🦋 Welcome to the MAIN() IRC channel of the Raku Programming Language (raku.org). Log available at irclogs.raku.org/raku/live.html . If you're a beginner, you can also check out the #raku-beginner channel!
Set by lizmat on 6 September 2022.
antononcube Happy Thanksgiving, tbrowder! I am working on making Raku to have vision. 00:21
xinming patrickb: it works now, I tried when I replied to get the error output. probably caching issue. 00:35
tbrowder__ antononcube: & happy thanksgiving to you, too! raku with vision sounds exciting. 01:21
antononcube Hopefully, I will make a post tomorrow... 02:39
nemokosch Curious to see it! 08:41
librasteve o/ 13:32
back on the topic of DateTime::Parse, I have not seen any reply from sergot and there are PRs pending since Oct20, so I think we must accept that it is an orphan module ... please can we adopt it as a raku community module (and then accept the pending PRs)? [failing this, I will fork it and fez under librasteve as author] 13:35
@antononcube - I have looked into DateTime::Grammar (your recent fork) and there are a couple of blockers for me around composability ... so I would like to steer the DateTime::Parser to a better place also 13:38
tbrowder__ tonyo: hi, i'm having one major problem with CSV::Parser. The show-stopper for me is how to use .trim on all columm contents (including in header contents). The test example for the Callable doesn't seem to actually do anything except increment a counter, and I can't get consistent results with my tests. What is the intent of the "header" arg in the Callable test example? Thanks. 13:42
Xliff I'm getting this when attempting to build rakudo. is there a workaround? 13:52
Entry '3rdparty/nqp-configure/doc/NQP-Config.md' not uptodate.
nemokosch is there a particular reason you are still using CPAN? 13:54
teatime .w
Xliff Who? Me? 13:57
App::Mi6
<nemokosch> Also, the reason I want RakuAST to serialize into ByteCode is that I want to run the AST once I've loaded it. I'll take what I can get, though. 13:58
nemokosch App::Mi6 can be used with fez 14:02
Xliff I'll have to figure out how to register with fez, then. Will need some assistance on that. 14:02
I'm fighting various time crunches as it is. 14:03
nemokosch I think it really was just fez register
Xliff I'd rather not worry about that though. I need to rebuild Rakudo and am getting the error:
Entry '3rdparty/nqp-configure/doc/NQP-Config.md' not uptodate.
How can I fix?
nemokosch can't help with that, sorry
Xliff OK, thanks.
antononcube @librasteve Can you list the blockers you have with "DateTime::Grammar" ? 14:32
tbrowder__ .tell tonyo see ref to CSV::Parser above ^^^ 14:44
tellable6 tbrowder__, I'll pass your message to tonyo
librasteve sure (i) you have copied the original DateTime::Parse code as role DateTime::Grammarish which strands all consumers of the original DateTime::Parse module and makes it rather unmaintainable. I sympathize with the desire for DateTime::Parse to be (a) composable role, but I would solve this by your new module being a wrapper that goes 'use DataTime::Parse' and having sergot (or the new agreed main fork/clone) change 16:32
class DateTime::Parse to a role and then stick in some tests so that it won't regress, (ii) you have split the Grammar and Actions and I think that composable Grammar module files should come as the Grammar/Action code together and (iii) you have named the Actions as Actions/Raku.rakumod whereas, in general I would expect a set of sub Grammar roles that consist of Grammar-Action sets in each file and have that named as a
consistent, evolvable module tree eg. imagine /DateTime/Parse/Localized/[RU|JP|US|UK|DE...] or maybe we evolve Parse.rakumod to be a parameterized Role via country code? It looks like you have good reasons for the above (eg. you want to have a structure of Actions to generate raku and python and wolfram code ultimately) but I do not want this feature: the original concept of DateTime::Parse is exactly what I need.
antononcube @librasteve Thank you for your detailed response! 19:08
melezhik o/ 19:53
tellable6 2023-11-23T21:24:52Z #raku <tbrowder__> melezhik thanks, florida, where did you live in us?
melezhik .tell tbrowder__: TX, the woodlands 20:18
tellable6 melezhik, I'll pass your message to tbrowder__
melezhik .tell tonyo: CVS will work in US, lol 😂 20:19
tellable6 melezhik, I'll pass your message to tonyo
avuserow A note from my compilation speed improvement quest, disabling the JIT makes it significantly slower (14.7s disabled vs 11.3s enabled, for the current test) 21:00
MasterDuke avuserow: same results with MVM_JIT_DISABLE or MVM_JIT_EXPR_DISABLE? 21:57
avuserow I only tried both. Let me try them one at a time 21:59
MasterDuke yeah, MVM_JIT_DISABLE is a superset of MVM_JIT_EXPR_DISABLE 22:01
avuserow Disabling the whole JIT makes it 20% slower. The expr part doesn't seem to make a difference. 22:03
MasterDuke not surprised, but thought it might be worth trying 22:05
avuserow I found a good file to optimize though. The one defining all my Cro routes probably could be divided up to avoid some imports. 22:13
If I modify my main App.rakumod file, it takes 4s to load. If I modify App.rakumod and load the Routes file, it takes 15s. 22:15
coleman rebooting docs site. back soon. 22:32
should be back 22:36