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.
Geth roast: 63f5308449 | (Zoffix Znet)++ | S05-match/make.t
[v6.d REVIEW] Don't .say into TAP

I checked and this still triggers the original bug
Orig: github.com/perl6/roast/commit/ceb0380ba
00:07
roast: 9115df7050 | (Zoffix Znet)++ | S03-binding/attributes.t
[v6.d REVIEW] Localize class definitions
00:09
roast: 00d17fae6a | (Zoffix Znet)++ | S32-hash/delete.t
[v6.d REVIEW] Precisify slice :delete return value

Orig: github.com/perl6/roast/commit/f5454af87
00:11
Geth roast: 51f5acbcd5 | (Zoffix Znet)++ | S32-hash/map.t
[v6.d REVIEW] Fix typo in desc
00:17
roast: 3f93e661e0 | (Zoffix Znet)++ | S03-junctions/misc.t
[v6.d REVIEW] Don't rely on Junction.perl order

Orig: github.com/perl6/roast/commit/8cfdd2f81
00:20
6.d-prep: 37a0ff81c5 | (Zoffix Znet)++ (committed using GitHub Web editor) | ChangeLog.md
[CL] =defn
00:39
roast: 77949f497f | (Zoffix Znet)++ | S02-literals/numeric.t
[v6.d REVIEW] Test we actually get results in .:<...> tests

Orig: github.com/perl6/roast/commit/494705473
00:47
roast: 963b08fe0b | (Zoffix Znet)++ | S29-any/isa.t
Remove trailing whitespace
01:25
roast: 6414bc201b | (Zoffix Znet)++ | S29-any/isa.t
[v6.d REVIEW] Remove .isa(Str:D) proptest fudge

That feature is slated for deprecation:
  github.com/rakudo/rakudo/issues/23...-425721762
01:26
roast: dd175d352e | (Zoffix Znet)++ | S06-other/main-refactored.t
[v6.d REVIEW] Loosen hidden-from-USAGE test

So it doesn't spec exact generated usage message
Orig: github.com/perl6/roast/commit/8da39b89e
01:40
Zoffix rrrrrrrrrrrraaaaaaaaaaa𝐚𝐚𝐚𝐚𝐚𝐚𝐚𝐖𝐖𝐖𝐖𝐖𝐖𝐖𝐖𝐖𝐖𝐖𝐖𝐖𝐖𝐑𝐑𝐑𝐑𝐑𝐑𝐑𝐑𝐑𝐑𝐑𝐑𝐑𝐑𝐑𝐑𝐑𝐑𝐑𝐑𝐑𝐑! 01:41
I finished my 6.d spec review \o/ 01:42
🎸🎉🍞🎸🎉🍞🎸🎉🍞🎸🎉🍞🎸🎉🍞🎸🎉🍞🎸🎉🍞🎸🎉🍞🎸🎉🍞🎸🎉🍞🎸🎉🍞🎸🎉🍞🎸🎉🍞🎸🎉🍞🎸🎉🍞
Geth 6.d-prep: e46b3821f5 | (Zoffix Znet)++ (committed using GitHub Web editor) | TODO/README.md
Mark Zoffix reviewer as completed
01:43
Zoffix goes to spend the remaining 20 minutes of the weekend playing ESO :)
/close
[Coke] zoffix++ review, awesome! 01:47
japhb CONGRATS, ZOFFIX! 01:58
AlexDaniel Zoffix++ you are awesome 02:05
yoleaux 14 Oct 2018 19:22Z <Zoffix_> AlexDaniel: can you describe the process for tagging in a branch and then merging it to master? Basically, I'm approaching a point where I want to split 6.d spec into a branch and then do finishing touches there, then tag it with 6.d tag and then merge it to master. Any idea what the git commands are? I don't wanna mess it up.
AlexDaniel Zoffix: so it's actually relatively simple except for one detail. You tag normally (e.g. the way the release guide says, or whatever), tags are ok to be on non-master branches so at that point there's nothing to mess up. When merging there's one small thingie – “direction” of the merge. See this SO answer for details: stackoverflow.com/a/45973607/2698019 02:20
Zoffix: Now, if you checkout `master` and merge another branch into it, then it's simple – `git checkout master; git merge another-branch` and that's it. The diff will look as if your branch changes landed on top of whatever was there in master. I don't do that for rakudo releases.
Zoffix: Instead I want it to look as if changes on `master` landed on top of the release (which makes more sense IMO). So `git checkout another-branch; git merge master`. However, now you end up with no merge commit (and other commits) on master, because we just committed a merge into another branch. So we should bring that stuff back into master: `git checkout master; git merge another-branch`. It sees that there are no additional
merge commits needed so it fast-forwards (and there are flags to ensure that, but we kinda know what's going to happen).
Zoffix: whether you should worry about it or not depends on how you want tools to behave, but it's really just a tiny detail and IMO either way is fine and there's not much to mess up.
Zoffix: do we have a release guide for roast? I don't see it. Maybe we should? Checklists generally help 02:23
or maybe skip the documentation and write automation directly
given how often language releases happen maybe it doesn't need to be automated. I'm just throwing out some thoughts :) 02:24
.tell Zoffix is your bot still working? In other words, do you receive messages without .tell? Anyway: colabti.org/irclogger/irclogger_log...-10-15#l62 . Let me know if there's anything else I can help with, I feel like I didn't answer your question but I really don't do anything special 02:26
yoleaux AlexDaniel: I'll pass your message to Zoffix.
Zoffix AlexDaniel++ 02:28
Yes, my bot still works.
Thanks for the description
AlexDaniel Zoffix: oh yeah, after that I remove the branch completely, and never even push it. Do you plan to keep the branch on roast available?
Zoffix I wasn't gonna, but I see there is a 6.c branch, so I guess yeah, I'm gonna keep it 02:29
AlexDaniel what's the point of that branch, actually 02:32
it has 12 commits in it
Zoffix: it's possible that we didn't have v6.c-errata branch at that point so people pushed something there? 02:33
so basically right now we have both “6.c” and “6.c-errata” branching out from a commit that was tagged “6.c” 02:34
confusing.
[Coke] yup, that's probably my fault.
Zoffix Yeah, I didn't expect a 6.c branch. I would've thought master is tagged with 6.X and then when first errata occurs, we create 6.X-errata branch
[Coke] ISTR I suggested we remove the tag at some point (and not use tags for that, only branches), but I don't think it got traction 02:35
AlexDaniel also what the hell would git even do when you say `git checkout 6.c`. Does it prioritize tags or branches?
[Coke] Having the tag be the first and the branch be the updates is fine; I'm happy if you want to standardize on that.
AlexDaniel [Coke]: I'm not strongly suggesting, but more like being curious: why not release more versions of say 6.c from the errata? 02:38
like 6.c.1 or something
I understand that roast is unlikely to be packaged, but if that happens, packagers like stuff to be tagged 02:39
and generally it makes sense, no? so .c .d .e is a major version, but we can have some minor releases or something like that 02:40
Zoffix Yeah, most of the 6.d spec is non-breaking changes, so going forward, we could make minor releases of stuff like that and keep breaking changes until major verson 02:41
Zoffix & 02:42
[Coke] the more tagged versions, the more the compiler has to be configured for it. I think that's fine going forward. 02:43
AlexDaniel [Coke]: what do you mean? What should be tweaked in the compiler? 02:44
[Coke] use 6.d ; use 6.d.1 ; use ... 02:50
also: what does 6.d mean? the tagged release? Probably not, probably the latest 6.d; but then how to get the original 6.d? 02:51
AlexDaniel yes but currently `use v6.c` means, uh… latest 6.c-errata commit? It's a bit LTA 02:57
[Coke] Yup. this would force us to be more strict. I'm not saying that's bad :) 02:58
but I think when starting out, we needed that slack more. 02:59
lizmat Files=1257, Tests=76293, 341 wallclock secs (15.65 usr 5.33 sys + 2384.65 cusr 247.54 csys = 2653.17 CPU) 06:44
|Tux| Rakudo version 2018.09-433-g7d99bf1b6 - MoarVM version 2018.09-125-g8c2dd3ad2
csv-ip5xs0.907 - 0.938
csv-ip5xs-207.380 - 7.442
csv-parser20.469 - 20.712
csv-test-xs-200.428 - 0.428
test7.846 - 8.256
test-t1.659 - 1.719
test-t --race0.794 - 0.795
test-t-2029.297 - 29.733
test-t-20 --race10.473 - 10.910
09:12
20181011 1.680❙20181001 1.674❙20180928 1.662❙20181015 1.659❙20181011 1.647❙ 09:17
Geth nqp: a768ea84b0 | (Paweł Murias)++ | tools/build/Makefile-Moar.in
[moar] Add missing dependencies to Makefile-Moar.in
09:43
pmurias if I touch src/HLL/Compiler.nqp and make m-install, building rakudo with --with-nqp doesn't work 09:44
travis-ci NQP build failed. Paweł Murias '[moar] Add missing dependencies to Makefile-Moar.in' 10:03
travis-ci.org/perl6/nqp/builds/441573642 github.com/perl6/nqp/compare/3816e...68ea84b0c6
Geth nqp: a1bca652ca | (Paweł Murias)++ | tools/build/Makefile-Moar.in
[moar] Add some more missing Makefile deps
10:03
pmurias how can I set $?FILES outside of the setting? 10:05
Geth nqp: fbd00dbd4a | (Zoffix Znet)++ | tools/build/MOAR_REVISION
[MoarVM Bump] 8c2dd3a [JIT] Be more silent abo […]

MoarVM bump brought: github.com/MoarVM/MoarVM/compare/2...5-g8c2dd3a
11:21
¦ nqp: version bump brought these changes: github.com/MoarVM/MoarVM/compare/2...5-g8c2dd3a
rakudo: fde779de4c | (Zoffix Znet)++ | tools/build/NQP_REVISION
[NQP Bump] Brings 4 commits

NQP bump brought: github.com/perl6/nqp/compare/2018....9-gfbd00db fbd00db [MoarVM Bump] 8c2dd3a [JIT] Be more silent abo […] a1bca65 [moar] Add some more missing Makefile deps a768ea8 [moar] Add missing dependencies to Makefile-Moar.in 3816e22 Cleanup indentation.
MoarVM bump brought: github.com/MoarVM/MoarVM/compare/2...5-g8c2dd3a
¦ rakudo: version bump brought these changes: github.com/perl6/nqp/compare/2018....9-gfbd00db
roast: 1ca966392a | (Zoffix Znet)++ | MISC/bug-coverage.t
Cover JIT label bugs

Closes github.com/rakudo/rakudo/issues/2347 R#2347 Closes github.com/rakudo/rakudo/issues/2346 R#2346 Fix: github.com/MoarVM/MoarVM/commit/8c2dd3ad22
   github.com/perl6/nqp/commit/fbd00dbd4a
   github.com/rakudo/rakudo/commit/fde779de4c
11:25
synopsebot R#2347 [open]: github.com/rakudo/rakudo/issues/2347 [Moar][⚠ blocker ⚠] JIT ERROR: Negative offset for dynamic label 32
synopsebot R#2346 [open]: github.com/rakudo/rakudo/issues/2346 [Moar][⚠ blocker ⚠] JIT ERROR: Negative offset for dynamic label 33
travis-ci NQP build errored. Zoffix Znet '[MoarVM Bump] 8c2dd3a [JIT] Be more silent abo […] 11:38
travis-ci.org/perl6/nqp/builds/441607182 github.com/perl6/nqp/compare/a1bca...d00dbd4a07
dogbert2_ yay, my $work address is no longer banned 12:13
Zoffix Can't build :( R#2376 12:27
synopsebot R#2376 [open]: github.com/rakudo/rakudo/issues/2376 Crash during build: Directive f not applicable for type BOOTNum
Zoffix .tell pmurias Maybe you should know what should be in that condition? github.com/rakudo/rakudo/issues/2377 12:28
yoleaux Zoffix: I'll pass your message to pmurias.
Geth rakudo: 0fb95aef2a | (Zoffix Znet)++ (committed using GitHub Web editor) | src/Perl6/Metamodel/BOOTSTRAP.nqp
Add missing `js` condition in #?if

Guessing that's what it's meant to be based on commit that introed that conditional:
  github.com/rakudo/rakudo/commit/3c...014124f003
Closes github.com/rakudo/rakudo/issues/2377
12:31
¦ rakudo: zoffixznet self-assigned Roast failure with typechecks in Hash[Str] github.com/rakudo/rakudo/issues/2375 12:33
pmurias Zoffix: sorry, it's js, I noticed to lack of the condition and forgot to push the fix :( 12:40
yoleaux 12:28Z <Zoffix> pmurias: Maybe you should know what should be in that condition? github.com/rakudo/rakudo/issues/2377
Zoffix cool. I commit a fix for it 12:41
Geth roast: f66c64e9ea | (Zoffix Znet)++ | S11-modules/export.t
Fix/unfudge parametarized Hash in modules proptests

Closes github.com/rakudo/rakudo/issues/2375 R#2375
The inner hashes in the initializers were not parametarized. The only reason these proptests were passing in the past is because we didn't do typechecks in that place, which is fixed now[^1]
  [1] github.com/rakudo/rakudo/commit/9a...716f532008
12:49
synopsebot R#2375 [open]: github.com/rakudo/rakudo/issues/2375 [⚠ blocker ⚠] Roast failure with typechecks in Hash[Str]
nqp: a011f985f9 | (Paweł Murias)++ | src/vm/js/nqp-runtime/core.js
[js] Fix evalling JavaScript in browser
12:54
nqp: 54889c3e16 | (Paweł Murias)++ | src/vm/js/nqp-runtime/browser.js
[js] Implement STDOUT hijacking for use in 6pad
nqp: f836084a0b | (Paweł Murias)++ | tools/build/Makefile-Moar.in
[moar] Add more missing dependencies
Zoffix Did something change with time_n op? 12:58
nqp: nqp::sprintf("%7.3f", [nqp::time_n() - nqp::time_n()])
camelia ( no output )
Zoffix On one of my boxes that crashes now: github.com/rakudo/rakudo/issues/23...-429843206
And it's blocking my rakudo build 12:59
or maybe with the way Nums are handled in NQP 13:01
pmurias Zoffix: I want to remove --encoding but a Makefile error is preventing me from changing the needed file :/ 13:07
Zoffix :( 13:08
travis-ci NQP build passed. Paweł Murias '[moar] Add more missing dependencies' 13:11
travis-ci.org/perl6/nqp/builds/441644954 github.com/perl6/nqp/compare/fbd00...36084a0b9f
Zoffix nm on R#2376.. After I nuked my install, the problem was gone. 13:18
synopsebot R#2376 [closed]: github.com/rakudo/rakudo/issues/2376 [⚠ blocker ⚠] Crash during build: Directive f not applicable for type BOOTNum
Geth rakudo: bea08ec6e4 | (Paweł Murias)++ | 2 files
Remove --encoding support
14:02
rakudo: 31b19d0949 | (Paweł Murias)++ | 3 files
Stop using --encoding in the Makefiles, it's utf8 anyway
lucasb Hello o/ My thanks to all of you working on P6 :) 14:17
So 6.d is happening? Before it ships, can I request a little love to a few simple issues:
#2131 14:18
GH #2131
GH#2131
synopsebot GH#2131 [open]: github.com/rakudo/rakudo/issues/2131 [@LARRY][consistency] Behavior of match occurrence adverbs: :nth(*) and :nth(*-1)
lucasb and the rename of the json exception class: 14:19
github.com/rakudo/rakudo/blob/mast...SON.pm6#L1
GH#1842 14:20
synopsebot GH#1842 [closed]: github.com/rakudo/rakudo/pull/1842 Revert "JSONException -> Rakudo::Internals::JSONException"
Zoffix .tell lucasb R#2131 is blocked by R#1289, so it's deferred until 6.e or later. And R#1842 is a Rakudo issue, not a language issue (neither of the classes involved are part of the spec). 15:14
yoleaux Zoffix: I'll pass your message to lucasb.
synopsebot R#2131 [open]: github.com/rakudo/rakudo/issues/2131 [@LARRY][consistency] Behavior of match occurrence adverbs: :nth(*) and :nth(*-1)
R#1289 [open]: github.com/rakudo/rakudo/issues/1289 [6.e] Implement a Way to Know Caller's Language
R#1842 [closed]: github.com/rakudo/rakudo/pull/1842 Revert "JSONException -> Rakudo::Internals::JSONException"
Zoffix .tell lucasb to clarify, R#2131 is blocked by R#1289, because current behaviour is part of 6.c spec ( github.com/perl6/roast/blob/6.c-er...#L218-L220 ) and any modifications—if agreed on—would have to be behind a language version pragma 15:17
yoleaux Zoffix: I'll pass your message to lucasb.
synopsebot R#2131 [open]: github.com/rakudo/rakudo/issues/2131 [@LARRY][consistency] Behavior of match occurrence adverbs: :nth(*) and :nth(*-1)
R#1289 [open]: github.com/rakudo/rakudo/issues/1289 [6.e] Implement a Way to Know Caller's Language
lizmat is seeing some spectest fallout re eval ? 15:56
S29-context/evalfile.t , 3 and S29-context/eval.t 25 15:57
are these known ?
Zoffix I didn't have them this morning 15:59
They pass for me on This is Rakudo version 2018.09-435-g0fb95ae built on MoarVM version 2018.09-125-g8c2dd3a
lizmat: ohhh
lizmat: yeah, that's likely the --encoding proptests. pmurias tossed it in the commit above 16:00
Zoffix pulls and rebuilds and will take care of it
lizmat is on 2018.09-437-g31b19d0
Zoffix++
Geth roast: 365a46df78 | (Zoffix Znet)++ | 2 files
[v6.d REVIEW] Remove --encoding EVAL[FILE] respectation

This feature did not make it[^1] and was removed from Rakudo[^2][^3]
  [1] colabti.org/irclogger/irclogger_log...10-05#l272
  [2] github.com/rakudo/rakudo/commit/31...8e8a971ec5
  [3] github.com/rakudo/rakudo/commit/be...8ffeb7f3cd
16:07
lizmat hmmm... looks like Perl6-Mathx-Stat by holyghost is not getting indexed 16:52
same for Perl6-Bayes-Learn looks like holyghost is doing something wrong in the upload 16:54
Zoffix lizmat: yeah, the archive just contains 2 source files and two backup files for them. Not even a META6.json file 16:56
lizmat :-(
well, then it won't make it to this week's new module list :)
Geth rakudo: 9b23f613c2 | (Tom Browder)++ (committed using GitHub Web editor) | docs/rakudo-nqp-notes.md
add info
17:54
Geth rakudo: c52c410a1c | (Tom Browder)++ (committed using GitHub Web editor) | docs/rakudo-nqp-notes.md
Update rakudo-nqp-notes.md
18:07
AlexDaniel oh it's Monday again 18:25
AlexDaniel lost track of time 18:26
weekly: reportable: gist.github.com/293fe9858cea405e29...0586741bac 18:29
notable6 AlexDaniel, Noted!
AlexDaniel lizmat: here's the same thing for exactly one year: gist.github.com/Whateverable/eb016...79585ee753 18:33
Geth rakudo: 8bfacf5f30 | (Tom Browder)++ (committed using GitHub Web editor) | docs/rakudo-nqp-notes.md
Update rakudo-nqp-notes.md
18:34
travis-ci Rakudo build failed. Tom Browder 'add info' 18:40
travis-ci.org/rakudo/rakudo/builds/441788297 github.com/rakudo/rakudo/compare/3...23f613c2ea
buggable [travis build above] ☠ All failures are due to: failed make test (1 failure). Across all jobs, only t/05-messages/01-errors.t test file failed. 18:40
Geth rakudo: 62f9b69ca3 | (Tom Browder)++ (committed using GitHub Web editor) | docs/rakudo-nqp-notes.md
Update rakudo-nqp-notes.md
18:45
brrt \o 18:55
timotimo o/ 18:56
brrt leont: I didn't use pthread_atfork, essentially because it means that rather than the OS interrupting the thread in the middle of processing, you have to fixup something, and that is Hard 18:57
releasable6 Next release in ≈4 days and ≈23 hours. 2 blockers. Please log your changes in the ChangeLog: github.com/rakudo/rakudo/wiki/ChangeLog-Draft 19:00
travis-ci Rakudo build failed. Tom Browder 'Update rakudo-nqp-notes.md' 19:21
travis-ci.org/rakudo/rakudo/builds/441794260 github.com/rakudo/rakudo/compare/9...2c410a1c3e
buggable [travis build above] ☠ Did not recognize some failures. Check results manually. 19:21
travis-ci Rakudo build passed. Tom Browder 'Update rakudo-nqp-notes.md' 19:56
travis-ci.org/rakudo/rakudo/builds/441812148 github.com/rakudo/rakudo/compare/8...f9b69ca3e7
Geth rakudo: 75ed0857c5 | (Tom Browder)++ (committed using GitHub Web editor) | docs/rakudo-nqp-notes.md
Update rakudo-nqp-notes.md
20:01
rakudo: 5f0bcfc96b | (Tom Browder)++ (committed using GitHub Web editor) | docs/rakudo-nqp-notes.md
Update rakudo-nqp-notes.md
20:18
rakudo: 4039b447eb | (Tom Browder)++ (committed using GitHub Web editor) | docs/rakudo-nqp-notes.md
Update rakudo-nqp-notes.md
20:19
rakudo: e1ebffeb89 | (Tom Browder)++ (committed using GitHub Web editor) | docs/rakudo-nqp-notes.md
Update rakudo-nqp-notes.md
20:23
lizmat And another Perl 6 Weekly hits the Net: p6weekly.wordpress.com/2018/10/15/...tom-for-6/ 21:17
travis-ci Rakudo build passed. Tom Browder 'Update rakudo-nqp-notes.md' 21:28
travis-ci.org/rakudo/rakudo/builds/441859951 github.com/rakudo/rakudo/compare/4...ebffeb89ff