Inside registerServiceWorker, we were originally using the nullash coalescing
operator to check if options.base was null or undefined. However, I realized
this check is not necessary.
If you look at getOptions' return value, we return an object with a key "base"
which is of type "string". We get that value by calling resolveBase which always
returns a string.
As a result, we didn't need to check if options.base was null or undefined
because it never can be.
- Moved everything I could into the class itself.
- Improve the logging situation a bit.
- Switch some trace logs to debug.
- Get debug port from message arguments.
This changes adds a new method called `.canTry` to the rate limiter to check if
there are tokens remaining in the bucket.
It also adds suggestions from @oxy to make sure the user can brute force past
the rate limiter.
Before, we weren't checking if a login was successful before counting it
against the rate limiter.
With this change, we only count unsuccessful logins against the rate limiter.
We did this because this was a bug but also because it caused problems with our
e2e tests hitting the rate limit.
It seems reaching into lib/vscode for the types caused tsc to establish
watches that caused it to restart over and over while vscode was
building.
The strategy used here is to symlink it instead which is the same thing
we do for the proxy agent.
Remove the Mac directory copy instead of refactoring it since we've had
this for a long time now and I think it's safe to assume that users
running code-server on Mac don't have the old directory anymore.
After thinking about it some more it's probably mostly only useful to
see the output when the tests are hanging. Otherwise there's a lot of
noise about Jest child processes and pipes.
