Tag: Automation

Auto Detect Chrome version and auto-download compatible chrome version !!!

Introduction:

Thanks to @webdriverio , i came to know about the chromedriver npm package. It’s a cool library that allows you to detect the installed chrome version and download the compatible chromedriver version automatically

https://www.npmjs.com/package/chromedriver

Pre-requisite:

install nodejs >v 10 : install

you can also download just the binary file instead of installing nodejs

now just run the below command

<oath_to_npm>/npm install -g chromedriver --detect_chromedriver_version --scripts-prepend-node-path

you can see the supported command-line flags at :

https://www.npmjs.com/package/chromedriver

Note: — scripts-prepend-node-path should be last argument, this flag is mandatory only if you are using node binaries than installation

if you want to download the binary to a specific folder then use the below repo instead:

I have added features to set target directory and download driver when used as a library. I have created a pull request for these but till that gets merged use the below repo instead

https://www.npmjs.com/package/chromedriver

download the zip or clone the files, unzip the content and then use the below command

C:\Users\test\Downloads\node-v14.17.0-win-x64\npm install -g "C:\Users\test\Downloads\chromedriver-main\chromedriver-main" --chromedriver_download_to_target_folder="C:\Users\test\Downloads\chromedriver-main\output" --scripts-prepend-node-path

Note: — scripts-prepend-node-path should be given as the last argument if you are running nodejs binary zip directly instead of installing nodejs

This will add node to PATH automatically for the scripts to detect

Now let see how we can run this library from nodejs , Note I have given information about alternative simple native libraries that does the same :

Running from Java:

You can run this nodejs library from java as:

package test_suites;
import java.io.IOException;
import java.util.regex.Matcher;
import java.util.regex.Pattern;
public class Testcase1 {
public static void main(String[] args) throws IOException, InterruptedException {
// just donwload the nodejs binary zip file and extract it
    //point to the npm.cmd inside the extracted folder
//if you already have nodejs installed then point to the npm.cmd of the installed library
    ProcessBuilder pb = new ProcessBuilder("C:\\Users\\test\\Downloads\\node-v14.17.0-win-x64\\npm.cmd",
      "install",
      "-g",
      "C:\\Users\\test\\Downloads\\chromedriver-main\\chromedriver-main",
      "--detect_chromedriver_version",
//  "--chromedriver-force-download",  // uncomment to force chromedriver dwonload even if it exists 
      // "--chromedriver_version=88.0.4324.96", //comment detect version and hard code the version
// "--chromedriver_download_to_target_folder=\"C:\\Users\\test\\Downloads\\temp_2\\github_chromedriver\\output\"" // available only for praveendvd/chromedriver repo
      "--scripts-prepend-node-path"); //"--scripts-prepend-node-path" should be always the last argument
Process p = null;
    String result = null;
    String error = null;
p = pb.start();
System.out.println("Started download ###############################");
//wait for download to complete
    p.waitFor();
System.out.println("Output  ###############################");
System.out.println("Output stream ###############################");
result = new String(p.getInputStream().readAllBytes());
    System.out.println(result);
System.out.println("Error stream ###############################");
error = new String(p.getErrorStream().readAllBytes());
    System.out.println(error);
System.out.println("Output end ###############################");
Pattern pattern = Pattern.compile("ChromeDriver binary available at(.*?)\n", Pattern.DOTALL);
    Matcher matcher = pattern.matcher(result);
    matcher.find();
    String chromedriverpath = matcher.group(1).trim();
System.out.println("****The driver is copied to**** : " + chromedriverpath  );
System.out.println("Completed download ###############################");
}
}

Output:

Or use :

GitHub — bonigarcia/webdrivermanager: Automated driver management for Selenium WebDriver

Mvn:

<dependency>
<groupId>io.github.bonigarcia</groupId>
<artifactId>webdrivermanager</artifactId>
<version>4.4.3</version>
</dependency>

Code:

import io.github.bonigarcia.wdm.WebDriverManager;
import io.github.bonigarcia.wdm.config.DriverManagerType;
public class Testcase1 {
public static void main(String[] args) throws IOException, InterruptedException {
WebDriverManager.getInstance(DriverManagerType.CHROME).forceDownload();
WebDriverManager.chromedriver().setup();
System.out.println(WebDriverManager.getInstance(DriverManagerType.CHROME).getDownloadedDriverPath());
}
}

Running from Python:

import subprocess
import sys
import re
command = subprocess.run(
[
"C:\\Users\\Downloads\\node-v14.17.0-win-x64\\npm.cmd",
"install",
"-g",
"C:\\Users\\Downloads\\chromedriver-main\\chromedriver-main",
"--detect_chromedriver_version",
# "--chromedriver-force-download",  # uncomment to force chromedriver dwonload even if it exists
# "--chromedriver_version=88.0.4324.96", #comment detect version and hard code the version
# "--chromedriver_download_to_target_folder=\"C:\\Downloads\\temp_2\\github_chromedriver\\output\"" # available only for praveendvd/chromedriver repo
"--scripts-prepend-node-path",
],
capture_output=True,
)
# --scripts-prepend-node-path" should be always the last argument'''
print("Started download ###############################")
print("Output  ###############################")
print("Output stream ###############################")
sys.stdout.buffer.write(command.stdout)
print("Error stream ###############################")
sys.stderr.buffer.write(command.stderr)
print("Output end ###############################")
output = command.stdout
result = re.search("ChromeDriver binary available at(.*?)\n", output.decode('utf-8') )
chromedriverpath = result.group(1)
print("****The driver is copied to**** : " + chromedriverpath  )
print(("Completed download ###############################"))
sys.exit(command.returncode)

Output:

Or use:

library: chromedriver-autoinstaller · PyPI

pip install chromedriver-autoinstaller

code:

import chromedriver_autoinstaller
# Check if the current version of chromedriver exists
# and if it doesn’t exist, download it automatically,
# then add chromedriver to path
driver = webdriver.Chrome(chromedriver_autoinstaller.install())

Running from Nodejs:

Clone or download and extract :

https://github.com/praveendvd/chromedriver

Now install the library to your npm project as :

C:\Users\test\Downloads\node-v14.17.0-win-x64\npm install "C:\Users\test\Downloads\chromedriver-main\chromedriver-main" --chromedriver_download_to_target_folder="C:\Users\test\Downloads\chromedriver-main\output" --scripts-prepend-node-path

Now you can use this library to autodownload compatible chrome driver version as :

var chromedriver =  require('chromedriver');
process.env.detect_chromedriver_version = true
process.env.chromedriver_force_download = true    
process.env.chromedriver_download_to_target_folder="C:/Users/output" 
chromedriver.download() //returns a promise

Sending text to fields for which no known locators.

Question:

https://sqa.stackexchange.com/questions/42251/how-to-interact-with-ngx-monaco-editor

Answer:

If you are not sure about the locator, then you can use the action class sendKeys method to interact with the field.

Here, it interacts with the active (currently focused ) element.

So the first step is to bring the element to focus, this can be done by just clicking it:

await browser.get('https://stackblitz.com/edit/ngx-monaco-editor-example')      
await browser.sleep(10000)
await  $('[class="view-line"]').click()
await browser.sleep(4000)

Now you can see the cursor is at the below place:

Now you can interact with the element using browser.actions():

await browser.actions().sendKeys('This is test').perform();

this will send input to the currently active element:

Now let us look deeper to find out the locator:

We now know that the sendKey using action works, so we can find the locator from the active element:

The outerHTML of the active element gives the locator:

await  $('[class="view-line"]').click()
let test = await browser.driver.switchTo().activeElement()
console.log("outer");
console.log(await test.getAttribute('outerHTML'))
//await test.sendKeys("a=1;c=a+10;") if you try this you can see even this sends data

Output:

<textarea data-mprt="6" class="inputarea" wrap="off" autocorrect="off" autocapitalize="off" autocomplete="off" spellcheck="false" aria-label="Editor content;Press Alt+F1 for Accessibility Options." role="textbox" aria-multiline="true" aria-haspopup="false" aria-autocomplete="both" style="font-size: 1px; line-height: 18px; top: 0px; left: 562px; width: 1px; height: 1px;"></textarea>

So the input element is the text area, and you can send data to this element. Try

$('textarea[class="inputarea"]').sendKeys('something');

Note: you can use this approach of getting outer HTML of the active element in cases where you are not sure about the element but browser actions work.

Summary:

So you can use two approaches:

1:

await elem.click()
await browser.actions().sendKeys('This is test').perform();

2:

await elem.click()
let field= await browser.driver.switchTo().activeElement()
await field.sendKeys("HI");

you can find the locator or element as:

await field.getAttribute('outerHTML');

Opening the chrome driver and browser on a specific port, and connecting to already opened chroem…

Introduction:

This article helps you in understanding how to start chrome driver on a specific port, connect to an existing browser session and opening chrome browser with debug port opened through selenium scripts.

You should use ChromeDriverService for starting chrome in a different port:

import org.openqa.selenium.chrome.WebDriver;
      import org.openqa.selenium.chrome.ChromeDriverService;
      import org.openqa.selenium.chrome.ChromeOptions;
      import org.openqa.selenium.remote.RemoteWebDriver;     

      WebDriver browser = null;
  	  ChromeDriverService service = new ChromeDriverService.Builder()
      .usingDriverExecutable(new File("C:\\chromedriver.exe"))
      .usingPort(4444) 
      .build();
  	  service.start(); 
  	  browser = new RemoteWebDriver(service.getUrl(), new ChromeOptions());

Once your tests are done, make sure to close the browser and the service:

browser.quit()
  service.stop()

Else the port remains open.

Output:

Add alt text

Connecting to an existing chrome browser from selenium:

close all chrome browsers and then, start the Chrome browser in debug mode by opening cmd and running chrome.exe with — remote-debugging-port argument

"<path>\chrome.exe" --remote-debugging-port=1559

Now open any other browser and navigate to localhost:1559 to check that the browser was indeed opened with that port

Now in selenium code use chrome experimental option to connect to this chrome option

System.setProperty("webdriver.chrome.driver", "C:\\chromedriver.exe");
    ChromeOptions options = new ChromeOptions();
    options.setExperimentalOption("debuggerAddress", "127.0.0.1:1557");
    WebDriver browser=new ChromeDriver(options);

Running Chrome on specific debug port from selenium:

System.setProperty("webdriver.chrome.driver", "C:\\chromedriver.exe");
    ChromeOptions options = new ChromeOptions();
    options.addArguments("--remote-debugging-port=1557");
    WebDriver browser = new ChromeDriver(options);
    options.setExperimentalOption("debuggerAddress", "127.0.0.1:1557");
    browser=new ChromeDriver(options);

Output:

Writing first test script and chaining outputs between scripts.

Till now , we have executed requests but have not written any testscripts. Meaning, there were no validation steps to verify whether the request passed or failed.

Now let us see who to do this. Let us write our first test script.!!!! 🙂

Automation tests with Postman:

There are two parts were we can write the scripts for our tests.

  1. Pre-request script
  2. Tests

Here pre-request script act as the test setup script, meaning, it will get executed before the tests. So if you have any pre-requisite like creating user variable or validating whether a user variable value exists, you can do it here.

Tests are were we write the actual tests, it get executed after we get the response.

Writing first Test script:

The syntax of writing tests are as follows:

pm.test("Test_case_description", function () {
    <Validation steps!> ( The steps are available as test snippet you just need to select it )
});

Note:

We can write tests at both the places ( Pre-request or Tests) but as tests are meant to be validating the user action, it should be written at the ‘Test’ tab so that it will be executed only after the action is completed.

Postman pm.test function expects different exceptions and passes or fails the test accordingly. You can try running just the below command without the ‘pm.test’ enclosed and observe that on sending the request an assertion is thrown.

(The below script is to validate whether the text ‘test’ as sub-string ‘b’ in it.)

pm.expect('test').to.have.string('b');

Now enclose the command in pm.test as below on both pre-request and Tests tab:

pm.test("Check string", function () {
   pm.expect('test').to.have.string('b');
});

Now send the request and you will observe that two tests were executed.

Run the same using collection runner (Make sure to click save before running runner):

Chaining the response together:

Get the data from response body and store it in an environmental variable:

pm.test("Get email", function () {
    var jsonData = pm.response.json();
    var email= jsonData.data[2].email;
    pm.environment.set("email", email);
});

Use this environmental variable in next test.

{
    "email": "{{email}}",
    "password": "cityslicka"
}

Till now , we have executed requests but have not written any testscripts. Meaning, there were no validation steps to verify whether the request passed or failed.

Now let us see who to do this. Let us write our first test script.!!!! 🙂

Automation tests with Postman:

There are two parts were we can write the scripts for our tests.

  1. Pre-request script
  2. Tests

Here pre-request script act as the test setup script, meaning, it will get executed before the tests. So if you have any pre-requisite like creating user variable or validating whether a user variable value exists, you can do it here.

Tests are were we write the actual tests, it get executed after we get the response.

Writing first Test script:

The syntax of writing tests are as follows:

pm.test("Test_case_description", function () {
    <Validation steps!> ( The steps are available as test snippet you just need to select it )
});

Note:

We can write tests at both the places ( Pre-request or Tests) but as tests are meant to be validating the user action, it should be written at the ‘Test’ tab so that it will be executed only after the action is completed.

Postman pm.test function expects different exceptions and passes or fails the test accordingly. You can try running just the below command without the ‘pm.test’ enclosed and observe that on sending the request an assertion is thrown.

(The below script is to validate whether the text ‘test’ as sub-string ‘b’ in it.)

pm.expect('test').to.have.string('b');

Now enclose the command in pm.test as below on both pre-request and Tests tab:

pm.test("Check string", function () {
   pm.expect('test').to.have.string('b');
});

Now send the request and you will observe that two tests were executed.

Run the same using collection runner (Make sure to click save before running runner):

Chaining the response together:

Get the data from response body and store it in an environmental variable:

pm.test("Get email", function () {
    var jsonData = pm.response.json();
    var email= jsonData.data[2].email;
    pm.environment.set("email", email);
});

Use this environmental variable in next test.

{
    "email": "{{email}}",
    "password": "cityslicka"
}