#nix-darwin on 2018-09-12

00:16 nD5Xjz has quit [Ping timeout: 246 seconds]

00:17 carlosdagos has joined #nix-darwin

00:18 nD5Xjz has joined #nix-darwin

00:50 johnw- has joined #nix-darwin

00:50 johnw_ has quit [Quit: Textual IRC Client: www.textualapp.com]

00:52 johnw has quit [Ping timeout: 252 seconds]

01:00 nD5Xjz has quit [Ping timeout: 245 seconds]

01:20 nD5Xjz has joined #nix-darwin

01:31 philr has joined #nix-darwin

01:39 philr has quit [Ping timeout: 252 seconds]

03:04 {^_^} has quit [Read error: Connection reset by peer]

03:04 nikivi has quit [Quit: ZNC is awesome]

03:05 {^_^} has joined #nix-darwin

03:06 nikivi has joined #nix-darwin

03:57 andreabedini has joined #nix-darwin

04:17 andreabedini has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]

04:19 hamishmack has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]

04:27 andreabedini has joined #nix-darwin

05:35 Ericson2314 has quit [Ping timeout: 250 seconds]

07:06 andreabedini has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]

07:36 philr has joined #nix-darwin

07:56 __Sander__ has joined #nix-darwin

08:24 copumpkin has quit [Ping timeout: 244 seconds]

08:25 contrapumpkin has joined #nix-darwin

08:53 __Sander__ has quit [Ping timeout: 240 seconds]

08:53 __Sander__ has joined #nix-darwin

11:19 philr has quit [Quit: WeeChat 2.2]

12:31 jtojnar has quit [Ping timeout: 252 seconds]

12:37 jtojnar has joined #nix-darwin

12:39 andreabedini has joined #nix-darwin

13:27 andreabedini has quit [Quit: Textual IRC Client: www.textualapp.com]

13:46 carlosdagos has quit [Quit: Connection closed for inactivity]

14:01 contrapumpkin has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]

14:03 copumpkin has joined #nix-darwin

14:03 <copumpkin> I figured out what was wrong with the python, but haven't fixed it yet

14:03 <copumpkin> was pretty subtle

14:03 <LnL> ah nice

14:04 <LnL> what was the problem

14:04 <copumpkin> it happens if your native python extension links against a different libpython from the one your python is using

14:04 <copumpkin> dtruss showed me nothing

14:05 <copumpkin> eventualy DYLD_PRINT_LIBRARIES_POST_LAUNCH=1 did

14:05 <copumpkin> I could see the bad one loading an extra libpython

14:05 <copumpkin> and the good one would just load libxml2

14:07 <LnL> oh neat, that's not in the manpage

14:07 <copumpkin> it's in some version of the manpage

14:08 <LnL> :p

14:08 <copumpkin> not sure if it's meaningfully different from DYLD_PRINT_LIBRARIES=1 but I set both just in case

14:08 <copumpkin> lots of printing :)

14:08 <copumpkin> those dyld variables are pretty handy

14:08 <copumpkin> it's also how I figured out the missing CF initializer the other day

14:08 <LnL> yes, that prints everything that gets loaded including Libsystem, etc.

14:09 <copumpkin> yup

14:10 <LnL> but POST_LAUNCH only other stuff (like dlopen calls?)

14:10 <copumpkin> that's what I gathered

14:14 <LnL> huh, DYLD_FALLBACK_FRAMEWORK_PATH By default, it is set to /Library/Frameworks:/Network/Library/Frameworks:/System/Library/Frameworks

14:14 <LnL> I wonder what /Network is doing in there

14:14 <copumpkin> yeah, that's annoyed me for a while

14:14 <copumpkin> because it'll use stuff we don't want it to use

14:14 <copumpkin> not much to do about it though

14:15 <copumpkin> no idea about /Network

14:15 <LnL> yeah

14:15 <copumpkin> perhaps netboot does that?

14:15 <LnL> I think this also explains why stuff keeps working if the CoreFoundation rpath entry is wrong

14:16 <LnL> got me pretty confused while I was adding that

14:17 <copumpkin> yeah, I wish we had a goods solution for it

14:17 <copumpkin> in our builds we could set the fallback paths env

15:57 __Sander__ has quit [Quit: Konversation terminated!]

16:36 nD5Xjz has quit [Ping timeout: 240 seconds]

16:54 nD5Xjz has joined #nix-darwin

17:25 philr has joined #nix-darwin

18:34 <LnL> urgh, is there a launchd log I don't know about?

18:58 <copumpkin> doesn't it all flow into the mac logging infrastructure?

18:58 <copumpkin> e.g., type "log show"

18:58 <copumpkin> and do your thing

19:01 Ericson2314 has joined #nix-darwin

19:11 <LnL> right, maybe

19:20 Ericson2314 has quit [Ping timeout: 240 seconds]

20:17 philr has quit [Ping timeout: 245 seconds]

21:07 jtojnar has quit [Remote host closed the connection]

21:07 jtojnar has joined #nix-darwin

21:20 jtojnar has quit [Quit: jtojnar]

21:21 jtojnar has joined #nix-darwin

21:40 contrapumpkin has joined #nix-darwin

21:43 copumpkin has quit [Ping timeout: 244 seconds]

22:52 <LnL> going over hydra failures is so exhausting

22:53 <contrapumpkin> oh shit I forgot that was happening

22:53 <LnL> btw, let me know if I should rebuild the wip branch

22:53 <contrapumpkin> you mean for 18.09?

22:53 <contrapumpkin> nah I think Hydra actually woke up

22:53 <contrapumpkin> and is making progress now :)

22:53 <contrapumpkin> but thanks!

22:54 <LnL> yeah https://github.com/NixOS/nixpkgs/issues/45961

22:54 <{^_^}> #45961 (by samueldr, 1 week ago, open): 18.09 Zero Hydra Failures (Darwin)

22:57 <contrapumpkin> damn, I'll try to take a look this weekend

22:58 <LnL> making progress on CF is more important tho

22:58 <contrapumpkin> yeah, gonna try to get that done sooner :)

22:58 <contrapumpkin> they're being awfully receptive now btw https://github.com/apple/swift-corelibs-foundation/pull/1690

22:58 <{^_^}> apple/swift-corelibs-foundation#1690 (by copumpkin, 52 minutes ago, open): Fix CoreFoundation.h header in standalone CF build

22:59 <LnL> and I'm annotating stuff that we should look at afterwards, eg. https://github.com/NixOS/nixpkgs/commit/642d69b5273ebec16747ea5ce87c18f058faeb03

22:59 <contrapumpkin> nice!

23:00 <LnL> no clue if it's related, but looks very suspicious :)

23:00 <contrapumpkin> yup

23:03 <LnL> also these https://search.nix.gsc.io/?q=FIXME.*CoreFoundation&i=nope&files=&repos=

23:05 <LnL> right now using a weird combination of darwin.apple_sdk.frameworks.CoreFoundation and LDFLAGS can work around certain issues

23:05 <LnL> but we probably don't want to keep those