jgraham

Do you have logs?

I'm not sure why it's trying to build it there; maybe something happened to the wheel package. In any case passing `--prefer-binary` or `--only-binary=:all:` to the pip command seems like...

It means that it won't pause until the browser is closed after each test. You should supply it because by default we'll pause if we are only running one tests...

wptrunner already provides an abstraction layer here. Each `executor` has to implement the interface at `https://github.com/w3c/web-platform-tests/blob/master/tools/wptrunner/wptrunner/executors/base.py#L100`. That more or less just requires a `run_test` method that loads the test and...

I wonder if it is possible to add the firefox prefs to http://searchfox.org/mozilla-central/source/testing/profiles/prefs_general.js; we already apply those testing prefs when running wptrunner. Otherwise, if you only want to apply these...

So, re: `ERROR`, I think there are two cases: * An error resulting from an in-browser condition (e.g. an uncauche exception) * An error resulting from a harness problem. Currently...

FWIW I would strongly prefer we run with a real WM running, so that we aren't testing for bugs in the headless mode.

We can't always store the bitmap because we don't always generate a bitmap (notably in the Firefox implementation one of the key optimisations is to only generate a PNG when...

Sure, the dashboard could/should store it. But it could/should store the full run logs for debugging which could/should also contain this data, so I'd fix it that way. You probably...

Results are extracted from the logs. Raw logs for a full run are rather large (maybe ~1Gb total, uncompressed), but of course they are incredibly compressible (maybe 24:1 or something,...