I want to automate tests for a website using Robotframework and Selenium. My test works well with Firefox. However, I would like to use Chromium instead. According to the documentation for "Open Browser", it should work for Chrome. But on my Ubuntu machine I can't use that browser out of the box.
Here is a minimal example:
*** Settings ***
Library SeleniumLibrary
*** Test Cases ***
Open Browser And Check Title
Open Browser about:blank chromium
Title Should Be ${EMPTY}
It fails with the following error:
ValueError: chromium is not a supported browser.
Is there a way to set up chromium for use with Selenium in Robotframework?
EDIT: This was originally asked for Selenium2Library. When moving to version 3, they dropped the version number from the name. The question has been updated accordingly.
The problem is that "chromium" isn't a name that the selenium library knows about. You can't use just any name. Use "chrome" rather than "chromium", and make sure that the ChromeDriver is installed in some folder in your path -- it doesn't get installed by default.
Try this:
${options}= Evaluate sys.modules['selenium.webdriver'].ChromeOptions() sys
${options.binary_location} Set Variable /var/blah/chromium
Create Webdriver Chrome my_alias chrome_options=${options}
Go To http://www.robotframework.org/
Related
I am successfully using the Conductor framework to scrape data off a website. I use the Chrome browser and therefore I have installed chromedriver.exe in the root of my project.
To speed things up I want to replace Chrome with the headless PhantomJS browser. I installed PhantomJS as explained in the answer to this Stackoverflow question: PhantomJS & Conductor Framework and have changed the browser to Browser.PHANTOMJS in #Config.
Whatever I do however, I get no results.
I found no documentation how to setup PhantomJS for use with Selenium or PhantomJS.
The question How to Implement Selenium WebDriver with PhantomJS and Can we Use Sikuli with PhantomJS? did not help either.
How to replace Chrome with PhantomJS for use with Selenium and Conductor?
The trouble you are having appears to be due to an older version of the PhantomJS library included in Conductor. The error when running PhantomJS can be found in this imported issue on the Selenium Github. The remedy is to import a fork of PhantomJS which works with the newer releases of Selenium.
You can easily implement this by editing the pom.xml file and swapping
<groupId>com.github.detro</groupId>
<artifactId>phantomjsdriver</artifactId>
<version>${phantomjs_version}</version>
with
<groupId>com.codeborne</groupId>
<artifactId>phantomjsdriver</artifactId>
<version>1.2.1</version>
You can use a headless chrome driver with selenium, as showed here:
https://duo.com/decipher/driving-headless-chrome-with-python
I had redhat Linux server command line but there is no GUI for that and I need to run selenium code on this server with firefox. As I am new to selenium so I am not sure whether it will work or not.
For achieving this I had install Firefox on my Redhat linux machine successfully but not able to trigger that as it is showing no display found.
Is there any other way to achieve this for headless browsing, where virtually a firefox will be opened and selenium code can be executed without GUI. Is it possible to do??
You can use HtmlUnitDriver of selenium to use headless browsing but it will not open firefox and may be not loading all content to it's cache as same as firefox
Refer Below:-
https://code.google.com/p/selenium/wiki/HtmlUnitDriver
Another thing you can use is Phantomjs with selenium Webdriver. This is most using pattern by industries for headless browsing
Refer below:-
http://www.guru99.com/selenium-with-htmlunit-driver-phantomjs.html
Yes you can trigger selenium file using command line
http://www.tutorialspoint.com/testng/testng_executing_tests.htm
Bash file is running fine in windows for testng but it is not working in linux/mac
Now if still your firefox is not opening then it is basically issue of some setting with your OS with firefox
Refer that too :-
https://serverfault.com/questions/279036/open-terminal-via-ssh-run-firefox-display-not-found
May be you have problem with some permission in red hat
Hope it will help you :)
When you attempt to run Selenium tests in a scripting context whose origin is at "safari-extension://...", the session hangs. I believe the problem is caused by extension sandboxing, which means if I get Selenium running in the same origin as the extension I am testing, it should be possible to execute selenium commands in my extension.
Question: Can I incorporate the Safari Driver into my Safari extension during testing to circumvent sandboxing?
We modified the Safari driver to navigate extension:// pages, and the problem persists. It looks like Apple blocks Selenium at the extension level :(
Has the robot framework support for IExplorer or only for Firefox and Chrome?
(If yes, how to configure it?)
Thanks!
Robot Framework does not, in itself, support any particular browser, so I am guessing you are referring to either SeleniumLibrary or Selenium2Library which use selenium and selenium 2 respectively. The browser support of these is well documented at seleniumhq and there is much support out there. It is recommended for new projects to use Selenium2Library as this will receive ongoing support.
Please check the driver compatiblity for browser.
You might have already known of IE driver.
Apart from that you also need to check Python version- Selenium2 version - IE Driver version - IE browser version compatibility.
In addition to #theheadofabroom 's answer, I should add that Internet Explorer does not play well with Robot Framework. Your test might not work for any number of reasons on IE while it may work just fine on FireFox and Chrome, but the most common is timing. IE is just slow enough that when Robot Framework goes to click on the next element, it searches the page for it, but it hasn't loaded in yet. As long as you have the Selenium webdriver for IE installed correctly and have written your Robot Framework code correctly, I'd recommend adding some Sleep keywords between actions to slow your code down and increase the probability that the element you want to click will load before Robot Framework searches the page for it. This is especially true if you're writing for Chrome and want to send it to either Firefox or IE.
Open Browser ${WEBAPPURL} ${BROWSER} is the keyword to open the browser.
For Firefox you can use firefox/ff instead of ${BROWSER}
For Google Chrome you can use googlechrome/gc/chrome instead of ${BROWSER}
For Internet Explorer you can use internetexplorer/ie instead of
${BROWSER}
For Firefox you don't need any driver but IE and Chrome you need to install the drivers
You can find the installers in and info here for Chrome and here for IE
Download IEdriver exe from here and put this exe file in Scripts folder of your Python installation directory. For eg, in my case it is C:\Python27\Scripts.
Ride will now launch IE for you.
Robot class supports keyboard inputs regardless of the browser. It is a class from the java.awt package and not specific to any browser. It is used in automation for performing operations on the web browser(stand alone application) in which a web-page is being automated
Note that it cannot perform operations directly on the web browser as it's a stand alone application, but can make use of keyboard shortcuts to indirectly perform the operation.
For example, if you want to open a new tab in a browser, you can use the Robot class to press Ctrl+t instead of trying to click on the new tab.
Code to use it to open a new tab in your program
Webdriver driver = new ChromeDriver(); //FirefoxDriver(), IntrrnetExplorerDriver();
driver.get("......");
//code goes here
//to open a new tab
Robot rob = new Robot();
rob.keyPress(Keys.VK_CTRL);
rob.keyPress(Keys.VK_t);
rob.keyRelease(Keys.VK_CTRL);
rob.keyRelease(Keys.VK_t);
//itetator to switch between the tabs
When I use the Selenium RC 1.0.3 on windows everything works fine.
When I config the client to use selenium on Ubuntu 10.10 what happens is that the RemoteRunner is opening and showing Command History as if everything works fine, But the other window of firefox does not open hence the test does not run.
I"m pretty sure I misconfiged something on the server, But what is it?
Thanks.
You need to check what happens in RC's command line. If it opens firefox window, but doesnt navigate further - then you need to set firefox to use proxy with the same address and port as your RC. And if it doesn't open at all, then it probably doesn't know where firefox is located, so you need to pass path along with browser definition. This is done either in test case file, or if you use selenium grid, in grid_configuration.yml
*firefox /usr/lib/firefox-5.0/firefox.sh
upd. some say that shell scripts are not good with selenium