01:30 sxmx left 01:32 kvw_5_ joined 01:35 kvw_5 left 03:20 lucasb left 04:40 squashable6 left 04:42 squashable6 joined 04:44 squashable6 left, squashable6 joined 05:53 Kaiepi left 05:54 Kaiepi joined 05:58 maggotbrain left 06:00 maggotbrain joined 06:12 Kaiepi left, Kaiepi joined 06:21 domidumont joined 06:56 domidumont left 07:10 domidumont joined
MasterDuke anyone have an opinion on my nqp::time series of PRs? merge and get a blin run? wait for more people to review them? 07:33
07:44 eaterof joined 07:46 domidumont left 08:06 sena_kun joined 08:33 frost-lab joined 08:37 MasterDuke left 09:15 Kaiepi left 09:16 Kaiepi joined 09:49 sivoais joined
lizmat MasterDuke: I'd be in favour of merging now and a blin run 10:12
tellable6 lizmat, I'll pass your message to MasterDuke
10:15 domidumont joined 10:41 lizmat_ joined 10:44 lizmat left 10:45 lizmat_ is now known as lizmat
lizmat just pushed github.com/rakudo/rakudo/commit/fb...6d28f3ff44 Make Int.unival about 30% faster 11:05
dogbert17 lizmat++ lots of cool optimizations 11:32
11:43 domidumont left 11:49 MasterDuke joined
lizmat m: dd (-1).unival # yuck 12:10
camelia Internal Error: MVM_codepoint_to_row_index call requested a synthetic codepoint that does not exist.
Requested synthetic 1 when only 6 have been created.
in block <unit> at <tmp> line 1
12:13 patrickb joined 12:15 patrickb left
MasterDuke i'll start the sequence of merging moarvm, updating the nqp pr with a bump, merge nqp, update the rakudo pr with a bump, merge rakudo 12:31
[Tux] Rakudo v2021.03-25-gbcfa54898 (v6.d) on MoarVM 2021.03-14-g16fe88fdf
csv-ip5xs0.806 - 0.828
csv-ip5xs-207.789 - 7.791
csv-parser27.706 - 28.995
csv-test-xs-200.377 - 0.390
test7.517 - 8.039
test-t1.948 - 1.965
test-t --race0.885 - 0.983
test-t-2033.625 - 34.653
test-t-20 --race9.867 - 10.081
MasterDuke ok, all three PRs merged 13:06
sena_kun: do you mind doing a blin run to see if those PRs caused any new problems? 13:37
sena_kun MasterDuke, I think the available option now is to cast vrurg 13:38
MasterDuke ok, thanks 13:39
13:41 klapperl joined
vrurg MasterDuke: I'll do it. 13:41
MasterDuke thanks 13:43
vrurg sena_kun: still to be done on ugexe-patch-1 branch?
sena_kun vrurg, yes. I think it's time to just merge it, by the way
vrurg sena_kun: it's up to you, I just run it. :) 13:44
sena_kun vrurg, yup. Sorry to burden you with this, I lost my ryzen machine some time ago. :S
vrurg sena_kun: no prob. That's why I bought this workstation a while ago - to help with development. 13:46
sena_kun: BTW, do you have any tricks for tests which pop-up windows? Like a gaming one. 13:47
sena_kun vrurg, I know about this one, but no tricks. I mean, the best bet is to patch the module to not do this, but the author is not available IIRC.
vrurg Ok. Perhaps we shall consider excluding modules from blin testing if an author doesn't respond to a PR for 3-6 months. 13:49
sena_kun vrurg, I think the motivation is "they are still used as SOME code to test rakudo with, even if the code is bogus, any differences in behavior between releases spotted can be regressions". 13:51
vrurg sena_kun: Most of the time – yes. Perhaps, if the rule is considered, it must not be a mandatory regulation but just give the right to do so if necessary. 13:53
sena_kun vrurg, yes, some modules are excluded from testing, so we can add github.com/FROGGS/p6-Games-BubbleB...r/issues/9 to the exclused list. 13:55
vrurg sena_kun: Perhaps, it worth considering. It just crossed my mind that I can actually start doing nightly blin runs. But if the module just silently bypasses if there is no DISPLAY then it's no problem. 13:59
sena_kun vrurg, the idea is cool, but IMO the development tempo does not require nightly. 14:00
vrurg sena_kun: I agree, but it might be helpful on occasion. I would only need your help to setup sharing of the results. I guess blin can auto-create a gist, can't it? 14:02
sena_kun vrurg, it can't, but it can be automated I believe. Once I get a better machine I'll take a look into it. 14:04
vrurg Wow, Pastebin::Gist! That's all I need. 14:10
MasterDuke: BTW, it already runs. Can't promise I'll be able to post it ASAP though. 14:11
MasterDuke well, hopefully there will be nothing to post... 14:12
14:29 frost-lab left 14:46 linkable6 joined
jdv79 MasterDuke: thanks for doing the time thing. i wanted to do something like that a few months ago but other things... 15:03
MasterDuke sure, turned out to not be too bad 15:04
i guess after the release it's time to submit PRs to the ecosystem modules to change `nqp::time_n` to `now` 15:08
b2gills Speaking of which, did that change break tests? t/02-rakudo/03-corekeys-6[cde].t t/02-rakudo/04-settingkeys-6[ce].t 15:11
I just ran my build bash script, and those tests prevented it from installing (a simple cd to the build directory and make install worked) 15:12
MasterDuke oh. whoops 15:16
b2gills FYI. I wrote my own build script so that it would purposefully be more fragile, for the sheer fact that it is different to everybody else. 15:17
That way I catch things that others might not catch 15:18
MasterDuke looks like i ran a spectest after my last change, but not a test 15:19
b2gills I just thought of something cool I could add to my build script. 15:23
It keeps track of the previous commit, so I could potentially make it generate a GitHub link to the diff. Which would help with such debugging.
MasterDuke i just directly pushed a fix to the tests 15:27
16:13 jmerelo joined
jmerelo Anyone know what happens with Geth? Can it be fixed? 16:25
tellable6 2021-03-24T21:17:12Z #raku-dev <[Coke]> jmerelo - hadn't heard anything about TRF, so was just curious, I trust you on the timing. I cannot guarantee I have any time to support the project, sorry
jmerelo .tell Will thanks :-)
tellable6 jmerelo, I haven't seen Will around, did you mean tyil?
jmerelo .tell [Coke] thanks :-)
tellable6 jmerelo, I'll pass your message to [Coke]
16:34 patrickb joined
lizmat m: dd unival(42.2) 16:40
camelia 5===SORRY!5=== Error while compiling <tmp>
Calling unival(Rat) will never work with any of these multi signatures:
(Str:D $str)
(Int:D $code)
at <tmp>:1
------> 3dd 7⏏5unival(42.2)
lizmat m: dd (42.2).unival
camelia NaN
lizmat that feels inconsistent
16:40 Geth joined
Geth nqp: 376e5627a2 | (Elizabeth Mattijsen)++ | docs/ops.markdown
Add time_i|n back as DEPRECATED forms
lizmat whee! Geth's back :-)
Geth nqp: patrickbkr++ created pull request #707:
Remove Travis badge
roast: c22b7e9b68 | (Elizabeth Mattijsen)++ | S06-signature/errors.t
unival(42.2) shouldn't fail if (42.2).unival doesn't

Built-in sub versions should just pass on to their method counterparts.
  (42.2).unival returns NaN, unival(42.2) should as well, therefore this
test is incorrect.
rakudo: f0a345dfb3 | (Elizabeth Mattijsen)++ | 3 files
Several unival fixups

  - make sub version just pass on whatever it gets
  - mark sub / method as "is pure" for better optimizability
  - make method "unival" a multi
rakudo: patrickbkr++ created pull request #4276:
Disable Travis, AppVeyor and CircleCI
roast: JJ++ created pull request #725:
Changes to test new predicate `wait`
patrickb @Whoever has admin permissions to our GitHub Rakudo/Rakudo repo: Can you please have a look at github.com/rakudo/rakudo/pull/4276 ? Thank you! 17:21
lizmat patrickb: webhooks removed, should I merge now ? 17:23
patrickb lizmat: Please do!
Thank you! 17:24
Geth rakudo: d272f4fda0 | (Patrick Böker)++ | 4 files
Disable Travis, AppVeyor and CircleCI

They have been superseded by AzureCI.
rakudo: 76022f6f7d | (Elizabeth Mattijsen)++ (committed using GitHub Web editor) | 4 files
Merge pull request #4276 from patrickbkr/disable-obsolete-CIs

Disable Travis, AppVeyor and CircleCI
rakudo: usev6++ created pull request #4277:
Unbreak 'use NativeCall' on JVM
lizmat yw
patrickb btw. any reason sena-kun shouldn't get admin access to the rakudo project? To me it feels he is one of the suspects that should have access. 17:25
sena_kun patrickb, the lower number of people the better? :) 17:26
patrickb On the one hand less people having access reduces chances of misuse and accidental breakage, but on the other hand also helps with the bus factor. 17:28
I also think one should have a look from time to time to make sure no people long gone from the project are in the admin list. 17:30
lizmat looks like masak and pmichaud are still admins 17:40
and moritz, jnthn and me
17:44 linkable6 left
patrickb How high is the potential of hurt feelings when removing people? 17:44
17:45 linkable6 joined
patrickb I really think it makes sense to clean up from time to time. But I'd hate giving people the impression of kicking them out. Contacting them first is probably wise... 17:46
lizmat moritz_ : wanna chime in ? 17:48
I'll ping jnthn, masak and pmichaud 17:49
17:50 jmerelo left
patrickb lizmat++ 17:50
moritz_ I'm fine with people removing my admin access to the rakudo project. 17:54
lizmat well, that wasn't actually the question :-) 17:55
because you're still around most of the time 17:56
masak and pmichaud are not
moritz_ yes, but I'm not actively admining, so there's no reason for me to keep the privs 17:59
my mental model of masak++ says that he won't object
18:20 Kaiepi left 18:21 Kaiepi joined 19:46 MasterDuke left 19:48 MasterDuke joined
MasterDuke lizmat: do you remember when the error referenced in github.com/rakudo/rakudo/commit/cd...dR175-R177 would happen? because i just changed it to `nqp::mod_I($hours, 24, Int)` and it built and passed a spectest 19:53
hm. did a quick grep for XXX in moarvm, nqp, and rakudo. 65, 80, and 103 respectively 20:24
some look like they might be easily removed, others less so 20:25
heh. github.com/rakudo/rakudo/blob/mast...#L449-L452 "# XXX Parrot compat hack.". from 7 years ago, i suspect that can be removed... 20:32
20:37 patrickb left 20:40 MasterDuke left 20:41 MasterDuke joined
MasterDuke ha, fascinating. if i remove that if, i get: 20:44
===SORRY!=== Error while compiling:
Circular module loading detected trying to precompile /home/dan/Source/perl6/install/share/perl6/site/sources/3E687CA5BFB386524F923BAC2F986CC3C66BD24F
when running rakudo-debug-m
huh, and that file doesn't even exist 20:55
20:55 MasterDuke left 20:56 MasterDuke joined
MasterDuke ugexe, nine: does gist.github.com/MasterDuke17/3854e...edff704514 look normal/expected to you? this is what i get after removing github.com/rakudo/rakudo/blob/mast...#L449-L452 21:28
Geth rakudo/master: 5 commits pushed by (Elizabeth Mattijsen)++ 22:01
lizmat more work on that tomorrow
Geth ¦ rakudo: tbrowder self-assigned The Dateish.daycount value should be a real number if intended to be the MJD github.com/rakudo/rakudo/issues/4279 22:02
22:38 cog_ joined, samcv_ joined 22:39 samebchase-6 joined 22:40 nebuchad` joined, tobs` joined 22:41 gfldex_ joined, samcv left 22:42 nebuchadnezzar left, tobs left, tobs` is now known as tobs 23:05 samcv_ is now known as samcv 23:06 samcv left, samcv joined
ugexe MasterDuke: i dont know why its doing that, but note how the non-debug file has a signature @ARGS and not @*ARGS github.com/rakudo/rakudo/blob/2d5e...in.nqp#L51 23:27