|
🦋 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 | rakudo/main: 8815167856 | (Elizabeth Mattijsen)++ | src/Perl6/Metamodel/Explaining.nqp Streamline Metamodel::Explaining |
10:50 | |
| rakudo/main: 975b663428 | (Elizabeth Mattijsen)++ | src/Perl6/Metamodel/Finalization.nqp Streamline Metamodel::Finalization |
|||
| roast: dde864e917 | (Elizabeth Mattijsen)++ | S12-introspection/can.t Get rid of warning during testing |
11:42 | ||
| lizmat | odd finding just now: there seems to be a pattern of building the "all_method_table" for classes during core compilation | 12:18 | |
| with a little debug message in all_method_table and invalidate_method_cache, I e.g. get: | |||
| invalidate: Array | |||
| all_method_table: Array (263) | |||
| invalidate: Array | |||
| all_method_table: Array (264) | |||
| invalidate: Array | |||
| all_method_table: Array (264) | |||
| invalidate: Array | |||
| all_method_table: Array (264) | |||
| invalidate: Array | |||
| all_method_table: Array (264) | |||
| this feels *hugely* sub-optimal | 12:19 | ||
| suggestions / comments / reasoning welcome! | |||
| I wonder if it's invalidating the cache for each multi method being added (when it shouldn't, as the only thing in the cache will be the proto) | 12:21 | ||
| also: perhaps .can and .find_method should just use the all_method_table cache, instead of walking the MRO each time | 12:23 | ||
| and e.g. have "add_method" just add to the cache | |||
| vrurg nine jnthn ^^ | 12:34 | ||
| Geth | rakudo/main: 1dfd7b4bc0 | (Elizabeth Mattijsen)++ | src/Perl6/Metamodel/MROBasedMethodDispatch.nqp Streamline Metamodel::MROBasedMethodDispatch There seems to be a pattern of building the "all_method_table" for classes during core compilation. I wonder if it's invalidating the cache for each multi method being added (when it shouldn't, as the only thing in the cache will be the proto). Also: perhaps .can and .find_method should just use the all_method_table cache, instead of walking the MRO each time |
12:51 | |
|
14:13
MasterDuke joined
14:38
sena_kun joined
15:20
sena_kun left
|
|||
| Geth | rakudo/main: 9f0d361b91 | (Elizabeth Mattijsen)++ | src/Perl6/Metamodel/MROBasedTypeChecking.nqp Streamline Metamodel::MROBasedTypeChecking |
16:48 | |
|
18:02
MasterDuke left
18:16
MasterDuke joined
|
|||
| MasterDuke | lizmat: btw, i noticed that the nqp jvm build broke recently, `Can not invoke object '&reverse' in publish_method_cache (gen/jvm/stage1/nqpmo.nqp:1579)` | 18:37 | |
| Geth | nqp/main: e3225825d5 | (Daniel Green)++ | 11 files Unbreak JVM build |
18:53 | |
| releasable6 | Next release in ≈4 days and ≈23 hours. There are no known blockers. Please log your changes in the ChangeLog: github.com/rakudo/rakudo/wiki/ChangeLog-Draft | 19:00 | |
|
19:35
MasterDuke left
|
|||
| Geth | rakudo/main: 80ac433643 | (Elizabeth Mattijsen)++ | src/Perl6/Metamodel/MetaMethodContainer.nqp Streamline Metamodel::MetaMethodContainer Also make sure that adding a meta method, and composing meta methods into the the meta object, are threadsafe. |
19:39 | |
|
20:02
finanalyst joined
21:00
sena_kun joined
22:56
sena_kun left
|
|||