01:01
jtojnar has quit [Quit: jtojnar]
01:37
<
samueldr >
I'm getting less busy, and updating to 19.03; what is our current stance on dropping attribute names in nixpkgs?
01:37
<
samueldr >
or, more to the point, moving them around?
01:38
<
samueldr >
a6d15b831c19be01dcba68e9626431c15a993ce9 dropped `nixpkgs.elementary-icon-theme` since it was integrated into the broader pantheon bits
01:40
init_6 has joined #nixos-dev
01:56
<
worldofpeace >
samueldr In that particular attribute name change, I thought it wouldn't be intrusive to people
01:58
<
samueldr >
(sorry, didn't want to harp on you)
01:58
<
samueldr >
I was asking generally before the specifics though
02:00
<
worldofpeace >
I think attributes being removed without any nice way to tell people isn't great though
02:09
<
worldofpeace >
atm we use just `aliases.nix` right? Though you don't know you're using an alias.
02:11
<
samueldr >
aliases.nix has the nice property of not being used withing ofborg and hydra
02:13
<
worldofpeace >
hmm, then I have no idea how eval-package-list-no-aliases check works
02:18
<
samueldr >
that's probably what I had in mind
02:48
init_6 has quit [Read error: Connection reset by peer]
02:53
init_6 has joined #nixos-dev
02:54
init_6 has quit [Read error: Connection reset by peer]
03:00
orivej has quit [Ping timeout: 245 seconds]
03:03
init_6 has joined #nixos-dev
03:07
init_6 has quit [Read error: Connection reset by peer]
03:13
init_6 has joined #nixos-dev
03:14
drakonis1 has quit [Quit: WeeChat 2.3]
03:24
tv has quit [Ping timeout: 244 seconds]
03:24
tv has joined #nixos-dev
03:24
phreedom_ has joined #nixos-dev
03:25
phreedom has quit [Ping timeout: 256 seconds]
04:42
pie__ has joined #nixos-dev
04:43
asymmetric has joined #nixos-dev
04:45
pie___ has quit [Ping timeout: 250 seconds]
05:22
asymmetric has quit [Ping timeout: 240 seconds]
05:48
drakonis_ has joined #nixos-dev
05:51
drakonis has quit [Ping timeout: 255 seconds]
06:37
drakonis has joined #nixos-dev
06:40
drakonis_ has quit [Ping timeout: 258 seconds]
06:42
asymmetric has joined #nixos-dev
07:09
lopsided98 has quit [Quit: No Ping reply in 180 seconds.]
07:10
lopsided98 has joined #nixos-dev
08:17
arianvp has quit [Remote host closed the connection]
08:21
johanot has joined #nixos-dev
08:34
asymmetric has quit [Ping timeout: 252 seconds]
08:40
asymmetric has joined #nixos-dev
08:56
johanot has quit [Ping timeout: 255 seconds]
09:00
johanot has joined #nixos-dev
09:01
arianvp has joined #nixos-dev
09:11
jtojnar has joined #nixos-dev
09:50
lopsided98 has quit [Remote host closed the connection]
09:52
lopsided98 has joined #nixos-dev
10:19
init_6 has joined #nixos-dev
10:38
orivej has joined #nixos-dev
11:01
edwtjo has quit [Ping timeout: 240 seconds]
11:22
orivej has quit [Ping timeout: 252 seconds]
11:23
edwtjo has joined #nixos-dev
11:40
teehemkay has quit [Ping timeout: 257 seconds]
11:41
teehemkay has joined #nixos-dev
12:15
orivej has joined #nixos-dev
12:27
orivej has quit [Ping timeout: 240 seconds]
12:30
<
gchristensen >
iirc aliases are checked that they all exist, and also nixpkgs is checked that it works without aliases
12:51
asymmetric has quit [Ping timeout: 255 seconds]
13:02
asymmetric has joined #nixos-dev
13:02
johanot has quit [Quit: WeeChat 2.2]
13:26
drakonis_ has joined #nixos-dev
13:27
orivej has joined #nixos-dev
13:29
drakonis has quit [Ping timeout: 252 seconds]
14:02
drakonis has joined #nixos-dev
14:16
asymmetric_ has joined #nixos-dev
14:17
asymmetric has quit [Remote host closed the connection]
14:17
asymmetric_ is now known as asymmetric
14:45
drakonis1 has joined #nixos-dev
14:49
drakonis_ has quit [Ping timeout: 245 seconds]
14:50
drakonis_ has joined #nixos-dev
14:51
johanot has joined #nixos-dev
14:53
drakonis1 has quit [Ping timeout: 258 seconds]
15:05
johanot has quit [Quit: WeeChat 2.2]
15:07
orivej has quit [Ping timeout: 252 seconds]
15:30
asymmetric has quit [Remote host closed the connection]
15:30
asymmetric has joined #nixos-dev
15:50
<
LnL >
is there a way to define your own fetchurl mirrors outside of nixpkgs?
15:55
asymmetric has quit [Ping timeout: 244 seconds]
16:17
<
andi- >
well you can probably override `fetchurl` and pass in another `mirrorsFile` but that is probably not what you wanted?
16:21
<
gchristensen >
how about a hashed mirror?
16:46
drakonis has quit [Quit: WeeChat 2.3]
16:49
drakonis has joined #nixos-dev
16:52
drakonis_ has quit [Ping timeout: 240 seconds]
17:17
drakonis_ has joined #nixos-dev
17:21
drakonis has quit [Ping timeout: 240 seconds]
17:25
asymmetric has joined #nixos-dev
17:44
drakonis has joined #nixos-dev
17:46
jtojnar_ has joined #nixos-dev
17:48
drakonis_ has quit [Ping timeout: 258 seconds]
17:52
yl has joined #nixos-dev
18:51
drakonis_ has joined #nixos-dev
18:55
drakonis has quit [Ping timeout: 252 seconds]
19:03
orivej has joined #nixos-dev
19:56
pie__ has quit [Remote host closed the connection]
19:57
pie__ has joined #nixos-dev
20:29
drakonis has joined #nixos-dev
20:32
drakonis_ has quit [Ping timeout: 252 seconds]
20:38
<
infinisil >
Hmm, the 19.03 milestone is due 2019-03-01, shouldn't this be 2019-04-01?
20:40
drakonis has quit [Read error: Connection reset by peer]
20:40
drakonis has joined #nixos-dev
20:42
<
samueldr >
you want this to go out in may? ;)
20:43
<
samueldr >
I don't think the date due is really useful tbf :)
20:43
Cale has quit [Ping timeout: 250 seconds]
20:43
<
samueldr >
it should be 1970-01-01 00:00:00 :D
20:43
<
samueldr >
(like the store paths)
20:43
<
samueldr >
(or is it 00:00:01?)
20:44
<
gchristensen >
but isn't 04-01 the real deadline?
20:45
<
samueldr >
I guess it is
20:50
drakonis_ has joined #nixos-dev
20:54
drakonis has quit [Read error: Connection reset by peer]
20:56
Cale has joined #nixos-dev
21:35
<
gchristensen >
98.9% r13y.com
21:38
<
tilpner >
man-db could be another easy win, but the other ones look harder
21:38
<
gchristensen >
python -- there is a hook clever was telling me about which should fix this
21:39
<
samueldr >
.32731.d smells like a tempdir
21:39
<
clever >
i remember seeing it, but havent been able to find it since mentioning it
21:39
<
samueldr >
(on the linux build)
21:39
<
tilpner >
Yeah, just delete that one
21:39
<
clever >
samueldr: its weird that its present on both builds, but absent on my local build
21:39
<
samueldr >
clever: -dev?
21:39
<
clever >
samueldr: pretty sure i checked the same output
21:39
<
gchristensen >
samueldr: wouldn't it be cool if this also uploaded the build directory?
21:40
<
clever >
gchristensen: ./pkgs/development/interpreters/python/cpython/2.7/nix-store-mtime.patch
21:40
<
clever >
./pkgs/development/interpreters/python/cpython/2.7/deterministic-build.patch
21:40
<
clever >
gchristensen: i think the env var DETERMINISTIC_BUILD modifies python to not save timestamps?
21:41
<
clever >
doc/languages-frameworks/python.section.md:When the environment variable `DETERMINISTIC_BUILD` is set, all bytecode will have timestamp 1.
21:41
<
clever >
its even in the docs! :O
21:41
<
samueldr >
right, realised the same derivation and the build folder doesn't have that temp dir
21:41
<
samueldr >
no it does have it
21:41
<
samueldr >
I was mistaken, this is legit
21:41
<
gchristensen >
really, you cracked it?
21:42
<
gchristensen >
interesting, clever ... hmm
21:44
* clever
heads off to bed
21:45
<
gchristensen >
thanks clever :)
21:46
<
clever >
pkgs/development/interpreters/python/setup-hook.sh:export DETERMINISTIC_BUILD=1;
21:46
<
clever >
ahh, the setup hook on python does it!
21:46
<
clever >
so if you ever have python in your build inputs, it fixes itself!
21:47
<
gchristensen >
well, that can't be true
21:47
<
clever >
something weird is happening
21:47
<
clever >
will need somebody more awake then me to investigate :P
21:48
<
clever >
nix show-derivation /nix/store/8b922yghx2lmjs1hvazphinxp9v03ayh-python2.7-pytest-3.10.1.drv
21:48
<
clever >
gchristensen: oh, ive seen python build systems before, that have a --pure mode in them
21:48
<
clever >
NIX_CFLAGS_COMPILE vanishes, and then everything breaks :P
21:49
* clever
heads off to bed
21:49
<
gchristensen >
see you :)
21:52
<
gchristensen >
DETERMINISTIC_BUILD didn't make a difference when explicitly set
21:53
drakonis has joined #nixos-dev
21:57
drakonis_ has quit [Ping timeout: 245 seconds]
21:58
drakonis has quit [Read error: Connection reset by peer]
22:02
disasm has joined #nixos-dev
22:02
asymmetric has quit [Ping timeout: 240 seconds]
22:02
drakonis has joined #nixos-dev
22:08
drakonis has quit [Ping timeout: 258 seconds]
22:10
disasm has quit [Quit: WeeChat 2.0]
22:14
disasm has joined #nixos-dev
22:18
<
LnL >
oh neat! echo "Hello World" | nix-instantiate --eval -E 'builtins.readFile /dev/stdin'
22:18
disasm has quit [Client Quit]
22:19
disasm has joined #nixos-dev
22:20
<
gchristensen >
oh cool
22:20
<
LnL >
hmm, only works on darwin :/
22:22
<
LnL >
guess I'll have to write a tempfile then
22:24
disasm has quit [Client Quit]
22:38
disasm has joined #nixos-dev
22:47
disasm has quit [Quit: WeeChat 2.0]
22:48
disasm has joined #nixos-dev
22:52
disasm has quit [Client Quit]
22:55
disasm has joined #nixos-dev
23:08
disasm has quit [Quit: WeeChat 2.0]
23:08
disasm has joined #nixos-dev
23:09
<
infinisil >
You just want the `disabled = ! python36Packages.pythonOlder "3.7" ;`?
23:15
<
matthewbauer[m] >
Does anyone know why nixpkgs-19.03-darwin isn't getting published? something in the channel scripts maybe...
23:17
<
gchristensen >
matthewbauer[m]: mind sending a PR?
23:20
<
{^_^} >
nixos-org-configurations#78 (by matthewbauer, 10 seconds ago, open): Add nixpkgs-19.03-darwin to update channels
23:20
<
gchristensen >
thanks
23:20
<
gchristensen >
I'll deploy after supper
23:23
<
samueldr >
tracked down to the command generating the stray temp (looking) dir for the kernel build
23:23
<
gchristensen >
ooh!
23:23
<
gchristensen >
samueldr++
23:23
<
samueldr >
waiting on a build which straces the make invocation
23:23
<
{^_^} >
samueldr's karma got increased to 58
23:24
<
samueldr >
see if there's a better way than rm a pattern matching glob
23:52
yl has quit [Ping timeout: 246 seconds]
23:56
Cale has quit [Ping timeout: 252 seconds]
23:58
<
gchristensen >
well I'll be
23:58
<
samueldr >
a `cc` invocation with -o .348.tmp will make `cc1` produce .348.d from a -MD call
23:59
<
samueldr >
looks like the only sane way to deal with this is to delete .*.d as they're not handled by the cleanup as they're not expected to be cleaned up