🦋 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.
releasable6 Next release in ≈6 days and ≈15 hours. There are no known blockers. Please log your changes in the ChangeLog: github.com/rakudo/rakudo/wiki/ChangeLog-Draft 03:00
Geth roast: 24f41fd423 | (Elizabeth Mattijsen)++ | 2 files
Move .subst test to .subst testing file

Instead of being out of place in a .substr testing file		 08:59
lizmat 825! :-)
Geth rakudo/main: 8039722595 | (Elizabeth Mattijsen)++ | 4 files
RakuAST: issue a worry over leading declarator doc

If at the end of compilation no declarand could be found for the leading declarator doc that was collected.
Also turn the X::Syntax::Doc::DeclaratorTrailing::MissingDeclarand class into a more general X::Syntax::Doc::Declarator::MissingDeclarand class, defaulting to "trailing".		 10:05
lizmat nine: is the $!context of compunit always correct? 14:09
I'm trying to fix the CHECK phasers, and two new test fails could be explained by either the resolver or the context not being correct 14:10
nine: sanity check: is QAST produced during CHECK time, or after that? 14:29
nine: seeing anything explicitely wrong with this? gist.github.com/lizmat/9ca50060210...ab5b89d096 14:47
the current implementation really treats CHECK phasers as INIT phasers, so they get pre-compiled and run every time 14:50
instead of just after compile time 14:51
nine QAST comes after CHECK 14:54
I would expect CHECK phasers to be run by a node's IMPL-CHECK 14:57
or by extension PERFORM-CHECK
lizmat hmmmm
interesting.... ok, lemme try taht 14:58
nine Since CHECK phasers can only be on blocks or comp units, those ought to do CheckTime and have a PERFORM-CHECK that runs those phasers 14:59
lizmat which sorta implies they don't need to be collected in the compunit either
nine No, only INIT needs that as it's part of the runtime 15:00
BEGIN is as early during compilation as possible, CHECK as late during compilation as possible and INIT is as early in runtime as possible with END being as late in runtime as possible. 15:01
Ha! That's even what the docs say :) 15:02
lizmat it all makes sense now :-) 15:03
testing now 15:04
Geth rakudo/main: a3f0f36584 | (Elizabeth Mattijsen)++ | 2 files
RakuAST: make CHECK phasers really run at CHECK time

Rather than that they basically act as INIT phasers		 15:12
lizmat sadly, no additional test files
which implies there are definitely no precomp tests for CHECK phasers 15:13
lizmat nine: I just found out that in RakuAST::Compunit.check, $!mainline is always a block with an empty statementlinst 15:28
that doesn't seem right ?
aahh..there's also a statementlist attribute... what's their relation? 15:34
would have expected the statement list to be living inside the mainline?
m: CHECK { say $*CU.statement-list } 15:42
camelia ===SORRY!=== Error while compiling <tmp>
An exception X::Dynamic::NotFound occurred while evaluating a CHECK: Dynamic variable $*CU not found
at <tmp>:1
Exception details:
Dynamic variable $*CU not found
in block at <tmp> line 1…
lizmat that's weird? works locally on that commit ? 15:43
aaah... duh 15:44
nine: do we agree that CHECK phasers should not live in the statementlist ? 15:45
aka: CHECK { say $*CU.statement-list }
should show an empty statement list\
nine As with all oddities, compunit's $!mainline weirdness is due to BEGIN time stuff 15:47
See commit 676aa173fbd47a971dc56068e65d871995c952f6
m: say CHECK { 1 } 15:48
camelia 1
nine AT least their result has to be available
lizmat I get that, but shouldn't $!statement-list be the statement list of $!mainline ? 15:49
yeah, that's all working already 15:50
nine If it helps, I guess yes.
lizmat hmmm... that breaks imports 16:33
guess that will need further investigation
Geth nqp/main: db7d603cd7 | (Elizabeth Mattijsen)++ | tools/templates/MOAR_REVISION
Bump MoarVM to get the latest libuv version
 18:25
rakudo/main: fac61c28ed | (Elizabeth Mattijsen)++ | tools/templates/NQP_REVISION
Bump NQP to get latest libuv version
 18:41
releasable6 Next release in ≈5 days and ≈19 hours. There are no known blockers. Please log your changes in the ChangeLog: github.com/rakudo/rakudo/wiki/ChangeLog-Draft 23:00

Powered by Cro and the Raku® Programming Language.

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

Thank you!