you know if i was as incompetent as karolat claims, i doubt that i would have had all this responsibility delegated to me đ
Notices by :abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy: (kaniini@pleroma.site), page 30
-
:abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy: (kaniini@pleroma.site)'s status on Sunday, 21-Jul-2019 21:48:13 EDT :abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy:
-
:abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy: (kaniini@pleroma.site)'s status on Sunday, 21-Jul-2019 21:46:48 EDT :abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy:
@Conan_Kudo @hellcp
yeah but we can do better, by learning from APK ;)
i've been wading through DNF source and all of that. -
:abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy: (kaniini@pleroma.site)'s status on Sunday, 21-Jul-2019 21:21:31 EDT :abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy:
@ben nothing, it's just fragile. -
:abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy: (kaniini@pleroma.site)'s status on Sunday, 21-Jul-2019 21:21:28 EDT :abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy:
@sjw @1iceloops123 @astheroth @no @riking
at any rate, the regression was solved yesterday in commit bc6c5c513ae69e7a868c63f878a009dce8dd8c63 and has very specific underlying causes, namely, Pleroma 1.0 abuses `sharedInbox` in a dubious way. i am more than happy to write a blog about what happened, why it happened, and how the entire class of faults are being eliminated from future releases.
i plan on fixing the `sharedInbox` abuse issue tomorrow and will backport it to 1.0.2 release after it cooks for a while.
this will permanently eliminate the issue by ensuring that AP messaging is done in a completely correct way.
but either way, assuming everyone who updates develop daily has updated, it will resolve itself within the next 24 hours as the broken cached profiles are refreshed. -
:abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy: (kaniini@pleroma.site)'s status on Sunday, 21-Jul-2019 21:14:13 EDT :abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy:
sharedInbox was a mistake honestly
-
:abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy: (kaniini@pleroma.site)'s status on Sunday, 21-Jul-2019 21:12:23 EDT :abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy:
and besides, follows between the Pleroma develop tree with regressions and Pleroma 1.0 wouldnât have been broken if Pleroma used sharedInbox appropriately.
donât worry, thatâs being fixed too.
-
:abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy: (kaniini@pleroma.site)'s status on Sunday, 21-Jul-2019 21:08:43 EDT :abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy:
@1iceloops123 @astheroth @no @riking
this is fascinating and all, but untag me. -
:abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy: (kaniini@pleroma.site)'s status on Sunday, 21-Jul-2019 21:04:24 EDT :abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy:
@sjw @karolat
and realistically, this issue will be completely resolved within the next 24 hours anyway. -
:abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy: (kaniini@pleroma.site)'s status on Sunday, 21-Jul-2019 20:59:13 EDT :abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy:
@karolat @roka @no
because, again, smugloli is sending the reply to the wrong place.
i can give you a workaround that makes the wrong place operate like the right place.
do you want it or not? -
:abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy: (kaniini@pleroma.site)'s status on Sunday, 21-Jul-2019 20:58:00 EDT :abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy:
@riking @no
i mean, i'm really not trying to be a dick here, but the manchild already exceeded my patience.
we already have all of this kind of automated testing.
we're going to have to agree to disagree on this point because i'm not arguing with a manchild and you at the same time. -
:abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy: (kaniini@pleroma.site)'s status on Sunday, 21-Jul-2019 20:54:51 EDT :abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy:
@karolat @no
then ask smugloli admin to invalidate your user so it will refresh, or alternatively wait for it to relearn your user.
either way, the underlying cause of the failure has been fixed. i can give you a workaround, or not, but i'm not going to be able to sit here and argue with you and code your workaround at the same time. -
:abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy: (kaniini@pleroma.site)'s status on Sunday, 21-Jul-2019 20:53:22 EDT :abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy:
yeah i just pulled a drepper, because I REALLY DONâT GIVE A SHIT
-
:abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy: (kaniini@pleroma.site)'s status on Sunday, 21-Jul-2019 20:51:26 EDT :abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy:
@riking @no
unless you're paying me, i really don't care to hear your opinions at this point. -
:abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy: (kaniini@pleroma.site)'s status on Sunday, 21-Jul-2019 20:49:14 EDT :abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy:
@karolat @no
*was* broken, the underlying cause is already fixed. do you want a workaround or not? if you do, let me code it and stop screaming at me. -
:abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy: (kaniini@pleroma.site)'s status on Sunday, 21-Jul-2019 20:47:06 EDT :abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy:
the underlying issue is that some people want the latest code, but they donât want the risk exposure that QA brings.
well, no, the underlying issue is that Pleroma does not have a QA team.
ideally, we would have a process where
develop
is automatically composed from topic branches that have had QA approval. -
:abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy: (kaniini@pleroma.site)'s status on Sunday, 21-Jul-2019 20:42:38 EDT :abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy:
we have integration testing. automated testing doesnât catch everything.
the solution, ultimately, is to make
develop
less bleeding-edge. -
:abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy: (kaniini@pleroma.site)'s status on Sunday, 21-Jul-2019 20:41:02 EDT :abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy:
@karolat i mean, if you can find some other people to help with QA, i'll talk with the rest of the core devs to come up with a way to "cook" the tree. trust me, i want real QA, because i know automated testing doesn't find everything. -
:abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy: (kaniini@pleroma.site)'s status on Sunday, 21-Jul-2019 20:37:26 EDT :abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy:
@karolat
you're right!
following users, ideally, should never break, and i'm sorry that they did.
for what it's worth, I agree with you that the current workflow is suboptimal, but changing the workflow is not something I can unilaterally do.
i would much prefer to have an integration tree where features graduate to develop after they are QA'ed.
unfortunately, we do not have the resources to do this, and such a process will harm the people who are paid to contribute to Pleroma (since, as far as I know, most of them are paid by completed task), so it's unlikely to ever happen.
i really do not need you shitting on me for trying to do the best i can with what i have to work with. -
:abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy: (kaniini@pleroma.site)'s status on Sunday, 21-Jul-2019 20:30:01 EDT :abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy:
@karolat
if you want to complain about instability, run the stable tree.
develop is for people who are willing to take risks in order to have the bleeding edge code.
but you already know this.
accordingly, i see no reason to argue with you, since your motives are simply to harass me in order to convince me to leave. -
:abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy: (kaniini@pleroma.site)'s status on Sunday, 21-Jul-2019 20:25:48 EDT :abunhdhappyhop: :abunhdhappy: :abunhdhop: :abunhd: :abunhdhappyhop: :abunhdhappy:
@karolat
so, do you want a workaround or not?