🦋 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 reportable6 left 00:03 reportable6 joined 00:46 japhb joined 03:28 linkable6 left, statisfiable6 left, notable6 left, bisectable6 left, committable6 left, evalable6 left, tellable6 left, nativecallable6 left, greppable6 left, shareable6 left, sourceable6 left, unicodable6 left, releasable6 left, benchable6 left, coverable6 left, reportable6 left, quotable6 left, squashable6 left, bloatable6 left, sourceable6 joined 03:29 statisfiable6 joined, squashable6 joined, releasable6 joined, shareable6 joined, benchable6 joined, linkable6 joined, greppable6 joined, unicodable6 joined 03:30 evalable6 joined, bisectable6 joined, tellable6 joined, coverable6 joined 03:31 nativecallable6 joined, quotable6 joined, reportable6 joined, notable6 joined, bloatable6 joined, committable6 joined 04:58 Xliff left 06:00 reportable6 left 06:03 reportable6 joined 07:06 reportable6 left, greppable6 left, nativecallable6 left, unicodable6 left, statisfiable6 left, coverable6 left, shareable6 left, bloatable6 left, committable6 left, evalable6 left, bisectable6 left, sourceable6 left, squashable6 left, tellable6 left, quotable6 left, linkable6 left, releasable6 left, quotable6 joined, shareable6 joined 07:07 linkable6 joined, coverable6 joined, unicodable6 joined, bloatable6 joined, greppable6 joined 07:08 committable6 joined, evalable6 joined, reportable6 joined, squashable6 joined, bisectable6 joined 07:09 sourceable6 joined, releasable6 joined, nativecallable6 joined, statisfiable6 joined, tellable6 joined 07:52 sena_kun joined 08:41 ab5tract left 08:55 epony left
Geth rakudo/main: 6235256a56 | (Elizabeth Mattijsen)++ | 3 files
RakuAST: add a lot of scaffolding for handling pod

  - add RakuAST::Doc::Paragraph.from-string method
   This will parse the given string for X<> markup sequences and
   process them into the proper RakuAST::Doc::Markup hierarchy.
   This will replace the cumbersome process in the legacy grammar
   which created a Match object for *every* *single* character in
... (8 more lines)
lizmat tonyo: looks like fez upload is not indexing any modules anymore, for possibly 4 days already 09:12
lizmat just uploaded 0.17 of String::Utils, and it hasn't appeared 09:13
a new install will install 0.16 of the module 09:14
09:48 sena_kun left 11:25 epony joined
nine Btw. I'm surprised that in all the discussion around defined vs. undefined List and .elems, no one has brought up that being able to call .elems on any object as if it was a list is quite odd. 11:44
Without this design decision List could easily have a method elems(List:U:) returning 0 11:45
11:47 ab5tract joined
Nemokosch I think at some point, it was a spelt-out design paradigm to "deliberately mistake single items for one-element lists" (it was expressed something like this) 11:49
Any delegates basically the complete interface of a List docs.raku.org/type/Any 11:55
and as things stand, this backs up most listy calls of Buf and Blob 11:59
12:00 reportable6 left 12:02 reportable6 joined
ab5tract nine: did you encounter this while working on $?PACKAGE? 12:26
m: module M {}; dd GLOBAL::.kv; dd ::("GLOBAL")::.kv; dd $?PACKAGE::.kv
camelia ("M", M).Seq
ab5tract is it fair for me to expect that these would be equivalent? 12:27
MasterDuke hm, the fails in github.com/MoarVM/MoarVM/pull/1745/checks seem like they might not be related to the PR. but i don't like the sheer number of them... 12:44
nine ab5tract: looks fishy 12:51
ab5tract m: module M { module N {}}; GLOBAL::.values.map: { dd $_::.kv }; 12:59
camelia ()
nine I'm not sure about $_:: doing what you think it's doing 13:01
same for $?PACKAGE:: btw
The latter may be interpreted as a lookup in a package called "PACKAGE" with sigil and twigil getting added 13:02
or maybe even dropped
ab5tract My only conscious assumption here is that a trailing '::' is the way to get to the stash 13:04
m: module M { module N {}}; GLOBAL::.values.map: { dd ::($_.^name)::.kv };
camelia ()
ab5tract m: module M { module N {}}; GLOBAL::.keys.map: { dd ::($_)::.kv };
camelia ()
ab5tract if I'm just holding it wrong, than I apologize for wasting time. but it seems to me that there should be some way to access a stash when given either a type object or a package name 13:10
nine .WHO
m: module M {}; dd GLOBAL::.kv; dd ::("GLOBAL").WHO.kv; dd $?PACKAGE.WHO.kv 13:11
camelia ("M", M).Seq
("M", M).Seq
("M", M).Seq
nine Sorry, took me a while to remember this even though I've used it a hundred times during development and debugging... 13:12
ab5tract No worries 13:14
I think somewhere in the docs it suggests that a trailing '::' is equivalent to a call to .WHO
I guess it's just unclear to me whether the discrepancy is by design or not 13:15
thanks for giving me the fix though :)
nine The trailing :: is part of the name. .WHO is a macroish call on an object 13:16
ab5tract right, but even when I go into "name-mode" with ::($module-name)::, it doesn't DWIM 13:19
If '::' were unimplemented at the moment, I would probably make it an infix that dispatches the RHS on the .WHO of the LHS 13:21
but that's probably difficult for the parser because '.kv' already means dispatch-on-the-topic variable 13:22
13:23 jgaz joined
lucs nine: When you mentioned "with sigil and twigil getting added", I thought to myself "Hmm... Isn't a sigil a one character prefix and a twigil, a two-character one?" 14:14
So I went to the docs, and no, your usage was correct.
Otherwise, we would have needed to for example talk about the '$*', '@*', '%*', and '&*' twigils to refer to dynamic variables instead of just the '*' twigil. 14:15
So my mind stands corrected.
nine I can say that the whole RakuAST implementation effort has been highly educational so far :) 14:20
lizmat also, I wouldn't be surprised if the new pod grammar would be orders of magnitude faster than the current one 14:39
nine We're gonna need a bit of a speedup since right now compilation is about twice as slow :D 14:41
lizmat I hope to be able to rip out the old pod parts from the new grammar 14:45
this week
and then we'll know 14:46
although I don't think it will matter for any code without pod much
such as the setting
nine Well I'm not surprised that we are much slower as compilation performance has not been a priority and there is some low hanging giant fruit
15:08 [Coke] left 16:03 epony left
lizmat PSA: module developer: any dist upload to fez after Wed, 19 Apr 2023 17:34:36 GMT have silently gone into a black hole 16:55
uploads are now working again: any tar files will need to be uploaded again to get published 16:56
japhb lizmat: What actually broke? 17:40
lizmat tonyo can tell
japhb Ah, fair enough
lizmat and yet another Rakudo Weekly News hits the Net: rakudoweekly.blog/2023/04/24/2023-...aseperiod/ 17:43
is there some kind of holiday in the States today? 17:44
or is everybody celebrating Tucker Carlson's breakup with Fox News ?
17:51 sena_kun joined
nine lizmat: typo: "will by hybrid" 17:52
lizmat thanks fixed!
18:00 reportable6 left 18:02 reportable6 joined
japhb lizmat: No US holiday that I'm aware of. 18:02
nine yet 18:03
lizmat :-) 18:04
tonyo japhb: i put in a corruption check and i think something is wrong with the library i used for the lambda 19:04
apologies for the delay fixing, have been travelling for a while
19:29 linkable6 left, evalable6 left, linkable6 joined 19:31 evalable6 joined 19:55 [Coke] joined 21:09 sena_kun left 21:16 jgaz left 22:59 linkable6 left, evalable6 left, evalable6 joined 23:00 linkable6 joined 23:30 jgaz joined