|
00:42
eseyman left
00:44
eseyman joined
01:47
hulk joined
01:48
kylese left
02:15
hulk left,
kylese joined
03:30
dicit joined
05:50
dicit left
05:56
Sgeo left
06:06
abraxxa joined
06:08
abraxxa left
08:02
vrurg_ joined
08:04
vrurg left
08:19
grondilu left
08:36
grondilu joined
|
|||
| disbot | <demanddystopia> Is there a reason that is/can't? be a thing openly documented) | 09:26 | |
|
09:28
DemandDystopia joined
|
|||
| lizmat | if it's about the release process, it *is* documented afaik ? | 09:29 | |
| DemandDystopia | Though the way it's chained rakudo ← nqp ← moarvm got obvious when I once wanted to see how exactly `.is-prime` was implemented which seems to go through the entire chain & ends in a c file | 09:32 | |
| lizmat | yeah, but that's pretty rare | 09:33 | |
| nqp::time is another one like that | |||
| Voldenet | heh, I remember needing process id before it was returned from Proc::Async, so I made an op for extracting that info from libuv, which made me cut across all the layers too | 09:35 | |
| lizmat | ah, yes, nqp::getpid | ||
|
09:36
DemandDystopia13 joined,
DemandDystopia left
|
|||
| Voldenet | implementing a new op is unique experience, that's for sure | 09:36 | |
|
09:37
DemandDystopia13 left
|
|||
| Voldenet | but also makes one grasp how compilation is done and also all the layers | 09:37 | |
|
09:38
DemandDystopia joined
|
|||
| Voldenet | (though if I implemented this today, I'd probably use nativecall and quick and dirty lib linked to libuv from moar…, definitely less work) | 09:38 | |
| DemandDystopia | Definitely wanna get some understanding of how the entire conpiler clicks today one day. Picked up Raku like 2–3 months ago and became really enamored with the introspection & "I can't believe this works" type stuff, which leads to eventually the best documentation being the actual source or grammar | 09:47 | |
|
09:59
DemandDystopia left
|
|||
| disbot | <antononcube> What do you do or want to do with Raku's introspection? | 11:44 | |
|
11:50
dakkar joined
11:52
dakkar left
11:57
human-blip left
11:59
human-blip joined
12:33
dicit joined
14:06
gugod left
14:07
gugod joined
14:15
Sgeo joined
14:51
[Coke]_ joined
14:53
[Coke] left
14:54
sorenson left
14:56
sorenson joined
15:11
abraxxa joined
15:18
abraxxa left
15:19
abraxxa joined
15:44
[Coke]_ is now known as [Coke]
17:02
human-blip left
17:14
human-blip joined
|
|||
| disbot | <demanddystopia> It's more exploration / experimentation with the language for which introspection is sometimes useful, like .^methods is neat, but often it's just that with introspection stuff you kinda get some data to just try out applying stuff to, so not-too-complex things like grabbing all Types that can be mutated, or just trying to grab as much information to store in reasonable ways. kinda interesting though idk how useful was | 18:35 | |
| finding out that by using callframe & .callable_for_phaser() one can like arbitrarily replay phasers or finding out that ==> can make at least prefix operators take at least 2 positional arguments (idk if that's fully intended but it's been there since it seems like 2022 so) | |||
| <demanddystopia> though the prefix ops thing is kinda a logjcal consequence of operators just being subs | 18:37 | ||
|
19:00
human-blip left
19:02
human-blip joined
|
|||
| ab5tract | demanddystopia: can you share a ==> example? And are you using RakuAST or the legacy compiler frontend? | 19:35 | |
|
19:38
DemandDystopia joined
|
|||
| DemandDystopia | m: sub prefix:<§>(*@a) {say "⟦{@a}⟧"}; 5 ==> §2 | 19:38 | |
| camelia | ⟦2 5⟧ | ||
|
19:42
DemandDystopia left
20:52
abraxxa left
|
|||