Lots of things depend upon it... I don't really think manifest generation is broken... it just takes a REALLY LONG TIME. We have a number of ways to optimize it. Just need to pick one and implement it. I have a branch somewhere that I could dig back into. On Wed, Oct 18, 2017 at 11:42 AM, Philip Jägenstedt <foolip@google.com> wrote: > > > On Mon, Oct 16, 2017, 18:51 James Graham <james@hoppipolla.co.uk> wrote: > >> On 16/10/17 17:30, John Jansen wrote: >> > I was trying to use the test runner[1] over the weekend. Unfortunately, >> during the 'updating and loading test manifest...' stage, it seems to hang. >> I've tried on multiple machines now and waited for over an hour. >> > >> > Is there an active bug on this? Anyone know what's going on with it? >> >> The manifest generation is rather broken and quite often hangs. No one >> is really working on it; ideally you either volunteer or use the >> wptrunner infrastructure (i.e. wpt run) that is actually maintained and >> more reliable :) >> > > Are there bugs about this? Should we remove the broken mechanism or does > anyone successfully depend on it? > >> -- Shane McCarron Projects Manager, Spec-Ops
Received on Wednesday, 18 October 2017 16:46:56 UTC