🦋 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.
timo m: sub circumfix:<{ }.is>(&b) { return %( value => &b() ) }; { 5 + 9 }.is<value>.say 00:01
camelia Type check failed in binding to parameter '&b'; expected Callable but got Int (14)
in sub circumfix:<{ }.is> at <tmp> line 1
in block <unit> at <tmp> line 1
timo ah, it won't parse that as a block, obviously
gfldex m: use Test; is (TR/\#/a/ with '#'), 'a', 'transliteration of # works'; 00:29
camelia not ok 1 - transliteration of \# works
# Failed test 'transliteration of \# works'
# at <tmp> line 1
# expected: 'a'
# got: '#'
gfldex Should we be able to transliterate "#" with tr/// and TR/// ? 00:30
RT#2118 may be related 00:32
releasable6 Next release in ≈2 days and ≈11 hours. 3 blockers. Please log your changes in the ChangeLog: github.com/rakudo/rakudo/wiki/ChangeLog-Draft 07:00
Geth rakudo: 1ef7019bea | (Christian Bartolomäus)++ | src/core.c/REPL.pm6
[JVM] Don't let signal(SIGINT) break the REPL

The support for signals on the JVM backend is only rudimentary. The recently introduced (cf7b4f18da) call to signal(SIGINT).tap doesn't work on the JVM backend:
   $ ./rakudo -e 'signal(SIGINT).tap'
... (5 more lines)
07:29
Geth rakudo: 7772b737b0 | (Daniel Green)++ | src/Perl6/Actions.nqp
Restore larger limit for using string_to_int on 64bit systems

The limit was reduced because NQP used to use doubles to represent ints and only numbers up to 2**52 can be exactly represented by doubles. As of github.com/Raku/nqp/pull/518, NQP uses integers for ints and these limits can be increased again.
08:08
rakudo: 0851a253ad | MasterDuke17++ (committed using GitHub Web editor) | src/Perl6/Actions.nqp
Merge pull request #4670 from MasterDuke17/restore_larger_limit_for_using_string_to_int_on_64bit_systems
timo who wants to have fun with this idea: some terminal emulators (probably most of them?) can tell you when the user has pasted a whole block of text. the REPL could react in a special way to that, for example by offering to put the stuff in as a properly escaped string literal or immediately paste it into a variable even 10:00
lizmat interesting: twitter.com/lestrrat/status/1470535419985297410 11:54
gfldex lizmat: That is a lenghty version of: Be polite or be blocked. :) 13:04
lizmat yeah, indeed. But I really like point 4 :-) 13:06
an open source version of no brown M&M's clause :-) 13:07
Geth ¦ problem-solving: Xliff assigned to codesections Issue A Mechanism for Adding Hooks into CURI for Custom Installation Output github.com/Raku/problem-solving/issues/309 14:41
japhb lizmat: Oh yeah, I've used point 4 at @day-jobs before. It's fairly effective (not perfect though, people who are following a complex template often leave it there while filling in the requested info, and then forget to remove the template header at the very end). 15:36
lizmat japhb: well, very simple solution: close the issue and tell them why (not having removed point 4 :-) 15:37
if it's important, they'll re-do the issue and have learned something
fwiw, I'll probably also fall into that "trap" :-) 15:38
hopefully only once :-)
japhb Yeah, the problem is when you have a LOT of customers, each of which contacts you usually only once. You can train individual customers for the general benefit of all the surrounding teams, but not as much for your own benefit. 15:40
lizmat well, if they're paying customers, it could be different of course 15:42
paying customers either would have a different template (lacking #4), or would be exempt from having to remove it
japhb timo: Terminal::LineEditor has the raw info to detect a paste event (github.com/japhb/Terminal-LineEdit...131-L135), but I'm not making use of that information since Rakudo doesn't know what to do with it. But I'm happy to support it if someone figures out exactly what UX they want.
Yeah, auto-closing paid customer tickets is a fast way to lose said customer. 15:43
I'd only consider deploying #4 for internal teams -- where my experience is that basically everyone is just trying to get things done, and thus constantly trying to shuffle effort around. 15:44
Geth rakudo/Pair.ACCEPTS: 711f378af6 | (Elizabeth Mattijsen)++ | src/core.c/Pair.pm6
Fix the behaviour of Pair ~~ Pair

Apparently, this was borked since d8f7e8b8d6ea25b about 6 years ago.
See discussion at logs.liz.nl/moarvm/2021-12-15.html#16:32-0003
This breaks test #177 of t/spec/S02-types/pair.t
17:07
rakudo: lizmat++ created pull request #4671:
Fix the behaviour of Pair ~~ Pair
[Tux] Rakudo v2021.10-149-g0851a253a (v6.d) on MoarVM 2021.10-126-g83c53580a
csv-ip5xs0.846 - 0.959
csv-ip5xs-205.243 - 5.516
csv-parser4.128 - 4.143
csv-test-xs-200.406 - 0.408
test7.051 - 7.084
test-t1.636 - 1.708
test-t --race0.954 - 1.007
test-t-2023.885 - 24.050
test-t-20 --race7.345 - 7.398
17:33
Geth DBIish: 885b9267cc | (Rod Taylor)++ | 2 files
Change README filename references

The file was converted to Markdown in 3c5e7ec.
19:08