Can I execute multiple test cases in parallel through Nightwatch ?Is it Possible?
I am searching for ability of threading capability in java for parallel test case execution.
Also what do you guys think about moving from Selenium to Nightwatch?
You can see the thread for parallelism: nightwatchjs also take a look into parallel run
Nightwatch is using the same selenium webdriver protocol but with some extra additions.
Yes you can leverage the parallel mode of nightwatch js using following configuration:
test_workers: {
enabled: true,
workers: 'auto'
}
To execute tests in multiple browsers, you need to add the desired capabilities of the browsers and Test_worker configurations in nightwatch.json file.
For eg. if you want to use Opera you have to add this config:
"cli_args": {
//path to Opera Webdriver File
"webdriver.opera.driver": "bin/operadriver"
}
"opera": {
"desiredCapabilities": {
"browserName": "opera"
}
}
For Test_Worker Configuration you should add:
"test_workers": {
"enabled": true,
"workers": "auto"
}
For example if you want to execute tests in three browsers parallely - Chrome, Firefox and Opera, your nightwatch.json should something like this.
{
"src_folders": [
"tests"
],
"output_folder": "reports",
"selenium": {
"start_process": true,
"server_path": "bin/selenium-server-standalone-3.12.0.jar",
"log_path": "",
"port": 4444,
"cli_args": {
"webdriver.chrome.driver": "bin/chromedriver",
"webdriver.gecko.driver": "bin/geckodriver",
"webdriver.opera.driver": "bin/operadriver"
}
},
"test_workers": {
"enabled": true,
"workers": "auto"
},
"test_settings": {
"default": {
"launch_url": "http://localhost",
"selenium_port": 4444,
"selenium_host": "localhost",
"silent": true,
"screenshots": {
"enabled": false,
"path": ""
},
"desiredCapabilities": {
"browserName": "chrome"
}
},
"firefox": {
"desiredCapabilities": {
"browserName": "firefox",
"marionette": true
}
},
"opera": {
"desiredCapabilities": {
"browserName": "opera"
}
}
}
}
For more info, you can look into this article: How To Execute Tests In Multiple Browsers Parallely With NIGHTWATCH JS.
Related
I've been trying to browser.actions().mouseMove(element(by.xpath('some_xpath'))).perform with protractor to move mouse to the element on page. And geted error when run test on CI\CD. Locally it works fine.
Chrome version 80
chromedriver version 80
here my protractor.config file where i turn w3c to false
multiCapabilities: [{
browserName: 'chrome',
platform: 'ANY',
version: 'ANY',
//Disable or Enable multiCapabilities
shardTestFiles: true,
maxInstances: 8,
// maxSession: 10,
splitTestsBetweenCapabilities: true,
chromeOptions: {
w3c: false,
args: [
'incognito',
'disable-extensions',
'disable-infobars',
// "--headless",
// "--no-sandbox",
"--disable-gpu",
"--window-size=1920,1080",
'show-fps-counter=true',
// '--use-fake-device-for-media-stream',
// '--use-fake-ui-for-media-stream'
"--auto-select-desktop-capture-source=Entire screen",
"--aggressive-cache-discard",
"--disable-cache",
"--disk-cache-size=0"
],
prefs: {
download: {
prompt_for_download: false,
directory_upgrade: true,
default_directory: downloadsPath
}
}
}
}],
I didn't find any information on how to properly setup the nightwatch config using Safari technology preview.
I've tried something like this:
module.exports = {
"src_folders": [
"suites"
],
"output_folder": "reports",
"custom_commands_path": "commands",
"page_objects_path": "pageObjects",
"globals_path": "globals.js",
"selenium": {
"start_process": true,
"server_path": seleniumServer.path,
"host": "127.0.0.1",
"port": 4444,
"log_path": "logs",
"cli_args": {
"webdriver.chrome.driver": chromedriver.path,
"webdriver.firefox.driver": '/usr/bin/geckodriver',
"webdriver.safari.driver": '/usr/bin/safaridriver'
}
},
"test_settings": {
"default": {
"launch_url": "http://google.com",
"screenshots": {
"enabled": true,
"on_failure" : true,
"on_error" : true
},
"globals": {
"waitForConditionTimeout": 5000,
"retryAssertionTimeout": 5000
},
"desiredCapabilities": {
"browserName": "safari",
"browserVersion": "12",
"technologyPreview": true
},
"skip_testcases_on_fail": false,
"end_session_on_fail": false
}
}
};
But after execution I get this error:
{ value:
{ message: 'Could not create a session: A browser with name \'safari\' version \'12\' could not be found on the system.\nBuild info: version: \'3.13.0\', revision: \'2f0d292\', time: \'2018-06-25T15:32:19.891Z\'\nSystem info: host: \'Boostas-MacBook-Pro-5.local\', ip: \'fe80:0:0:0:100a:70f9:d6d7:cbfc%en0\', os.name: \'Mac OS X\', os.arch: \'x86_64\', os.version: \'10.13.6\', java.version: \'10.0.1\'\nDriver info: driver.version: unknown\nremote stacktrace: ',
error: 'session not created' },
status: 33 }
Note: If I set technologyPreview to false the test will be run in Safari browser.
Hope you guys can help me!
Those two options worked in my case in nightwatch.json to launch Safari Technology Preview :
"desiredCapabilities": {
"browserName": "safari",
"safari.options": {"technologyPreview": true},
}
Useful reference : https://macops.ca/using-safari-technology-preview-with-selenium-webdriver/
Hope this helps !
In my react project, I would like to use nightwatch as testing tool. I'm actually using Nightwatch v1.0.4 and selenium-server-standalone-3.9.1.jar on windows.
And this is my config (nightwatch.json) :
{
"src_folders": [
"tests"
],
"output_folder": "reports",
"custom_commands_path": "",
"custom_assertions_path": "",
"page_objects_path": "",
"globals_path": "",
"disable_colors": false,
"test_workers": false,
"selenium": {
"start_process": true,
"host": "localhost",
"port": 4444,
"server_path": "./bin/selenium-server-standalone-3.9.1.jar",
"log_path": "./logs",
"cli_args": {
"webdriver.chrome.driver": "./bin/chromedriver"
}
},
"desiredCapabilities": {
"browserName": "chrome",
"acceptSslCerts": true
},
"test_settings": {
"default": {
"webdriver": {
"server_path": "./bin/chromedriver",
"cli_args": [
"--log",
"debug"
]
},
"disable_colors": false,
"screenshots": {
"enabled": false,
"path": ""
},
"request_timeout_options": {
"timeout": 60000,
"retry_attempts": 5
},
"default_path_prefix" : "",
"desiredCapabilities": {
"browserName": "chrome",
"acceptInsecureCerts": true
}
},
"chrome": {
"webdriver": {
"port": 9515,
"default_path_prefix": "",
"server_path": "./bin/chromedriver",
"cli_args": [
"--verbose"
]
},
"desiredCapabilities": {
"browserName": "chrome",
"loggingPrefs": {
"driver": "INFO",
"server": "OFF",
"browser": "INFO"
}
}
}
}
}
And my test file (tests/index.js):
const host = 'http://localhost:3000'
module.exports = {
'Demo test' : function (browser) {
browser
.url('http://google.com')
.waitForElementVisible('body', 1000)
.pause(5000)
.end();
}
}
When I run nightwatch, it gives me the following error:
Can anyone tell me what I'm doing wrong please ?
Your ChromeDriver version -which you've mentioned as v2.9 is pretty old.
As of the ChromeDriver download page, as per your Chrome version, you should use ChromeDriver version above 2.36 - I would recommend the latest 2.38
Here is my problem: I want be abble to set firefox in specific language to run our e2e tests, and be abble to distribute them.
all verions:
Mozilla Firefox 53.0.3
Selenium 3.4.0 (from npm)
geckodriver 0.17.0 (from npm)
nightwatch.conf.js:
const seleniumServer = require('selenium-server');
module.exports = {
"src_folders": [
"test/functional/features"
],
"custom_commands_path": [
"test/functional/commands/commun",
"test/functional/commands/project"
],
"output_folder": false,
"selenium": {
"start_process": true,
"server_path": seleniumServer.path,
"port": 4444,
"cli_args": {
"webdriver.gecko.driver": "./node_modules/.bin/geckodriver"
}
},
"test_settings": {
"default": {
"selenium_port": 4444,
"selenium_host": "localhost",
"silent": true,
"screenshots": {
"enabled": false
},
"detailed_output": false,
"desiredCapabilities": {
"javascriptEnabled": true,
"acceptSslCerts": true,
"marionette": true,
"browserName": "firefox"
}
}
}
};
In general I tryed two way:
dynamically create a firefox profile, and use it.
In theorie, I can use a before(Each) and use firefox-profile. In practice it create my profile but doesn't use it.
const FirefoxProfile = require('firefox-profile');
function setProfile(browser, profile, callback) {
profile.encoded(function (encodedProfile) {
browser.options.desiredCapabilities['firefox_profile'] = encodedProfile;
callback();
});
}
function setPrefToFrench(browser, done) {
var FFprofile = new FirefoxProfile();
FFprofile.setPreference('general.useragent.locale', 'fr');
FFprofile.setPreference('intl.accept_languages', 'fr');
setProfile(browser, FFprofile, done);
}
module.exports = {
'default' : {
isLocal : true
},
beforeEach: function(browser, done) {
setPrefToFrench(browser, done);
done();
}
};
Save a profile (why not ziped) and include it in project
The same, in theorie, i just have to add it on the nightwatch.conf.js
"selenium": {
"start_process": true,
"server_path": seleniumServer.path,
"port": 4444,
"cli_args": {
"webdriver.gecko.driver": "./node_modules/.bin/geckodriver",
"webdriver.firefox.profile": "path/to/my/profile(or base64 zip)"
}
},
But all time the same, my firefox is still in english.
Does anyone whould have a solution, or maybe I made something wrong?
I am attempting to set up a Sencha extjs 6 project. I would like to be able to have both the modern and classic packages available. I'm following this set-up tutorial, but am not finding it very helpful.
If my understanding is correct, I should be able to have one app that accessible as both a desktop and mobile site. This is great and works for the uncompiled path (my app is in the folder root/sencha) mysite.com/sencha. However, if access the root site it also gives me the uncompiled path instead of using the build app.js.
My index page is at root/layouts/index.html. My extjs app is in root/sencha.
The relevant parts of my app.json are:
...
"indexHtmlPath": "../layouts/index.html",
"production": {
"output": {
"page": {
"path": "../../../../layouts/index.html",
"enable": false
},
"appCache": {
"enable": true,
"path": "cache.appcache"
},
"microloader": {
"path": "microloader.js",
"embed": false,
"enable": true
}
},
"loader": {
"cache": "${build.timestamp}"
},
"cache": {
"enable": true
},
"compressor": {
"type": "yui"
}
},
"output": {
"base": "${workspace.build.dir}/${build.environment}/${app.name}",
"page": "index.html",
"manifest": "${build.id}.json",
"js": "${build.id}/app.js",
"appCache": {
"enable": false
},
"resources": {
"path": "${build.id}/resources",
"shared": "resources"
}
},
...
As pagep suggested you just deploy on your production server the content of the build/production directory and not the source code. That way you only deploy the minimized and compressed files (classic/app.js, classic/resources/YourApp.css, modern/app.js, modern/resources/YourApp.css, etc...).