🦋 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.
lizmat nine: could it be that True / False in src/Raku/ast land, are not a True / False in Raku land ? 09:52
[Tux] Rakudo v2023.04-47-g2e3690bda (v6.d) on MoarVM 2023.04-1-g965be0af5
csv-ip5xs0.823 - 1.016
csv-ip5xs-205.545 - 6.421
csv-parser3.662 - 4.030
csv-test-xs-200.388 - 0.396
test6.695 - 7.541
test-t1.424 - 1.467
test-t --race0.859 - 0.974
test-t-2021.988 - 22.147
test-t-20 --race6.170 - 6.997
15:23
www.perlmonks.org/?node_id=11151954 ← I answered with what I think is the answer to the original quest, but I have no explanation to the trouble that hits his fan 15:38
[Coke] what's the name of the ? in Any.?does-not-exist() 15:44
(trying to search for it in docs)
ugexe safe call operator 15:47
docs.raku.org/language/operators.h...ethodop_.? 15:48
(the ? isn't a thing by itself, its tied to . ala .?)
[Coke] masterduke, ugexe: Ended up adding a way to pick a working dir for my $dayjob project so I can set it to "C:\g" to get me a little more breating room on long path. 17:22
ugexe (.?) Danke. 17:23
*breathing
I know I still need to test the masterduke long path patch, but I expect, as ugexe noted, that it still needs some raku work as well. 17:25
Geth rakudo/main: ae1856db2c | (Elizabeth Mattijsen)++ | 4 files
RakuAST: properly handle configuration of pod blocks

  - use the default <colonpair> token instead of a custom <doc-colonpair>
  - allow for :foo(1,2,3) in configuration
  - hack around NQP's True and False not being HLL True / False for :bar
  - make sure doc blocks are not eaten by statements that won't be CHECKed
  - allow configuration for =for doc blocks
  - make sure that empty strings are not added to legacy pod
17:32
[Coke] Want to cut over Moarvm from master to main this weekend so we match on all three repos? 17:42
I can press the actual button, probably needs a note to Tux and a ticket on rakubrew. 17:43
(and the weekly, and #raku) 17:48
lizmat ack
[Coke] ack protocal ACK or ack bill the cat ACK? 17:51
..........o
lizmat weekly ack
[Coke] or should we wait a week to pre-announce?
Just want to make sure we eventually press the button 17:52
lizmat nah, not a lot happening with MoarVM atm
so it won't break anything
[Coke] m: "THUMBS UP SIGN".uniparse.say 17:56
camelia 👍
nine lizmat: True/False should be the same and I haven't seen any indication that this isn't the case 18:50
lizmat hmmm... ok, then I will need to revisit that part 18:51
Geth rakudo/main: 6f3c535281 | (Elizabeth Mattijsen)++ | 2 files
RakuAST: add special handling for =begin comment

Specifically with regards to invalid pod before the =end comment
19:08
lizmat 803!
nine: alas, when binding an NQP True or False to a hash. passing that to an argument to a HLL method, the resulting hash can not be .rakued 19:12
No such method 'raku' for invocant of type 'NQPMu'
so I think somehow there's some automatic HLLizing going on
that being in a hash is circumventing
calling nqp::hllizefor(True,'Raku') doesn't help either 19:14
so keeping the hack in there for now 19:15
(Actions/extract-config)
nine There is no NQP True or False. True or False are the objects defined in the BOOTSTRAP. 19:32
Note that that's only true for src/Raku/ast/ not Grammar or Actions 19:34
Oh you are talking about Actions? There you'd need to look up these symbols
Or better let RakuAST node code handle that 19:36
vrurg There'd be a lot of errors avoided would NQP complain about missing symbols instead of using NQPMu.
lizmat thing is, I want to be able to build the entire contents of the RakuAST::Doc tree *without* needing a CHECK run 19:40
nine so? 19:45
lizmat ok, I think I figured it: $*R.resolve-lexical-constant-in-setting('True').compile-time-value;
right?
nine Looks good
Geth rakudo/main: d1ca18f17e | (Elizabeth Mattijsen)++ | 2 files
RakuAST: remove ginormous True/False hack

Thanks to nine++ and vrurg++ it occurred that the symbols as such weren't known in the Actions, and so silently NQPMu was inserted.
20:02