Capture browser console logs with capybara

Question

I need to capture the console logs (category: info) of a browser using Ruby & Capybara. Until now I have tried using driver.manage.logs.get(:browser) or (:client) but, using this, the result is not what I want. It gives out the interaction results between selenium and browser where I can see my javascript statements sent for execution, but the resulting output fails to get captured.

Answer 1

Whether or not logs are available when using selenium depends on what browser you are using with Selenium. If you were using Firefox you'd be out of luck since it doesn't support the log retrieval API, however since you're using Chrome they are accessible. The issue you're having is that, by default, only WARN or ERROR level logs are captured. You can change this in the driver registration through the loggingPrefs capability

Selenium 3

    Capybara.register_driver :logging_selenium_chrome do |app|
      caps = Selenium::WebDriver::Remote::Capabilities.chrome(loggingPrefs:{browser: 'ALL'})
      browser_options = ::Selenium::WebDriver::Chrome::Options.new()
      # browser_options.args << '--some_option' # add whatever browser args and other options you need (--headless, etc)
      Capybara::Selenium::Driver.new(app, browser: :chrome, options: browser_options, desired_capabilities: caps)
    end

Selenium 4

   Capybara.register_driver :logging_selenium_chrome do |app|
     options = Selenium::WebDriver::Chrome::Options.new
     options.add_option("goog:loggingPrefs", {browser: 'ALL'})

     browser_options = ::Selenium::WebDriver::Chrome::Options.new()

     Capybara.register_driver :chrome do |app|
     Capybara::Selenium::Driver.new(app,
                                 capabilities: options,
                                 browser: :chrome)
     end
   end

and then specify to use :logging_selenium_chrome as your driver

 Capybara.javascript_driver = :logging_selenium_chrome # or however else you're specifying which driver to use

which should then allow you to get the logs in your tests with

page.driver.browser.manage.logs.get(:browser)

Answer 2

Thomas Walpole answer is correct but it seems that nowadays if you are using chrome as your driver you should use

Selenium::WebDriver::Remote::Capabilities.chrome( "goog:loggingPrefs": { browser: 'ALL' } )

Notice goog:loggingPrefs instead of loggingPrefs only with this solution i was able to get console.log printed in the log.

Took me a while and got it from here https://intellipaat.com/community/5478/getting-console-log-output-from-chrome-with-selenium-python-api-bindings after several frustrating attempts.

Answer 3

November 2022 update, use:

page.driver.browser.logs.get(:browser)

Answer 4

Not sure that this is what you want, but take a look at https://github.com/dbalatero/capybara-chromedriver-logger.

It helps me identify the problem with dynamic modules import(''). Works both locally and in Github Actions / Circle CI by displaying failed loads of assets (which i believe outputs as console.error).

Answer 5

Building on the answers from @thomas-walpole and @0ds0, rummaging through various other posts, and following-up on deprecation warnings led me to this code (working as of July 2023 w/ Chrome v114). It also reproduces the code that Capybara uses to build the original selenium_chrome driver.

Capybara.register_driver :selenium_chrome do |app|
  browser_options = Selenium::WebDriver::Chrome::Options.new.tap do |opts|
    # Workaround https://bugs.chromium.org/p/chromedriver/issues/detail?id=2650&q=load&sort=-id&colspec=ID%20Status%20Pri%20Owner%20Summary
    opts.add_argument('--disable-site-isolation-trials')

    opts.add_option('goog:loggingPrefs', browser: 'ALL')
  end

  Capybara::Selenium::Driver.new(app, browser: :chrome, options: browser_options)
end

Capybara.javascript_driver = :selenium_chrome

I store the logs like so:

config.prepend_after(js: true) do |example|
  Selenium::WebDriver.logger.info "Finishing Capybara test: #{example.location}\n"

  browser_logs = page.driver.browser.logs.get(:browser)
  log_str = SystemConfigurationHelpers.make_browser_log_string(browser_logs)

  Selenium::WebDriver.logger.info "** Browser logs for #{example.location}:\n\n#{log_str}\n"

  driver_logs = page.driver.browser.logs.get(:driver)
  log_str = SystemConfigurationHelpers.make_browser_log_string(driver_logs)

  Selenium::WebDriver.logger.info "** Driver logs for #{example.location}:\n\n#{log_str}\n"
end

the code that makes the log entries look nice is:

module SystemConfigurationHelpers
  # Convenience method used for displaying browser log statements extracted from Selenium
  # @param [Array<String>] collection of browser log statements
  # @return [String] newline-delimited list of formatted log messages
  def self.make_browser_log_string(logs)
    logs.inject(+'') do |result, log|
      result << "* #{log.time} #{log.level}: #{log.message}\n"
    end
  end
end

Answer 6

I was running into the deprecated warning about Selenium::WebDriver::Remote::Capabilities.

This is working for me, and seems inline with what the documentation suggests:

    ...
    # this gives us access to the javascript logs using `page.driver.browser.logs.get(:browser)` in our test
    # see https://www.selenium.dev/documentation/webdriver/browsers/chrome/
    service = Selenium::WebDriver::Service.chrome
    service.args << '--log-level=ALL'

    Capybara::Selenium::Driver.new(
      app,
      browser: :chrome,
      options: options,
      service: service
    )