01:38
johnw_ has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
01:50
philr has joined #nix-darwin
04:22
copumpkin has quit [Ping timeout: 272 seconds]
04:24
copumpkin has joined #nix-darwin
06:08
hamishmack has quit [Ping timeout: 272 seconds]
06:45
hamishmack has joined #nix-darwin
07:33
philr has quit [Quit: WeeChat 2.2]
07:42
<
Enzime >
LnL: for new modules to nix-darwin
07:43
<
Enzime >
should we be creating .nix files inside modules/services
07:43
<
Enzime >
or create a new folder with a default.nix inside?
07:44
<
LnL >
default.nix is a bit nicer, then you don't have to rename stuff if you need an extra file in the future
07:44
<
LnL >
good morning btw :)
07:45
<
Enzime >
LnL: it's 5:45pm here ;p
07:45
<
Enzime >
but mornin'
07:46
<
Enzime >
LnL: the fact that the configuration system is currently broken up into options (specification) and config (implementation)
07:46
<
Enzime >
it'd be pretty cool if we could take the options from nixpkgs
07:46
<
Enzime >
and write a Darwin specific implementation
07:49
<
LnL >
that's not really possible with the current module system
07:50
<
LnL >
and I'm scared of accidentally pulling in an implementation of something that does dangerous things for a non nixos system
07:54
<
LnL >
but yeah, copy pasting modules isn't great either :/
07:55
<
Enzime >
remote: Resolving deltas: 100% (131/131), completed with 28 local objects.
07:55
<
Enzime >
remote: Create a pull request for 'add/synergy-service' on GitHub by visiting:
07:55
<
Enzime >
^ oh nice, GH started copying BitBucket
07:55
<
Enzime >
I actually really liked that feature
08:00
<
LnL >
I'll have a proper look after work, but looks good at first glance
08:02
<
Enzime >
is TravisCI on nix-darwin working?
08:02
<
Enzime >
appears to have broken (unrelated to my PR?)
08:03
<
LnL >
error: unsupported builtin function 'buildenv'
08:03
<
LnL >
that's a 2.1 client talking to a 2.0 daemon
08:04
<
LnL >
new and unrelated, no need to worry about that
08:30
copumpkin has quit [Ping timeout: 244 seconds]
08:42
copumpkin has joined #nix-darwin
09:16
nD5Xjz has quit [Ping timeout: 244 seconds]
09:45
hamishmack has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
09:46
carlosdagos has quit [Quit: Connection closed for inactivity]
09:47
hamishmack has joined #nix-darwin
09:50
nD5Xjz has joined #nix-darwin
10:39
hamishmack has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
10:43
hamishmack has joined #nix-darwin
10:48
nD5Xjz has quit [Ping timeout: 272 seconds]
10:53
hamishmack has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
10:54
hamishmack has joined #nix-darwin
11:11
nD5Xjz has joined #nix-darwin
11:16
nD5Xjz has quit [Ping timeout: 240 seconds]
11:20
nD5Xjz has joined #nix-darwin
11:25
nD5Xjz has quit [Ping timeout: 244 seconds]
11:30
nD5Xjz has joined #nix-darwin
11:37
<
Enzime >
LnL: is exposing service.<service>.package necessary?
11:38
<
Enzime >
also because the synergy service is separated into two parts service.synergy.client and service.synergy.server, if I expose .package should I be exposing it on service.synergy or on each service.synergy.client and service.synergy.server
11:39
<
Enzime >
also if I don't expose .package, does it make a difference to tests? (is the issue that the test will actually build synergy?)
11:41
nD5Xjz has quit [Ping timeout: 252 seconds]
11:44
<
LnL >
a package option is useful if you want to override something or use a different version
11:44
<
LnL >
it's also possible with overlays, but that's not as nice or flexible
11:48
nD5Xjz has joined #nix-darwin
12:07
nD5Xjz has quit [Ping timeout: 245 seconds]
12:08
nD5Xjz has joined #nix-darwin
12:13
nD5Xjz has quit [Ping timeout: 252 seconds]
12:18
nD5Xjz has joined #nix-darwin
12:18
jtojnar has quit [Ping timeout: 264 seconds]
12:19
jtojnar has joined #nix-darwin
12:22
nD5Xjz has quit [Ping timeout: 240 seconds]
13:08
nD5Xjz has joined #nix-darwin
13:13
nD5Xjz has quit [Ping timeout: 240 seconds]
13:22
jtojnar has quit [Remote host closed the connection]
13:34
nD5Xjz has joined #nix-darwin
14:23
jtojnar has joined #nix-darwin
14:48
copumpkin has joined #nix-darwin
14:49
copumpkin has quit [Client Quit]
14:59
<
cransom >
<- a big fan of exposing a package option for services, but no, not a requirement.
15:05
nD5Xjz has quit [Ping timeout: 264 seconds]
15:06
copumpkin has joined #nix-darwin
15:48
nD5Xjz has joined #nix-darwin
16:48
nD5Xjz has quit [Ping timeout: 244 seconds]
17:05
Ericson2314 has joined #nix-darwin
17:05
disasm_ has joined #nix-darwin
17:05
<
Ericson2314 >
LnL: really close with darwin -> linux?
17:05
<
Ericson2314 >
I just need elf.h on darwin
17:05
<
Ericson2314 >
for paxctl and patchelf
17:07
<
Ericson2314 >
if you google "elf.h on darwin" this is something jammed in dtrace and also some gist
17:07
<
Ericson2314 >
i just want to not to anything too weird
17:07
<
Ericson2314 >
though
*just* solving the problem, provenance of header be damend, is not too hard
17:08
disasm has quit [Ping timeout: 246 seconds]
17:08
<
LnL >
yeah, there's a sys/elf.h in dtrace
17:08
<
LnL >
the headers are just not copied over
17:09
nD5Xjz has joined #nix-darwin
17:12
<
Ericson2314 >
LnL: as that exposed anywhere in nixpkgs?
17:12
<
Ericson2314 >
*is that?
17:12
<
Ericson2314 >
or maybe not even installed on osx let alone put in nixpkgs, you mean, hmm
17:15
<
LnL >
we have darwin.dtrace, but it currently only includes some binaries
17:15
<
Ericson2314 >
ok cool
17:15
<
LnL >
the headers are in the sources tho
17:28
<
Ericson2314 >
LnL: on second thought those look a bit different but I'll pull out elf.h
17:28
<
Ericson2314 >
out of glibc
17:59
philr has joined #nix-darwin
18:00
<
copumpkin >
you talking about cross-compilation?
18:06
matthewbauer has joined #nix-darwin
18:12
<
nikivi >
say I have a binary I want to run on my system that is not built with nix
18:12
<
nikivi >
where would you put it?
18:13
<
nikivi >
I mean that is in my PATH
18:45
<
copumpkin >
LnL: your eval looks pretty promising. Not quite done yet
18:45
<
copumpkin >
are those 10.12 boxes? 10.13?
18:46
<
LnL >
10.12 and llvm got stuck last night so I had to cancel/restart some stuff :/
18:47
<
copumpkin >
stuck for a concerning reason or stuck for random reason?
18:47
<
LnL >
I think I've had this before
19:08
philr has quit [Ping timeout: 252 seconds]
19:15
matthewbauer has quit [Ping timeout: 264 seconds]
19:41
matthewbauer has joined #nix-darwin
19:41
<
copumpkin >
omg it's matthewbauer
19:47
matthewbauer has quit [Ping timeout: 240 seconds]
19:54
matthewbauer has joined #nix-darwin
20:07
matthewbauer has quit [Ping timeout: 252 seconds]
20:08
<
LnL >
uh oh, "Duration: 12h 37m 47s"
20:13
matthewbauer has joined #nix-darwin
20:25
matthewbauer has quit [Ping timeout: 264 seconds]
20:33
<
LnL >
yeah nope, next build failure
20:38
matthewbauer has joined #nix-darwin
20:39
<
copumpkin >
so the branch is b0rked?
20:40
<
LnL >
oh no, this is for zhf
20:41
<
LnL >
this one does look a bit weird tho Fatal Python error: PyThreadState_Get: no current thread
20:42
<
copumpkin >
where you seeing that?
20:45
<
LnL >
that and libsecurity are the only actual errors so far
20:45
<
copumpkin >
that does look like a legit failure, thanks :) repro'd locally
20:45
<
copumpkin >
will take a look a bit later
20:46
<
LnL >
ah you can reproduce it?
20:49
<
matthewbauer >
The API_AVAILABLE preproc seems to be missing too in the swift-corefoundation branch. xnu has that under Availability.h IIRC
20:50
<
copumpkin >
oh yeah, thanks
20:50
hamishmack has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
20:50
<
copumpkin >
what package breaks because of that? so I can test easily
20:51
hamishmack has joined #nix-darwin
20:51
hamishmack has quit [Client Quit]
20:51
<
LnL >
darwin.libsecurity_utilities
20:51
<
matthewbauer >
darwin.libsecurity_utilities
20:51
<
copumpkin >
cool thanks :)
20:51
<
copumpkin >
will try to fix
20:52
<
copumpkin >
matthewbauer: not sure if you saw, but I also have LTO working
20:52
<
copumpkin >
so will likely get this in, then the LTO stuff in, then update bootstrap tools
20:53
<
matthewbauer >
yeah very exciting!
20:54
<
copumpkin >
basically have a bit more time and a bit more motivation again :)
20:54
<
matthewbauer >
on gnome-doc-utils: i would try doing patchShebangs on xml2po. it may be trying to use the system python through /usr/bin/python
20:54
<
copumpkin >
what were the major differences between the new libdispatch and the other one?
20:54
<
copumpkin >
oh okay
20:55
<
copumpkin >
looks like there's already a substituteInPlace attempting that
20:55
<
copumpkin >
perhaps not enough
20:57
<
copumpkin >
yeah you're right
20:58
<
copumpkin >
oh it did get patched already though
21:00
matthewbauer has quit [Ping timeout: 240 seconds]
21:03
<
copumpkin >
so xml2po is failing to import libxml2 properly
21:03
<
copumpkin >
anyway, will take a deeper look later
21:07
<
copumpkin >
libxml2Python = pkgs.buildEnv { # slightly hacky
21:07
<
copumpkin >
just a tad :P
21:07
<
copumpkin >
not under pythonPackages, all sorts of hackery
21:23
hamishmack has joined #nix-darwin
22:44
clever has quit [Ping timeout: 268 seconds]
23:07
johnw_ has joined #nix-darwin
23:41
nD5Xjz has quit [Ping timeout: 245 seconds]
23:59
nD5Xjz has joined #nix-darwin