This PR enables the exclusion of JavaScript and JSON source by `buildType`, and enables the running of `eslint` under LavaMoat. 80-90% of the changes in this PR are `.patch` files and LavaMoat policy additions.
The file exclusion is designed to work in conjunction with our code fencing. If you forget to fence an import statement of an excluded file, the application will now error on boot. **This PR commits us to a particular naming convention for files intended only for certain builds.** Continue reading for details.
### Code Fencing and ESLint
When a file is modified by the code fencing transform, we run ESLint on it to ensure that we fail early for syntax-related issues. This PR adds the first code fences that will be actually be removed in production builds. As a consequence, this was also the first time we attempted to run ESLint under LavaMoat. Making that work required a lot of manual labor because of ESLint's use of dynamic imports, but the manual changes necessary were ultimately quite minor.
### File Exclusion
For all builds, any file in `app/`, `shared/` or `ui/` in a sub-directory matching `**/${otherBuildType}/**` (where `otherBuildType` is any build type except `main`) will be added to the list of excluded files, regardless of its file extension. For example, if we want to add one or more pages to the UI settings in Flask, we'd create the folder `ui/pages/settings/flask`, add any necessary files or sub-folders there, and fence the import statements for anything in that folder. If we wanted the same thing for Beta, we would name the directory `ui/pages/settings/beta`.
As it happens, we already organize some of our source files in this way, namely the logo JSON for Beta and Flask builds. See `ui/helpers/utils/build-types.js` to see how this works in practice.
Because the list of ignored filed is only passed to `browserify.exclude()`, any files not bundled by `browserify` will be ignored. For our purposes, this is mostly relevant for `.scss`. Since we don't have anything like code fencing for SCSS, we'll have to consider how to handle our styles separately.
Static files have been added for the Flask build. This includes logos
of each size and variety that we use, and it includes the 3D model JSON
file.
Closes#12427
The new logo update includes the gradient feature.
The version of `jsdom` we're using via Jest has also been updated in
the lockfile. This was necessary to get unit tests to pass because the
version we were using previously didn't support `replaceChildren`,
which was added in `jsdom@6.6.0` [1].
[1]: 04f6c13f4a
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
On an M1 Mac, when running `yarn start`, CPU can spike to 100% CPU, and
sometimes a bunch of `mdworker` instances will get spawned. This seems
to be caused by the file-watching mechanism used in dev to automatically
regenerate the build when something is changed. More specifically, we
are using an older version of `watchify`, which uses an older version of
`chokidar`, which is the package that actually does the watching. v4.0.0
of `watchify` upgrades `chokidar` to v3.x ([1]), which comes with
"massive CPU & RAM consumption improvements" ([2]). After the upgrade,
CPU usage decreases to 20-40%.
[1]: https://github.com/browserify/watchify/blob/master/CHANGELOG.md#400
[2]: https://github.com/paulmillr/chokidar/releases/tag/3.0.0
* Connect ledger via webhid if that option is available
* Explicitly setting preference for webhid
* Use ledgerTransportType enum instead of booleans for ledger live and webhid preferences
* Use single setLEdgerTransport preference methods and property
* Temp
* Lint fix
* Unit test fix
* Remove async keyword from setLedgerTransportPreference function definition in preferences controller
* Fix ledgelive setting toggle logic
* Migrate useLedgerLive preference property to ledgerTransportType
* Use shared constants for ledger transport type enums
* Use constant for ledger usb vendor id
* Use correct property to check if ledgerLive preference is set when deciding whether to ask for webhid connection
* Update eth-ledger-bridge-keyring to v0.9.0
* Only show ledger live transaction helper messages if using ledger live
* Only show ledger live part of tutorial if ledger live setting is on
* Fix ledger related prop type errors
* Explicitly use u2f enum instead of empty string as a transport type; default transport type to webhid if available; use constants for u2f and webhid
* Cleanup
* Wrap ledger webhid device request in try/catch
* Clean up
* Lint fix
* Ensure user can easily connect their ledger wallet when they need to.
* Fix locales
* Fix/improve locales changes
* Remove unused isFirefox property from confirm-transaction-base.container.js
* Disable transaction and message signing confirmation if ledger webhid requires connection
* Ensure translation keys for ledger connection options in settings dropdown can be properly detected by verify-locales
* Drop .component from ledger-instruction-field file name
* Move renderLedgerLiveStep to module scope
* Remove ledgerLive from function and message names in ledger-instruction-field
* Wrap ledger connection logic in ledger-instruction-field in try catch
* Clean up signature-request.component.js
* Check whether the signing address, and not the selected address, is a ledger account in singature-request.container
* Ensure ledger instructions and webhid connection button are shown on signature-request-original signatures
* Improve webhid selection handling in select-ledger-transport-type onChange handler
* Move metamask redux focused ledger selectors to metamask duck
* Lint fix
* Use async await in checkWebHidStatusRef.current
* Remove unnecessary use of ref in ledger-instruction-field.js
* Lint fix
* Remove unnecessary try/catch in ledger-instruction-field.js
* Check if from address, not selected address, is from a ledger account in confirm-approve
* Move findKeyringForAddress to metamask duck
* Fix typo in function name
* Ensure isEqualCaseInsensitive handles possible differences in address casing
* Fix Learn More link size in advanced settings tab
* Update app/scripts/migrations/066.js
Co-authored-by: Mark Stacey <markjstacey@gmail.com>
* Update ui/pages/settings/advanced-tab/advanced-tab.component.test.js
Co-authored-by: Mark Stacey <markjstacey@gmail.com>
* Add jsdoc comments for new selectors
* Use jest.spyOn for mocking navigator in ledger webhid migration tests
* Use LEDGER_TRANSPORT_TYPES values to set proptype of ledgerTransportType
* Use LEDGER_TRANSPORT_TYPES values to set proptype of ledgerTransportType
* Fix font size of link in ledger connection description in advanced settings
* Fix return type in setLedgerTransportPreference comment
* Clean up connectHardware code for webhid connection in actions.js
* Update app/scripts/migrations/066.test.js
Co-authored-by: Mark Stacey <markjstacey@gmail.com>
* Update ui/ducks/metamask/metamask.js
Co-authored-by: Mark Stacey <markjstacey@gmail.com>
* Add migration test for when useLedgerLive is true in a browser that supports webhid
* Lint fix
* Fix inline-link size
Co-authored-by: Mark Stacey <markjstacey@gmail.com>
We were using an outdated version of the package `yazl` in our build
system, resulting in a Buffer warning during the production and test
builds about the use of the deprecated Buffer constructor.
`yazl` has been updated to the latest version, and no longer uses the
deprecated Buffer constructor that caused this warning.
The warning looked like this:
```
(node:52293) [DEP0005] DeprecationWarning: Buffer() is deprecated due to security and usability issues. Please use the Buffer.alloc(), Buffer.allocUnsafe(), or Buffer.from() methods instead.
```
`immer` has been updated to v9. This didn't require any changes on our
part; the only breaking changes are to the TypeScript types [1].
The `@reduxjs/toolkit` library has been updated as well, to ensure that
it's using the updated version of Immer internally as well. This update
makes our patch of that package obsolete, as the problematic pattern
that were were patching out is no longer present.
[1]: https://github.com/immerjs/immer/releases/tag/v9.0.0
* lavamoat - add lavamoat to webapp background
* test:e2e - add delay to resolve failure
* test:e2e - add delay to resolve failure
* build - add a switch for applying lavamoat, currently off for all
* test/e2e - remove delays added for lavamoat
* Revert "test/e2e - remove delays added for lavamoat"
This reverts commit 79c3479f15c072ed362ba1d4f1af41ea11a17d63.
The warnings about use of the unsafe Buffer constructor have been
addressed by package updates and patches.
The updates were:
* `gulp-sourcemaps` was updated from v2 to v3, and was patched to
replace remaining uses of the `Buffer` constructor
* Upstream PR: https://github.com/gulp-sourcemaps/gulp-sourcemaps/pull/388
* The transitive dependency `yazl` was updated from v2.4.3 to v2.5.1
in the lockfile.
* The abandoned packages `combine-source-map` and `inline-source-map`
were patched.
The dependency `caniuse-lite` has been updated using a Yarn resolution,
because it was pinned to a specific version by some dependencies. All
versions requested in our dependency tree are 1.x so this did not
introduce any breaking changes.
This resolves a frequent console warning that shows up during builds,
and when running tests and the linter.
* bump @metamask/controllers to v15.0.1 and remove AbortController workaround in e2e tests
* remove old abortcontroller polyfill
* bump @metamask/controllers to v15.0.2
There are a few issues encountered when running `yarn setup` on new
Apple Silicon (aka M1, aka arm64) Macs:
* The script halts when attempting to run the install step for
the `chromedriver` package with the message "Only Mac 64 bits
supported". This is somewhat misleading as it seems to indicate that
chromedriver can only be installed on a 64-bit Mac. However, what I
think is happening is that the installation script for `chromedriver`
is not able to detect that an arm64 CPU *is* a 64-bit CPU. After
looking through the `chromedriver` repo, it appears that 87.0.1 is the
first version that adds a proper check ([1]).
Note that upgrading chromedriver caused the Chrome-specific tests to
fail intermittently on CI. I was not able to 100% work out the reason
for this, but ensuring that X (which provides a way for Chrome to run
in a GUI setting from the command line) is available seems to fix
these issues.
* The script also halts when attempting to run the install step for
the `electron` package. This happens because for the version of
`electron` we are using (9.4.2), there is no available binary for
arm64. It appears that Electron 11.x was the first version to support
arm64 Macs ([2]). This is a bit trickier to resolve because we don't
explicitly rely on `electron` — that's brought in by `react-devtools`.
The first version of `react-devtools` that relies on `electron` 11.x
is 4.11.0 ([3]).
[1]: 469dd0a6ee
[2]: https://www.electronjs.org/blog/apple-silicon
[3]: https://github.com/facebook/react/blob/main/packages/react-devtools/CHANGELOG.md#4110-april-9-2021
* EIP-1559 - Provide support for Ledger
* Update ui/selectors/selectors.js
Co-authored-by: Mark Stacey <markjstacey@gmail.com>
* Add shared constants for hw types
* bump eth-ledger-bridge-keyring to v0.7.0
Co-authored-by: David Walsh <davidwalsh83@gmail.com>
Co-authored-by: Mark Stacey <markjstacey@gmail.com>
Co-authored-by: Alex <adonesky@gmail.com>
Adds the latest version of `@metamask/controllers`, and updates our usage of the `ApprovalController`, which has been migrated to `BaseControllerV2`. Of [the new `controllers` release](https://github.com/MetaMask/controllers/releases/tag/v15.0.0), only the `ApprovalController` migration should be breaking.
This is the first time we use events on the `ControllerMessenger` to update the badge, so I turned the messenger into a property on the main `MetaMaskController` in order to subscribe to events on it in `background.js`. I confirmed that the badge does indeed update during local QA.
As it turns out, [MetaMask/controllers#571](https://github.com/MetaMask/controllers/pull/571) was breaking for a single unit test case, which is now handled during setup and teardown for the related test suite (`metamask-controller.test.js`).
This update includes just configuration updates. There are no
functional changes. The updated config is only used as a fallback in
case the config update fails for some reason.
