<MichaelRaskin>
Profpatsch: now you just need to find enough tests that can be created without an RFC on «no NixOS tests are not good enough for testing GUI packages»?
<Profpatsch>
MichaelRaskin: The interesting parts are headless anyway ;)
<Profpatsch>
One generic tests you can do is binary --version and then parse the output to see whether it contains ${version}
<MichaelRaskin>
That was indeed in my rejected PR
<Profpatsch>
You also test basic execution that way.
<MichaelRaskin>
Basic execution depends on the definition (I did write a few versions of exit code filter there, though)
eadwu has joined #nixos-dev
<Profpatsch>
MichaelRaskin: I had one where I tested a beets plugin, using a special version of beets with the plugin enabled in the test.
<Profpatsch>
dunno where I put it.
<Profpatsch>
I wasn’t sure how to pass the arguments.
<Profpatsch>
In a nice way and without too much boilerplate.
init_6 has joined #nixos-dev
eadwu has quit [Ping timeout: 252 seconds]
eadwu has joined #nixos-dev
eadwu has quit [Read error: Connection reset by peer]
<Profpatsch>
It also inits a local store and calls `nix eval` from within a nix build. :P
eadwu has joined #nixos-dev
<MichaelRaskin>
Well, sure, streaming data crunchers can be tested easily that way
<Profpatsch>
Which is, like, most stuff.
<MichaelRaskin>
(of course, one could hope that make test does it)
<MichaelRaskin>
Most of the stuff re: number of packages — maybe, although complicated
<MichaelRaskin>
Most of the stuff re: number of breakage — I have doubts
<Profpatsch>
The bazel build takes around 15–20 minutes.
<Profpatsch>
And then a trivial test breaks at the last possible moment.
<Profpatsch>
There’s nothing more infuriating.
<Profpatsch>
If the test is run in a different derivation (that gets the new bazel package as input), at least the build is cached.
<MichaelRaskin>
Oh there is. Two letters: LO
<Profpatsch>
LO?
<MichaelRaskin>
LibreOffice
<Profpatsch>
ah, heh
<Profpatsch>
Anyway, need to go to bed.
<Profpatsch>
gn8
<MichaelRaskin>
Actually me too
<MichaelRaskin>
good night
<{^_^}>
Night!
<samueldr>
sphalerite: if you didn't want to (yet) be on that team, you can remove yourself, I github's UI said "invite to team" and not "add to team" :/
eadwu has quit [Ping timeout: 250 seconds]
eadwu has joined #nixos-dev
eadwu has quit [Quit: WeeChat 2.3]
eadwu has joined #nixos-dev
orivej has quit [Ping timeout: 250 seconds]
worldofpeace has joined #nixos-dev
eadwu has quit [Ping timeout: 272 seconds]
eadwu has joined #nixos-dev
lassulus_ has joined #nixos-dev
lassulus has quit [Ping timeout: 268 seconds]
lassulus_ is now known as lassulus
init_6 has quit [Ping timeout: 268 seconds]
eadwu has quit [Ping timeout: 250 seconds]
init_6 has joined #nixos-dev
worldofpeace has quit [Ping timeout: 240 seconds]
emily has joined #nixos-dev
MichaelRaskin has quit [Quit: MichaelRaskin]
<sphalerite>
samueldr: team? I don't seem to have any relevant notifications or emails
<sphalerite>
samueldr: ah, I saw from a PR that I'm in the backports team now, I'm guessing that's what you meant
<sphalerite>
all good :)
orivej has joined #nixos-dev
pepesza has quit [Ping timeout: 250 seconds]
pepesza has joined #nixos-dev
pepesza has quit [Ping timeout: 244 seconds]
pepesza has joined #nixos-dev
jtojnar has joined #nixos-dev
FRidh has joined #nixos-dev
orivej has quit [Ping timeout: 245 seconds]
eadwu has joined #nixos-dev
obadz has quit [Ping timeout: 250 seconds]
obadz has joined #nixos-dev
aristid has quit [Quit: WeeChat 2.2]
orivej has joined #nixos-dev
niksnut has quit [Quit: Lost terminal]
niksnut has joined #nixos-dev
eadwu has quit [Ping timeout: 250 seconds]
eadwu has joined #nixos-dev
eadwu has quit [Ping timeout: 250 seconds]
eadwu has joined #nixos-dev
eadwu has quit [Read error: Connection reset by peer]
eadwu has joined #nixos-dev
<timokau[m]>
Does somebody here have direct access to the hydra builder "packet-epyc-1" and can try to re-run the suitesparse build there? The compiler has now segfaulted twice on that machine and I'd like to know if it is reproducible.
<timokau[m]>
infinisil: Thanks for the pointer. Probably not related since it already segfaults during build and that only touches install.
<timokau[m]>
There is a new suitesparse version though. My ideal outcome would be that the failure is deterministically reproducable on that machine *and* the new version already happens to fix it
tilpner has joined #nixos-dev
tilpner has quit [Quit: WeeChat 2.3]
tilpner has joined #nixos-dev
<timokau[m]>
gchristensen: ping
<gchristensen>
is it 30min already? :')
<gchristensen>
send me the command you want me to run?
<gchristensen>
had to restore a few more secrets from backup before I could SSH in.
<timokau[m]>
gchristensen: Glad I'm giving you an opportunity to test your backups :)
<thoughtpolice>
If I want to author a python module for upstream, what's the correct way to specify "python 3 only"? e.g. exclude it from python2Packages
<timokau[m]>
thoughtpolice: `disabled = !isPy3` or `diabled = isPy2`, depending on what you think about pypi
<thoughtpolice>
That did it, thanks timokau[m] !
<thoughtpolice>
('isPy3k' actually rather than 'isPy3')
<gchristensen>
timokau[m]: worked
<timokau[m]>
Mh, I would have preferred it to fail :/
<timokau[m]>
How much ram does the machine have? Is it likely the build ran out?
<timokau[m]>
But I'm running two suitesparse build at the same time and it only takes 4G, so that really is unlikely
<timokau[m]>
Doesn't look like there was anything special going on there
<timokau[m]>
But then again I've re-build suitesparse 14 times now without reproducing the error, so its probably at least somewhat machine dependent
<gchristensen>
timokau[m]: I wish I had that much ram, too. this and several other servers are extremely graciously provided by Packet.... I haven't calculated how much the monthly bill would be, but... it'd be a lot.
<timokau[m]>
Looks like we couldn't wish for a better sponsor :)
<timokau[m]>
Regarding the suitesparse issue I don't know how to proceed. Probably will do the 5.4.0 update (#53128) and pray that it resolves the issue somehow