🦋 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. |
|||
00:10
finanalyst left
|
|||
Geth | rakudo/lizmat-post2025.01: 82b055b0a1 | (Elizabeth Mattijsen)++ | src/core.c/Date.rakumod Make sure cloning a Date works for subclasses As found in the research for #1238. Also re-worked some of the internals to make it more performant in the new-disp situation (as most of that code was optimized for the pre-new-disp situation). Makes Date creation (except Date.today) between 25% faster (basic cases) and 3.5x as fast (for subclasses of Date) |
14:19 | |
lizmat | on to DateTime :-) | ||
Geth | rakudo/lizmat-post2025.01: aa0a16efbf | (Elizabeth Mattijsen)++ | 2 files Remove SET-DAYCOUNT private method It no longer serves a purpose, as the "daycount" check nowadays depends on it not being 0 (rather than not being nqp::null). Part of the #1238 fixes for DateTime |
15:55 | |
[Coke] | ... I see why you wanted a potentially earlier release. :) | 16:54 | |
lizmat | heh :-) | ||
no worries :-) | |||
[Coke] | lizmat: will the code coverage tools you wrote also work on core? | 17:05 | |
wondering if we could do a spec test run to find things that might be unused. | |||
lizmat | not yet, but it should be possible soon | ||
dinner& | 17:06 | ||
[Coke] | ooooh | ||
lizmat | meh, looks my latest Date changes trigger some dispatch issue: | 17:47 | |
assign requires a concrete object (got a Callable[Numeric] type object instead) | |||
at src/vm/moar/dispatchers.nqp:222 | |||
wtf is a Callable[Numeric] type object ? | |||
Geth | rakudo/lizmat-post2025.01: 4ba95d3fa4 | (Elizabeth Mattijsen)++ | src/core.c/Date.rakumod Fix issue with Date.today and Date.clone That were somehow missed in a previous spectest :-( Date.today called without a :formatter argument, would initialize the &!formatter attribute without a container, which would cause a very obscure issue in the dispatcher when trying to .clone such ... (6 more lines) |
18:44 | |
18:57
finanalyst left
|
|||
ab5tract | lizmat++ | 19:47 | |
That commit sounds like it was a bit of an adventure | |||
lizmat | yeah, it was | 19:51 | |
it would only fail on the Date.today fast path | 19:52 | ||
which would not check values for year / month / day, as we can be sure that nqp::localtime produces valid values | |||
greppable6: BUILDALL | 19:56 | ||
greppable6 | lizmat, 23 lines, 13 modules: gist.github.com/53deece8ab620118ba...b53dec2b84 | 19:57 | |
lizmat | meh | 19:59 | |
20:12
finanalyst joined
|
|||
[Coke] | I see some comments, a jnthn, a zef, ... | 20:13 | |
ugexe | that is searching old code, not just every modules current version | 20:16 | |
for instance there is not a single .rakumod file listed | 20:17 | ||
ab5tract | Following the link for zef, I landed at a copy of that repository inside of lizmat’s raku-all-modules | 20:35 | |
github.com/lizmat/raku-all-modules...on.pm6#L59 | |||
That’s true for a few other links as well. but not all of them. My spot check revealed the OO::Monitors to point to the community module repo | 20:37 | ||
ugexe | that repo hasnt been updated in 4 years | ||
oh nevermind, the link was from an old tree | 20:38 | ||
22:06
coleman left
22:07
coleman joined
|
|||
ab5tract | apparently that’s the repo greppable6 is designed to search in, according to its help page | 22:34 | |
greppable6: help | 22:35 | ||
greppable6 | ab5tract, Like this: greppable6: password # See wiki for more examples: github.com/Raku/whateverable/wiki/Greppable | ||
ab5tract | Seems like this bot could stand for a few more knobs and/or buttons | ||
22:48
coleman left,
coleman joined
|
|||
lizmat | wow, I completely had forgotten about that repo | 22:58 | |
yeah, looks like I inherited that from moritz | 22:59 | ||
my only commit is to adapt the README :-) | |||
releasable6 | Next release in ≈19 hours. There are no known blockers. Please log your changes in the ChangeLog: github.com/rakudo/rakudo/wiki/ChangeLog-Draft | 23:00 | |
lizmat | fwiw, I think a subset of the REA would be a more appropriate haystack | ||
will mull about that :-) | |||
23:48
finanalyst left
|