🦋 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.
Geth ¦ problem-solving: 2colours assigned to codesections Issue Handling of non-breaking spaces when splitting to words github.com/Raku/problem-solving/issues/357 00:24
nqp/main: 0724fd6ab1 | (Will Coleda)++ | docs/ops.markdown
these ops are not in js
00:46
Geth rakudo: vrurg++ created pull request #5135:
Implement a reliable approach to experimental RakuAST
05:11
vrurg lizmat: ^^^ 05:12
lizmat: Without this PR something like `say RakuAST::Node` would still work.
Xliff Timings for the p6-GLib suite have been absent for the past few weeks due to an error in the dependency resolution code.This has now been fixed. 08:27
tellable6 2022-12-13T20:09:38Z #raku <vrurg> Xliff make use you have rakudo correctly installed. call.t passes for me.
2022-12-13T20:59:50Z #raku <vrurg> Xliff Correction. For test to work with RAKUDO_RAKUAST env RakuAST must fully or nearly fully complete which is far the case. Tests in 12-rakuast are for already implemented nodes only.
Xliff P6-GLib suite timing statistics for Raku 2022.12.1.gd.52342.eb.0 08:28
Total number of projects: 33 (502427 loc)
Total non-parallel compile times: 8754.460 (265.287 avg)
Total parallel compile times: 1748.927 (52.998 avg) 5.006x speedup
lizmat vrurg: that's untrue: 09:32
say RakuAST::Node'
Must do a 'use experimental :rakuast' to access RakuAST::Node
m: say RakuAST::Node
camelia Could not find symbol '&Node' in 'GLOBAL::RakuAST'
in block <unit> at <tmp> line 1
lizmat ah, looks like camelia is stuck 09:34
probably the master -> main change
m: say RakuAST::Node 10:52
camelia Must do a 'use experimental :rakuast' to access RakuAST::Node
in block <unit> at <tmp> line 1
lizmat m: use experimental :rakuast; say RakuAST::Node
camelia (Node)
lizmat moritz++
vrurg ^^
Xliff Hmmm 12:57
lizmat: Error while importing from 'experimental': no such tag 'rakuast'
Xliff OK, found out I was on the wrong rakudo vversion. Switched to the right one and 12-rakuast is working. 13:23
lizmat Xliff: *phew* 15:33
Xliff P6-GLib suite timing statistics for Raku 2022.12.762.g.7.d.4.c.4.ae.07 15:43
Total parallel compile times: 1631.155 (49.429 avg)
So the latest raku lost almost 120 seconds in parallel seconds to the one from the week before. 15:44
Xliff You can see the raw tracking data here: github.com/Xliff/p6-GLib/blob/main...s?raw=true 15:47
On the sheet "Parallel Timints 2", last 2 columns.
Geth rakudo/main: 7e69a6de5d | (Elizabeth Mattijsen)++ | src/core.c/List.pm6
Make List.head() about 2.4x as fast

Which now makes @foo.head 1.5x as fast as @foo[0], whereas before
  @foo.head was 1.5x as *slow* as @foo[0]
Done by adding a dedicated List.head candidate.
Curious observation: about 15% of performance gained by marking the method as "is raw" and doing the decont explicitely.
19:13
lizmat fg
mmeh 19:14