These are chat archives for rosshinkley/nightmare

Jun 2016
Rob Brackett
Jun 16 2016 00:47
hmmmmm, it usually (or used to?) fire did-finish-load after did-fail-load, which would have set isReady back to true regardless of the result of the navigation (which is what we’d want)
Wonder if that’s a bug with attachments or a change in behavior… need to test, I guess
anyway, I think you’d set isReady back to true regardless of the success or failure
So it really wouldn’t be dirty to do it on did-fail-load. I don’t think it should matter what the code is, but you’d want to make sure isMainFrame === true
Although, hmmm, I suppose there is a race condition when a goto is aborted by another call to goto.
Rob Brackett
Jun 16 2016 00:56
Not sure if there's a guaranteed event order there
need to make sure what you’re observing (did-fail-load on response w/ Content-Disposition: attachment and no accompanying did-finish-load) is actually what’s going on, since I think that would be an Electron bug. But working around it if so shouldn't be too bad
Minjong Chung
Jun 16 2016 06:12
Does Nightmare work smoothly on Ubuntu as well (headlessly)?