to evaluate the state of JQUery: http://stackoverflow.com/questions/20414729/webdriver-wait-for-jquery-to-finish But of course if you use another library, it becomes tricky. Additionally you can try to wait for the change you expect at the element, such as the added or removed class or the added content ...


0

You can try making a custom ExpectedCondition that will repeat whatever process you find unreliable. It works just like any other wait; it just repeats some function until it returns true or times out. Don't be intimidated by how much code there is. Like most java, it's mostly all just boilerplate you copy/paste :P In your case you want to click the ...


1

In order to keep the same build for testing as production there needs to be some sort of toggle in the application. Preferable in a configuration file (or database setting). By default the useCaptcha setting is enabled, but in the test environment you disable this in the configuration file. This means the code will have two paths One that does not load ...


2

Cleanest solution is to get it removed for QA environment (render it conditionally) - you have one right? You do test BEFORE deploying to production? So don't use CAPTCHA in environment where you run automated tests. If you cannot, use automated humans: there are companies which will solve CAPTCHA for you - it is quite cheap, 1000 for a buck, with decent ...


0

From what I understand, you have two options: If it's a local instance of reCaptcha, then you can look up the image ID in the database to find the code it's tied to. http://www.softwaretestingclub.com/forum/topics/how-to-break-recaptcha-using-selenium-webdriver?commentId=751045%3AComment%3A243339 Have your Dev install a backdoor. So that when a specific ...


0

From what I understand they compare the DOM of different build versions. If you expect the DOM to be same this could be a very fast way to verify you didn't change anything. Also see this page that explains the DOM, since there is some confusion about what it actually is. You can use dom-compare to compare the DOM's, they also show some examples. It seems ...


0

Se2 experts suggest NOT USE IMPLICIT WAIT at all. Especially don't mix them with explicit waits (Jim Evans is Se2 core team member). So @kirbycope advice is against opinion of experts (and my own experience). But he correctly advises to search StackOverflow (where I found link above). What works well for me, is to (explicitly) wait for some other element ...


0

Explicit Waits are not ideal. Look into using an Implicit Wait. Official documentation, here. Here is a copy+paste should the URL change or something: An implicit wait is to tell WebDriver to poll the DOM for a certain amount of time when trying to find an element or elements if they are not immediately available. The default setting is 0. Once set, ...


1

This load test article is a great walkthru for your scenario - http://www.mkyong.com/unittest/testng-selenium-load-testing-example/


2

WebDriverJs is just the driver to control the browser. For the tests itself I recommend a JavaScript based testing framework like QUnit or Mocha. The test framework take care of the test reports. Mocha is also mentioned on the WebDriverJs wiki page. You can find more information about how the reporting works on the official websites. QUnit: Bottom left of ...


0

Issue is resolved by correcting "app-package". I was using old app "app-package".


0

If not protractor, why not use C# implementation of Selenium. I am not familiar with TFS or MTM, but I am sure they will be having some easy integration options with Selenium Webdriver running with C# client. Before going to the costly options, try Selenium which is a very efficient and commonly used web application test automation tool.


0

As you want to drive different technologies end-to-end I think not many solutions will work. The browser, the web-app to remotely control a mobile device and the device it self. Have a look at the open-source Sikuli its one of test frameworks which can interact with any application as long as you can see it. It uses partial screenshots to interact with ...


0

depending on what remote-control means and on network connectivity, look for web server app for [your platform] I tried i-jetty on Android.


0

More Tools Linkcheck - http://www.websitegoodies.com/tools/linkcheck.php Broken Links - http://www.linktiger.com/ Other resources - http://freenuts.com/top-10-free-online-spell-checkers/


0

If you are using WordPress there is a cool plugin called 'Broken Link Checker (WP Plugin)' that can crawl all the pages and build a report for broken or redirecting urls. You can also trust some online tools like 'Broken Link Checker' from WTH.


0

Use UIAutomator. This will give you package name of currently active app. UIAutomator can be found at Android/sdk/tools/ .After launching UIAutomator, click on device screenshot button. But make sure, your device is connected to your system and debug mode is enabled. For Camera app, it gives package name as com.sec.android.app.camera


1

The question is focussing on the disadvantages of SilkTest I'd say. I hope the following answer is not just a personal opinion and I can provide proof for my reasons. I'm using SilkTest (mainly Silk4J together with SAP, i.e. the Premium license) for 6 months now and I'd say the main reason is that the product does not provide enough added value. It simply ...


1

Good luck in your intention. The most popular proprietary/paid performance testing tools as for now are: HP LoadRunner Borland Silk Performer Microsoft Visual Studio Load Testing Framework Each tool has free trial so you should be able to perform your own assessment, Loadrunner seems to be the most popular one. In regards to free/open source tools, ...


0

When the computer locks the main screen is showing the login buttons, the Sikuli tests cannot see the application anymore. You need to disable the locking. Since you have to lock the physical machine I think your only option is to install a virtual machine on a physical machine. That way you can have the physical machine locked and the virtual-machine ...


1

Try again with the latest version of the maven-surefire-plugin. This is the config I am using which works fine <plugin> <groupId>org.apache.maven.plugins</groupId> <artifactId>maven-surefire-plugin</artifactId> <version>2.18.1</version> <configuration> ...


0

Thanks for your suggestions Guys I have solved my problem by using the javascript executor. Below is the code: var element = driver.findElement(webdriver.By.xpath(xpath)); driver.executeScript("arguments[0].scrollIntoView(true);", element);


-1

Best practices recommend to use XPath as locator of last resort. Did you tried locating element by ID or name? Or even CSS?


0

U should try scroll to element. it is in api. Element should be visible to interact with it. also check: Did it occurs when windows is maximized? Did it work with css locator? Is it browser dependent?(on all drivers or specific one).


0

Its very hard to answers questions like this, without being able to debug and run the test code it self, but here is a try going over your code top to bottom: Why are you using a sleep? Thread.sleep(1000*4); Here you want to wait for the element to be visible as describe in the post: ...


3

Test Automation is an extremely broad subject. It doesn't just mean performing regression testing or load testing with a specific tool, although that's often what employers want (it's easy to identify). The ability to create specific programmatic solutions to help you test is hugely valuable. Over the years automation tool vendors have tried to make the ...


4

Recording tests often leads to hard to maintain tests, therefore most experienced testers prefer hand coding automated tests over recording. Automated tests often cover the same area multiple times, in order to setup data, or for example navigation. When a central part of the application changes this could result in multiple tests failing, all these tests ...


0

record-playback is extremely limited - in example, if you recorded today's date, how you will run your test tomorrow? Good test can detect different situations and react accordingly - it is way beyond record-playback. So, good test are programs. A tip for you: seems that many manual testers can feel the inevitable change to increase use of automated ...


1

I'm seeing two distinct questions here: How do you write automated tests that don't generate false positives or false negatives? How do you check existing automated tests for false positives or false negatives? Both have somewhat different answers: Dale's answer is a good one for writing tests in a way that protects against false positives and false ...


1

This looks like a position description for automated testing, which means writing automation code to run tests against the company's software. It's possible in your area there's a bigger emphasis on automation than I'm accustomed to, along with the understanding that good test automation requires programming skill. This is a good thing, especially if, as ...


1

A typical job-post for a testing position. This job description seems to be describing someone who has more programmatic and automation experience than someone with testing experience. That's not to say the best person for this job won't have both, but the JD seems to focus on desired automation skills. I can't comment on this being "typical" of job ...


0

I think the best way to learn is to "scratch your own itch" - this has the advantage that you're immediately applying what you learn, which helps you learn. For example, if you're currently testing a login page, spend time figuring out how to automate the checks you're doing. Checking user permissions? Figure out how to read the permissions from the DB, ...


1

My question is, why do so many (nearly 90%) job lists post programming as a requirement? Without speaking to every job poster, this question is not answerable. I would guess that it's because test automation is more in vogue than manual testing, based on my on experiences of testing scalability. I can only assume others have come to the same conclusion ...


2

In addition to the techniques Victor mentioned, which introduce errors into the program code, you can also introduce errors into the test code. I do this all the time as I'm test-driving new code. Make the test wrong, run it, and observe the results. Then make the test right... which might differ from how it was before.


3

Most automated testing frameworks require the user to either write in scripting language or in an actual programming language. Thus test automation equals writing code to run tests against the application. Maybe some frameworks work fully with recording the tests, but experience learns that recorded tests are harder to maintain. As an automated test ...


1

Position is test automation, not manual testing. So obviously programming is required. You will NOT be developing production code, but automated test. You need different skill set from a developer, use different tools for different goals, use different design patterns, need different communication skills, because you will encounter different problems than ...



Top 50 recent answers are included