|
🦋 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. |
|||
|
00:00
reportable6 left,
reportable6 joined
01:15
uzl[m] joined
01:38
samcv left,
samcv joined
01:45
epony left
03:27
epony joined
03:31
epony left
03:36
epony joined
03:47
epony left
04:13
lucs left,
lucs joined
06:00
reportable6 left
06:02
reportable6 joined
07:39
sena_kun joined
08:54
sena_kun left
09:16
sena_kun joined
09:30
sena_kun left
10:57
NemokoschKiwi joined
12:00
reportable6 left,
reportable6 joined
12:04
NemokoschKiwi left
12:14
NemokoschKiwi joined
|
|||
| lizmat | m: dd enum Foo <A B C> # TIL that an enum definition returns its Map, rather than the type | 12:20 | |
| camelia | Map.new((:A(0),:B(1),:C(2))) | ||
| lizmat | m: dd Foo | ||
| camelia | ===SORRY!=== Error while compiling <tmp> Undeclared name: Foo used at line 1 |
||
| lizmat | m: dd enum Foo <A B C>; dd Foo | ||
| camelia | Map.new((:A(0),:B(1),:C(2))) Foo |
||
| lizmat | this sorta feels wrong ? | 12:26 | |
|
12:27
NemokoschKiwi left
|
|||
| nine | It's certainly surprising | 13:17 | |
|
14:00
MasterDuke joined
|
|||
| MasterDuke | . | 14:01 | |
| tellable6 | 2023-04-13T14:50:49Z #raku-dev <nine> MasterDuke: have you ever gotten somewhere with the migrate_colonpairs madness? | ||
| MasterDuke | nine: don't remember doing anything with migrate_colonpairs, but still working on making the deftrap (e.g., `sub one { 1 }; one;` error only happen for core stuff | 14:02 | |
| i have *something* happening in PERFORM-CHECK, but once i get the *correct* thing to happen i guess i need to not throw the exception in the grammar anymore | 14:05 | ||
|
14:05
japhb joined
|
|||
| [Coke] | (enum) is the surprising part that the Map "leaks" or that they're different, or that we get no indication it's an enum in one or both cases? | 14:21 | |
| [Coke] reads back a few lines and sees lizmat explicitly say what the surprise is, oops. | 14:23 | ||
| nine | MasterDuke: ah, ok. I have started working on that colonpair thing in any case. FWIW it's about {:a:b:c} being equivalent to {:a, :b, :c} | 14:39 | |
|
14:43
Xliff left
17:07
evalable6 left,
linkable6 left
17:08
linkable6 joined,
evalable6 joined
|
|||
| tbrowder__ | hi | 17:26 | |
| having just had trouble with loading a module, i would like to see a mod to Raku so the user can see errors when a module isn't loaded. | 17:28 | ||
| i know one can do the docs thing for each module, but a pragma or a raku run flag or env var would be very handy during development. | 17:30 | ||
| lizmat | what errors do you mean? | ||
| tbrowder__ | hi, liz. i mean something like forgetting a "use lib..." when a module isn't in the zef places, etc. | 17:32 | |
| i get no specific error, just fallout errors like unfound methods which fool me into thinking it's another problem | 17:34 | ||
| lizmat | so you're saying that a "use Foo" doesn't bomb in some cases ? | ||
| tbrowder__ | when it's just a dumb mistake and a missing lib | ||
| yes | |||
| lizmat | I thought it would always bombs? | 17:35 | |
| m: use Foo | |||
| camelia | ===SORRY!=== Error while compiling <tmp> Could not find Foo in: /home/camelia/.raku /home/camelia/rakudo-m-inst-1/share/perl6/site /home/camelia/rakudo-m-inst-1/share/perl6/vendor /home/camelia/rakudo-m-inst-1/share/per… |
||
| lizmat | m: try use Foo; dd "alive" | ||
| camelia | ===SORRY!=== Error while compiling <tmp> Could not find Foo in: /home/camelia/.raku /home/camelia/rakudo-m-inst-1/share/perl6/site /home/camelia/rakudo-m-inst-1/share/perl6/vendor /home/camelia/rakudo-m-inst-1/share/per… |
||
| lizmat | m: BEGIN try use Foo; dd "alive" | ||
| camelia | ===SORRY!=== Error while compiling <tmp> Could not find Foo in: /home/camelia/.raku /home/camelia/rakudo-m-inst-1/share/perl6/site /home/camelia/rakudo-m-inst-1/share/perl6/vendor /home/camelia/rakudo-m-inst-1/share/per… |
||
| tbrowder__ | ok, maybe it's because i'm upgrading a module and it's not seeing the upgraded one. i guess you can't cure stupid ;-( | 17:37 | |
| lizmat | well, if you can produce a golf, than we can look at fixing it :-) | 17:38 | |
| tbrowder__ | no, i know that's the problem. sorry for the false alarm | 17:39 | |
| MasterDuke | in RakuAST, what's the best/right way to say that two compile-time-value's are the same? | 17:44 | |
|
17:45
sena_kun joined
|
|||
| lizmat | how the same would you like to test? | 17:45 | |
| MasterDuke | in this particular case, that the sub i'm calling is the same as the sub of the same name in the core | 17:46 | |
| lizmat | nqp::eqaddr? | ||
| MasterDuke | =:= doesn't seem to work | ||
| yeah, was wondering if that was the right now | 17:47 | ||
| lizmat | then they're not the same? | ||
| MasterDuke | *one | ||
| lizmat | =:= is syntactic sugar for nqp::eqaddr though | ||
| MasterDuke | they are though | ||
| lizmat | if they're not the same with nqp::eqaddr, then they're not the same? decont needed maybe ? | 17:48 | |
| MasterDuke | hm, will give that a try | ||
| getting closer. seems like maybe actually getting the right exception thrown is the next step | 18:00 | ||
|
18:00
reportable6 left,
reportable6 joined
|
|||
| Geth | rakudo/main: b817812bfb | (Elizabeth Mattijsen)++ | 5 files RakuAST: fix some edge cases when setting up $=pod - make sure missing declarand worry is issued only if it is really correct. The grammar can call the "add-trailing" method multiple times, sometimes *before* the declarand is actually set. Introduce a $*DECLARAND-WORRIES hash that keeps potential issues, and issue the worries in there when a declarand is set, and the declarand is ... (13 more lines) |
19:06 | |
|
19:22
NemokoschKiwi joined
19:23
linkable6 left,
linkable6 joined
20:05
sena_kun left
20:06
sena_kun joined
20:24
NemokoschKiwi left
|
|||
| Geth | nqp/configure_warn_p5mod_missing: a895e6029b | (Timo Paulssen)++ | Configure.pl Warn about missing perl5 libs in Configure instead of just crashing On a system where some of the standard library is chopped off into modules, Configure --help would crash with the not-too-scrutible error message that perl5 gives when a module isn't installed, so i'm thinking a friendly message would be nice |
21:17 | |
| nqp: timo++ created pull request #801: Warn about missing perl5 libs in Configure instead of just crashing |
|||
| rakudo/configure_warn_p5mod_missing: 7ee407ebf3 | (Timo Paulssen)++ | Configure.pl Warn about missing perl5 mods instead of crashing There even was already a bit of code in here to get an earlier crash if FindBin, an essential module for this script, is missing, but with this patch there's a nicer error message, and earlier bailout if stuff is missing |
21:19 | ||
| rakudo: timo++ created pull request #5244: Warn about missing perl5 mods instead of crashing |
|||
| timo1 | even put a little comment in the PRs that show the pain that is trying to get Configure to run on systems with this particular oddity | 21:31 | |
|
21:31
timo1 is now known as timo
22:31
sena_kun left
|
|||