|
00:15
lizmat joined
01:06
gcd left
01:46
hulk joined
01:47
kylese left
02:15
hulk left,
kylese joined
02:52
johnjay joined
03:03
simcop2387 left
03:09
oodani left
03:11
oodani joined
03:12
simcop2387 joined
03:47
johnjay left
03:53
johnjay joined
03:59
johnjay left
04:01
johnjay joined
04:08
johnjay left
04:10
johnjay joined
04:19
johnjay left
04:21
johnjay joined
04:29
johnjay left
04:31
johnjay joined
04:38
johnjay left
04:41
johnjay joined
04:49
johnjay left
04:51
johnjay joined
04:58
johnjay left
05:04
johnjay joined
05:11
johnjay left
05:13
johnjay joined
05:20
johnjay left
05:22
johnjay joined
05:30
johnjay left
05:33
johnjay joined
05:42
johnjay left
05:44
johnjay joined
05:52
johnjay left
05:54
johnjay joined
06:02
johnjay left
06:04
johnjay joined
06:15
johnjay left
06:17
johnjay joined
06:25
johnjay left
06:27
johnjay joined
06:33
johnjay left
06:34
johnjay joined
07:33
johnjay left
07:36
johnjay joined
07:45
johnjay left
07:47
johnjay joined
07:54
johnjay left
07:57
johnjay joined
08:04
johnjay left
08:06
johnjay joined
08:14
johnjay left
08:16
johnjay joined
08:24
johnjay left
08:26
johnjay joined
08:33
johnjay left
08:36
johnjay joined
08:45
johnjay left
08:47
johnjay joined
08:49
johnjay left
08:51
johnjay joined
08:57
johnjay left
09:03
johnjay joined
09:10
johnjay left
09:12
johnjay joined
09:18
johnjay left
09:21
johnjay joined,
Sgeo left
09:27
johnjay left
09:29
johnjay joined
09:42
johnjay left
09:44
johnjay joined
09:51
johnjay left
09:54
johnjay joined
10:02
johnjay left
10:04
johnjay joined
|
|||
| lizmat | weekly: andrewshitov.com/2026/10/03/using-...in-python/ | 10:11 | |
| notable6 | lizmat, Noted! (weekly) | ||
| lizmat | weekly: andrewshitov.com/2026/10/03/raku-a...ce-part-2/ | ||
| notable6 | lizmat, Noted! (weekly) | ||
|
10:14
johnjay left
10:16
johnjay joined
10:23
johnjay left
10:26
johnjay joined
|
|||
| Voldenet | huh, very unobvious conclusion, but it's possible to run rakupp from rakudo via nativecall too | 10:26 | |
| I can't think of any use case instantly though | 10:32 | ||
|
10:38
johnjay left
10:41
johnjay joined
10:48
johnjay left
10:50
johnjay joined
10:58
johnjay left
11:00
johnjay joined
11:07
johnjay left
11:10
johnjay joined
11:17
johnjay left,
johnjay joined
|
|||
| disbot | <antononcube> I have been asking for this. | 12:17 | |
| <antononcube> Of course we should eat our dog food in a variety of ways. | |||
| <antononcube> For example, the Raku package "Proc::ZMQed" supports ZMQ connections to Raku. | 12:19 | ||
| <antononcube> So, Rakupp can be used in Rakudo via "Proc::ZMQed" and vice versa. At least in principle, I have not tried it yet. | 12:20 | ||
| <antononcube> This works: use Proc::ZMQed::Raku; my Proc::ZMQed::Raku $rakuProc; $rakuProc .= new(url => 'tcp://127.0.0.1', port => '5556', cli-name => $*HOME.add(<.local bin rakupp>).Str); $rakuProc.start-proc(setup-lines => Empty, :!proclaim); say '$*RAKU.compiler.id : ', $rakuProc.evaluate('$*RAKU.compiler.id'); say '$*RAKU.compiler.codename : ', $rakuProc.evaluate('$*RAKU.compiler.codename'); say | 12:47 | ||
| $rakuProc.evaluate('(30,40,2)>>.sqrt'); $rakuProc.terminate; | |||
| <antononcube> Gives: $*RAKU.compiler.id : "5.2.1" $*RAKU.compiler.codename : "Raku++" $(5.477225575051661e0, 6.324555320336759e0, 1.4142135623730951e0) | 12:48 | ||
| Voldenet | well yeah ofc, zmq is over simple tcp | 12:49 | |
| though it's rather heavy-handed solution for something so simple | 12:51 | ||
| disbot | <antononcube> I am not sure what do you mean by "simple". One advantage of using ZMQ is that connection to multiple languages are both possible and easy to setup. | 12:58 | |
| <antononcube> I mentioned this before, in order to compare Rakudo and Rakupp in the "same space" I use Wolfram notebooks. | 13:01 | ||
| <antononcube> In Wolfram notebooks the connection to Rakudo is via ZMQ and the connection to Rakupp is via FFI. | |||
| Voldenet | by simple tcp I meant that it doesn't do a lot of reliability work for you | 13:02 | |
| disbot | <antononcube> Yeah!! Often there are problems with it/ | 13:03 | |
| Voldenet | messages usually will arrive and arrive once, but they can be duplicated or arrive twice for instance | ||
| otoh native call would be 100% reliable | |||
| disbot | <antononcube> Yes, that is some special connection patterns are used when connecting to Jupyter via ZMQ. | 13:04 | |
| <antononcube> I talked to Andrew about this -- you can file an issue at Rakupp's GitHub. | 13:05 | ||
| Voldenet | I've used various patterns of peer-to-peer networking with multinodes at work around 15 years ago and reliability problems were very much impossible to debug :( | ||
| disbot | <antononcube> 💯 | 13:06 | |
| <antononcube> That is why I give up on figuring certain notebook connection problems. | |||
| Voldenet | though back to what I said earlier, I don't know if I have the use case for rakudo -> rakupp nativecall | ||
| disbot | <antononcube> Also, the Raku ZMQ implementation is/was unfinished last time I checked. | ||
| <antononcube> I think that is needed for comparison purposes. But it is going to be an instructive thing to do/have, regardless. | 13:07 | ||
| <antononcube> I might bring this up to today's Raku Study Group meeting. (If it is still on.0 | 13:09 | ||
| <antononcube> Or, if you prefer, I can file one. | 13:18 | ||
| Voldenet | hm, doesn't Net::ZMQ work? | 13:20 | |
| eh, I see some issues so probably not quite | 13:21 | ||
| disbot | <antononcube> It works. I use it all the time, actually. | 13:23 | |
| <antononcube> (Via "Jupyter::Chatbook".) | 13:28 | ||
| Voldenet | ah ok | ||
|
14:00
johnjay left
14:58
mc2 left
16:39
belluzj joined
17:04
Sgeo joined
18:39
belluzj left
18:50
abraxxa joined
20:32
abraxxa left
20:57
gcd joined
|
|||