Ericson2314 has quit [Ping timeout: 244 seconds]
<
copumpkin>
LnL: so I can't quite get rid of cf-private yet, but I can make it a lot nicer
<
Enzime>
copumpkin: :(
<
Enzime>
is this to do with the Swift CoreFoundation stuff?
hamishmack has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<
copumpkin>
yup, I almost have that all working
<
copumpkin>
just fixing some of the stragglers
jtojnar has joined #nix-darwin
philr has joined #nix-darwin
philr has quit [Client Quit]
Ericson2314 has joined #nix-darwin
hamishmack has joined #nix-darwin
nikivi has quit [Quit: ZNC is awesome]
nikivi has joined #nix-darwin
jtojnar has quit [Ping timeout: 252 seconds]
jtojnar has joined #nix-darwin
jtojnar_ has joined #nix-darwin
jtojnar has quit [Read error: Connection reset by peer]
jtojnar_ is now known as jtojnar
<
LnL>
Enzime: the travis failure you asked about before
<
LnL>
nix edit can't find appleDerivations :/
trcc has joined #nix-darwin
<
LnL>
oh, look at this: man darwinup
<
LnL>
I think that might be what manages the immutable base system stuff
trcc has quit [Remote host closed the connection]
trcc has joined #nix-darwin
trcc has quit [Ping timeout: 240 seconds]
jtojnar has quit [Quit: jtojnar]
jtojnar_ has joined #nix-darwin
jtojnar_ has quit [Ping timeout: 272 seconds]
jtojnar_ has joined #nix-darwin
jtojnar_ has quit [Read error: Connection reset by peer]
<
copumpkin>
okay I think the number of failures is becoming acceptable
<
copumpkin>
to me at least :P
<
copumpkin>
I don't think it'd be realistic to try to fix everything with a change this big
<
LnL>
nah, but comparing to the correct trunk eval would help seeing most of the differences
<
LnL>
my last build still had a bunch of builds failing bacause of missing symbols
<
LnL>
did you fix those?
<
copumpkin>
how can I compare to a specific trunk eval?
<
copumpkin>
I "fixed" the missing objc symbols from CF
<
LnL>
you can change the id in the url
<
copumpkin>
hydra doesn't have a "find me earliest eval that includes commit X" secret feature, right? :P
<
LnL>
yeah, not that I know of
<
LnL>
but you can find the nixpkgs in the inputs
<
LnL>
but I guess the question is should we put this in staging or wait?
<
copumpkin>
I rebased on staging
<
copumpkin>
locally
<
copumpkin>
just in case it changed anything in linux
<
copumpkin>
okay I think I found the closest eval after
<
copumpkin>
makes me look even better :D
<
{^_^}>
#46704 (by copumpkin, 28 minutes ago, open): stdenv/darwin: integrate a new CoreFoundation
<
copumpkin>
rebuild-linux: 501
<
copumpkin>
guess it's good I submitted to staging :)
<
copumpkin>
I guess the python boot change might cause that
gchristensen has quit [Ping timeout: 260 seconds]
gchristensen has joined #nix-darwin
<
LnL>
python boot is only used on darwin
<
copumpkin>
might also be one of the other changes I made
<
copumpkin>
haven't bothered tracking it down
<
copumpkin>
you think I should just merge or wait a bit for feedback?
<
copumpkin>
I'll admit I mostly made the PR to get ofborg to take a look :)
<
copumpkin>
oh I broke qtbase, fix incoming
hamishmack has quit [Ping timeout: 246 seconds]
johnw- is now known as johnw
<
copumpkin>
LnL: once I finish verifying this qtbase fix locally, I think I'm just going to merge it unless you see some major issues with it
<
copumpkin>
or matthew speaks up soon :)
<
LnL>
I'm still a bit nervous about the changes, but don't know how to test them better so that's not a good reason to wait
<
copumpkin>
same, but 18.09 is already branched
<
copumpkin>
so I think we'll have time to fix stuff
<
copumpkin>
unless you do want me to wait until after I get back from China
<
copumpkin>
assuming anyone actually uses those named releases on darwin :)
<
LnL>
well, worst case I can revert and put it back on a branch if there's some critical issue I can't fix
<
copumpkin>
you sure? I really can wait :)
<
copumpkin>
oh whoops just realized I forgot to delete the old CF :D
<
copumpkin>
okay qtbase is happy again :)
<
copumpkin>
okay I think I'm as ready as I'll get to merge this
<
copumpkin>
make some noise if you don't want me to merge it, otherwise I will :P
<
copumpkin>
LnL: ^ :)
<
copumpkin>
why are we still supporting qt4?
<
copumpkin>
it uses a deprecated API in CF that has since disappeared
<
copumpkin>
merged and now building the LTO junk locally to make sure it all still works sensibly
<
copumpkin>
that should be far less risky :) it just means that if you ask ld for LTO, it now supports it
nikivi has quit [Quit: ZNC is awesome]
nikivi has joined #nix-darwin