I am using chrome 56, chrome driver 2.27(latest release) with selenium web driver 3.1.0. Referring to the issue(https://github.com/seleniumhq/selenium-google-code-issue-archive/issues/1811) where chrome closes all the instances once the program finishes and it does not give me a chance to debug. I just want to know if this is fixed then why it is still happening ? or i am missing something ?
I am using the following code. Any help is appreciated.
require "uri"
require "net/http"
require 'watir-webdriver'
require 'selenium-webdriver'
#b = Watir::Browser.new :chrome
#b.goto 'http://www.google.com'
Firstly, watir-webdriver gem is deprecated. The updated code is in the watir gem. Also, you shouldn't need to require any of those other gems directly.
The chromedriver service is stopped when the ruby process exits. If you do not want the browsers that were started by chromedriver to close as well, you need to use the detach parameter. Currently this is done like so:
require 'watir'
caps = Selenium::WebDriver::Remote::Capabilities.chrome
caps[:chrome_options] = {detach: true}
#b = Watir::Browser.new :chrome, desired_capabilities: caps
Declare these
caps = Selenium::WebDriver::Remote::Capabilities.chrome("chromeOptions" => {'detach' => true})
browser = Watir::Browser.new :chrome, desired_capabilities: caps
On a side note! this might give a problem when you are running multiple scenario tests, chromedriver will actively refuse connection in case an other test initiates in the same chrome session. Ensure you have browser.close whenever required.
Related
I am very new at BDD testing and I am trying to figure out how to open a chrome browser in a specific URL using these technologies. I already installed bddfire gem and already ran it.
Now I have a feature:
#openingChrome
Scenario: Opening Chrome on facebook page
Given I open chrome and write "url"
bddfir_steps.rb
Given(/^I open chrome and write "([^"]*)"$/) do |arg1|
$session.visit("https://www.facebook.com.br")
end
In the hooks file I wrote
Before do
$session = Capybara::Session.new(:selenium, browser: :chrome)
end
And in the env.rb file that came with bddfire gem there is this
Capybara.register_driver :chrome do |app|
Capybara::Selenium::Driver.new(app, browser: :chrome)
end
So chrome is already added and I already installed chrome webdriver. Unfortunately this line of code is not working: $session = Capybara::Session.new(:selenium, browser: :chrome)
it throws this error: The second parameter to Session::new should be a rack app if passed. (TypeError)
does anyone know why?
Session#new takes the name of the driver and optionally a rack app instance to start. Since you don't seem to be starting an app and you're registering your driver with a name of :chrome you want
Before do
$session = Capybara::Session.new(:chrome)
end
Using Ruby 2.0.0 p481 in RubyMine and chromedriver 2.10
When Chrome starts it displays a message in a yellow popup bar: "You are using an unsupported command-line flag: --ignore-certificate-errors. Stability and security will suffer." This simple example reproduces the problem.
require "selenium-webdriver"
driver = Selenium::WebDriver.for :chrome
driver.navigate.to login_url
This question has been answered for Java and Python. I have looked everywhere for a Ruby analog. Does anyone have a suggestion or know how to translate the Python answer (Unsupported command-line flag: --ignore-certificate-errors) to Ruby? Thank you!
The Ruby selenium-webdriver API doesn't expose a separate Chrome options object like Java/Python but you can set the options via "Capabilities".
The Capabilities web page provides a Ruby example and the table of recognized capabilities that you can inject. Plugging those together with excludeSwitches:
caps = Selenium::WebDriver::Remote::Capabilities.chrome("chromeOptions" => {"excludeSwitches" => [ "--ignore-certificate-errors" ]})
driver = Selenium::WebDriver.for :chrome, desired_capabilities: caps
Take a look at Watir too, it's a front end for WebDriver.
Their examples show how you can send a :switches array which is passed straight through to the web driver so you can do the same. That makes adding other switches a bit easier rather than going through capabilities.
There is a chromedriver issue on the topic as well. There are posts detailing that you can add a --test-type argument to work around the certificate issue and ruby code examples like above.
I adjusted:
driver = Selenium::WebDriver.for :chrome
to read:
driver = Selenium::WebDriver.for :chrome, :switches => %w[--test-type]
...and the script ran successfully without the yellow flag. Clearly, other command-line switches could be added.
Thank you to Nguyen Vu Hoang and mtm.
I donot know ruby, however my approach is set mode "test-type" to ChromeDriver capabilities
Here's my sample code in Java
DesiredCapabilities capabilities = new DesiredCapabilities();
capabilities = DesiredCapabilities.chrome();
ChromeOptions options = new ChromeOptions();
options.addArguments("test-type", "start-maximized",
"no-default-browser-check");
capabilities.setCapability(ChromeOptions.CAPABILITY, options);
With Capybara:
Capybara.register_driver :chrome do |app|
Capybara::Selenium::Driver.new(app, browser: :chrome, switches: ['--test-type'])
end
This error caused my rspec tests to fail. I think it was causing Chrome to slow down so the above fixes did remove the error messages but did not resolve my problem of my tests failing.
If you are using chromedriver-helper then this should fix the problem. Run this from the command line
chromedriver-update
This is described more in chromedriver-helper, "This might be necessary on platforms on which Chrome auto-updates, which has been known to introduce incompatibilities with older versions of chromedrive"
Once I ran this I was able to remove the fixes described elsewhere and Chrome ran with out warnings.
I've been playing with Watir and Tor browser and I can't get them to work.
The correct browser opens, however, I'm unable to open a website.
My code:
Selenium::WebDriver::Firefox.path = '\path\Tor Browser\Start Tor Browser.exe'
driver = Selenium::WebDriver.for :firefox
browser = Watir::Browser.new :firefox, :driver => driver
This results in a message box popping up, which says:
Your Firefox profile cannot be loaded. It may be missing or inaccessible.
The correct browser also opens up, however, the website doesn't.
I've also tried:
Selenium::WebDriver::Firefox.path = '\path\Tor Browser\Browser\firefox.exe'
driver = Selenium::WebDriver.for :firefox
browser = Watir::Browser.new :firefox, :driver => driver
With this, the message box doesn't appear, but still, the website is not loaded.
I'm not sure, if the problem is in my code or in the browser.
Thank you very much for any help!
Start Tor then run this, assuming your socks port is 9050
require 'watir-webdriver'
profile = Selenium::WebDriver::Firefox::Profile.new
profile['network.proxy.socks'] = "127.0.0.1"
profile['network.proxy.socks_port'] = 9050
profile['network.proxy.type'] = 1
$browser = Watir::Browser.new :firefox, :profile => profile
$browser.goto "whatsmyip.org"
Firefox and Tor are not the same browser. Watir uses webdriver API to control the browser. You can't use Firefox webdriver API to control another browser.
In your second example, browser don't open the site, but you don't tell him to. You should add :
browser.goto('http://my.example.com')
And don't forget to put a browser.quit at the end of your code to close the browser.
I am trying to automate some IE browser tests using Ruby/Selenium WebDriver.
When I run the following code, it opens a new IE browser with the url but it always tells that 'There is a problem with this website's security certificate.'
Is there any way to set the IE profile/capabilities using Ruby similar to the ones used in Java?
require 'selenium-webdriver'
driver = Selenium::WebDriver.for :ie
driver.get "https://xxxxxxxxxxxxxxxxxx.com"
There's no way to set it by any capabilities, even if you were using Java. If you have found any approaches to achieve it in Java, please post it and see if it can be translated into Ruby.
But you can always simulate the clicking to bypass it.
# Tested under Windows 7, IE 10, Ruby 2.0.0
require 'selenium-webdriver'
driver = Selenium::WebDriver.for :ie
driver.get "https://xxxxxxxxxxxxxxxxxx.com"
driver.get("javascript:document.getElementById('overridelink').click()");
I am trying to scrape a web page with a lot of javascript. with the help of pguardiano i have this piece of code in ruby.
require 'rubygems'
require 'watir-webdriver'
require 'csv'
#browser = Watir::Browser.new
#browser.goto 'http://www.oddsportal.com/matches/soccer/'
CSV.open('out.csv', 'w') do |out|
#browser.trs(:class => /deactivate/).each do |tr|
out << tr.tds.map(&:text)
end
end
The scraping is done recursively in background with a sleep time of 1 hour approximatively. I have no experience of ruby and in particular of web scraping, so i have a couple of questions.
How can i avoid that every time a new firefox session is opened with a lot of cpu and ram consumption?
Is it possible to use a firefox engine without using his GUI?
You can try a headless option.
require 'watir-webdriver'
require 'headless'
headless = Headless.new
headless.start
b = Watir::Browser.start 'www.google.com'
puts b.title
b.close
headless.destroy
An alternative is to use the selenium server. A third alternative is to use a scraper like Kapow.