Perl 6 language and compiler development | Logs at colabti.org/irclogger/irclogger_log/perl6-dev | For toolchain/installation stuff see #perl6-toolchain | For MoarVM see #moarvm
Set by Zoffix on 27 July 2018.
01:06 Kaiepi left, Kaiepi joined 01:09 Kaiepi left 01:12 Kaiepi joined 02:39 go|dfish left 02:40 go|dfish joined 02:54 squashable6 left 02:57 squashable6 joined 04:39 tyil left 04:40 tyilanmenyn joined 04:41 tyilanmenyn is now known as tyil 06:10 squashable6 left 06:11 squashable6 joined, ChanServ sets mode: +v squashable6
masak timotimo: hm, why does that happen? 07:04
08:32 robertle joined
|Tux| Rakudo version 2019.07.1-237-g0455f42e3 - MoarVM version 2019.07.1-136-g204eb28f0
csv-ip5xs0.657 - 0.660
csv-ip5xs-204.909 - 5.037
csv-parser21.366 - 22.231
csv-test-xs-200.428 - 0.460
test6.907 - 6.999
test-t1.723 - 1.725
test-t --race0.776 - 0.795
test-t-2029.153 - 29.318
test-t-20 --race9.321 - 9.503
 10:29
11:46 kiwi_7 joined 12:11 Kaypie joined 12:12 Kaiepi left 12:21 lucasb joined 13:31 epony left 13:32 epony joined 13:36 vrurg left 13:41 vrurg joined 13:55 rba[m] left 13:58 samcv left 14:00 rba[m] joined 14:02 kiwi_7 left 14:05 samcv joined 14:09 vrurg left 14:22 vrurg joined 15:13 ufobat__ joined 15:16 ufobat_ left 15:33 robertle left 16:15 robertle joined
kawaii AlexDaniel: special delivery usercontent.irccloud-cdn.com/file/...-29.tar.gz 16:15
AlexDaniel holy moly 16:17
that doesn't look good at all
vrurg: we have an emergency lol xD 16:18
japhb AlexDaniel: Not looking at the tarball, but if you're seeing a lot of failed installs, it might be because YAMLish won't install right now. Author is aware.
github.com/Leont/yamlish/issues/29
16:19 ufobat__ left
AlexDaniel haha that svg output 16:19
japhb: github.com/rakudo/rakudo/issues/31...-526260444 16:22
ahhh these are not even related to yamlish
because they bisected to a different commit
ahhhh ignore that, it's probably because of multiple issues in the same range 16:25
kawaii Sorry to be the bearer of bad news ;) 16:29
AlexDaniel ok that's about it, not that bad now that I look at it 16:33
vrurg AlexDaniel: will check. I wish I could run blin on my side... 16:40
16:44 MasterDuke joined
AlexDaniel vrurg: so what's the reason you can't? 17:04
vrurg: set up a VM, install perl6 and blin, run it
vrurg: also you can download that .tar.gz mentioned above and see the output of each module 17:05
tobs: btw this one is really annoying: github.com/perl6/whateverable/issues/45 17:06
tobs: I can explain why it doesn't work right now. Or at least my limited understanding of how it works at all 17:07
vrurg AlexDaniel: oh, without #45 resolved blin would be useless anyway. I'll wait when branches are supported then. 17:13
tobs AlexDaniel: that sounds like a really cool feature indee 17:19
d
oh, it's already implemented but buggy? Anyway I have to go afk for a bit 17:20
AlexDaniel tobs: it doesn't work 17:30
vrurg: yeah, thanks, blin is useless, alright xD
:D
I mean, I get it, but there's indeed something about the idea that contributions are not appreciated enough 17:31
vrurg AlexDaniel: oops... Sorry, fixing the issue, skipping "extra" words. ;) useless for me, apparently.
AlexDaniel (in open-source projects in general)
tobs: it's possible that it is a one-line patch to make it work. So, here's what's going on. Bots have a rakudo repo (REPO-CURRENT in build.p6) which they use to look up commits and stuff 17:37
tobs: which means that running something on HEAD will get you the most recent commit on master of that repo 17:38
so far so good
when a new commit lands in rakudo, build.p6 script will `git pull` and start building stuff 17:39
however, we don't want the bots to know about these new commits, otherwise running something on HEAD will fail because there's no build yet
for that reason there's a separate git repo (REPO-LATEST in build.p6) that is used for building stuff, and when done, it just pulls in REPO-CURRENT from REPO-LATEST 17:40
here: github.com/perl6/whateverable/blob...ld.p6#L152
and that's exactly the line where it doesn't work 17:41
because git pull like that won't pull the branches
well not only that line, because this script itself also doesn't produce the builds for branches, so yeah. But without that the bots won't see stuff too 17:45
well the good thing is that that's the only file we'll need to tweak
17:46 Kaypie is now known as Kaiepi
AlexDaniel so somewhere within 240 lines lies the answer 17:46
tobs: also fwiw I'd prefer to turn it into a service just like any other bot, so that we can then resolve this also: github.com/perl6/whateverable/issues/207 17:57
tobs: I'm a bit worried about touching that script, because in the past rakudo used to break apart if I did something differently… 17:58
like, this stuff for example is against memory leaks: github.com/perl6/whateverable/blob...ld.p6#L157 17:59
but all that is no longer relevant, I guess?
and if not we'll see
18:53 chloekek joined
Geth roast: b5b8c5c9b1 | usev6++ | S32-container/cat.t
Remove another reference to placeholder ticket
 19:29
roast: c5d724c891 | usev6++ | S03-operators/adverbial-modifiers.t
Remove another reference to placeholder ticket
 19:47
bartolin closed (rejected) about 160 RT tickets that didn't had relevant information. 19:51
20:00 chloekek left 20:11 a3f left, a3f joined
AlexDaniel bartolin++ 20:21
20:51 gdonald joined, toddr joined 20:53 toddr left
AlexDaniel “The “1.0” for the Perl 6 spec (6.c, because reasons) was released in 2015” 21:03
yeah, fixable or not, I think it is stupid
sorry :)
22:51 dogbert17 left
AlexDaniel yay 23:34
jnthn AlexDaniel: One slight issue with the voting thing: I think any change to a PR invalidates existing approvals. 23:36
AlexDaniel jnthn: yeah, it's unfortunate, I'm not sure what would be the solution 23:37
jnthn: for this PR in particular maybe we shouldn't push any new changes after we start voting, but for future stuff some solution is needed…
github can automatically invalidate reviews after new stuff is pushed, should I enable that? 23:38
but then people will have to approve more than once, which is annoying
jnthn Hm, I hadn't realized it was an option; I thought pushing new stuff always invalidated the existing approvals
AlexDaniel it's currently disabled 23:39
jnthn I'd rather not leave the door open to "I approved, but not of this change"
AlexDaniel jnthn: which means?
jnthn Somebody reads it and approves. Then there's a change. If their approval sticks, rather than being invalidated by the change and needing re-approval, they could quite legitimately argue that they didn't actually approve the final resulting PR that was merged. 23:40
AlexDaniel so I should turn it on?
jnthn My feeling would be yes. I guess it's been turned on on other repos I've worked on and I never realized it was an option. :)
AlexDaniel it's a thing that's available for protected branches only, we don't use it anywhere else that much 23:41
jnthn Ah, ok 23:42
AlexDaniel alright, jnthn++ 23:43
jnthn Even with that, I'm not really sure how we handle things best, in so far as presumably if somebody requsets changes, we want to encourage a dialog, not just immediately consider that there was no consensus and so the PR is doomed. :) 23:44
AlexDaniel that's exactly what I had in mind
maybe it should be clarified in the README, or something like that
well, I'll definitely write a comment if somebody rejects :) 23:45
jnthn Maybe we want to say there's a 2 week window by default, but also the last N days of that are a change freeze to give folks chance to get their votes in, and if we really need to change it inside those N days, it has to be stretched. Maybe that has its own problems.
AlexDaniel ooo that'd be nice 23:46
jnthn It's not how things are, but I do wish we'd had chance to polish the problem-solving process a bit more before it had to deal with such a substantive issue. :) 23:47
AlexDaniel I'm sure that problem-solving repo will be tweaked even more now that'll see things in action
yeah, I didn't see the naming issue coming that soon…
jnthn otoh, I'm happy with the overall way it's worked so far, in that most folks have been civil, and GitHub has the means to deal with the exceptions to that
haha...that latest post on #81 makes me want to to go Kyoto to ride the Raku bus :D 23:50
Time for some rest...'night o/
AlexDaniel jnthn: should I ping the reviewers to encourage them to suggest changes to the PR? 23:51
lizmat is currently away, but once she's back she'll be able to adjust the PR, I think 23:52

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!