Skip to content

datadope-io/synthetics-recorder

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Elastic Synthetics Recorder

Installation and Usage

Download and Install

We publish releases of the Script Recorder on its GitHub repo. You can find downloadable installers there for a variety of platforms.

Note that the Script Recorder is in a Tech Preview phase at the moment, and not supported.

Download and unpack the appropriate installer for your platform, and install it.

Usage

This section describes the basic usage of the Script Recorder. It explains how to start a journey and record steps, as well as testing and outputting a completed script.

After starting up the application, you may input a URL. This URL will be the starting point of the journey script Elastic Synthetics will create.

Once you start a journey, Elastic Synthetics will record actions based on your interaction with the browser window. This includes clicking on text, navigation, focusing on inputs like buttons and text fields, and more. As you complete your journey, you will see the actions you are generating populate in the Script Recorder's window.

You can also add assertions to your journey. Use these to make determinations about the state of the page you are testing. Assertions can include checks for things like the visibility of an element, or the contents of a text field.

The Script Recorder also includes a Pause feature. When you pause the recording session, you may click around the browser window without recording any actions. Un-pause the Recorder to continue recording journey actions.

At any point during or after the recording process concludes, you may test your script. When you click the Test button, Elastic Synthetics will run the journey you have defined. After the test concludes, the Recorder will display results on a per-step basis.

When you are satisfied with the script you have generated, you can save it to file using the Export script button. Additionally, you can display the generated JavaScript code using the Show script feature.

Develop

Note: the recorder is intended to be run against a specific version of node/npm. If you use nvm to manage your versions, you can simply run nvm use to switch to the appropriate version. If not, you can view the current supported version in the .nvmrc file. You can see potential error outputs as a result of using the incorrect version below in the troubleshooting section.

Install the dependencies

npm install

Run the recorder app in dev mode.

npm run dev

Managing Playwright dependency

When updating the version of @elastic/synthetics(called Synthetics agent hereafter), it is important to align the version of Playwright that the Synthetics agent uses and the forked Playwright that is installed by the recorder.

Summary

You are going to create a custom version of the released Playwright. Here is a condensed explanation of this process:

  1. Pull latest Playwright changes from Microsoft -> Elastic fork.
  2. Rebase the synthetics-recorder branch off of latest main and resolve conflicts.
  3. Identify the Playwright version(s) we will need to build.
  4. Spawn custom branch(es) off of the release tag(s) for the version(s) used by the target version of Synthetics.
  5. git cherry-pick our custom commit off of the synthetics-recorder branch into the branches you created from Playwright release tags.
  6. Build Playwright and commit the updated src and lib files, push to Elastic remote.
  7. Update the dependencies in the Recorder's package.json to reference the branch(es) you have pushed.
  8. Build the Recorder and test to make sure the browsers are working and you can export generated code.
Detailed steps
Fetch latest Playwright commits

Go to @elastic/playwright, fetch upstream microsoft:main into main if needed. We keep our modifications in synthetics-recorder branch. It is supposed to have only one extra commit(84309bf)) compared to the main branch. If main branch has new commits, fetch the changes into synthetics-recorder branch by pulling it with rebase option

Prepare to update the Elastic Playwright fork

Pull the remote changes to your machine. If necessary, set the remote as follows:

git remote add upstream https://github.com/microsoft/playwright.git
git remote add elastic https://github.com/elastic/playwright.git
git remote -v
// prints:
// upstream	https://github.com/microsoft/playwright.git (fetch)
// upstream	https://github.com/microsoft/playwright.git (push)
// elastic  https://github.com/elastic/playwright.git (fetch)
// elastic  https://github.com/elastic/playwright.git (push)
Identify which versions of Playwright and Playwright-Core are needed

The Recorder depends on custom versions of the playwright and playwright-core packages. Confirm the Playwright version from Synthetics agent's package.json:

// @elastic/synthetics's package.json
    "playwright": "=1.38.0",
    "playwright-core": "=1.38.1",
Create branches for the target Playwright packages

If playwright and playwright-core are the exact same version, you only need to create one new branch.

What does this mean?

If you look at the Synthetics package.json and see that the version for playwright and playwright-core are the same, you only need to branch off of the release tag for that version. Example:

  // you only need to branch off of tag v1.38.0
  "playwright": "=1.38.0",
  "playwright-core": "=1.38.0"

If these do not match, like in the first example shown in the section, where Synthetics depends on 1.38.0 and 1.38.1, you'll need to build both of those tags. If we fail to provide corresponding custom builds for both versions, the @elastic/synthetics package will pull in the vanilla version of the missing package when installed in the Recorder, and the app will break when it runs because it is missing functionality. Fetch the tags from upstream, checkout to the version, and create a new branch:

git fetch upstream --tags
git checkout -b 1.38.0-recorder v1.38.0
git checkout -b 1.38.1-recorder v1.38.1
Overlay our custom changes to your new release branches

Cherry-pick the final commit from the synthetics-recorder branch. It is likely you will need to resolve some merge conflicts. You can reference older release branches if you have any confusion with conflicts.

Once you have finished the cherry-pick, run npm run build. This will run the script Playwright uses to generate its lib files.

After the build runs, uncomment the package ignore lines in the .gitignore. Depending on which packages your branch needs (playwright, playwright-core, or both), you'll need to uncomment the corresponding lines. Here's a link to an example: !packages/playwright-core/lib. If you don't see the package lib directory for your target package in the .gitignore, you must add it, as all lib directories are ignored by default.

After uncommenting/adding the ignore lines, git status should now show the lib files you built for your target packages. Check all the lib files are staged, and commit/push it to elastic remote:

# example:
git cherry-pick 84309bf
# solve conflicts if necessary
# if building for `playwright-core`, uncomment/add `!packages/playwright-core/lib/` in .gitignore
# if building for `playwright`, uncomment/add `!packages/playwright/lib/` in .gitignore
npm run build
git add .
git commit -m "feat: generate libs"
# if you're reusing these commands, replace the semver with the version you're targeting
git push --set-upstream elastic 1.38.0-recorder
Build the Recorder and test your changes

Test new changes in the Recorder by updating Recorder's package.json. Update the playwright and playwright-core dependencies with the branch you pushed from above step:

// @elastic/synthetics-recorder's package.json
// replace the semver in the fields with the one that matches your package targets.
// (they should correspond to the versions used by the Synthetics version you are using)
{
  "dependencies": {
    // whichever versions of `playwright` and `playwright-core` used by Synthetics must match the versions specified below
    "@elastic/synthetics": "^1.5.2",
    // these links will cause our custom builds of Playwright and Playwright-Core to be used instead of the official releases
    "playwright": "https://gitpkg.now.sh/elastic/playwright/packages/playwright?1.38.0-recorder",
    "playwright-core": "https://gitpkg.now.sh/elastic/playwright/packages/playwright-core?1.38.0-recorder"
  }
}
# run your install to update `package-lock.json` and install your target synthetics and playwright packages
npm install

# you can now run the Recorder and make sure that you can record and test journeys
npm run dev
Javascript Codegen

We generate tailored javascript files for synthetics agent, and we keep that piece of code in the recorder(temporarily), so this can be outdated even though we've updated the Playwright fork. When we update Playwright, it is also advisable to review the JavascriptLanguageGenerator and update our codegen code inherited the class, especially generateAction function that we override.

Build

Build and Package the app for all platforms

npm run pack -- -mwl

About

Record synthetic user journeys

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • TypeScript 84.9%
  • JavaScript 14.0%
  • Other 1.1%