How to take partial screenshot (frame) with Selenium WebDriver?

Question

Is it possible to take a screenshot with WebDriver of only one frame (not the complete window) within a frameset?

Alternatively, is it possible to define coordinates of a window for the screenshot or crop the image afterwards?

Franz Ebner · Accepted Answer · 2014-03-20T22:33:35.393

This should work:

import java.awt.image.BufferedImage;
import java.io.File;

import javax.imageio.ImageIO;

import org.openqa.selenium.By;
import org.openqa.selenium.OutputType;
import org.openqa.selenium.Point;
import org.openqa.selenium.TakesScreenshot;
import org.openqa.selenium.WebDriver;

public class Shooter{

    private WebDriver driver;

    public void shootWebElement(WebElement element) throws IOException  {

        File screen = ((TakesScreenshot) this.driver).getScreenshotAs(OutputType.FILE);

        Point p = element.getLocation();

        int width = element.getSize().getWidth();
        int height = element.getSize().getHeight();

        BufferedImage img = ImageIO.read(screen);

        BufferedImage dest = img.getSubimage(p.getX(), p.getY(), width,   
                                 height);

        ImageIO.write(dest, "png", screen);

        File f = new File("S:\\ome\\where\\over\\the\\rainbow");

        FileUtils.copyFile(screen, f);

    }
}

Great! That is exactly, what I was looking for. Just the "-" at the height variable was wrong, which led to a RasterFormatException. Thank you very much. — timomeinen, Jun 01 '12 at 20:42
Works great. But what is the p.moveBy for? I can't see a difference in using it and just removing the line. — timomeinen, Jun 05 '12 at 19:33

score 8 · Answer 2 · answered Oct 06 '14 at 21:27

Python solution (dependency: PIL or Pillow):

from PIL import Image
from selenium import webdriver

def element_screenshot(driver, element, filename):
    bounding_box = (
        element.location['x'], # left
        element.location['y'], # upper
        (element.location['x'] + element.size['width']), # right
        (element.location['y'] + element.size['height']) # bottom
    )
    return bounding_box_screenshot(driver, bounding_box, filename)

def bounding_box_screenshot(driver, bounding_box, filename):
    driver.save_screenshot(filename)
    base_image = Image.open(filename)
    cropped_image = base_image.crop(bounding_box)
    base_image = base_image.resize(cropped_image.size)
    base_image.paste(cropped_image, (0, 0))
    base_image.save(filename)
    return base_image

if __name__ == '__main__':
    driver = webdriver.Firefox()
    driver.get('https://www.google.com/?gws_rd=ssl')
    element = driver.find_element_by_id('body')

    screenshot = element_screenshot(driver, element, 'element.png') # Screenshot the '#body' element

    bounding_box = (100, 100, 600, 600)
    screenshot = bounding_box_screenshot(driver, bounding_box, 'box.png') # Screenshot the bounding box (100, 100, 600, 600)

score 6 · Answer 3 · answered Sep 19 '13 at 11:40

Ruby solution

Setup

Install xvfb

apt-get install xvfb -y

Install headless browser and image manipulation gem

gem install chunky_png
gem install headless
gem install selenium-webdriver

Install chrome driver

wget http://chromedriver.googlecode.com/files/chromedriver_linux64_<latest>.zip
apt-get install unzip
unzip chromedriver_linux64_<latest>.zip
cp chromedriver /usr/local/bin

More info can be found here

Code

#!/usr/bin/env ruby

require "headless"
require "selenium-webdriver"
require 'chunky_png'

headless = Headless.new
headless.start

site = "?_some_site_?"

driver = Selenium::WebDriver.for :chrome
driver.navigate.to site
sleep 1
driver.save_screenshot('screenshot.png')

el= driver.find_element(:xpath, '?_some_xpath_?')

image = ChunkyPNG::Image.from_file('screenshot.png')

image.crop!(el.location.x + 1, el.location.y + 1, el.size.width, el.size.height)
image.save('croped.png')

driver.quit
headless.destroy

score 2 · Answer 4 · answered Sep 06 '16 at 14:12

Just posting a solution based on C# on taking a specific Html element's screenshot in selenium:

First we need to take the screen shot of entire web page using the GetScreenShot method of selenium web driver as below.

Screenshot screenshot = ((ITakesScreenshot)this.driver).GetScreenshot(); 
screenshot.SaveAsFile(filename, System.Drawing.Imaging.ImageFormat.Jpeg);

Then create a rectangle from the location , height and width of the specified html element. (This has to be obtained using FindElement() method of selenium by providing id or class name).

Image img = Bitmap.FromFile(uniqueName);
Rectangle rect = new Rectangle();

if (element != null)
{
    // Get the Width and Height of the WebElement using
    int width = element.Size.Width;
    int height = element.Size.Height;

    // Get the Location of WebElement in a Point.
    // This will provide X & Y co-ordinates of the WebElement
    Point p = element.Location;

    // Create a rectangle using Width, Height and element location
    rect = new Rectangle(p.X, p.Y, width, height);
}

Using this we are going to crop the the screenshot as below and the result will be the screenshot specific web element.

Bitmap bmpImage = new Bitmap(img);
var cropedImag = bmpImage.Clone(rect, bmpImage.PixelFormat);

Full code as a method below:

/// <summary>
/// Captures the element screen shot.
/// </summary>
/// <param name="element">The element.</param>
/// <param name="uniqueName">Name of the unique.</param>
/// <returns>returns the screenshot  image </returns>
public Image CaptureElementScreenShot(HTMLElement element, string uniqueName)
{
    Screenshot screenshot = ((ITakesScreenshot)this.driver).GetScreenshot();
    screenshot.SaveAsFile(filename, System.Drawing.Imaging.ImageFormat.Jpeg);

    Image img = Bitmap.FromFile(uniqueName);
    Rectangle rect = new Rectangle();

    if (element != null)
    {
        // Get the Width and Height of the WebElement using
        int width = element.Size.Width;
        int height = element.Size.Height;

        // Get the Location of WebElement in a Point.
        // This will provide X & Y co-ordinates of the WebElement
        Point p = element.Location;

        // Create a rectangle using Width, Height and element location
        rect = new Rectangle(p.X, p.Y, width, height);
    }

    // croping the image based on rect.
    Bitmap bmpImage = new Bitmap(img);
    var cropedImag = bmpImage.Clone(rect, bmpImage.PixelFormat);

    return cropedImag;
}

score 1 · Answer 5 · answered Dec 01 '13 at 21:15

Scala solution:

import javax.imageio.ImageIO
import java.awt.image.BufferedImage

// phone is a case class with witdth and height Int fields

def producer {       
 [..]
 processScreenshot(driver.getScreenshotAs(OutputType.FILE), ph)
}

def processScreenshot(file: File, phone: Phone) = {
 val img: BufferedImage = ImageIO.read(file)
 val w = math.min(img.getWidth, phone.width)
 val h = math.min(img.getHeight, phone.height)
 val dest = img.getSubimage(0, 0, w, h)
 ImageIO.write(dest, "png", new File(s"/tmp/${un}_${phone.file}.png"))
}

How to take partial screenshot (frame) with Selenium WebDriver?

5 Answers5

Ruby solution

Linked