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
* Add `--leave-running` flag to E2E test script
The `--leave-running` flag has been added to the E2E test runner. This
ensures the browser, ganache, and everything else stays running upon
test failure. This is useful for local debugging, for investigating
what state the extension was in when it failed.
* Add `--leave-running` support to `metamask-ui.spec.js`
* warn users when they attempt to add a network that is already configured
* clean up validation logic
* fixing up e2e tests
* Update test/e2e/helpers.js
Co-authored-by: Mark Stacey <markjstacey@gmail.com>
Co-authored-by: Mark Stacey <markjstacey@gmail.com>
The metrics e2e test would fail if the segment events still weren't
dispatched when the page loaded. The Segment events are sent on a set
interval, so it isn't abnormal for them to lag behind the page load
itself. The `waitUntilCalled` utility has been used to wait until all
required events have been dispatched.
The `wait-until-called` module was converted to an ES5 module, so that
it could be used from an e2e test. The optional `callCount` parameter
has also been added, to allow waiting for more than one call.
The `segmentSpy` had to be converted to a `segmentStub`, to allow the
`waitUntilCalled` utility to be used.
An e2e test has been added that uses the new mock Segment server to
verify that the three initial page metric events are sent correctly.
Using the mock Segment server requires a special build with this mock
Segment server hostname embedded, so a distinct job for building and
running this test was required. As such, it was left out of the
`run-all.sh` script.
`@metamask/eslint-config` has been updated to v4.1.0. This update
requires that we update `eslint` to v7 as well, which in turn requires
updating most `eslint`-related packages.
Most notably, `babel-eslint` was replaced with `@babel/eslint-parser`,
and `babel-eslint-plugin` was replaced by `@babel/eslint-plugin`. This
required renaming all the `babel/*` rules to `@babel/*`.
Most new or updated rules that resulted in lint errors have been
temporarily disabled. They will be fixed and re-enabled in subsequent
PRs.
The `withFixtures` helper function now has the option of starting the
test dapp as well. It will wait to ensure it has started up correctly,
and it'll shut it down when the test ends.
The e2e test helper function `withFixtures` now includes verbose
reporting on failure. Whenever a test fails, debugging information will
be saved to disk, just as with the other e2e test modules.
e2e test that use the `withFixtures` helper now check for console
errors after each successful test. If any errors are found, the test
fails.
It's currently enabled for Chrome only, because the Firefox driver
throws an error when you attempt to get the browser logs. Not sure why
exactly, but it's a long-standing problem.
The `withFixtures` helper will instantiate ganache, a web driver, and
a fixture server initialized with the given set of fixtures. It is
meant to facilitating writing small, isolated e2e tests.
The first example test has been added: simple-send. It ensures that the
user can send 1 ETH to another account.
These new e2e tests will run during the normal e2e test run.
Closes#6548
The Selenium webdriver is difficult to use, and easy to misuse. To help
use the driver and make it easier to maintain our e2e tests, all driver
interactions are now performed via a `driver` module. This is basically
a wrapper class around the `selenium-webdriver` that exposes only the
methods we want to use directly, along with all of our helper methods.
Typically the fullscreen UI will open upon installation, though this
behaviour was suppressed in development. This was dealt with in the e2e
tests by waiting for it to open, then closing it.
Instead this behaviour is now suppressed for test builds as well.
* Improve `openNewPage` helper function
The two delays were removed, and the window handle for the new page is
now returned. This was made possible with the new `newWindow` function
added in `v4.0.0-alpha.3` of `selenium-webdriver`.
* Replace recursion with loops
This should result in far more pleasant stack traces for any
exceptions in these functions. It might also be faster. These functions
seem easier to understand as loops as well.
* Remove unused string parameter
The `closeAllWindowHandlesExcept` function has been simplified by
removing a branch that handles the case where the `exceptions`
parameter given is a string. That parameter is never a string.
Update `selenium-webdriver` to v4.0.0-alpha.5. Despite the fact that
this version has "alpha" in the name, the maintainer of
`selenium-webdriver` has described this release as stable [1].
A few APIs were removed or changed in v4, which required changes to our
Firefox webdriver.
The port used for webdriver communication can now be specified
manually. This was required to ensure the threebox tests kept working,
because they used two different driver instances. This new version of
`selenium-webdriver` now uses the same port for each instance of the
webdriver (unlike the old version, which generated a new port for each
one), so it was necessary to manually specify the port to prevent the
same port from being used for both instances.
`chromedriver` required an update, as the version we were using was not
compatible with the new W3C WebDriver protocol. I've updated
`geckodriver` as well, just to bring it in line with the version of
Firefox we are using.
[1]: https://github.com/SeleniumHQ/selenium/issues/5617#issuecomment-373446249
