🦋 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.
00:00 evalable6 joined, reportable6 left 00:01 linkable6 joined 00:03 reportable6 joined
Geth Template6/main: e53f468f98 | (Polgár Márton)++ | 8 files
01:20 jgaz left 01:25 jgaz joined 02:25 bloatable6 left, sourceable6 left, greppable6 left, shareable6 left, coverable6 left, quotable6 left, linkable6 left, notable6 left, unicodable6 left, benchable6 left, committable6 left, nativecallable6 left, statisfiable6 left, evalable6 left, tellable6 left, reportable6 left, releasable6 left, squashable6 left, evalable6 joined 02:26 releasable6 joined, coverable6 joined, shareable6 joined 02:27 sourceable6 joined, tellable6 joined, notable6 joined, committable6 joined, statisfiable6 joined, linkable6 joined, unicodable6 joined, bloatable6 joined 02:28 benchable6 joined, squashable6 joined, quotable6 joined, nativecallable6 joined, reportable6 joined, greppable6 joined 03:28 notable6 left, statisfiable6 left, evalable6 left, shareable6 left, nativecallable6 left, bloatable6 left, linkable6 left, unicodable6 left, greppable6 left, reportable6 left, committable6 left, coverable6 left, tellable6 left, squashable6 left, quotable6 left, benchable6 left, releasable6 left, sourceable6 left 03:29 bloatable6 joined, linkable6 joined, coverable6 joined, shareable6 joined 03:30 tellable6 joined, greppable6 joined, committable6 joined, notable6 joined, evalable6 joined, reportable6 joined, benchable6 joined, unicodable6 joined 03:31 statisfiable6 joined, squashable6 joined, quotable6 joined, nativecallable6 joined, sourceable6 joined, releasable6 joined 05:25 discord-raku-bot left 06:00 reportable6 left, reportable6 joined 06:13 nebuchadnezzar joined 07:51 sena_kun joined 08:16 heartburn left 08:39 heartburn joined
Geth rakudo/main: 205fb661b5 | (Elizabeth Mattijsen)++ | 2 files
RakuAST: some more actions/grammar streamlining
09:56 evalable6 left, linkable6 left 09:59 linkable6 joined, evalable6 joined
Geth rakudo/main: 11334377a4 | (Elizabeth Mattijsen)++ | 2 files
RakuAST: streamline language version setup

  - only lookup dynamic vars once where possible
  - all dynamic vars are UPPERCASE
  - use more meaningful variable names
  - lots of comments added / reformatted / edited
  - kekab-case where possible
rakudo/main: 34a6c01dde | (Elizabeth Mattijsen)++ | 4 files
RakuAST: add ::resolver.setting-constant method

Mostly to improve readability of:
12:00 reportable6 left 12:02 reportable6 joined 12:42 sena_kun left
Geth rakudo/main: 375d04e3ac | (Elizabeth Mattijsen)++ | 3 files
Make X::Syntax::Confused more logical

What's the point of having a default value for the reason, to use that later to have it return 'Confused'. Make 'Confused' the default and be done with it.
Also apply this to some of the call sites.
jdv lizmat: are you changing professions to blogger? 13:40
nice series though 13:41
13:41 donaldh joined
lizmat jdv: no, it was just something I had already spent about 2 months on that was just laying there 13:41
and thank you :-)
only 9 episodes left unpublished 13:42
jdv did you pick over the advent stuff? it all feels like echos of that, to me. 13:43
lizmat in 2019 there was one advent post that contained one chapter of the book
I forget which
and then there were the 2018 posts on opensource.com of course 13:44
jdv ah
which book?
lizmat the book that I was working on late 2019 / early 2020 13:45
the book that won't get published :-)
13:49 donaldh left
jdv :) 13:51
guess i missed that bit 13:52
lizmat it's in the first paragraph of the first post :-) 13:53
jdv anything large/breaking this release cycle?
i skim...:)
lizmat nah, adding adverbs to .min / .max and :k to sort, are the biggies I guess
jdv cool 13:54
why didn't you do the book, maybe a a mini thing with some place like leanpub or whatever like bdfoy does for "perl news" 14:16
lizmat many reasons.. some will be revealed in the final blog posr 14:17
jdv should i get popcorn? i can't wait!
lizmat no, it's more nostalgia 14:18
jdv hmm, i didn't know KEEP and LEAVE depended on the return value 14:40
that's not what the docs say, imho
i had assumed it had to do with whether an exception was thrown or something like that 14:41
lizmat KEEP / UNDO depend on return value 14:44
jdv the docs just say "successful/unsuccessfull block exit" but ok 14:46
lizmat LEAVE is run always 14:48
jdv oh, i wrote LEAVE when i meant UNDO 14:51
sorry - dooing too many things at once
lizmat yeah.. that happens :-)
jdv also, any reason you left out the other phasers - async/exceptions/control?
lizmat CATCH / CONTROL have their own blog post 14:53
DONE will not be handled
jdv cool 14:54
🥞🥞 No regressions found, dot file not saved 15:13
lizmat whee! pancakes!
jdv no fun this round:( i guess the release is on for Monday as planned.
lizmat cool!
Geth rakudo/main: c7ad58de8c | (Elizabeth Mattijsen)++ | 2 files
RakuAST: Streamline / comment statements section

  - a lot more comments
  - some token / method renaming for clarity
  - some code moving into the proper section
15:44 NemokoschKiwi joined 17:19 jgaz left 18:00 reportable6 left 18:03 reportable6 joined 18:24 Geth left 18:30 Geth joined
[Coke] jdv: JVM is broken. 18:57
but I think it's been broken for several releases now. :|
jdv yup 19:06
afaik thats not a blocker, unless someone believes otherwise 19:07
[Coke] lizmat: gist.github.com/coke/cd6adb22f7193...91dfb837d3 - is this expected behavior, or a rakuast bug?
jdv: I can't say it's a blocker, no
<dad>I'm just disappointed</dad> 19:08
lizmat well, it's not a valid table, but the error is LTA
[Coke] I believe it's the same error with a valid table. 19:09
one sec, will try harder.
(are you sure? it looks very similar to some tables in the raku/doc repo
updated gist with the exact table giving me trouble in the repo 19:10
lizmat well, it's a table with two rows and 1 column 19:11
[Coke] I updated the gist.
So if that's an invalid table, we should be probably consider throwing an error at --doc time. (can you give me an example of a valid table that shouldn't throw an error?) 19:12
(--doc works fine on the file containing that table, it's me trying to call .paragraphs that's the problem)
lizmat yeah, looking at it now 19:14
[Coke] lizmat++ 19:15
lizmat ok, I figured it out, now to find a solution 19:17
testing fix 19:22
[Coke]: you probably want to do map(*.paragraphs.Slip) if you want to iterate over all paragraphs 19:24
Geth rakudo/main: e6f33599f0 | (Elizabeth Mattijsen)++ | 3 files
RakuAST: make ::Statement::IfWith a proper base class

It was just called ::Statement::If, with ::Statement::With being a subclass of that. This was a bit confusing.
Also add "set-elsifs" and "set-else" methods, for convenience
rakudo/main: 28d266ff47 | (Elizabeth Mattijsen)++ | src/core.c/RakuAST/Fixups.pm6
RakuAST: fix issue with ::Doc::Row.raku

This one escapes the .raku dispatcher, and therefore doesn't set the
  $*INDENT dynamic var if it hasn't been set yet causing an execution
error with indenting.
It now sets a $*INDENT if there isn't one yet, so it should *always* work now, whether there is an $*INDENT set or not
lizmat [Coke] ^^
[Coke] thanks 19:34
afk 19:35
vrurg lizmat: instead of nqp::istype($*INDENT,Failure) it'd better to use nqp::isnull(nqp::getlexdyn('$*INDENT')). Faster and at least one allocation of Failure less. 19:36
lizmat hmmm... I guess I could do that if I'm sure it won't be set in PROCESS:: or SETTING:: right ?
still, to be handling threads correctly... one would have to look up $*PROMISE as well, no ? 19:38
vrurg Yes. But if it can be found there then, perhaps, DYNAMIC::.EXISTS-KEY would still do better.
lizmat hmmm... I think I'm gonna leave that for another day 19:40
also: the creation of the Failure would only happen on the first time 19:45
Geth rakudo/main: efb5866426 | (Elizabeth Mattijsen)++ | src/core.c/RakuAST/Raku.pm6
RakuAST: use "given" instead of "with" to topialize

As we can be sure now that $*INDENT will always be set
rakudo/main: 35467a2082 | (Elizabeth Mattijsen)++ | src/Raku/Actions.nqp
RakuAST: Streamline statement control actions

  - add helper methods to concentrate common logic
  - simplify if/with handling
  - add more comments
lizmat and that concludes my hacking for today&
20:43 vrurg_ joined 20:45 vrurg left, vrurg_ is now known as vrurg 20:57 NemokoschKiwi left