This tool measures any web application performances.
This project uses Puppeteer to make painless automation.
To install metrx :
npm install -g metrx
To run the application, just use :
npx metrx <url>
# Or if installed globally
metrx <url>
Several options are available to enhance metrics easily. Use -h (--help)
to display them.
➜ metrx -h
Usage: metrx [options] <url ...>
Measures web application loading metrics
Options:
-r, --repeat [n] The number of times the page metrics are measured (default: 5)
-w, --width [width] The viewport's width to set (default: 1920)
-H, --height [height] The viewport's height to set (default: 1080)
-c, --custom-path [custom-path] Path to custom path configuration file
-o, --output-format [output-format] The desired output format (default: table)
--output-file [output-file] Whether we want to export data in a file, and the desired path to the file
--wait-until [wait-until] The waitUntil value of the Page.reload options accepted by puppeteer
--with-redirects Whether we want to test the timings of the whole redirect chain
--no-headless Defines if we dont want to use puppeteer headless mode
--no-sandbox Disable chrome sandbox mode, mandatory in some systems
-h, --help Output usage information
A custom file path can be set in the cli options. That way, you can tell puppeteer what it should do before measuring any kind of metric.
This option can be useful if you need to be logged in before being able to access your application.
To include your file into the process, just use -c <relative path to your file>
option.
metrx localhost:8000 -c '../../custom-path.mjs'
The custom-path.mjs
file must be an ES module that exports an async function as default export.
// index.js: The custom path function is called like so :
if (customPath) {
const customPathFunction = import(customPath);
await customPathFunction.default(page, logInfo);
}
// custom-path.js: example of login process
const LOGIN_INPUT = 'input[type="login"]';
const PASSWORD_INPUT = 'input[type="password"]';
export default async (page, logInfo) => {
const login = 'my-secret-login';
const password = 'my-really-secret-password';
const loginUrl = 'http://localhost:8080/login';
logInfo(`Loading ${loginUrl}`);
// Go to the login page url, and wait for the selector to be ready.
await page.goto(loginUrl);
await page.waitForSelector(LOGIN_INPUT);
logInfo('Logging in...');
// Type creditentials.
await page.type(LOGIN_INPUT, login);
await page.type(PASSWORD_INPUT, password);
logInfo('Redirecting');
// The process will continue once the redirect is resolved.
return page.waitForNavigation();
};
Those functions have access to two arguments :
page
(ThePage
puppeteer object to be able to access the full puppeteer page instance API)logInfo
(To log custom informations)
You can choose to export to multiple formats and export formated data to a file. For now, only table
, raw
, json
and csv
are available.
table
and raw
data will be exported to a txt
file. To use it, just type :
metrx localhost:8000 --output-format json --output-file ~/results.json
If you don't provide any filename, a file will automatically be created in your current directory.
To make a page reload, metrx
does a Page.reload()
from puppeteer's Page
object. This object accepts a waitUntil
parameter, which defines when the page navigation has succeeded, and when the application should collect the metrics and reload the page. You can find more information about Page.reload()
right here.
To use, just add the --wait-until
flag and the desired options. Since Page.reload
accepts either a String
or an Array
of Strings
, if you want to add multiple values, just split them with a ,
For example:
metrx localhost:8000 --wait-until networkidle0,load
To contribute, just run the following commands :
git clone https://github.com/mcastagnetti/metrx.git
cd metrx
npm install
Then, to use metrx
just run it via cli.js
, for example :
./cli.js http://localhost:8000