🦋 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: ... | Logs can be inspected at colabti.org/irclogger/irclogger_log/raku-dev | For MoarVM see #moarvm
Set by lizmat on 26 April 2021.
Geth roast: 60c4cd35df | (Tom Browder)++ (committed using GitHub Web editor) | S32-temporal/baum-gregorian-data.t
Correct the comments

Show the correct value for the Modified Julian Date reference.
10:37
AlexDaniel` can somebody from IRC tell me if *able6 bots are connected through ipv4 or ipv6 addresses? 13:13
I guess ipv6 13:16
right?
lizmat I"m not sure Freenode even does IPv6 13:29
raydiak they do seem to be connecting from IPv6 addresses: "evalable6 (~eva-able@2001:41d0:2:5eb5::) has quit (Remote host closed the connection)" 13:34
AlexDaniel` I see, thanks 13:40
raydiak yw 13:45
AlexDaniel` OK, now that probably looks right 13:55
(bots coming from both ipv4 and ipv6 addresses)
raydiak now it appears that some of them are on IPv4, others on IPv6
AlexDaniel` yeah, good! Thanks!
so I think I fixed the issue with some bots not rejoining 😊 13:56
raydiak nice! what was the cause, ooc? 13:57
AlexDaniel` well, there are many bots, and freenode has a limit of 10 per IP I think. So years ago I came up with this really quick workaround that worked for a looong time: 13:59
:host(%*ENV<TESTABLE> ?? ‘127.0.0.1’ !! <chat.freenode.net 84.17.44.180>.pick)
look at the .pick part. The server is ipv6-capable, so it connects using ipv6 by default
and the ipv4 address is of some specific freenode server I think
raydiak ah, I see 14:01
AlexDaniel` my best guess is that that IP is no longer there doing its freenode thing, and you're not supposed to connect to specific servers anyway
changed the IP, hopefully will last a few more years :)
raydiak AlexDaniel`++ 14:04
[Coke] MasterDuke: I do have one, but it's an older version. 17:29
(haven't done an install in a few months)
MasterDuke [Coke]: cool, that should be recent enough. do you mind running something for me? 18:27
[Tux] Rakudo v2021.04-10-gbb069a99c (v6.d) on MoarVM 2021.04-7-g09c4c4d42
csv-ip5xs0.823 - 0.849
csv-ip5xs-207.754 - 8.437
csv-parser24.890 - 26.079
csv-test-xs-200.381 - 0.385
test7.661 - 7.888
test-t1.879 - 1.900
test-t --race0.848 - 0.894
test-t-2032.773 - 33.183
test-t-20 --race9.239 - 10.746
18:43
MasterDuke [Coke]: in a powershell command prompt, from your rakudo directory, can you run `prove -j2 -e '../install/bin/raku' -vlr t`? 18:48
hm, you might not have it installed in a place where you need the initial '..', but i think that's whats breaking 18:49
it might not matter, maybe just running prove with any random path starting with '..' will have the same problem i'm seeing 18:50
but it's in azure, and i'm not 100% sure where the problem is occuring
Geth nqp: 3ff1af5daa | (Christian Bartolomäus)++ | src/vm/jvm/runtime/org/raku/nqp/sixmodel/reprs/Uninstantiable.java
[JVM] Add debug name to error for Uninstantiable

This mimics github.com/MoarVM/MoarVM/commit/40987366fc and makes the corresponding tests from Rakudo's t/05-messages/01-errors.t pass (added with github.com/rakudo/rakudo/commit/4666505570).
19:02
Geth rakudo: 21100c918b | (Christian Bartolomäus)++ | t/05-messages/03-errors.t
Fix typo in test description
20:27
[Coke] MasterDuke: ok - that's work machine, will give it shot at some point. 22:37
MasterDuke [Coke]: thanks, but no need now. think i figured it out. this is/was for github.com/MoarVM/MoarVM/pull/1483 22:38
i'd combined jobs and accidentally left in the only-do-this-if-os-is-not-windows conditional to build rakudo. so i was using what should/would be the correct paths, but it was also correctly complaining that nothing was there 22:41
[Coke] aha 22:46
MasterDuke++
Geth rakudo: 45e8e8d987 | (Christian Bartolomäus)++ | t/02-rakudo/repl.t
Add test for REPL handling junction correctly

As described in github.com/Raku/old-issue-tracker/issues/2917 the REPL failed badly (segfaulting) when smartmatching against a junction that contains a set:
   $ echo -e 'my \\a = any set <1 2 3>; \n say 1 ~~ a' | perl6
... (6 more lines)
23:58
rakudo: cbf12d22e1 | (Christian Bartolomäus)++ | t/02-rakudo/repl.t
Add test for REPL handling incomplete regex

As described in github.com/Raku/old-issue-tracker/issues/3254 the following evaluation used to fail badly in the REPL (on the JVM backend) a couple of years ago:
   'asa' ~~ /:s ^a* %
... (9 more lines)
rakudo: 877fdd775e | (Vadim Belman)++ (committed using GitHub Web editor) | t/02-rakudo/repl.t
Merge pull request #4292 from usev6/old_issue_tracker_2917

Add tests for former REPL crashes