Skip to content

Latest commit

 

History

History
223 lines (158 loc) · 9.54 KB

File metadata and controls

223 lines (158 loc) · 9.54 KB

Development

Several files which are needed for developing on MetaMask.

Usually each file or directory contains information about its scope / usage.

Segment

Debugging with the Mock Segment API

To start the Mock Segment API:

  • Add/replace the SEGMENT_HOST and SEGMENT_WRITE_KEY variables in .metamaskrc
    SEGMENT_HOST='http://localhost:9090'
    SEGMENT_WRITE_KEY='FAKE'
    
  • Build the project to the ./dist/ folder with yarn dist
  • Run the Mock Segment API from the command line
    node development/mock-segment.js
    

Events triggered whilst using the extension will be logged to the console of the Mock Segment API.

More information on the API and its usage can be found here.

Debugging in Segment

To debug in a production Segment environment:

  • Create a free account on Segment
  • Create a New Workspace
  • Add a Source (Node.js)
  • Copy the Write Key from the API Keys section under Settings
  • Add/replace the SEGMENT_HOST and SEGMENT_WRITE_KEY variables in .metamaskrc
    SEGMENT_HOST='https://api.segment.io'
    SEGMENT_WRITE_KEY='COPIED_WRITE_KEY'
    
  • Build the project to the ./dist/ folder with yarn dist

Events triggered whilst using the extension will be displayed in Segment's Debugger.

Debugging Segment requests in MetaMask

To opt in to MetaMetrics;

  • Unlock the extension
  • Open the Account menu
  • Click the Settings menu item
  • Click the Security & Privacy menu item
  • Toggle the Participate in MetaMetrics menu option to the ON position

You can inspect the requests in the Network tab of your browser's Developer Tools (background.html) by filtering for POST requests to /v1/batch. The full url will be http://localhost:9090/v1/batch or https://api.segment.io/v1/batch respectively.

Sentry

Overview

MetaMask uses Sentry for error tracking and performance monitoring. Events are automatically categorized by environment based on build target and GitHub context.

Sentry Projects

Project Purpose DSN Variable
test-metamask Local dev and one-off QA testing SENTRY_DSN_DEV
metamask-performance CI/CD and automated performance testing SENTRY_DSN_PERFORMANCE
metamask-extension Production events only SENTRY_DSN

⚠️ Important: Never use SENTRY_DSN (production) for local development or CI/CD builds. Use SENTRY_DSN_DEV for local development and manual QA testing. Use SENTRY_DSN_PERFORMANCE for CI/CD pipelines to send performance metrics to Sentry.

Environments

Environment When Used Build Command Webpack Command Branch/Context
production Production releases yarn build prod yarn webpack --mode production --env production release/* branches
staging Main branch builds yarn build dist yarn webpack --mode production main branch
development Local development yarn start yarn webpack --mode development --watch Local
testing E2E test builds yarn build:test yarn webpack --mode production --test Any
pull-request PR builds yarn build dist yarn webpack --mode production pull_request event
release-candidate Release branches yarn build dist yarn webpack --mode production release/* branches
other Local dist builds yarn build dist yarn webpack --mode production Local builds

For non-main build types (beta, flask, experimental), the environment includes the build type suffix:

  • staging-beta, staging-flask, staging-experimental
  • development-beta, development-flask, etc.
  • testing-flask, etc.

Environment Determination Logic

Environments are determined by development/build/utils.js:getEnvironment():

  1. Build target (highest priority):

    • prodproduction
    • dev or testDevdevelopment
    • testtesting
  2. Branch name (for dist builds):

    • release/*release-candidate
    • mainstaging
  3. GitHub event (fallback):

    • pull_requestpull-request
  4. Default: other (local dist builds)

DSN Selection Logic

The Sentry target DSN is determined by app/scripts/lib/setupSentry.js:getSentryTarget():

if (IN_TEST && !SENTRY_DSN_DEV)     → SENTRY_DSN_FAKE (no events sent)
if (METAMASK_ENVIRONMENT !== 'production') → SENTRY_DSN_DEV (test-metamask)
if (METAMASK_ENVIRONMENT === 'production') → SENTRY_DSN (production project)

Note: SENTRY_DSN_PERFORMANCE is used by CI/CD pipelines to send performance metrics and benchmark data to Sentry. It is configured in GitHub Actions secrets.

Querying Sentry

Common Sentry search queries:

What you want Sentry Query
Development errors environment:development OR environment:pull-request
Staging errors environment:staging OR environment:staging-*
Test failures environment:testing OR environment:testing-*
Production only environment:production
Flask builds environment:*-flask
Beta builds environment:*-beta

Local Development Setup

To enable Sentry error reporting for local development:

  1. Copy the example config and set the DSN:

    cp .metamaskrc.dist .metamaskrc
    # Uncomment SENTRY_DSN_DEV in .metamaskrc
  2. Enable Sentry debug logs (optional): add DEBUG=metamask:sentry:* to .metamaskrc

  3. Enable MetaMetrics via Settings > Security & privacy > Participate in MetaMetrics

  4. To test Sentry errors:

    • Add ENABLE_SETTINGS_PAGE_DEV_OPTIONS=true to .metamaskrc and use Developer Options in Settings
    • Or call window.stateHooks.throwTestError() in the browser console

Debugging the Publish Release Flow

Sentry UI Setup

  1. Go to Sentry and Create an Organization, by clicking Account Menu > Switch organization > Create a new organization
  2. Create a Javascript Project, by clicking Projects > Create Project > Javascript
  3. Create a User Auth Token, by clicking Account Menu > User auth tokens
  4. Select your newly created project and grant all permissions to your token
  5. Copy your token to your clipboard

Sentry-Cli Setup

  1. Go to your terminal, inside the metamask-extension project
  2. Login to Sentry using the command line yarn sentry-cli login --auth-token YOUR_TOKEN
  3. List your organizations and copy the id for the organization you want to see yarn sentry-cli organizations list
  4. List your organization projects and copy the id for the you created yarn sentry-cli projects list --org YOUR_ORG_ID

Publish a Release to Sentry

  1. Build your desired MetaMask project. Examples:
    1. yarn dist to create an MV3 build
    2. yarn dist:mv2 to create an MV2 build
    3. (and so on)
  2. Move the build to its corresponding folder. Ie: mv dist dist-mv2 (skip this step, if you did the regular MV3 build)
  3. Publish the release to Sentry:
    1. If it's an MV3 build yarn sentry:publish --org YOUR_ORG_ID --project YOUR_PROJECT_ID
    2. If it's an MV2 build yarn sentry:publish --dist mv2 --org YOUR_ORG_ID --project YOUR_PROJECT_ID
  4. See build files and source maps are uploaded
  5. Go to Sentry
  6. Check the Source Maps have been uploaded correctly in Sentry: go to Settings > Projects > Project Name > Source Maps

Extra Note: if you already uploaded one version, you can change the package.json version and run again the publish step, to test the complete flow.

Source Maps

Debugging production builds using Source Maps

To unbundle the extensions compiled and minified JavaScript using Source Maps:

  • Open Chrome DevTools to inspect the background.html or home.html view
  • Click on the Sources tab in Chrome DevTools
  • In the Sources tab, click on the Page panel
  • Expand the file directory in the Page panel until you see the source files you're after
  • Select a source file in the Page panel
chrome-extension://{EXTENSION_ID}/common-0.js
  • Double click the source file to open it in the Workspace
  • Right click in the body of the source file and select Add source map...
  • Enter the path to the corresponding source map file, and Click Add
file:///{LOCAL_FILE_SYSTEM}/metamask-extension/dist/sourcemaps/common-0.js.map
  • Repeat the steps above as necessary adding all the relevant source map files
  • Your source maps should now be added to the DevTools Console, and you should be able to see your original source files when you debug your code