27 Oct 2021
Xliff I am now spending more time in precompilation than I am in compiling new code with the latest changes to rakudo. This really hurts projects with large dependency chains, as you are constantly precompiling compunits -THAT WERE ALREADY PRECOMPILED- seconds ago. 02:47
I have reported this once, and the onus was put upon me to prove the problem exists. I have no idea as to the best way to do that.
But this is the second week I have had to cancel timings due to this bug. All that I can ask is that someone PLEASE look into this. 02:48
It is seriously killing my drive to do things in Raku.
Something that used to take less than 5 seconds now takes over 3 minutes
samebchase Is it just me or are the code blocks looking weird in the Advent blog? raku-advent.blog/2020/12/04/day-4-...-grammars/ 05:35
japhb Xliff: Where's the original issue report?
samebchase Almost, unreadable 😐
japhb samebchase: I don't see unreadable, but I do see a mix between code blocks that look normal-ish, and code blocks that look like they're missing half the style info 05:37
Xliff japhB: It was reported here. Never made into an issue since it was not known how to best describe such a thing.
japhb Xliff: Hmmm. You said there was a regression. When did it last work quickly, and when did it first start being slow? 05:38
samebchase Ah, for me the font weight is so thin, I can barely read it. I'm on a recent FF btw 05:47
Xliff japhb: Last timings were made on the 10th. By the 17th, I could not get isolated precompiles of a single compunit done in most-efficient compile order. 05:49
elcaro I'm also on FF and agree the font weight could be heavier for me 05:53
also font size is inconsistent... small in some, larger in others. 05:54
Xliff japhb: Anecdotal evidence. I just precompiled a compunit. I then make changes in its DESCENDANT, try and compile it, and now I am precompiling the entire chain all over again.
It's like I never spent 327 seconds precompiling the first one.
At any rate, I literally just woke up for a bathroom break. I'm back to bed for another couple of hours.
japhb Ah, sleep well then. :-) 05:55
.tell Xliff If you can write down steps to recreate the timing difference, say between 2021.09 and 2021.10, that would probably be sufficient to start the issue. Next up would be tracing the module precomp for both versions by exporting the environment variable setting RAKUDO_MODULE_DEBUG=1 and rerunning. Then someone could compare what you're getting with what they are getting and start ruling out problems. 05:58
tellable6 japhb, I'll pass your message to Xliff
Xliff japhb: That will involve p6-GtkPlus. Historically, people will slow down tracing issues when I involve any of the p6-GLib software unless it is pre-golfed. I have plenty of log acces, but NOTHING conducive to replication steps. 06:16
tellable6 hey Xliff, you have a message: gist.github.com/08952f482efc489b73...e659d95c83
Xliff Actually, if I do write it up, you all may just have to accept a compile log to see the issue. It's perfectly illustrated there. 06:17
*sigh* -- I think I have a reporting strategy. It will take time to test. 06:23
So much for sleep. :/
Geth examples: terceranexus6++ created pull request #90:
malware analysis example
examples: f3bff1fddd | Paula++ (committed using GitHub Web editor) | categories/other/malware_analysis.raku
malware analysis example

This file, which is also explained in [here](dev.to/terceranexus6/raku-malware-analysis-jon) aims to mimic YARA malware detection standard using grammars. The whole idea is to use RAKU grammars for studying malware genes.
examples: 3694c3b033 | (Juan Julián Merelo Guervós)++ (committed using GitHub Web editor) | categories/other/malware_analysis.raku
Merge pull request #90 from terceranexus6/patch-1

malware analysis example
tbrowder samebchase: it does look inconsistent, but imho that's on wordpress so i'm not surprised--very fragile source-to-desired-html 10:22
Geth ¦ raku.org: c3579d1bc8 | (Tom Browder)++ (committed using GitHub Web editor) | source/community/index.html 10:46
¦ raku.org: Add new entry
¦ raku.org: review: github.com/Raku/raku.org/commit/c3579d1bc8
¦ raku.org: 9ce6e0140c | (Tom Browder)++ (committed using GitHub Web editor) | source/community/index.html 11:01
¦ raku.org: Update index.html
¦ raku.org: review: github.com/Raku/raku.org/commit/9ce6e0140c
¦ raku.org: 3935a8be08 | (Tom Browder)++ (committed using GitHub Web editor) | source/community/index.html 11:04
¦ raku.org: Update index.html
¦ raku.org: review: github.com/Raku/raku.org/commit/3935a8be08
tbrowder .tell melezhik see the Raku.org Community page for the new item (it will probably need some attention to get it as you desire) 11:06
tellable6 tbrowder, I'll pass your message to melezhik
ugexe Xliff: i know whats wrong, no need to come up with a repro 11:43
gfldex CIAvash: sadly add_dispatchee doesn't work with Block, pointy or not. 15:21
[Coke] boingboing.net/2021/10/27/macbook-...ow-is-this - can't remember where we talked about getting new macs once they had enough memory, but the new laptop design appears to have a bug. 15:40