🦋 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.
Kaiepi i can't compile CORE.d.setting on the jvm 07:40
# A fatal error has been detected by the Java Runtime Environment:
#
# Internal Error (metaspaceArena.cpp:86), pid=18933, tid=18934
# guarantee(requested_word_size <= chunklevel::MAX_CHUNK_WORD_SIZE) failed: Requested size too large (558709) - max allowed size per allocation is 524288.
#
# JRE version: OpenJDK Runtime Environment (18.0.1+10) (build 18.0.1+10-suse-1.4-x8664)
# Java VM: OpenJDK 64-Bit Server VM (18.0.1+10-suse-1.4-x8664, mixed mode, tiered, compressed oops, compressed class ptrs, g1 gc, linux-amd64)
ah github.com/rakudo/rakudo/issues/4952 07:41
AlexDaniel what's wrong with bisectable 10:26
bisectable6: old=2019.11 say Date.new('2015-11-24').last-date-in-month;
bisectable6 AlexDaniel, Bisecting by exit code (old=2019.11 new=aa5e680). Old exit code: 1
AlexDaniel, bisect log: gist.github.com/9b30d8bd628a42a87d...a1b5804c16 10:27
AlexDaniel, (2020-04-17) github.com/rakudo/rakudo/commit/2f...78951717fe
AlexDaniel 🤷
lizmat AlexDaniel++ 10:40
AlexDaniel I didn't do anything though 🤔 10:41
it just… works now? I don't understand the previous error
it could be that github returned an error while it was trying to gist the output 10:42
and the error was: “The 'For…” the rest we can't see😆
lizmat i guess... anyways, I was able to answer Mohammad's tweet :-) 10:43
and because of that got an idea for another feature in rak 10:44
releasable6 Next release in ≈3 days and ≈7 hours. 1 blocker. Please log your changes in the ChangeLog: github.com/rakudo/rakudo/wiki/ChangeLog-Draft 11:00
Geth rakudo/lizmat-smartcase: 249f2a05cc | (Elizabeth Mattijsen)++ | src/core.c/Str.pm6
Introduce :smartcase named argument

In principle for all string related functions that also have :ignorecase. In this commit, only "starts-with" has been equipped with it. Basically it operates as :ignorecase if the needle does **not** have any uppercase characters. Otherwise it is ignored.
... (5 more lines)		 14:11
rakudo: lizmat++ created pull request #5005:
Introduce :smartcase named argument
AlexDaniel OK I have a request :) 16:30
please remove review requests from me on these two pulls: github.com/rakudo/rakudo/pull/1884 github.com/Raku/raku.org/pull/138 16:31
and perhaps do something with this PR: github.com/Raku/roast/pull/527
reason being is that I keep seeing these PRs on github.com/pulls, but I can do absolutely nothing about them 16:32
I can't unassign myself, I can't work on these PRs
oh, and that too, please: github.com/MoarVM/MoarVM/pull/659 16:33
sena_kun AlexDaniel, done! 16:34
AlexDaniel good, thanks, if you can also remove the “@” sign in this comment it'd be perfect: github.com/Raku/roast/pull/527#iss...-481468718 16:37
it's OK to mention me but that specific mention with @ in a long-standing PR just makes it show up for me indefinitely
sena_kun wow, didn't know that 16:38
done doo. :)
AlexDaniel thank you!
sena_kun AlexDaniel, you're welcome! 16:42
Geth rakudo/lizmat-died-naturally: 376b0a7055 | (Elizabeth Mattijsen)++ | src/main.nqp
s/DIED/DONE/

  vrurg++ for the suggestion		 18:18
Geth rakudo: 4fc3d455b0 | (Elizabeth Mattijsen)++ | lib/CompUnit/Repository/Staging.rakumod
Revert "Add CUR::Staging.candidates method"

This reverts commit 50025c1a48dd194696d275acadb9c7fd5b82f9a2.
In response to #4949, will revisit this at a later time, most likely post RakuAST merge		 18:42
melezhik o/ 18:54
in a light of upcoming release, I let a comment for this issue - github.com/rakudo/rakudo/issues/4278 , but have not received any feedback ... 18:55
github.com/rakudo/rakudo/issues/42...1193486846
lizmat I saw you comment , so I marked issue as open in R3 list , thanks 19:08
Geth rakudo: 621d69e0e2 | (Elizabeth Mattijsen)++ | src/core.c/DateTime.pm6
Make DateTime.ACCEPTS a multi, as it should be

Fixes DateTime(Str).ACCEPTS(True), which was the new reason why the code of github.com/rakudo/rakudo/issues/4278 failed.		 19:15
lizmat jdv: please make sure that github.com/rakudo/rakudo/commit/4fc3d455b0 and github.com/rakudo/rakudo/commit/621d69e0e2 are included in the release 19:21
bisectable6: old=2022.01 "2019.02-06".DateTime 19:27
bisectable6 lizmat, Cannot find revision “2022.01” (did you mean “2022.06”?)
lizmat bisectable6: old=2022.02 "2019.02-06".DateTime
bisectable6 lizmat, On both starting points (old=2022.02 new=621d69e) the exit code is 1 and the output is identical as well
lizmat, Output on both points: «Invalid DateTime string '2019.02-06'; use an ISO 8601 timestamp (yyyy-mm-ddThh:mm:ssZ or yyyy-mm-ddThh:mm:ss+01:00) instead␤ in block <unit> at /tmp/LC090Ph3Ow line 1␤␤»
lizmat bisectable6: old=2022.01 say DateTime.new: "2019.02-06" 19:29
bisectable6 lizmat, Cannot find revision “2022.01” (did you mean “2022.06”?)
lizmat bisectable6: old=2022.02 say DateTime.new: "2019.02-06"
bisectable6 lizmat, On both starting points (old=2022.02 new=621d69e) the exit code is 1 and the output is identical as well
lizmat, Output on both points: «Invalid DateTime string '2019.02-06'; use an ISO 8601 timestamp (yyyy-mm-ddThh:mm:ssZ or yyyy-mm-ddThh:mm:ss+01:00) instead␤ in block <unit> at /tmp/DQDEm8uIZL line 1␤␤»
lizmat bisectable6: old=2022.02 say DateTime.new: "2019-02-06"
bisectable6 lizmat, Bisecting by exit code (old=2022.02 new=621d69e). Old exit code: 1
lizmat, bisect log: gist.github.com/7947f15e8845e2fa75...818bfb9ede 19:30
lizmat, (2022-06-09) github.com/rakudo/rakudo/commit/aa...fc857cc493
jdv lizmat: i mean its in master, no? 19:32
lizmat yes
jdv so very probably:)
haha
lizmat ok... just wanted to be sure :-)
jdv thank you
lizmat ++jdv for doing a release!
Geth Test-Stream: cbc5ed988d | (Elizabeth Mattijsen)++ | 2 files
Remove tests for now for 2022.07+
 19:50
melezhik Hi lizmat - you have some comment on r3 report format in github.com/rakudo/rakudo/issues/42...1195943508 , we can discuss then there or if you want - drop a ticket in r3 repo, anyway I feel like I  need some informal explanation here ... 20:31
lizmat sure, go ahead 20:32
melezhik ok, give me some time ...
lizmat no hurry :-) 20:34
melezhik "[bash: main.build.pre] :: " - r3 tools runs tests as Sparrow tasks, even thought from .spakyci.yaml point of view there are not - github.com/melezhik/r3tool/blob/ma...i.yaml#L14 , so every Sparrow task output line is prepended by some meaningful header which more or less describes a task  , like in case of this code 20:37
github.com/melezhik/r3tool/blob/ma...ml#L13-L19 it's converted into Sparrow task with a header ("bash: main.build.pre") which basically means execute bash code for sparkyci `main.pre.build` section - github.com/melezhik/sparkyci/blob/...uild-hooks , does it help ?
it's possible to change such a header , however I am just explaining why a header appears in output ... 20:38
lizmat Q: who need to look at these reports if something goes wrong? 20:40
melezhik on this - github.com/rakudo/rakudo/issues/42...1195946178 - the same logic applies - one just sees an output from "main.per.build" section , and this section is just a simple Bash code that sets RAKUBIN variable, installs Tomty - underlying test framework and runs tests ... - it's pretty much described here btw - 20:41
github.com/melezhik/r3tool#usage
A: Rakudo developers and GH issues authors (remember r3 runs tests based on Rakudo GH bugs ?) 20:42
lizmat so, humans, right? 20:43
melezhik Now, I am ready to accept that outout format isa bit unusual for the first users, but I can't agree that it is very hard to understand ... Yes, human beings
melezhik first -> first time 20:43
lizmat well, the format is too verbose for first time users, from my perspective: it may be telling all the things it has done, and then *somewhere* in all of those lines, it actually tells what the error was 20:44
it may be clear to you quickly, but it took me at least 30 seconds just to find where to look for the actual error 20:45
and by that time, I'm annoyed and want to do something else :-)
and that means that first time users will never take the trouble of finding out how useful your tool is] 20:47
which is a pity
melezhik I don't know what to say, you are entitled to have this impression ... however on my view you exaggerate here - the report has maybe 15 Iines of output form test preparation section, which I can't find "too verbose", more over - it could be adjusted , every tools has it's first user threshold - well 30 seconds spent for the one time ( you are not 20:51
going to this time again and again one you understand the idea  )  - is not that bad result ...
lizmat "you can never get a second chance to make a first impression" :-) 20:53
melezhik that's all right - I just would not apply your first impression to everyone else , like I said you are entitle to have your own judgement , I just don't see your point of saying that tool is hard to use/read just because you've seen 15 lines of output from test preparation section, moreover this things are well documented and easy to track ... 20:55
lizmat as a core developer / module author, I just get the issues / error reports of the tool 20:56
I'm not interested in the tool itself per se
so maybe the problem is, that *you* created the tests for App::Rak to be helpful 20:57
but for me (at this moment) it just means extra work without any clear benefits
compare this to github.com/rakudo/rakudo/issues/5002 20:58
melezhik I see that , I still don't see the point - the tools btw already revealed some bugs in Rakudo and other raku modules, what other usefulness you need ?
lizmat that shows a clear summary of things that failed, with options to look at old and new output 20:59
melezhik | it just means extra work without any clear benefits - how does that relate to the tool? The tool just found bugs in App-Raku , I don't understand your point ...
lizmat well, it produced an error report that was hard to understand, and that I could not reproduce locally because it is still unclear what and how you're exactly testing 21:00
so to me, at this point, it has been extra work without any clear benefits
melezhik lizmat I am not sure it's productive to continue this discussion, you keep telling me that you dislike tool A just because you like tolls X, Y, Z , it's not productive ... You keep telling me that _you_ find hard it understand the tool report format, after I've tried to help you to understand , and you keep telling about you don't see benefits. 21:02
Well if you get bug reported in you software , what other benefits you need?
I can't and should not course you into anything , I am absolute OK if people like or dislike things I've don't I just don't get when you say some arguments in a scope of your preferences and try to "sell it" as technical arguments, meanwhile I am trying to get a discussion in pure technical scope of view , it's not fair 😉 21:05
lizmat melezhik: please, I appreciate what you're doing 21:06
melezhik course -> coerce , sorry for the typo
lizmat no worries 21:07
vrurg If it's the same tool, I was looking into, then must admit I had some hard time trying to get what it would like to tell me. Due to lack of time I did not try to analyze it more through.
lizmat melezhik: ^^ my point 21:08
melezhik: I'm trying to make you understand that your tools could be even better 21:09
melezhik vrurg what exactly you found hard to understand ? if this the same as Liz, I hope my answers gave you some clarity ...
lizmat they are great at finding issues
melezhik: if you need to teach your people how to read a report 21:10
you've lost 99% of your potential audience already
your tools are not great at *reporting* issues in a way in which a non-proficient reader of the reports, can work with
vrurg melezhik: As I mentioned, I didn't have much time on my hands. So, just looked into, noticed that it mentions DateTime, and then gave up because didn't know what to do with it. 21:12
melezhik lizmat  "I'm trying to make you understand that your tools could be even better" - saying I am not going to use your tools does not help at all , not sure why you convey this as a help, also being categorical in numbers does not help also, indeed it takes opposite effect
lizmat also: I'm getting the impression that you consider my comments as a personal attack
you are not your software 21:13
vrurg lizmat++
melezhik I am consider them as I no nice way to make your point, because they sound way too categorical
vrurg Perhaps what is needed is easy access to the test source code in first place. Then it needs more consideration, what else could be of use. Detailed reports with stack traces, perhaps. 21:14
melezhik lizmat -  am not, however , it does not feel nice as I said you ... 21:15
if you'd try not pose you opinion as everyone's else one, that what made your comments more productive and not categorical, however I feel like you don't bother to covey your comments that way ... 21:18
lizmat melezhik: I'm sorry to hear you feel that way 21:22
I was trying to convey that in its current state, it is not very appealing nor useful to *me* 21:23
and now I'm going to go away from this discussion and have some sleep over it
&
Kaipei i can make it a little bit further into a JVM build than CORE.d.setting, but only on alpine, weirdly 22:47
CompUnit::Repository::FileSystem runs into problems of some sort because of its use of atomics 22:48
in the final installation
(on the counter branch, need to try master)
likes to hang, but i say that because i've seen null exceptions in nqp::atomicload 22:50
originating from there 22:51

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!