Xliff \o 01:14
I know its late. Has there been any progress of getting Raku to pre-compile in parallel?
I still haven't gotten back to my work on removing the .precomp lock. So busy with other things. 01:15
I hope to get back to it before May ends.
jjmerelo Morning 06:19
AlexDaniel o/ 06:47
Geth rakudo: 5644c709a7 | (Jeremy Studer)++ (committed by Aleks-Daniel Jakimenko-Aleksejev) | src/core.c/Kernel.pm6
Simplify Kernel signals method using Signal values

The Kernel signals method previously had a lot of logic for determining signal values on the host system and backends. Now that that logic is encapsulated into the getsignals ops, this method can be greatly simplified.
Closes #1901.		 06:52
AlexDaniel I'd love to know why github doesn't know how to merge something even if locally there's no problem at all 06:54
is it using some different merge strategy or something?
[Tux] Rakudo version 2020.02.1-388-g3d2522eed - MoarVM version 2020.02.1-163-g8b109c68b
csv-ip5xs0.831 - 0.882
csv-ip5xs-207.818 - 9.508
csv-parser23.511 - 26.452
csv-test-xs-200.381 - 0.406
test7.746 - 7.786
test-t1.980 - 2.021
test-t --race0.851 - 0.969
test-t-2031.823 - 33.146
test-t-20 --race9.780 - 9.930
 06:58
MasterDuke didn't git change its default merge strategy recently? maybe github just hasn't caught up yet 07:01
Geth rakudo: 89e570c2b0 | cygx++ (committed by Aleks-Daniel Jakimenko-Aleksejev) | lib/NativeCall.rakumod
Let NC map int to longlong

While `int` in nativecall declarations might be somewhat confusing, mapping it to the same thing as p6-land native `int` would be consistent - in normal situation, `int` would just be an alias for
  `int64` and `num` for `num64`.
On 'mainstream' architectures, this patch should be enough to do it, possibly there will be a better solution in the future.
Closes #572.		 07:10
MasterDuke how do you view conflicts on github? 07:12
AlexDaniel MasterDuke: hahahahahahah 07:14
MasterDuke: most of the time you can't 07:15
MasterDuke huh
AlexDaniel but sometimes it even lets you fix conflicts on github via their web ui
jjmerelo MasterDuke only when the file name has not changed 07:16
MasterDuke oh, that's probably the only conflict with github.com/rakudo/rakudo/pull/1791 then 07:17
jjmerelo MasterDuke right. core -> core.c
Also, AppVeyor fails in many outstanding PRs. Should we pay any attention to it at all? 07:18
MasterDuke anyone mind if i do the merge manually then? and change the one test description, or is that bad form?
jjmerelo MasterDuke I don't mind... It's probably the best to save that PR. 07:19
AlexDaniel MasterDuke: lately I tend to do `wget github.com/rakudo/rakudo/pull/1791.patch` then git am -3 < 1791.patch 07:22
MasterDuke i just ran the command they showed 07:23
AlexDaniel honestly I don't care as long as you close these PRs 07:24
:)
MasterDuke the only difference is you're keeping the patch around?
AlexDaniel MasterDuke: well, this way I can easily just change the patch if it doesn't apply cleanly, and I'm also not merging a branch from two years back (so git log should remain pretty) 07:25
but there are many ways to do it, as long as you feel comfortable it'll be ok 07:26
MasterDuke: also, whatever you do it'll likely not close the original PR, so include `Closes #…` in your commit message or close it manually 07:31
MasterDuke k
jjmerelo AlexDaniel++ with the issue closing. 07:37
I mean, some are so simple that I don't really understand why they remained there for so long...
Geth rakudo: b02a741398 | vendethiel++ (committed by Daniel Green) | 2 files
GH#1758: Only suggest a submethod if it's local

This avoids us suggesting a submethod from a parent class, which would always result into an error.		 07:41
rakudo: cc10570da5 | (Daniel Green)++ | 2 files
Merge branch 'vendethiel-bugfix/gh-1758'

Fixes #1873 and closes #1791
Geth rakudo/sumanstats-master: 9a0aa98871 | (JJ Merelo)++ | README.md
Fixes conflict
 08:03
rakudo: JJ++ created pull request #3662:
Changes to README.md		 08:09
lizmat Files=1305, Tests=111241, 213 wallclock secs (28.94 usr 8.29 sys + 3002.44 cusr 276.02 csys = 3315.69 CPU) 08:10
Geth rakudo: 9a0aa98871 | (JJ Merelo)++ | README.md
Fixes conflict
rakudo: 0a3f4f95f6 | (JJ Merelo)++ | README.md
Small changes to README.md, fixes conflict

Closes #3125
rakudo: 2c9168b9ad | (Juan Julián Merelo Guervós)++ (committed using GitHub Web editor) | README.md
Merge pull request #3662 from JJ/master

Changes to README.md
AlexDaniel lizmat: take a look at github.com/rakudo/rakudo/pulls?q=i...dated-desc :) 08:12
lizmat: maybe there's something that is a LHF for you
lizmat are we ok to commit wrt the release ?
AlexDaniel lizmat: too late, we're already committing like crazy 08:13
lizmat *sigh*
AlexDaniel so releasing from a branch is the way
lizmat: come on, it's a rare occurance that people get motivated to work on old PRs 08:14
lizmat & 08:15
lizmat starts working on the RWN first and hopes to be able to mention a new release in it 08:50
lizmat m: my @a = ^5; @a[2]:delete; my @b is default(42) = @a.reverse; dd @b[2] # should this be Any or 42 09:08
camelia Any @b = Any
lizmat m: my @a = ^5; @a[2]:delete; dd @a.reverse 09:09
camelia (4, 3, Any, 1, 0).Seq
lizmat m: my @a = ^5; @a[2]:delete; $_ = 42 for @a.reverse; dd @a
camelia Array @a = [42, 42, 42, 42, 42]
lizmat seeing this, I'm starting to lean towards .reverse always deconting and returning Nil for holes 09:10
jnthn Why would reverse decont? 09:21
A deleted slot is ordinarily assignable
Not sure about the wisdom of 89e570c2b0 09:25
AlexDaniel m: use NativeCall; say nativesizeof(int) 09:31
camelia 8
AlexDaniel jnthn: see github.com/rakudo/rakudo/pull/572
IMO it's wrong either way
lizmat 89e570c2b0 09:32
AlexDaniel but now it's right for more people, I guess
lizmat: the bots are down, I'm working on it
lizmat ok
jnthn Well, the point of the original choice was that `int` was likely to get you what `int` means in C, for those just naively translitting signatures. 09:34
lizmat jnthn: in the example, @a contains an nqp::null. normally, AT-POS would return a container for it so you can assign into the array. Because .reverse does that now as well, the nqp::null does not get bound to @b, so its observed value becomes Any, rather than 42
jnthn e.g. a "pit of success" thing
lizmat if you think the current behaviour is ok, then we're fine
jnthn lizmat: ah, hm, it's kinda more complicated 09:35
AlexDaniel jnthn: if you feel strongly about it we can revert it, but IIRC you should never use `int` in signatures anyway
which also kinda means that you can never get your signatures right if you're dev-ing for multiple platforms…
that's something I stumbled upon years ago, maybe there were improvements for that meanwhile, I don't know
jnthn The "can never use `int` in signatures" was also never something I was that thrilled about, for the very reason you mention. The original idea was that int mapped to the current platform's notion of int. 09:36
AlexDaniel jnthn: what about nativesizeof(int) returning 8? 09:39
jnthn That's inconsistent, and probably what should be fixed.
AlexDaniel jnthn: okay, so what about `my int @a`, should that give you an array of 32-bit ints? 09:41
AlexDaniel 89e570c2b0 09:47
linkable6 (2015-11-04) github.com/rakudo/rakudo/commit/89e570c2b0 Let NC map int to longlong
AlexDaniel lizmat: fixed, thanks!
lizmat yw :-) 09:48
jnthn AlexDaniel: No, because that's nothing to do with nativecall 09:52
CArray[int], on the other hand, probably should. 09:53
AlexDaniel jnthn: in the PR people were making a point that maybe we should have a “cint” which can deconfuse things a little bit 09:55
jnthn: anyway, can you revert it then?
jnthn Um, well, my expectation is that people who merge PRs take responsibility for that action. 10:06
Thus why I only merge those I've had time to carefully look through.
And feel comfortable about.
AlexDaniel no problem, I can revert too 10:07
jnthn Thanks
I finally have a week with a load of time to work on implementing the MoarVM dispatcher stuff, and waking up to a flood of emails about PRs that I maybe should pay attention to was like...urgh. :) 10:08
I mean, I kinda appreciate the enthusiasm, but I worry it'll lead to hasty/bad decisions.
Geth rakudo: c362bd8295 | (Aleks-Daniel Jakimenko-Aleksejev)++ | lib/NativeCall.rakumod
Revert "Let NC map int to longlong"

This reverts commit 89e570c2b0bdec925aa758cca35bbebe6b0811f1.
See the discussion on
  colabti.org/irclogger/irclogger_lo...05-04#l146
Basically, we'll probably need a problem-solving ticket to figure out what `int` should mean in different contexts and whether we need something like `cint` to express what `int` means in C.		 10:09
AlexDaniel jnthn: ↑ I hope this makes sense
jnthn Yes, that summarizes it well. Thanks.
MasterDuke jnthn: btw, did you mean to comment on github.com/rakudo/rakudo/pull/2737 ? 10:10
jjmerelo ... and we're down to 90 PRs :-) github.com/rakudo/rakudo/pulls Thanks everyone!
jnthn MasterDuke: haha, what happened there? I was just looking for a commit to link to... 10:11
MasterDuke heh
jjmerelo jnthn MasterDuke ran the benchmarks, it was no longer needed, it was closed 10:12
jnthn Yes, but since it was almost certainly changes I did a month or so back that led to the improvement, I thought I'd link them. :) Done now. 10:13
Geth ¦ problem-solving: AlexDaniel assigned to jnthn Issue What should be the size of `int`? (and other NativeCall type issues) github.com/Raku/problem-solving/issues/182 10:14
AlexDaniel jnthn: ↑ also this 10:15
jnthn Thanks.
AlexDaniel jnthn: thanks for the help, btw 10:16
Geth ¦ problem-solving: AlexDaniel assigned to jnthn Issue What should unicode subscripts do? (and should they do anything at all?) github.com/Raku/problem-solving/issues/183 10:24
AlexDaniel here's an interesting aspect regarding permissions: github.com/Raku/problem-solving/is...-623387149 10:35
jnthn: ↑ take a look perhaps
feel free to slap me if what I'm doing is wrong 10:36
it looks like this: github.com/rakudo/rakudo/settings/access
nine Oh, speaking of, AlexDaniel could you please give me admin permissions to rakudo/rakudo? Can't remember any specifics right now but from time to time I stumble over not having them 10:40
AlexDaniel nine: done 10:43
nine thanks :)
AlexDaniel github is now adding more levels so maybe eventually it'll be more fine-grained
Geth rakudo/master: 5 commits pushed by ribbon-otter++, (Juan Julián Merelo Guervós)++
AlexDaniel but webhooks and stuff is not available for the `Maintain` level and there's nothing in between
Geth ¦ problem-solving: AlexDaniel assigned to jnthn Issue Should ¿? and ¡! be an alternative to "" ? github.com/Raku/problem-solving/issues/184 11:23
AlexDaniel m: CATCH { when X::Y {} } 11:35
camelia 5===SORRY!5===
Function 'X::Y' needs parens to avoid gobbling block (or perhaps it's a class that's not declared or available in this scope?)
at <tmp>:1
------> 3CATCH { when X::Y {}7⏏5 }
Missing block (apparently claimed by 'X::Y')…
Geth rakudo: dac68b27b3 | ribbon-otter++ | CONTRIBUTING.md
Replace `RT #12345` with `#12345`. Explain `RT`

The previous version caused me to think that `RT` was just the prefix that goes before issue numbers. I only learned the true meaning from when zoffixznet kindly informed me of it. Hopefully, changing the examples and adding an explanation will prevent future misunderstandings of this sort.		 11:39
rakudo: 10a8f6a20f | (Juan Julián Merelo Guervós)++ (committed using GitHub Web editor) | CONTRIBUTING.md
Fix links

And update some text in the process.
rakudo: d6576888a1 | (Juan Julián Merelo Guervós)++ (committed using GitHub Web editor) | CONTRIBUTING.md
Merge pull request #2226 from ribbon-otter/contributing-fix

Update CONTRIBUTING.md with regards to prefix `RT`
jjmerelo Open PRs in rakudo/rakudo is down roughly 20% since yesterday. Can we clear the backlog? Yes we can! (y) <3 12:00
(at least a tiny bit)
lizmat notable6: weekly 12:07
notable6 lizmat, 3 notes: gist.github.com/ce722ac87f7bee6854...a50c6824eb
lizmat notable6: weekly reset 12:10
notable6 lizmat, Moved existing notes to “weekly_2020-05-04T12:10:10Z”
[Coke] I have received JJ Merelo's CLA. 12:40
AlexDaniel [Coke]: thank you 12:47
jnthn: can you add jmerelo then? ↑
jnthn: github handle is @JJ 12:48
jjmerelo (y) <3
lizmat and yet another Rakudo Weekly News hits the Net: rakudoweekly.blog/2020/05/04/2020-...community/ 12:56
[Tux] nothing on the slowdown in this weekly? 13:03
lizmat first item in core developments 13:10
AlexDaniel nine: so regarding your comment wrt symlinks, did you delete it? It left me and lizmat a bit confused. Do I understand it right that there's no issue with symlinks because .f will actually be True anyway?
nine: I have to admit that for a moment it made me go “oh shit” too :) 13:11
[Tux] Ah, I didn't make the link 13:13
nine Oh, sorry about that. That comment was plain wrong
lizmat *phew*
nine I did test it, but of course the symlink I tested with was pointing at a directory :) 13:14
AlexDaniel nine: ok! Good attention to detail anyway, thanks! 13:15
Geth rakudo: 4278e4c28f | (Ray Perry)++ | 2 files
Issue #2092 | Clarify detached method error messaging.

Originally, this error said "only alphabetic methods may be detached" but those aren't the only methods that can be detached. The new messaging is a bit clearer on valid detachment candidates.		 13:23
rakudo: ee22c2ef48 | (Aleks-Daniel Jakimenko-Aleksejev)++ (committed using GitHub Web editor) | 2 files
Merge pull request #3204 from RayMPerry/enhancement/2092-detached-method-error-messaging
linkable6 RAKUDO#2092 [open]: github.com/rakudo/rakudo/issues/2092 [Hacktoberfest][LTA][easy to resolve][good first issue] "only alpha methods may be detached" message could be a bit clearer/accurater
lizmat afk for a few hours& 13:28
AlexDaniel > I can't imagine contemplating such an invasive refactoring inside MoarVM as proposed here unless you had stellar certainty the finished product would be rock solid. 13:38
interesting comment
MasterDuke it's true though. and it makes experimenting a lot more fun (especially since my new cpu runs a spectest so much faster), since i'm assured of getting good feedback on my changes 13:41
jnthn Between all the tests + being able to run them with spesh stressing + blin, we're in a fairly decent place to detect regressions. 13:42
sena_kun AlexDaniel, link?
AlexDaniel sena_kun: www.reddit.com/r/rakulang/comments...rthington/
jnthn vrurg: Hmm, I'd not noticed until now but when I make a change in NQP, it only seems to recompile stage 2 from the point of the change I made onward. It used to recompile the whole of stage 2 unconditionally, so you could spot if you broke the code-gen... 13:43
nine Well the comment is in the context of having a fantastic test suite 13:44
AlexDaniel btw if anybody feels like helping me with getting bisectable to dive into nqp and moarvm bumps, please don't hesitate :) It'd make hacking on moarvm maybe even more fun
vrurg jnthn: It was intentional implementation but seemingly wrong. Ok, noted. 13:51
jnthn: Just to make sure: its ok to rebuild stage 1 partially but always full rebuild of stage 2. 13:53
nine vrurg: yes
vrurg Heh... nine – thanks!
nine vrurg: the point of the stages is to end up with a compiler that is compiled by itself. Stage 0 is the old compiler, stage 1 is the new compiler compiled with the old compiler, stage 2 is the new compiler compiled by the new compiler.
jnthn Indeed, it's a kind of "I didn't screw up compiling the compiler" validation :) 13:54
tbrowder nine: i have a short test case that fails when using Inline::Perl5. it is using the perl Excel::Writer::XLSX module which a pretty big one.
[Tux] so in fact it is not really a small test case :) 13:55
vrurg It just never crossed my mind that stage 2 is a tests itself.
nine tbrowder: that reminds me.... I think there is a known issue related to AUTOLOAD on that module.
tbrowder well, the script is small, but the perl is deep ;-) 13:56
but i have been able so far to use much of the module. the current problem is in setting format for font name and size. either one causes a core dump. i have tried using the Scalar as first arg but no help. 13:59
nine That does sound like the AUTOLOAD issue 14:01
vrurg jnthn: BTW, does anybody test Comma on macOS? When I'm trying to debug with a breakpoint the thing I get reproducibly is it hits the bp for the first time, then I resume and get 'marking thread 6 unblocked, but its status is already NONE.'
tbrowder i don't see an issue posted about that.
i need to check closed though..l 14:02
nada 14:06
nine I think I've so far gotten two emails and now an IRC message about this ;) 14:09
jnthn vrurg: Hm, that messages comes from the debug server in MoarVM (plus it'd surprised me if it were macOS-specific, though stranger things have happened...) 14:10
vrurg jnthn: I just cannot try it anywhere else yet. 14:11
tbrowder nine: yep, i see autoload being used in the format modue. is there any way to handle that with inline::perl5? 14:13
nine tbrowder: how about a deal? You write a reduced test case that doesn't rely on any installed module and I write a fix? 14:14
vrurg jnthn: And another thing. comma fails to access the marketplace. Don't know if its on purpose, but I think the cause is in the request url it is using: plugins.jetbrains.com/api/search?o...&max=9 – the server replies with empty list and I think this is due to it not knowning about such build 'CT-2020.4'. 14:16
sena_kun vrurg, that's a sad known bug. :S 14:19
tbrowder hm, it sounds like a good deal if i knew how to start. so the perl code has to originate in the test script itself... i'll try to give it a shot. thanks!
vrurg sena_kun: a tolerable one. Wish I have the debugging work as this is the only thing holding me back from buying comma. 14:22
Geth nqp/new-disp: 29f2695491 | (Jonathan Worthington)++ | src/vm/moar/QAST/QASTOperationsMAST.nqp
Basic compilation of new dispatch operation
 14:35
vrurg jnthn++ Yes! :D 14:36
jnthn vrurg: Is the debugging issue reproducible with something publicly available (or that can be privately shared)? 14:40
vrurg jnthn: yes, but I'd have to commit latest changes eventhough they're incomplete. Will submit later today. 14:42
jnthn vrurg: Thanks.
Geth rakudo/awesome-malformed-loops: d3c3e5346e | (Aleks-Daniel Jakimenko-Aleksejev)++ | src/Perl6/Grammar.nqp
Awesomify “malformed loop spec” errors

This is a questionable fix for RT #127857.
Makes each parse of a loop construct 0.04 ms slower. So if you have one hundred ｢loop｣-s in your code, that's already about four milliseconds slower.
Also breaks one 6.c-errata test that checks for “loop spec” text.		 14:48
tbrowder nine: i just noticed if i use the alternative format setting syntax i don't get a core dump. does that still mean an autoload problem? 14:57
i'll show the working syntax in the issue description. 14:58
tbrowder nine: is it ok to put a very stripped down version of Excel::Writer::XLSX in your test dir where the other perl modules are? 15:46
Altai-man_ By the way, can anyone check github.com/rakudo/rakudo/wiki/ChangeLog-Draft for sanity? I am not really sure if changes to e.g. Rakudo::Iterator should be internal or public or if VM.request-garbage-collection is public or not. 15:47
Any suggestions are welcome.
nine tbrowder: if it's stripped down to only the parts really needed to reproduce the issue, then it's certainly welcome to t/lib 15:48
tbrowder okey dokey, thnx 15:49
jjmerelo Altai-man_ it looks fine.
Geth rakudo: d3c3e5346e | (Aleks-Daniel Jakimenko-Aleksejev)++ | src/Perl6/Grammar.nqp
Awesomify “malformed loop spec” errors

This is a questionable fix for RT #127857.
Makes each parse of a loop construct 0.04 ms slower. So if you have one hundred ｢loop｣-s in your code, that's already about four milliseconds slower.
Also breaks one 6.c-errata test that checks for “loop spec” text.		 15:55
rakudo: 1a536ca9db | (Aleks-Daniel Jakimenko-Aleksejev)++ (committed using GitHub Web editor) | src/Perl6/Grammar.nqp
Merge pull request #1188 from rakudo/awesome-malformed-loops
roast/6.d-errata: 1b0b309937 | (Aleks-Daniel Jakimenko-Aleksejev)++ | S32-exceptions/misc.t
More tests for malformed `loop`s

See rakudo/rakudo#1188.		 15:57
linkable6 RAKUDO#1188 [closed]: github.com/rakudo/rakudo/pull/1188 [has conflicts] Awesomify “malformed loop spec” errors
Geth roast/6.c-errata: 2a74a21955 | (Aleks-Daniel Jakimenko-Aleksejev)++ | S32-exceptions/misc.t
Make malformed loop tests less specific

Recent rakudo versions print some extra info on what went wrong.		 15:59
AlexDaniel` whoops, I did it, folks! 16:01
I pushed into 6.d-errata what I meant to push into master
nevermind will fix soon… 16:02
Geth roast: 212042d578 | (Aleks-Daniel Jakimenko-Aleksejev)++ | S32-exceptions/misc.t
More tests for malformed `loop`s

See rakudo/rakudo#1188.
linkable6 RAKUDO#1188 [closed]: github.com/rakudo/rakudo/pull/1188 [has conflicts] Awesomify “malformed loop spec” errors
nine Nothing a force push can't fix... 16:03
Geth roast/6.d-errata: 71ed05cec3 | (Aleks-Daniel Jakimenko-Aleksejev)++ | S32-exceptions/misc.t
Revert "More tests for malformed `loop`s"

This reverts commit 1b0b3099371bbc46ac5fe94e58d974b4129f86e7.
roast/6.d-errata: bc25d4b50f | (Aleks-Daniel Jakimenko-Aleksejev)++ | S32-exceptions/misc.t
Make malformed loop tests less specific

Recent rakudo versions print some extra info on what went wrong.
AlexDaniel` nine: xD
AlexDaniel Altai-man_: btw how do you keep track of the roast version that is supposed to be used for the release? 16:15
I used to like… rely on my bash history… :S /o\ 16:16
Altai-man_ AlexDaniel, I am not sure I'm getting your question. Do you like if there is some cool mechanism for that I should invent and encourage people to use? Nope. I know the commit I tested many times against and that's all. 16:17
s/like/ask/
Engrish is da stronkest today.
AlexDaniel Altai-man_: ah okay
Altai-man_ In the meanwhile, we still have no changelog for moar. 16:18
Also, one who will do it, it will be really helpful to add changes in a branch deriving from github.com/MoarVM/MoarVM/commit/3d...54433d5fb2 16:20
As we don't want to take the last commit in the release.
AlexDaniel who will be doing it, by the way? 16:22
samcv?
Altai-man_ No idea yet, asked in #moarvm this morning and nobody was particularly enthusiastic. OTOH, I'm wrong for doing it on MONDAY, so no surprise people are busy. 16:23
I also suspect I can do it myself with some reviewing.
AlexDaniel jnthn: among all the notification spam, I think PS#184 is a relatively easy opportunity to say “no” :)
linkable6 PS#184 [open]: github.com/Raku/problem-solving/issues/184 [language] Should ¿? and ¡! be an alternative to "" ?
jnthn Indeed :) 16:39
AlexDaniel thanks :) 16:51
Geth nqp/new-disp: e19e513763 | (Jonathan Worthington)++ | t/moar/53-dispatch.t
An initial test for the new dispatchers

Failing at time of commit.		 17:38
nine tbrowder: looks like I've got a fix. If you've got a test, I can commit and push :) 17:40
tbrowder sounds good, but i’m still trimming module. down to .5mb 17:49
it will be a while, dr appt soon 18:19
sena_kun How horrible are perl: warning: Setting locale failed. Falling back to a fallback locale ("en_US.utf8").? 20:04
Geth rakudo/2020.05-release: 2a17339306 | Altai-man++ | 3 files
2020.05 release
 21:17
vrurg .tell jnthn here is detailed description of the problem with debugging in Comma: gist.github.com/vrurg/47a7ea373eb7...5a96c215cd 21:39
tellable6 vrurg, I'll pass your message to jnthn
vrurg sena_kun++! 21:41
tbrowder nine: it's still about 428 mb but it uses no Perl modules outside the core Perl dist. i commented out the one critical one it uses Archive::Zip so it can't produce an Excel file BUT I have a perl script that works with it while the raku script core dumps. 21:45
i can work up a PR from that unless you want me to keep trimming... 21:46
Geth rakudo/2020.05-release: b7f2f46dbc | Altai-man++ | 3 files
Update changelog + announcement

Deliberately not logged:
1d11885 bf2a542 26f4a36 3134ec7 358c5a8 80dde85 a10140f b9105f1 9b66980 feb0367 fa0da6f 65f18a9 18d083e 48d3a7f 4f63764 979d384 a47372a 35ac7c4 8f015aa a4680db a501754 9ebbad3 e8e2d3c 82b3249 ... (6 more lines)		 21:54
rakudo/2020.05-release: 2d13f4ff5d | Altai-man++ | tools/templates/NQP_REVISION
[release] Bump NQP revision to 2020.05
rakudo/2020.05-release: 40a82d8723 | Altai-man++ | VERSION
[release] Bump VERSION to 2020.05
Kaiepi sena_kun++ 22:30
lizmat, is github.com/rakudo/rakudo/pull/3589 good to go now with the release made? 22:34
Geth nqp: 2f5d5b7540 | Altai-man++ | tools/templates/MOAR_REVISION
[release] Bump MoarVM revision to 2020.05
 23:31
nqp: a04f4985fb | Altai-man++ | VERSION
[release] Bump VERSION to 2020.05
nqp: vrurg++ created pull request #625:
Make stage 2 always fully rebuild upon changes in sources		 23:33
rakudo: Altai-man++ created pull request #3663:
2020.05 release		 23:36
rakudo/master: 4 commits pushed by Altai-man++

Powered by Cro and the Raku® Programming Language.

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

Thank you!