| KGB-0 | rhythmbox ubuntu/master 3770658 Jeremy Bicha debian/ changelog control control.in * Drop Build-Depends on gir1.2-webkit2-4.0 * https://deb.li/3qTtc | 00:17 |
|---|---|---|
| bandali | lissyx, even that didn't help either. starting to suspect this might be a red herring | 00:42 |
| bandali | lissyx, how do you prefer i proceed? should i push the commit to beta to at least unblock the builds for now? or..? | 00:42 |
| lissyx | bandali, pushing with a broken test is not going to be super useful ... | 05:39 |
| lissyx | bandali, you have not updated D140803 ? the version I download is still based on dec 12 :( | 05:41 |
| lissyx | bandali, and it does not rebase properly | 05:42 |
| lissyx | I tried something not sure if it's the correct one but I'll see if I can run it locally | 05:50 |
| lissyx | bandali, https://treeherder.mozilla.org/jobs?repo=try&revision=11ef8062a0cc3811d25d82e63b600fd6cf142dce&selectedTaskRun=Z2lQWv8qRNOFY6FjKelS4A.0 | 08:16 |
| lissyx | bandali, so fixing the ChromeUtils.import() with ChromeUtils.importESModule() seems OK ? | 08:17 |
| bandali | bandali, sorry, didn't realize you wanted me to push to phab right away, i thought you meant i should do that eventually | 13:00 |
| bandali | lissyx, what diff does that treeherder try link correspond to? https://hg.mozilla.org/try/rev/11ef8062a0cc3811d25d82e63b600fd6cf142dce doesn't seem very illuminating ? | 13:01 |
| bandali | err, pinged myself in the first message lol. clearly haven't woken up yet | 13:01 |
| bandali | lissyx, so, are you confirming that replacing the one occurrence of ChromeUtils.import() with ChromeUtils.importESModule() looks good to you and everything passes? | 13:02 |
| bandali | lissyx, ah it was the other commit. so does that one correspond to my commit to firefox-snap.git's nightly branch for refreshing the patch + change to importESModule as you mentioned? | 13:09 |
| lissyx | i dont know | 13:17 |
| lissyx | see my patch | 13:17 |
| lissyx | https://hg.mozilla.org/try/rev/ce14270d1e49fe6e52e0fa25140d1b8199fb935f | 13:17 |
| lissyx | bandali, as I said I had to take the phab one, and rebase all the way to now | 13:36 |
| bandali | lissyx, ok thanks. i might've found the culprit. i'll test to make sure | 13:47 |
| bandali | fwiw, it seems the patch we carry in firefox-snap.git vs the one in phab have diverged a bit. but perhaps it won't be too hard to consolidate them in the nearish future | 13:47 |
| lissyx | whatever diverged you need to make the one on phab match | 13:51 |
| bandali | yeah | 13:56 |
| bandali | hm actually i think the phab one might have some more things than the git one. for example, in toolkit/components/extensions/NativeMessagingPortal.h in phab it has an OnGetManifestDone function that our git one doesn't have | 13:58 |
| bandali | or like there's some docs in the phab that we don't have. so i think the phab one is mostly more 'forward' than our git one, but yeah | 13:59 |
| lissyx | building the snap using --destructive-mode (I dont want to depend on lxd/multipass), I get a weird message about missing libpulse? | 14:03 |
| lissyx | https://treeherder.mozilla.org/logviewer?job_id=418573239&repo=try&lineNumber=21131 | 14:03 |
| lissyx | bandali, probably because olivier used his patch as soon as possible and in parallel kept updating the phab one? | 14:03 |
| bandali | lissyx, re patch and olivier: yeah, probably so. | 15:29 |
| bandali | lissyx, re building the snap in destructive mode, i've never tried building firefox that way myself. perhaps you're missing some of the dependencies? is libpulse-dev installed? | 15:29 |
| lissyx | shouldn't snapcraft deal with this? | 15:30 |
| lissyx | or destructive mode alleviate? | 15:30 |
| bandali | i'd think so? maybe try adding it to snapcraft.yaml manually? | 16:14 |
| lissyx | i've added to the docker image and it worked in the end | 16:18 |
| lissyx | bandali, cf https://treeherder.mozilla.org/jobs?repo=try&revision=b0b434cdf456239ca0112bb5f2e1ca86df56ba9e&selectedTaskRun=FC7Gg_C5TvyunER8merF1w.0 | 16:30 |
| bandali | cool cool | 16:39 |
| bandali | for some reason mach bootstrap doesn't seem to install python dbusmock in its virtualenv, so i'm having trouble running the tests locally | 16:39 |
| bandali | i'll need to step out to leave shortly, but i hope i'll be able to come back online later tonight to investigate more | 16:40 |
| bandali | err, i didn't mean mach bootstrap specifically, but rather the build system in general | 16:41 |
| bandali | i hope i'll be able to get the tests to finally properly run, but if not, i might push what i have to nightly and probably beta as well anyway | 16:42 |
| lissyx | it's likely taken care in the ci setup itself? | 16:44 |
| lissyx | weird build error on beta-core22 branch https://treeherder.mozilla.org/logviewer?job_id=418600887&repo=try&lineNumber=128 | 16:45 |
| lissyx | but I'm out for today | 16:45 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!