🦋 Welcome to the IRC channel of the core developers of the Raku Programming Language (raku.org #rakulang). This channel is logged for the purpose of history keeping about its development | evalbot usage: 'm: say 3;' or /msg camelia m: ... | log inspection situation still under development | For MoarVM see #moarvm
Set by lizmat on 22 May 2021.
vrurg greppable6: 'is\s+dynamic' 01:04
greppable6 vrurg, Found nothing!
vrurg greppable6: 'is dynamic' 01:05
greppable6 vrurg, Found nothing!
vrurg greppable6: CALLER::MY 01:28
greppable6 vrurg, 1 line, 1 module: gist.github.com/5b60799a97a2b304f4...e8420051b6
vrurg greppable6: CALLER::LEXICAL 01:29
greppable6 vrurg, 5 lines, 1 module: gist.github.com/953da7bbffcb61eb09...8601a06c8a
vrurg greppable6: LEXICAL 01:45
greppable6 vrurg, 25 lines, 3 modules: gist.github.com/f1821330f160e6e259...0ad88622ad
sjn quick question; why are the *able6 bots reset their connection so often? 08:13
resetting*
MasterDuke AlexDaniel: ^^^ 08:17
Geth rakudo/exit-with-note: e7be747338 | (Elizabeth Mattijsen)++ | src/core.c/Rakudo/Internals.pm6
Allow a string to be specified with exit()

This came up recently in a blog post: having an easy way to exit a CLI script with a message and an error, but *without* any stack trace.
In Perl, this is achieved with die() if the message does not have a newline. ... (10 more lines)		 08:21
rakudo: lizmat++ created pull request #4494:
Allow a string to be specified with exit()		 08:22
lizmat FWIW, I've been getting disconnected the past days several times as well
it seems that Libera.chat is doing some maintenance / upgrading 08:23
[06:38:06] -glguy-[Global Notice] We'll be rerouting 4 servers and restarting 2 servers shortly as part of scheduled maintenance. Updates will continue at the rate of a couple servers a week in this time slot as we upgrade the network to the new Debian release.
sjn lizmat: did you mean "..if the message _has_ a newline" in the commit message above? 08:24
lizmat no, *not* is my impression?
aah... I *did* get it backwards!
sjn perl -e 'die "foo"'; perl -e 'die "foo\n"'
lizmat fixing it in the PR 08:25
AlexDaniel sjn: I don't know! There's a rather agressive watchdog which kills them if nothing is received for a while
sjn: sometimes it gets triggered when nobody says anything :D 08:26
it used to be that IRC::Client would get stuck and not reconnect
which is why I added a watchdog
it's a total hack, but it worked 08:27
lizmat sjn++ # keeping an eye on thngs!
sjn are you able to run a tcpdump (or equivalent) to see if something happens on the network?
AlexDaniel sjn: not sure how that'd help. I think this is the bug report: github.com/raku-community-modules/.../issues/27 08:39
sjn: the watchdog is configured here github.com/Raku/whateverable/blob/...ervice#L45 and is notified here github.com/Raku/whateverable/blob/...at.pm6#L25 08:41
(haha at “TODO below” but there is no todo)
sjn: and here's where it is called: github.com/Raku/whateverable/blob/...#L143-L148 08:42
so basically, if somebody sends a message, then the watchdog is notified and the bot is kept alive
if it's too quiet, then the bots are killed
I can increase the timeout but then bots will be “stuck” reconnecting because of that bug in IRC::Client 08:43
sjn killed by... who?
AlexDaniel does that make sense?
systemd
oh also github.com/Raku/whateverable/issues/276 08:44
releasable6 Next release in ≈4 days and ≈3 hours. 1 blocker. Please log your changes in the ChangeLog: github.com/rakudo/rakudo/wiki/ChangeLog-Draft 15:00
lizmat weekly: phoenixtrap.com/2021/08/17/perl-ra...frenemies/ 18:59
notable6 lizmat, Noted! (weekly)
Geth nqp/new-disp: 86 commits pushed by (Jonathan Worthington)++, Altai-man++
review: github.com/Raku/nqp/compare/6c28ff...fcaa61e530		 20:31
rakudo/new-disp: 105 commits pushed by (Jonathan Worthington)++, (Stefan Seifert)++, Altai-man++
review: github.com/rakudo/rakudo/compare/9...d5d99e7154		 20:32
nqp/new-disp: 4 commits pushed by (Stefan Seifert)++ 20:34
rakudo/new-disp: 27f37a6b98 | (Stefan Seifert)++ | src/Perl6/Actions.nqp
Fix argument to nqp::hllize inadvertently getting a name

The value of a Pair in an arglist gets a .named annotation. The p6callmethodhow desugar re-used the existing QAST::Op node, modified it and used it as argument to nqp::hllize. The new dispatcher based implementation doesn't cope with the name though. Fix by creating a new un-named QAST::Op and copying over the data we need.
rakudo/new-disp: 48ddac6ce7 | (Stefan Seifert)++ | src/Perl6/bootstrap.c/BOOTSTRAP.nqp
Add an raku-hllize dispatcher to replace nqp::hllize(for) on MoarVM

Powered by Cro and the Raku® Programming Language.

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

Thank you!