Learn Selenium - Advanced Browser Actions

#Introduction

After mastering click, type, and submit in episode 4, you're ready for more challenging actions: advanced browser actions. Real applications have JavaScript alerts, iframes that hide content, new tabs that open, elements that must be dragged, files that must be uploaded, and long pages that must be scrolled. Episode 8 equips you with all of that control.

These actions are often the trickiest part of automation, because many of them can't be done with plain find_element and click. You'll need special objects like Alert, ActionChains, and an understanding of window handles. Let's break them down one by one.

#Alerts, Popups, and Windows

#Handling JavaScript Alerts

Browser alerts are native dialogs that block the page. Selenium manages them through the Alert object:

Accepting and reading alerts

from selenium.webdriver.common.alert import Alert
 
Alert(driver).accept()

For alerts with input, use send_keys on the alert object then accept:

Alert with input

alert = driver.switch_to.alert
alert.send_keys("nama pengguna")
alert.accept()

driver.switch_to.alert switches context to the active dialog, send_keys fills the prompt, and accept() confirms it. If a cancel button appears, dismiss() is the opposite.

#Working with Multiple Windows

When a click opens a new tab, the browser has more than one window handle:

Switching between tabs

sebelum = driver.current_window_handle
driver.find_element("id", "buka-tab").click()
 
for handle in driver.window_handles:
    if handle != sebelum:
        driver.switch_to.window(handle)
        break
 
print("Judul tab baru:", driver.title)
driver.close()
driver.switch_to.window(sebelum)

driver.window_handles returns the list of all tabs. The pattern above switches to the new tab, works there, closes it, then returns to the original tab.

#Frames and iframes

Content inside <iframe> can't be found directly from the main page context. You must switch context first:

Entering and exiting an iframe

driver.switch_to.frame("payment-frame")
driver.find_element("id", "nomor-kartu").send_keys("4111111111111111")
driver.switch_to.default_content()

driver.switch_to.frame("payment-frame") moves context into the iframe by id, name, or element. When done, default_content() returns context to the main page. Don't forget to exit the iframe before looking for elements on the main page.

#Drag and Drop, Hover, and Keyboard Actions

#ActionChains for Chained Actions

ActionChains enables complex sequential actions:

Drag and drop with ActionChains

from selenium.webdriver.common.action_chains import ActionChains
 
sumber = driver.find_element("id", "sumber")
target = driver.find_element("id", "target")
ActionChains(driver).drag_and_drop(sumber, target).perform()

drag_and_drop(sumber, target) drags an element to a target. For other actions — hover to trigger a menu, double click, and right click — ActionChains also provides move_to_element, double_click, and context_click.

#Keyboard Actions

Key combinations like Ctrl+A and Enter can be sent through the Keys class:

Keyboard combinations

from selenium.webdriver.common.keys import Keys
from selenium.webdriver.common.action_chains import ActionChains
 
inputan = driver.find_element("id", "editor")
inputan.send_keys(Keys.CONTROL, "a")
inputan.send_keys(Keys.BACKSPACE)
ActionChains(driver).send_keys(Keys.TAB).perform()

Keys.CONTROL and Keys.BACKSPACE are examples of keyboard modifiers. ActionChains send_keys(Keys.TAB) sends keys at the page level, not to a specific element.

#File Upload and Download

#Upload via send_keys

HTML file inputs hide the OS dialog, but Selenium can still send the path directly:

File upload

path = "/home/devnull/Documents/cv.pdf"
driver.find_element("id", "input-file").send_keys(path)

send_keys(path_lengkap) on a file-type input element is the most stable upload method. Native dialogs can't be automated directly by Selenium, so never try to click a "Browse" button.

#Download and Configuration

For downloads, set browser preferences so files go to the desired folder:

Chrome download configuration

from selenium import webdriver
 
opsi = webdriver.ChromeOptions()
prefs = {"download.default_directory": "/tmp/download-test"}
opsi.add_experimental_option("prefs", prefs)
driver = webdriver.Chrome(options=opsi)

prefs["download.default_directory"] sets the download destination folder. After triggering a download, wait for the file to actually appear in the folder before continuing with assertions.

#Scrolling and Viewport

Elements below the fold sometimes need scrolling before they can be interacted with. Either via JavaScript or a scroll action:

Scroll and viewport size

driver.execute_script("window.scrollTo(0, document.body.scrollHeight)")
driver.set_window_size(1440, 900)
driver.execute_script("arguments[0].scrollIntoView(true);", tombol)

window.scrollTo(0, document.body.scrollHeight) scrolls to the bottom of the page, while scrollIntoView(true) brings a specific element into view. set_window_size(1440, 900) sets the viewport size for responsive tests.

#Conclusion

Episode 8 extends the reach of your automation: page-blocking alerts, new tabs, iframes hiding content, drag and drop, hover, keyboard, file upload and download, and scrolling long pages — all now under your control.

Key takeaways:

• Alerts are handled via driver.switch_to.alert; accept, dismiss, and input.
• Switch tabs by comparing window handles, then return.
• iframe content requires switch_to.frame and returning via default_content.
• ActionChains for drag, hover, and chained actions; Keys for keyboard combinations.
• Upload uses send_keys to a file input; download is configured via browser prefs.

In episode 9 next, we'll cover data-driven and parameterized testing — using data providers for many test cases, parameterizing inputs and expected results, pulling data from CSV, JSON, and Excel, and expanding test coverage across many browsers and environments.