🦋 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 available at irclogs.raku.org/raku-dev/live.html | For MoarVM see #moarvm
Set by lizmat on 8 June 2022.
00:49 AlexDaniel left 00:55 MasterDuke joined 02:05 bisectable6 left, committable6 left 02:07 bisectable6 joined, committable6 joined 02:16 hythm joined 03:17 committable6 left, bisectable6 left 03:19 bisectable6 joined 03:20 committable6 joined 04:20 committable6 left 04:21 committable6 joined 04:41 vrurg_ joined 04:44 vrurg left 05:26 melezhik joined 05:27 melezhik left 06:29 hythm left 06:56 ab5tract joined 07:09 ab5tract left 07:11 ab5tract joined 07:17 kjp left 07:19 kjp joined 07:57 sena_kun joined 08:23 lizmat left, MasterDuke left 08:24 lizmat joined
ab5tract lovely... sinking doesnt seem to be guaranteed until check time. but I need to know about sunkenness at begintime :( 09:01
lizmat ? if it is not the last statement in a statement list, it is sunk, isn't it? And you can't know this at BEGIN time, as you may or may not have reached the end of the statement list yet ? 09:02
so, isn't it impossible to know at BEGIN time ? 09:03
ab5tract hmm, I guess you are right. Thought that sinking worked a bit differently than that (I thought it was dependent on the nature of the statements themselves) 09:08
m: Q| |((* > 1) xx 3) | .AST.EVAL 09:09
camelia WARNINGS for <tmp>:
Useless use of "|" in expression "((* > 1) xx 3) | .AST.EVAL" in sink context (line 1)
No such method 'CALL-ME' for string ' '
in block <unit> at <tmp> line 1
ab5tract m: Q| dd |((* > 1) xx 3) | .AST.EVAL
camelia WARNINGS for <tmp>:
Useless use of "|" in expression "((* > 1) xx 3) | .AST.EVAL" in sink context (line 1)
No such method 'CALL-ME' for string ' dd '
in block <unit> at <tmp> line 1
ab5tract anyway, I'm having trouble with that specific whatever form.. it keeps trying to curry the `~~` instead of the `>` and I don't have a good way of detecting when this situation has occurred (without special casing it to unknown fragility levels) 09:10
I'm literally 1 test case away from being able to replace the entire whatever stack with a single role. what we have now fits all the tests but it is really fragile :( 09:14
lizmat could it be some magic in EXPR ? 09:15
ab5tract you mean as a cause or as a solution? 09:18
lizmat solution actually
ab5tract oh, interesting! 09:19
lizmat I'm still trying to grok why the l10n things I'm doing don't work, and it seems that EXPR has something to do wiuth it
ab5tract :(
the l10n stuff is really interesting, btw 09:20
lizmat yeah, I got stuff like: say "zonder" zonder Int; already working 09:22
but as soon as there's something declarative, it reverts to assuming it's a sub being called 09:23
so: 'mijn $a' complains about the sub "mijn" not being found
the real problem is the 1 minute+ test cycle :-( 09:25
09:32 sena_kun left 09:51 tellable6 left 09:52 tellable6 joined
lizmat ab5tract: ok, I've come to the conclusion that github.com/rakudo/rakudo/issues/5387 is the reason l10n doesn't work 10:13
and the reason it works so far, is that EXPR is "manually" calling the associated Action methods 10:14
so I guess I will need to focus on #5387 first :-( 10:15
which probably means I will have to start debugging NQP :-( 10:17
ab5tract ouch! 10:25
lizmat well, at least it's not that the braid loses the actions... 10:31
so it's the !reduce that probably decides somehow there's nothing to call
so that would mean nqp::tryfindmethod($actions, $name) is failing ? 10:32
ab5tract that would be.. odd 10:33
but I guess that would make sense 10:35
10:35 ab5tract left
lizmat well, I won't be able to do much the coming days as I'll be mostly afk 10:49
another datapoint: adding a FALLBACK method to the actions class, it gets called if the requested method is absent, in the legacy grammar 10:57
in the Raku grammar, the FALLBACK method doesn't get called either
11:24 ab5tract joined
lizmat afk& 11:26
11:26 lizmat left
[Coke] all the Old Ones visible in the mist this week. 11:41
12:19 NemokoschKiwi joined 12:51 NemokoschKiwi left 13:04 ab5tract left 14:13 ab5tract joined 14:47 AlexDaniel joined 16:03 zostay left, zostay joined 16:07 elcaro left, elcaro joined 16:49 ab5tract left, ab5tract joined 18:01 sena_kun joined
ab5tract I've got four spectests passing with the the new whatevercode implementation. but I've got some tough gotchas in s/S02-types/whatever.t 18:21
19:21 committable6 left 19:24 committable6 joined 20:58 lizmat joined
lizmat . 21:10
22:10 committable6 left, bisectable6 left, bisectable6 joined 22:11 committable6 joined 22:42 AlexDaniel left 22:49 MasterDuke joined 22:56 sena_kun left