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 rakudo/ugexe-patch-1: f66b0097c4 | (Nick Logan)++ (committed using GitHub Web editor) | src/core/CompUnit/RepositoryRegistry.pm6
Update RepositoryRegistry.pm6

These are used as lookup keys, so path slashes need to be normalized. For instance -- on windows when trying to `use Linenoise;` it turns out `CompUnit::RepositoryRegistry.repository-for-name("site")` is `Nil` inside of `Distribution::Resource`.
This is a redo of github.com/rakudo/rakudo/commit/c3...cd6afc2ba5		 00:03
rakudo: ugexe++ created pull request #2212:
Normalize the path portion of repo specs		 00:04
rakudo/ugexe-patch-1: 28a14e23fe | (Nick Logan)++ | src/core/CompUnit/RepositoryRegistry.pm6
Normalize the path portion of repo specs

These are used as lookup keys, so path slashes need to be normalized. For instance -- on windows when trying to `use Linenoise;` it turns out `CompUnit::RepositoryRegistry.repository-for-name("site")` is `Nil` inside of `Distribution::Resource`.
This is a redo of github.com/rakudo/rakudo/commit/c3...cd6afc2ba5		 00:09
Geth rakudo/ugexe-patch-1: e65857f7a5 | (Nick Logan)++ (committed using GitHub Web editor) | src/core/CompUnit/RepositoryRegistry.pm6
Optimize path spec normalization

Steal the working bits from github.com/rakudo/rakudo/commit/c3...cd6afc2ba5
  @hoelzro++		 02:47
rakudo: aae0827f6b | (Nick Logan)++ (committed using GitHub Web editor) | src/core/CompUnit/RepositoryRegistry.pm6
Normalize the path portion of repo specs (#2212)

Steal the working bits from github.com/rakudo/rakudo/commit/c3...cd6afc2ba5
  @hoelzro++		 02:50
[Tux] Rakudo version 2018.06-432-gaae0827f6 - MoarVM version 2018.06-419-g829b8ee70
csv-ip5xs0.934 - 0.957
csv-ip5xs-207.469 - 7.677
csv-parser23.324 - 23.515
csv-test-xs-200.458 - 0.465
test9.377 - 9.418
test-t2.145 - 2.250
test-t --race0.921 - 0.924
test-t-2038.023 - 38.540
test-t-20 --race12.421 - 12.651
 06:59
Geth nqp: cb87b26787 | (Paweł Murias)++ | src/vm/js/Operations.nqp
[js] Implement nqp::istype_nd
 09:35
nqp: a39f5610f1 | (Paweł Murias)++ | t/nqp/067-container.t
Implement nqp::istype_nd
pmurias jnthn: I'm porting over the rescalar branch to the js backend 09:51
jnthn: It's great that it reduces the number of nqp::p6... ops as those are annoying to implement when creating a new backend 09:52
travis-ci NQP build failed. Paweł Murias 'Implement nqp::istype_nd' 09:52
travis-ci.org/perl6/nqp/builds/415838908 github.com/perl6/nqp/compare/1c60a...9f5610f1cc
Geth nqp: 1f5c9737a2 | (Jonathan Worthington)++ | src/vm/moar/QAST/QASTOperationsMAST.nqp
Fix thinko in getattr/bindattr code-gen opt
 09:59
jnthn pmurias: Yeah, the hope is to get rid of extops entirely 10:00
travis-ci NQP build failed. Jonathan Worthington 'Fix thinko in getattr/bindattr code-gen opt' 10:18
travis-ci.org/perl6/nqp/builds/415847983 github.com/perl6/nqp/compare/a39f5...5c9737a28c
lizmat and another Perl 6 Weekly hits the Net: p6weekly.wordpress.com/2018/08/14/...m-glasgow/ 10:30
brrt fwiw, I like Camelia. And I have a Camelia pluche, courtesy of lizmat++ and Woolfy++, so I'm invested 10:48
Geth nqp: 9fe83eb3e9 | (Jonathan Worthington)++ | src/vm/moar/QAST/QASTCompilerMAST.nqp
Revert "Disable the SC write barrier during load"

This reverts commit 1ca664ac66ebe9a9da180faa7d4a8c811f0f4a73. That approach won't work, since we rely on `our` symbols bound into a Stash at load time triggering repossession.		 12:44
Geth rakudo: 2b2f568a33 | (Jonathan Worthington)++ | src/Perl6/Metamodel/BOOTSTRAP.nqp
Mark Scalar as never being repossessed

This restores a somewhat dubious, but relied upon, pre-rescalar behavior. A previous attempt at fixing #2056 led to #2126. Together with the revert in NQP commit 9fe83eb3e9, the modules in both tickets now all work.		 12:52
synopsebot RAKUDO#2056 [open]: github.com/rakudo/rakudo/issues/2056 [regression][testneeded] “MVMHash representation requires MVMString keys” rescalar regression
RAKUDO#2126 [open]: github.com/rakudo/rakudo/issues/2126 [regression][⚠ blocker ⚠] Weird failure in Path::Router module
travis-ci NQP build failed. Jonathan Worthington 'Revert "Disable the SC write barrier during load" 13:01
travis-ci.org/perl6/nqp/builds/415902592 github.com/perl6/nqp/compare/1f5c9...e83eb3e9d8
releasable6 Next release in ≈4 days and ≈3 hours. 5 blockers. Please log your changes in the ChangeLog: github.com/rakudo/rakudo/wiki/ChangeLog-Draft 15:00
Geth nqp/hllbool: b3e9eaed9a | (Stefan Seifert)++ | src/vm/moar/QAST/QASTOperationsMAST.nqp
Map hllbool and hllboolfor ops
 15:59
rakudo/hllbool: 4 commits pushed by (Stefan Seifert)++
Geth nqp: 0e81fdf557 | (Zoffix Znet)++ | tools/build/MOAR_REVISION
[MoarVM Bump] Brings 6 commits

MoarVM bump brought: github.com/MoarVM/MoarVM/compare/2...5-g0daf99b 0daf99b Fix inlining of things with multiple `return`s 9861e80 Make neverrepossess on type apply to the type 8b32104 Use #if, not #ifdef 0fc1e17 Add an MVM_UNLIKELY to encoding name init cond f039e81 Avoid some code duplication 6217f82 Add repossession logging behind a #define		 16:28
¦ nqp: version bump brought these changes: github.com/MoarVM/MoarVM/compare/2...5-g0daf99b
rakudo: 66aa6892b3 | (Zoffix Znet)++ | tools/build/NQP_REVISION
[NQP Bump] Brings 5 commits

NQP bump brought: github.com/perl6/nqp/compare/2018....1-g0e81fdf 0e81fdf [MoarVM Bump] Brings 6 commits 9fe83eb Revert "Disable the SC write barrier during load" 1f5c973 Fix thinko in getattr/bindattr code-gen opt a39f561 Implement nqp::istype_nd ... (9 more lines)
¦ rakudo: version bump brought these changes: github.com/perl6/nqp/compare/2018....1-g0e81fdf 16:29
Geth perl6-Toaster: 9e970de622 | (Zoffix Znet)++ | 6 files
Update repo URL
 17:02
Geth rakudo: 7c75ab8e44 | (Elizabeth Mattijsen)++ | src/core/REPL.pm6
Add "quit" as a way to exit the REPL

Fixes R#2214. Seems like a simple thing to accommodate visitors from other places to the REPL.		 17:29
synopsebot R#2214 [open]: github.com/rakudo/rakudo/issues/2214 [REPL] Make `quit` work like `exit` in the REPL
lizmat . 17:36
yoleaux 13:40Z <brrt> lizmat: that part of why I wanted to inline the expression info node was to have better cache line support :-)
lizmat afk again&
Zoffix buggable: toast 17:47
buggable Zoffix, Between 2018.06-434-g66aa689 and 2018.06: 18 (1.60%) modules got burnt; 22 (1.96%) got unsucced; 244 (21.73%) out of 1123 modules appear unusable. See toast.perl6.party/ for details.
Zoffix .tell AlexDaniel fresh toast: toast.perl6.party/
yoleaux Zoffix: I'll pass your message to AlexDaniel.
AlexDaniel . 17:50
yoleaux 17:47Z <Zoffix> AlexDaniel: fresh toast: toast.perl6.party/
AlexDaniel o just in time
:)
Zoffix: thank you very much!
aw crap why is everything so red 17:51
jnthn It's a lot less red than the last one, no?
AlexDaniel possibly, yeah 17:52
jnthn I've glanced through it. The DBIish one doesn't obviusly look like a Rakudo issue (# DBDish::SQLite: Error: attempt to write a readonly database (8)) 17:53
AlexDaniel yeah there are always some floppers
jnthn DBI::Async is a dep on that
AlexDaniel HTTP::Parser seems to be broken though
and it brought down some other modules with it 17:54
jnthn Ditto for Config::TOML 17:55
AlexDaniel makes another table
jnthn btw, finally fixed that longstanding serialization blocker today 17:56
dinner & 17:58
AlexDaniel yeah, I've seen that
jnthn++
Geth ¦ rakudo: AlexDaniel self-assigned 2018-08-10 Toast results / 29 burns github.com/rakudo/rakudo/issues/2196 17:59
AlexDaniel R#2124 R#1976
synopsebot R#2124 [open]: github.com/rakudo/rakudo/issues/2124 [⚠ blocker ⚠] Failure in t/04-nativecall/02-simple-args.t with clang
R#1976 [open]: github.com/rakudo/rakudo/issues/1976 [Windows][⚠ blocker ⚠] 2018.06 build for JVM fail
AlexDaniel so R#1976 just needs someone with a windows machine to try it
and R#2124 is a bit more interesting… not sure for how long we had that issue, but having a failing test when you use a different compiler is kinda very LTA 18:00
AlexDaniel ok that doesn't look too bad: github.com/rakudo/rakudo/issues/21...-412966242 18:14
just need to figure out what's going on with Config::TOML and HTTP::Parser
Geth nqp/master: 5 commits pushed by (Paweł Murias)++ 19:22
jnthn AlexDaniel: Interesting that they're both parsing-related. :) If you manage to get a golf even better; I'll be able to take a look later on today. :) 20:04
Geth nqp: 558b39cf33 | (Paweł Murias)++ | src/vm/js/Operations.nqp
[js] Add no-op wantdecont for JS backend

We currently don't push information about things we'd like to have decontainerized down the tree on the JVM backend, so just make it compile into something that will give a usable result, if non-optimal.		 20:44
nqp: 8ed28ff8bd | (Paweł Murias)++ | src/vm/js/Operations.nqp
[js] Implement nqp::hllboxtype_i, nqp::hllboxtype_n, nqp::hllboxtype_s
nqp: df48ad285e | (Paweł Murias)++ | t/qast/01-qast.t
Test nqp::hllboxtype_i, nqp::hllboxtype_n, nqp::hllboxtype_s
rakudo: b7190f6497 | (Elizabeth Mattijsen)++ | src/core/REPL.pm6
Revert "Add "quit" as a way to exit the REPL"

This reverts commit 7c75ab8e44986b27a814a3c73e912d3b2588bfca.
R#2214 fixed without needing this patch		 20:45
synopsebot R#2214 [open]: github.com/rakudo/rakudo/issues/2214 [REPL] Make `quit` work like `exit` in the REPL
pmurias jnthn: Should p6reprname be replaced with nqp::reprname, nqp::box_s and nqp::hllboxtype_s? 20:47
dogbert11 c: 2018.06 my $blob = blob8.new([1, 2, 3]); $blob.subbuf(0, Any) 20:53
committable6 dogbert11, ¦2018.06: «»
dogbert11 m: my $blob = blob8.new([1, 2, 3]); $blob.subbuf(0, Any)
camelia Cannot resolve caller subbuf(Blob[uint8]: Int, Any); none of these signatures match:
(Blob:D: Range:D $fromto, *%_)
(Blob:D: Int:D $From, *%_)
(Blob:D: &From, *%_)
(Blob:D: Int:D $From, Int:D $Length, *%_)
(Blob:D: &From,…
dogbert11 HTTP::Parser failure I believe
jnthn pmurias: yes
I'd just register it as a desugar in Actions 20:54
pmurias I'll change it once I fix rakudo.js after the merge 20:55
dogbert11 bisect: good=2018.06 bad=HEAD my $blob = blob8.new([1, 2, 3]); $blob.subbuf(0, Any) 20:58
bisectable6 dogbert11, Bisecting by exit code (old=2018.06 new=b7190f6). Old exit code: 0
dogbert11, bisect log: gist.github.com/889e3d7c07810da219...fde641b198
dogbert11, (2018-08-05) github.com/rakudo/rakudo/commit/cd...a069ca3d8f
AlexDaniel dogbert11: cool, can you file a ticket? :) 20:59
dogbert11 on the Rakudo repo I guess :) 21:00
AlexDaniel yeah
dogbert11 will do
dogbert11 AlexDaniel: GH#2215 21:07
synopsebot GH#2215 [open]: github.com/rakudo/rakudo/issues/2215 Blob.subbuf regression since 2018.06
Geth rakudo: lizmat self-assigned Blob.subbuf regression since 2018.06 github.com/rakudo/rakudo/issues/2215
26b491693b | (Elizabeth Mattijsen)++ | src/core/Buf.pm6

  - added (Int,&),(&,&),(\from,*),(\from,Numeric) candidates		 21:08
AlexDaniel lizmat++ 21:09
lizmat dogbert11: what is subbuf(0,Any) supposed to do? 21:13
lizmat c: 2018.06 my $blob = blob8.new([1, 2, 3]); dd $blob.subbuf(0, Any) 21:14
committable6 lizmat, ¦2018.06: «Blob[uint8].new(1,2,3)␤»
lizmat act as a Whatever, apparently ?
c: 2018.06 my $blob = blob8.new([1, 2, 3]); dd $blob.subbuf(0, *) 21:16
committable6 lizmat, ¦2018.06: «Cannot resolve caller Real(Whatever: ); none of these signatures match:␤ (Mu:U \v: *%_)␤ in block <unit> at /tmp/GUAqYfYP5M line 1␤␤ «exit code = 1»»
lizmat m: dd Nil !~~ Nil 21:17
camelia Bool::False
lizmat m: dd Nil ~~ Nil
camelia Bool::True
dogbert11 Perhaps the tests in the module are incorrect 21:19
i.e. using undefined behavior
lizmat dogbert11: please check my latest comment 21:21
dogbert11 ah, so there might be something else at play perhaps 21:22
lizmat m: "foobar".substr(2,Any) 21:24
camelia Use of uninitialized value of type Any in numeric context
in block <unit> at <tmp> line 1
No such method 'Int' for invocant of type 'Any'
in block <unit> at <tmp> line 1
lizmat the char based equivalent of subbuf doesn't allow Any as a second parameter either
dogbert11 m: my $blob = blob8.new([0x47, 0x45, 0x54]); my $a = $blob.first(* > 127, :k); say $a 21:31
camelia (Any)
jnthn I think I have some good news about Config::TOML. Verifying first. 21:36
jnthn Yes. 21:48
github.com/rakudo/rakudo/issues/21...-413027124 :) 21:49
AlexDaniel jnthn: awesome! 22:20
.seen tokuhirom 22:23
yoleaux I saw tokuhirom 30 Oct 2015 02:34Z in #perl6: <tokuhirom> after: split('=', 'YYY=', 2) returns just 'YYY'
jnthn AlexDaniel: Yeah, glad this round of toaster is (so far) going rather smoother than last month's :)
AlexDaniel HTTP::Parser issue is a bit annoying
AlexDaniel three other modules depending on it and the PR is unlikely to be merged soon… eh 22:23
it's an issue with a test so that's ok 22:24
ah no
it's not
`lib/HTTP/Parser.pm6` :S
actually, tokuhirom is active on github, so maybe it'll be merged soon 22:26
Geth roast: 3b30b38330 | (Elizabeth Mattijsen)++ | S03-operators/buf.t
Add more thorough subbuf testing
 22:27
jnthn It's pretty non-annoying in so far as we understand what it is, and it looks like the worst case is that we decide to add an extra multi candidate to keep the module working.
lizmat jnthn: do you think subbuf(x,Any) should be valid ? 22:28
or maybe make it valid now with a DEPRECATION ? 22:29
AlexDaniel that'd be ideal, yeah
lizmat ok, will do 22:32
jnthn No, it wouldn't be valid; DEPRECATED and remove later seems like a reasonable way 22:34
*shouldn't
lizmat oki, am on it 22:40
jnthn lizmat++ 22:43
AlexDaniel lizmat++ indeed 22:49
lizmat: sorry for reopening issues annoyingly :) sometimes it needs to be done I guess… 22:50
lizmat AlexDaniel: I'll let you know if it gets annoying 22:52
Geth rakudo: 22f779a883 | (Elizabeth Mattijsen)++ | src/core/Buf.pm6
Add support for Buf.subbuf(x,Any)

And deprecate it at the same time		 23:11
lizmat good night, #perl6-dev! 23:17
timotimo good night lizmat and i hope your recovery is swift :)
lizmat thanks!
jnthn 'night, lizmat; get well :) 23:24
jnthn also calls it a night o/ 23:25
timotimo sometimes, a night is just a night 23:36
Geth ¦ rakudo: AlexDaniel self-unassigned 2018-08-10 Toast results / 2̶9̶  18 burns github.com/rakudo/rakudo/issues/2196 23:46

Powered by Cro and the Raku® Programming Language.

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

Thank you!