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.
AlexDaniel 6c: my @a = 1..2 Z 'a'..'b'; say @a[$_][*-1] // 42 for <0 -1>; 01:01
committable6 AlexDaniel, gist.github.com/406c77253a8cf26502...989f309889
AlexDaniel bisect: my @a = 1..2 Z 'a'..'b'; say @a[$_][*-1] // 42 for <0 -1>; 01:02
bisectable6 AlexDaniel, Bisecting by exit code (old=2015.12 new=dbf3638). Old exit code: 0
AlexDaniel, bisect log: gist.github.com/ff0bed6c3d220f1e06...eba5a95deb
AlexDaniel, (2019-03-16) github.com/rakudo/rakudo/commit/4f...d943651942
AlexDaniel c: 4ffb4082^,4ffb4082 my @a = 1..2 Z 'a'..'b'; say @a[$_][*-1] // 42 for <0 -1>;
committable6 AlexDaniel, ¦4ffb4082^: «a␤42␤» ¦4ffb408: «a␤Index out of range. Is: -1, should be in 0..^Inf␤ in block <unit> at /tmp/6VW30a8WaD line 1␤␤ «exit code = 1»»
01:55 vrurg left, vrurg joined 02:25 vrurg left, vrurg joined 03:11 AlexDani` joined 03:16 AlexDaniel left 03:38 vrurg left 03:39 vrurg joined 06:23 robertle joined 06:50 ufobat joined 06:51 ufobat_ left 07:09 patrickb joined
patrickb .tell Kaiepi I was the one introducing MVM_exepath. Please do ask if you need help with that or something related. I might be able to give a hint. 07:12
yoleaux patrickb: I'll pass your message to Kaiepi.
07:25 robertle left
Kaiepi thanks patrickb, i'll need some after my break 07:33
yoleaux 07:12Z <patrickb> Kaiepi: I was the one introducing MVM_exepath. Please do ask if you need help with that or something related. I might be able to give a hint.
lizmat Files=1255, Tests=90107, 390 wallclock secs (22.55 usr 6.85 sys + 2796.45 cusr 235.40 csys = 3061.25 CPU) 07:34
so yesterdays was a fluke
+ this test adds about 5 CPU seconds of new sprintf tests
07:47 ufobat left
kawaii releasable6: status 07:57
releasable6 kawaii, Next release in ≈5 days and ≈11 hours. 12 blockers. 0 out of 204 commits logged
kawaii, Details: gist.github.com/b30225bd9a34400224...304eb9b979
|Tux| Rakudo version 2019.03.1-204-gdbf3638ea - MoarVM version 2019.03-76-gc10fee65c
csv-ip5xs0.710 - 0.738
csv-ip5xs-205.886 - 6.010
csv-parser22.539 - 22.824
csv-test-xs-200.426 - 0.440
test7.669 - 7.701
test-t1.669 - 1.784
test-t --race0.869 - 0.881
test-t-2028.163 - 29.698
test-t-20 --race10.110 - 10.776
 08:37
jnthn Hm, 1.669 is one of the lowest we've seen for a while, I think? 08:42
timotimo in all of my scrollback at least 08:45
yoleaux 07:26Z <patrickb> timotimo: For the record: --no-relocatable does not change the runner to be a shell script again. The things it changes are:
07:27Z <patrickb> timotimo: - link libmoar with an absolute path.
07:28Z <patrickb> timotimo: - Set NQP_HOME and PERL6_HOME to an absolute path and do not try to determine them at run time.
|Tux| 2018-10-31 16:17:45 test-t 1.671 08:52
2019-04-15 10:31:40 test-t 1.669
2018-09-28 12:01:04 test-t 1.662
2018-10-15 11:01:41 test-t 1.659
2018-10-11 11:16:39 test-t 1.647
jnthn Damn, that's getting close. :) Maybe my next round of EA improvements will get a new low :) 08:53
timotimo the test-t times have startup time subtracted, right? 08:59
lizmat yes, afaik 09:09
or maybe not... doubting now
[Tux] ^^ ?
reportable6: 2019-04-08T12:00:00Z 2019-04-15T00:00:00Z 09:10
reportable6 lizmat, OK, working on it! This may take up to 40 seconds
lizmat, gist.github.com/603ff72e5ac3c20fdb...09224e7418 09:11
|Tux| lizmat. the times I post here *include* the startup time. (see tux.nl/Talks/CSV6/speed5.html) 09:32
timotimo cool, so work on that will also be reflected 09:33
not that i have anything for that at the moment 09:47
lizmat can someone explain me the importance of aur.archlinux.org/packages/comma-ide-community/ ? 09:52
09:57 ufobat joined
timotimo people who have the AUR ("arch user repository") activated in their arch package manager, which i think many do?, will now be able to just use pacman to install (and keep up to date) their comma 09:58
lizmat so that's pretty good and cool news, no? 09:59
timotimo the barrier to entry to the AUR is relatively low, so it's not quite as trusted as, for example, debian repositories might be ... perhaps?
would have to ask arch users for their opinion
but yeah, that's cool
tobs lizmat: yeah, it's community-maintained build scripts for software not in the main repositories. At least it goes through the package manager like that. 10:05
I just tried it, the script did its thing for 5 minutes and now I'm running comma \o/ 10:06
11:19 Guest88667 joined
Guest88667 lizmat: how is your sprintf rewrite progressing? (dogbert17) 11:25
lizmat Guest88667: I've decided to first implement extensive sprintf testing
I thought I understood sprintf, but found that I completely underestimated the intricacies of it 11:26
so it will most likely *not* be in this release
also: there's the question of making Perl 6 sprintf more logical instead of a maze of special cases
specifically with regards to the handling of 0, precedence of flags and how the "%o" is rendered 11:27
*format
Geth rakudo: 7931fc256b | (Elizabeth Mattijsen)++ | src/core/Rakudo/Internals/Sprintf.pm6
Remove new Sprintf implementation for now

This will be picked up again after the 2019.04 release.		 11:28
Guest88667 lizmat did you see my theory as to why 0 is used instead of , e.g. 0b0 ? 11:49
11:52 scovit joined
Guest88667 too bad Timtoady has gone missing, he probably knows the intricacies of (s)printf inside out 11:53
lizmat Guest88667: I'm pretty sure that was done that way, because '0' in Perl 5 is false, and '0b0' is true 11:54
we don't have that distinction in Perl 6 anymore
m: dd "0".Bool
evalable6 Bool::True
Guest88667 that's what I thought as well 12:09
lizmat Guest88667: also wrt to TimToady, feels to me he's more AWOL than MIA :-) 12:11
Guest88667 lizmat: btw gcc on my machine didn't like the 'b' flag
seems to be non standard 12:12
lizmat Guest88667: but it did like 'B' ? 12:14
Guest88667 test.c:4:15: warning: unknown conversion type character ‘B’ in format [-Wformat=]
perhaps b/B was never part of standard C (s)printf 12:15
12:22 pmurias joined
Geth nqp/truffle: be2f9e34aa | (Paweł Murias)++ | 6 files
[truffle] Implement P6int, P6num and P6str reprs
 12:23
12:26 AlexDani` is now known as AlexDaniel, AlexDaniel left, AlexDaniel joined
Geth nqp/truffle: c464c60145 | (Paweł Murias)++ | 4 files
[truffle] Remove trailing space and add missing files
 12:28
nqp/truffle: 8c7bd643e5 | (Paweł Murias)++ | 2 files
[truffle] Add Uninstantiable REPR
lizmat notable6: weekly 12:30
notable6 lizmat, 11 notes: gist.github.com/e32a190cdb68c818de...51ed0bab76
lizmat notable6: weekly reset 12:46
notable6 lizmat, Moved existing notes to “weekly_2019-04-15T12:46:47Z”
Geth nqp: b46d91b9fc | (Elizabeth Mattijsen)++ | tools/build/MOAR_REVISION
Bump MoarVM to get improved inlining of outer lexical binds
 12:47
¦ nqp: version bump brought these changes: github.com/MoarVM/MoarVM/compare/2...6-gc10fee6
pmurias lizmat: any news about what happened to TimToady or just wild speculations? :/ 12:52
lizmat nothing I can say out of first hand 12:53
he looked happy and healthy at the PuPPy event :-)
Geth rakudo: b7562d3b0e | (Elizabeth Mattijsen)++ | tools/build/NQP_REVISION
Bump NQP for latest optimizations and JVM backend fixes
 13:02
rakudo: version bump brought these changes: github.com/perl6/nqp/compare/2019....2-gb46d91b
7bbbebc0be | (Elizabeth Mattijsen)++ | 3 files

Part of fixing R#2788
lizmat and another Perl 6 Weekly hits the Net: p6weekly.wordpress.com/2019/04/15/...hrodinger/ 13:26
AlexDaniel please use github.com/rakudo/rakudo/wiki/ChangeLog-Draft to make things easier for kawaii :) 13:33
also, this looks scary: github.com/rakudo/rakudo/labels/BLOCKER 13:37
some of them were bisected nicely so maybe it looks scarier than it is
14:06 pmurias left 14:26 pmurias joined 15:16 pmurias left, Guest88667 left 15:26 patrickb left
Kaiepi i think i fixed the issue with $*EXECUTABLE and modules! 16:14
just needs some more testing so i can be positive 16:15
16:57 vrurg left, vrurg joined 17:06 vrurg left, vrurg joined 17:15 patrickb joined
patrickb How are the commit shas in the Changelog meant to work? For my relocatability work, do I put only the merge commit sha in there or all the later fixup commit shas as well? 18:30
ugexe put the merge commit under ignore 18:34
group the rest of the commits as you see fit and summarize
also in case you didn't notice: github.com/rakudo/rakudo/issues/2840
i'm not sure if that can cause a regression when e.g. upgrading an existing rakudo or not 18:35
im thinking no, but just in case
patrickb ugexe: I've seen it. I'll look into that soon. Thanks for the pointer! 18:39
ugexe: So I need list every single commit? 18:41
ugexe yes. there is a list somewhere of unlogged commits to make it a bit easier to put that list together (but i forget where) 18:47
but you don't have to put a description next to each commit. you can list multiple commits next to a single explaination 18:48
releasable6 Next release in ≈4 days and ≈23 hours. 12 blockers. Please log your changes in the ChangeLog: github.com/rakudo/rakudo/wiki/ChangeLog-Draft 19:00
19:22 vrurg left, vrurg joined 19:49 lucasb joined 20:28 vrurg left 20:29 vrurg joined
rakudo: 9ce87eeb9c | (Elizabeth Mattijsen)++ | src/core/array_slice.pm6
Make sure any Failures from .POSITIONS are passed on

Fixes R#2788
synopsebot R#2788 [open]: github.com/rakudo/rakudo/issues/2788 [BLOCKER][regression] Defined-or operator lets negative subscript cause a crash
MasterDuke lizmat can't repro R#2827 on her mac, but a bunch of different linuxes have reproed. could someone else on a mac or a different BSD try? 22:03
yoleaux 17:44Z <samcv> MasterDuke: can you open a bug ticket for this? i will have to get clang 8 and try to fix this.
synopsebot R#2827 [open]: github.com/rakudo/rakudo/issues/2827 [BLOCKER][regression] --target=parse causes a hang since sometime after 2018.10
22:13 evalable6 left 22:15 evalable6 joined, ChanServ sets mode: +v evalable6 23:27 squashable6 left 23:32 squashable6 joined, ChanServ sets mode: +v squashable6 23:41 patrickz joined 23:45 patrickb left 23:48 lucasb left 23:55 vrurg left, vrurg joined

Powered by Cro and the Raku® Programming Language.

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

Thank you!