🦋 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: ... | log inspection situation still under development | For MoarVM see #moarvm
Set by lizmat on 22 May 2021.
00:02 reportable6 left 00:05 reportable6 joined 00:35 squashable6 left 00:37 squashable6 joined
tbrowder so, gentlemen, new featre 00:40
feature on c, d, or e?
lizmat: ? 00:41
01:13 frost joined 01:38 frost left 02:01 nine left, nine joined
japhb I don't see why we'd put a new feature into c ... it's not even the current release, let alone a future one. 02:02
I guess people find it easier to add to core.c, but if that's actually a problem, we should fix that, because we shouldn't have a builtin incentive to change the past. 02:03
Geth rakudo: Kaiepi++ created pull request #4436:
Handle updates to curried role typecheck lists more lazily
03:03 frost joined 04:17 linkable6 left, evalable6 left, linkable6 joined 04:18 evalable6 joined 06:02 reportable6 left 06:03 reportable6 joined 07:43 releasable6 left, unicodable6 left, statisfiable6 left, bisectable6 left, quotable6 left, committable6 left, benchable6 left, coverable6 left, shareable6 left, nativecallable6 left, bloatable6 left, sourceable6 left, linkable6 left, reportable6 left, tellable6 left, evalable6 left, squashable6 left, greppable6 left, notable6 left 07:44 sourceable6 joined, bloatable6 joined, notable6 joined, evalable6 joined 07:45 tellable6 joined, linkable6 joined, shareable6 joined, quotable6 joined, greppable6 joined, squashable6 joined, nativecallable6 joined, benchable6 joined, committable6 joined 07:46 coverable6 joined, bisectable6 joined, statisfiable6 joined, reportable6 joined, unicodable6 joined, releasable6 joined
gfldex tbrowder: please consider the following case: gist.github.com/gfldex/784bc98adb6...e9d0166a57 08:31
japhb: maybe because Rakudo doesn't make it easy to spectest against .PREVIEW . 08:32
japhb gfldex: Something to fix, I suppose. :-) 08:35
gfldex A CLI option for rakudo to force a language version would be nice. Would make it easy to spectest and CI against multiple language versions. 08:37
raydiak we already support pragmas like strict and isms with -M, perhaps it could be extended to support versions as well 08:47
gfldex It would have to cover the case where a script already requests a version. That wont work with -M. In fact if there is a conflict we normally want a SORRY. 08:49
raydiak how does using -M vs making up some other switch change that complexity? 08:53
gfldex I would prefer a clear statement in wrappers. The intent of --override-version=v6.e.PREVIEW is clear. 08:55
raydiak I guess we're well past trying to parallel p5, but -M is what is used there, fwiw 09:06
either way, if it dies when a different version is requested, "override" isn't exactly accurate. perhaps "force" or so?
09:17 squashable6 left 09:19 squashable6 joined
tbrowder gfldex: i will try to put it in .e 10:07
when or if i get it working in c, then i will move it 10:08
12:03 reportable6 left, reportable6 joined 14:01 frost left
ugexe pretty much an impossible task 16:12
how much of the roast launches a process using `raku ...` 16:13
those don't carry their compiler arguments forward
on a similar note though -M also doesn't work with ver/auth/api 16:14
Remember: "as a developer I want to be able to do thing", not "as a developer we should implement X Y and Z to do thing" 16:20
nine Language versions just do not work exactly like people here seem to think 16:29
17:47 benchable6 left, coverable6 left, linkable6 left, committable6 left, notable6 left, bloatable6 left, tellable6 left, nativecallable6 left, quotable6 left, bisectable6 left, squashable6 left, statisfiable6 left, sourceable6 left, releasable6 left, shareable6 left, reportable6 left, greppable6 left, unicodable6 left, evalable6 left, notable6 joined 17:48 nativecallable6 joined, reportable6 joined, sourceable6 joined, linkable6 joined, tellable6 joined, releasable6 joined, greppable6 joined, quotable6 joined, statisfiable6 joined 17:49 committable6 joined, shareable6 joined, evalable6 joined, squashable6 joined, unicodable6 joined 17:50 coverable6 joined, benchable6 joined, bisectable6 joined, bloatable6 joined 18:02 reportable6 left 18:03 reportable6 joined 18:36 ChanServ sets mode: +o ugexe 18:40 ChanServ sets mode: -o ugexe 18:50 notna joined
Kaiepi t/02-rakudo/15-gh_1202.t failed on me during a test run. passes a second time around 20:16
during which t/09-moar/01-profilers.t failed, but passes when tested individually 20:20
bad update that may not reflect master, mb 20:25
20:39 notna left 22:14 LizBot joined, LizBot left, Geth left 22:15 Geth joined, TempIRCLogger joined
tbrowder i just looked around for any checklist or guidance for devs to follow when submitting code for a future release and didn't find too much help. 22:50
it looks like, considering operators.pm6 and Grammar.pm6, i would dup some parts of the pertinent files in core.c (class decls, etc.) and add just the new parts. 23:11
then add the new symbol to the core.e settings (but NOT the core.c and core.d settings). 23:13
the partial files would go only in the src/core.e dir 23:14
23:15 evalable6 left, linkable6 left
tbrowder does that sound about right? if so, i'll start the move now. 23:15
23:16 linkable6 joined, evalable6 joined
tbrowder *working on the move process 23:16
looking more 23:19
looking more closely, looks like the operators can be new additons only 23:40