Geth roast: skids++ created pull request #309:
Unskips and various roast cleanups		 01:00
roast/master: 16 commits pushed by skids++
review: github.com/perl6/roast/compare/bae...bcc599cc1b
AlexDaniel looks at github.com/perl6/roast/commit/a743...24f9ec925f and scratches his head 02:54
that's not the first time the behavior of state vars in code blocks is changed, and again I wonder how many people were relying on this behavior…
.tell samcv Hey. I'm looking at github.com/perl6/roast/commit/640d...082c4b03a5 and I can't quite understand it. To me it seems like the previous test was explicitly testing Malayalam combiners on a random character, but now it's two combiners on nothing. I guess the previous version wasn't really “unanchored” and the current version is, but the way behavior has changed is really confusing. So if 03:15
previously it was possible to match some combiner with <:smth>, now it's not? Is it something we want to change within 6c? (asking because it needs errata change)
yoleaux AlexDaniel: I'll pass your message to samcv.
AlexDaniel .tell samcv previously it was possible to match some combiner with <:smth>, now it's not? Is it something we want to change within 6c? (asking because it needs errata change)
yoleaux AlexDaniel: I'll pass your message to samcv.
samcv hello 03:16
yoleaux 03:15Z <AlexDaniel> samcv: Hey. I'm looking at github.com/perl6/roast/commit/640d...082c4b03a5 and I can't quite understand it. To me it seems like the previous test was explicitly testing Malayalam combiners on a random character, but now it's two combiners on nothing. I guess the previous version wasn't really “unanchored” and the current version is, but the way behavior has changed is really confusing.
So if
03:15Z <AlexDaniel> samcv: previously it was possible to match some combiner with <:smth>, now it's not? Is it something we want to change within 6c? (asking because it needs errata change)
AlexDaniel ha, yoleaux didn't manage to fit the first part in one message
samcv well it should match the base character so i'm not sure why it acted that way before. it changed when i updated the unicode database. so unicode may have changed things 03:17
i mean it doesn't make much sense to match against the combiners, unless the combiners aren't attached to any base character
but it used to test that InMalayam would match if it was a CJK character with a in malayam combining glyph 03:18
which didn't make much sense
i mean it's a pretty old file
m: say "á\c[combining caron]" ~~ /<:Mn>/ 03:19
camelia Nil
samcv like this doesn't match
even though combining caron is Mn
m: say "á\c[combining caron]" ~~ /<:InCombining_Diacritical_Marks>/ 03:20
camelia Nil
samcv this doesn't match either
neither did "á\c[combining caron]" ~~ /<:InCombining_Diacritical_Marks>/ match in older versions
i have no clue why that test passed before. but in no other cases does it work like that to my knowledge. it seems like it was a mistake 03:21
that just happened to work maybe
AlexDaniel indeed 03:22
samcv i'm guessing
i fixed it when i made property values distinct per property :P
AlexDaniel m: say "\c[combining caron]" ~~ /<:Mn>/
camelia ｢̌｣
samcv since the block didn't change at all
AlexDaniel m: say "\c[combining caron]" ~~ /<:InCombining_Diacritical_Marks>/
camelia ｢̌｣
Geth roast/6.c-errata: 592a8d8c47 | (Samantha McVey)++ (committed by Aleks-Daniel Jakimenko-Aleksejev) | S05-mass/properties-block.t
Fix InMalayalam test

Previously this test was a CJK base with a Malayalam extending character. Replace it so the base character is also Malayalam.
IRC discussion: irclog.perlgeek.de/perl6-dev/2017-...i_15144609		 03:27
AlexDaniel .tell Zoffix toaster plz 03:29
yoleaux AlexDaniel: I'll pass your message to Zoffix.
samcv i think i almost have fixed the problem where prepend doesn't work right. like "\c[arabic number sign]1" ~~ /:m 1/ 03:38
nice
yay 03:41
AlexDaniel is thinking that unicode support in rakudo would've been garbage without samcv 03:42
samcv :P
it was pretty good, now it's just more better 03:43
thanks for the compliment though :)
if all the codepoints are prepend i'll set it to 0 (the index of the base character) otherwise it's whatever codepoint is after the last prepend in the synthetic 03:48
u: { .uniprop('Prepend') } 04:07
unicodable6 samcv, U+0000 <control-0000> [Cc] (control character) 04:08
samcv, U+0001 <control-0001> [Cc] (control character)
samcv, Cowardly refusing to gist more than 5000 lines
samcv u: { .uniprop-bool('Prepend') }
unicodable6 samcv, U+0001 <control-0001> [Cc] (control character)
samcv, U+0000 <control-0000> [Cc] (control character)
samcv, Cowardly refusing to gist more than 5000 lines
ugexe m: while note ++$ { my $a := gather { loop { take 1 } }; for 0 { await start { $a[0] }, start { $a[0] } } } 04:50
camelia (signal SEGV)1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
…
ugexe m: while note ++$ { my $a := gather { loop { take 1 } }; await start { $a[0] }, start { $a[0] } } 04:52
camelia 1
Tried to get the result of a broken Promise
in block <unit> at <tmp> line 1

Original exception:
Cannot call method 'push' on a null object
in block at <tmp> line 1
in block at <tmp> line 1
ugexe debug symbols + gdb show the segfault at a few different array access points 05:02
MasterDuke i just ran it with perl6-gdb-m a couple times and it always pointed to src/6model/reprs/VMArray.c:148 05:03
ugexe yeah thats one. i've also gotten src/core/compunit.h:9, src/gc/wb.h:8, and src/6model/reprs/MVMMultiCache.c:418 05:06
MasterDuke oh, just got MVMSpeshLog.c:54 05:07
just did about 10 runs with MVM_SPESH_DISABLE=1, always src/6model/reprs/VMArray.c:148 05:09
[Tux] This is Rakudo version 2017.08-106-g7da0c2159 built on MoarVM version 2017.08.1-150-g0b81969d 06:23
csv-ip5xs 1.429 - 1.530
test 10.015 - 10.187
test-t 3.487 - 3.624
csv-parser 11.059 - 11.462
lizmat Files=1223, Tests=67636, 291 wallclock secs (11.13 usr 4.54 sys + 1971.26 cusr 207.84 csys = 2194.77 CPU) 06:56
Geth rakudo/nom: 0ca5ffa408 | (Elizabeth Mattijsen)++ | src/Perl6/Metamodel/BUILDPLAN.nqp
Use more idiomatic push rather than @a[+@a]

Should be faster as well, although that doesn't matter much at compile time.		 07:03
masak ok -- hi -- so I have another one. just golfed it. 07:24
m: constant C = hash(D => sub () { say "OH HAI" }); C<D>()
camelia OH HAI
masak if you put the `constant C = hash(D => sub () { say "OH HAI" })` bit in a module and import the module, you get an error.
`Cannot invoke this object (REPR: Null; VMNull)`
something BC-related, p'haps? 07:25
jnthn: ping ^
nine timotimo: Thinking about your "i find it bad enough that we start the compiler at run time when we use nativecall", that's only necessary because when the native trait is applied, we may not yet know about the returns trait. We only need to be able to run code _after_ the sub is fully declared. 07:27
masak m: constant C = sub () { say "OH HAI" }; (C)() 07:28
camelia OH HAI
masak shorter golf ^
"[BUG] Subs in constants don't survive through module boundary in Rakudo" <-- if I were in a position to submit this one to RT 07:32
geekosaur is it that, or compile vs. runtime fun? 08:10
hm, on second thought I suspect this is SC related 08:11
lizmat good *, #perl6-dev! 09:07
timotimo: what is the way to trigger the BUILDALLPLAN 12 case: container_initializer ? 09:08
timotimo: apparently, that is being set if %cont_info<build_ast> is set, by 09:09
but I can't find where *that* is set, so I wonder if this is a piece of dead code >
?
Skarsnik releasable6, status 09:13
releasable6 Skarsnik, Next release in 5 days and ≈9 hours. 1 blocker. Changelog for this release was not started yet
Skarsnik, Details: gist.github.com/5059886f6536c810d0...733e84f670
Geth rakudo: Gnouc++ created pull request #1154:
Supress line number if throw X::Package::Stubbed		 09:38
roast: Gnouc++ created pull request #310:
Add test for stubbed class		 09:39
Skarsnik Who is in charge of this release ? x⁾ 09:45
and how did I do this fancy parenthesis xD 09:46
geekosaur ⁾ for me is <compose> ^ ) 09:47
Skarsnik oh it's ^ + ) for me 09:51
interesting, ^ is a composer for stuff
geekosaur "dead key" 09:52
in X11 parlance at least
Skarsnik can only do ⁼⁾⁻⁽ and 2 others that does not show on this font/os 09:53
m: say 2² 09:56
camelia 4
lizmat timotimo: found it (re container_initializer) 10:02
Geth rakudo/nom: 760530a524 | (Elizabeth Mattijsen)++ | 2 files
Remove action 13 from BUILDALLPLAN

In BUILDALL this is a noop, so don't put it in the BUILDALLPLAN to begin with. And since we now removed that, we don't need to support this in Mu.BUILDALL either.
This should have some effect minor effect on memory usage, and maybe some noticeable effect in build times of objects with attributes that don't need initialization.		 10:09
jnthn masak: fwiw, I think that one already is in RT 10:30
lizmat: I think you got a patch in relating to rt.perl.org/Ticket/Display.html?id=132030 ? 10:58
lunch & 11:00
Skarsnik hm if I understand this fh are not closed when exiting the sub? 11:01
like sub foo { my $fh = open "blabl"} I need to close $fh at the end?
ilmari Skarsnik: yes, because perl6 doesn't use reference counting, so doesn't provide timely destruction 11:03
LEAVE { $fh.close } should do the trick 11:04
Skarsnik but should the gc close the handle when it will free $fh?
ilmari it will, but there's no guarantee as to when that will happen
lizmat jnthn: indeed RT #132030 has been fixed using proposed method one (keeping list of open files) 11:19
synopsebot6 Link: rt.perl.org/rt3/Public/Bug/Display...?id=132030
lizmat jnthn: mind you, not on the JVM, as nqp::filenofh is not implemented there 11:20
nine And cannot be implemented 11:49
At least not without rewriting large parts of I/O
jnthn We didn't implementing buffering on JVM either 12:11
*impleemnt 12:15
So it's not a big deal 12:16
AlexDaniel Skarsnik_: I am, why? 13:01
ilmari speaking of reference counting: xkcd.com/1888/ 16:22
TimToady nine: so maybe that's another good reason to get rid of the returns trait in favor of --> 17:52
yoleaux 8 Sep 2017 15:05Z <b2gills> TimToady: I would like your input on a document for implementors of Perl 6 dialects since there are currently 2 that I know of gist.github.com/b2gills/7e2781dfd7...59bd751115
TimToady b2gills: it is good to have those principles out where people can see them, in short, "Don't ever put the user or the compiler in a situation where they don't know what language they're running." 17:54
nine: or at least restrict nativecall to using -->
of course, we could always just say it's a DIHWIDT if 'native' comes too early too 17:55
nine We could add such a restriction in a NativeCall:api<2> 18:00
Except that use NativeCall; (as documented) doesn't specify an API version at all
Skarsnik I don't like the --> syntax x) 18:01
(yes, very useful input)
nine Skarsnik: well what use is the whole "multiple ways to do things" mantra when one cannot make use of it? 18:02
What we could do however is have the Native trait compile at compile time if the signature is already complete and fall back to runtime if it isn't. Then it's still backwards compatible. 18:03
Skarsnik I must be missing a bit of the conversation, what is the issue with returns ?
hm the code in the native trait is run at compile time. you mean the CALL-ME creation? 18:04
nine Skarsnik: the native call site in MoarVM is created through the nqp::nativecallbuild OP the first time a native sub is called. Doing that at the time the native trait is applied doesn't work in the case of sub foo() is native(Str) returns Int { * } 18:05
In that case the return type is not yet known at the time the native trait is applied.
Skarsnik hoo I though returns was just another way to write --> 18:06
so with --> you know the return type?
nine apparently :) 18:07
Skarsnik dunno when I wrote github.com/rakudo/rakudo/blob/nom/...l.pm6#L234 I remember having the return type at compile time
nine The returns trait just calls $target.signature.set_returns($type) 18:08
Skarsnik m: INIT { say "hello"}; use NativeCall; sub foo is native returns Int {*}; 18:09
camelia hello
Skarsnik that shoud have returned an error, interesting 18:10
m: INIT { say "hello"}; use NativeCall; sub foo is native(---> Int) {*};
camelia 5===SORRY!5=== Error while compiling <tmp>
Prefix - requires an argument, but no valid term found
at <tmp>:1
------> 3}; use NativeCall; sub foo is native(---7⏏5> Int) {*};
expecting any of:
prefix
Skarsnik m: INIT { say "hello"}; use NativeCall; sub foo is native(--> Int) {*};
camelia 5===SORRY!5=== Error while compiling <tmp>
Prefix -- requires an argument, but no valid term found
at <tmp>:1
------> 3"}; use NativeCall; sub foo is native(--7⏏5> Int) {*};
expecting any of:
prefix
Skarsnik duh, it's not the syntax for --> ? 18:11
nine m: INIT { say "hello"}; use NativeCall; sub foo(--> Int) is native {*}; 18:12
camelia Potential difficulties:
The returning type of 'foo' --> Int is erroneous. You should not return a non NativeCall supported type (like Int inplace of int32), truncating errors can appear with different architectures
at <tmp>:1
------> 3…
Skarsnik Oh yeah I am dumb 18:14
geekosaur is native(...) looks in ... for the symbol
kinda hard to distinguish that from a proto
er, a signature
Skarsnik nine, could you fix the test? (04/18-sig) and replace the returns with --> ? 18:15
timotimo nine: is the returns thing in place if you put it before is native perhaps? 18:36
AlexDaniel . 19:08
gfldex is IO::Handle.lines supposed to work on fifos? 19:53
i'm getting Original exception: 19:55
It is not possible to seek this kind of handle
in block <unit> at bin/fifo-test.p6 line 7
timotimo hm, why do we try to seek in there? 19:57
mst presumably because lines does a seek-to-start to get all the lines? 19:58
gfldex it throws after the first input line is delivered
timotimo hm, not sure if it should. i'd expect to be able to read a few bytes and then start consuming lines 19:59
gfldex source: gist.github.com/gfldex/7d89ad2b177...34c57bb6a8 20:00
I can't RT that because RT doesn't like my e-mails anymore
timotimo i suppose just "radio-fifo-in".IO.lines won't do it? 20:01
gfldex I can try 20:02
.emit "name".IO.lines; just quits after reading 2 lines 20:05
.emit for "name".IO.lines; # even
.emit for $fifo-in.lines; # quits after reading just one line 20:08
timotimo huh, that's certainly weird 20:09
i must admit i haven't used fifo ... i think ever?
ugexe maybe github.com/rakudo/rakudo/blob/7605...le.pm#L470 needs to have a conditional instead of a try 20:11
timotimo what would you condite on? 20:12
having an attribute "is this a fifo"?
it shouldn't give you the error though :\
ugexe in moarvm its here github.com/MoarVM/MoarVM/blob/41e1...ile.c#L102 20:13
timotimo is it really this specific sink-all? it shouldn't be sinking at all 20:21
geekosaur really, you should have a fallback for seek failed (terminals/ptys, pipes, fifos, sockets, character special files) 21:02
it's a common thing, ylou should not blindly assume seekable and then come up with ways to detect the exceptions beforehand 21:04
especially since you cannot distinguish a fifo except by moderately tricky open behaviors that require a cooperating second opener *and no non-cooperating ones* which means in practice you can't do those tests. 21:05
hm, actually I guess stat/fstat catches fifos. pipes are harder, ptys vs. ttys/serial ports vs. (say) parallel ports, ... just don't get into that business, it will only make you cry 21:06
lizmat and another Perl 6 Weekly hits the Net: p6weekly.wordpress.com/2017/09/11/...ng-sorted/ 21:18
samcv is there a way to `require` something that hasn't been installed with zef? 21:31
or impossible?
like something in a lib folder for example
geekosaur -Ilib ? 21:38
or use lib 'lib'
Geth roast/6.c-errata: c554b68d92 | (Jonathan Worthington)++ (committed by Aleks-Daniel Jakimenko-Aleksejev) | S04-declarations/state.t
Correct bad state expectation

Due to a compiler bug, the closure here was not being cloned by its outer scope. This caused various nasty thread safety bugs, and has now been rectified. However, this test depended on the buggy behavior.		 21:48
jnthn AlexDaniel++ 21:51
samcv Geth, that won't work 21:57
you can't use lib in modules 21:58
geekosaur oh, that was completion fail. was wondering... 21:59
samcv require works for thinks installed with zef. cause i'm guessing it's installed to compunits or something? 22:00
geekosaur I think if you are using modules "out of place" you're expected to use -I. but I'd have to know more details (and unfortunately would then likely hit the limits of what I know)
docs.perl6.org/language/modules#Finding_Modules ? 22:02
samcv it's just in a folder though 22:03
so the question is to get it to work without having it installed 22:04
geekosaur hm. I thought that worked for random uninstalled modules 22:05
lizmat good night, #perl6-dev! 22:06
geekosaur maybe we really do need the dummy CUR for not-properly-installed stuff. would have to error out on things like :ver and :auth since it can't deal with them, and would probably need to keep its own precomp and precomp on first use
samcv github.com/samcv/test
this is what i'm trying
geekosaur (we had someone around the other day who was whiny about not being able to arbitrarily hotpatch any file on the system, including having a rant about OS package managers being evil) 22:07
samcv the require fails in lib/main.pm6
night lizmat !
jnthn samcv: ooc, is it specifically require that's having issues, and use works? 22:09
samcv yes just require
i mean it actually doesn't work when i do use 22:11
because it can't find bar with ::( )
jnthn The require itself doesn't fail though?
samcv but if i comment out that line. and only leave the require. then yes that doesn't work
the require is what fails yes. if i change the rquire for use, then it fails on the line below
timotimo lizmat: i find it strange to have the "ipython perl6 notebook in docker" once as a blog post and once as a tweet
jnthn samcv: ah, I see the problem 22:12
samcv maybe it wants me to refer to it as something other than bar, i don't know? or maybe i need to add module unit or something?
jnthn sub bar ($str) is export {
That's a lexical sub bar being exported
samcv need a diff name?
jnthn I think if you add at the top 22:13
unit module bar;
And then make it our sub bar($str) {
Then it'll work out
::($var)::{"&$var"}($str); is just a package lookup
samcv ok cool
timotimo lizmat++ # weekly again 22:14
samcv hmm jnthn Cannot find method 'Any' on object of type Str 22:16
jnthn hmm
samcv ah it's not getting it back hold on
jnthn, how can i dump all symbols? 22:23
cause ::($var) is an object
i'm just getting any back when i try to do the lookup
timotimo you'll probably want ::.keys or maybe .WHO.keys 22:27
ugexe don't you have to `our` something you want to access with ::name? 22:28
timotimo oh, mhm 22:29
but we also have MY:: for example
and ::('$var') is MY:: i think
samcv not getting anything 22:30
oh wait nvm
::($var)::.WHO.keys.say; i got the sub 22:31
if i put `our` on the sub
jnthn 22:13 < jnthn> And then make it our sub bar($str) { 22:32
;)
samcv yeah i still can't use it though 22:33
::($var)::{"&bar-sub"} this doesn't work
do i need to use some other way to access it?
ah .WHO{ } works 22:34
jnthn ::("$var::&bar-sub") probably works too
samcv now i get undeclared routine bar-sub used 22:41
heh
oh oops that was my bad 22:42
ugexe why not import the symbol? require ::("bar") <&bar> 22:43
samcv yay got it
can <&bar> be a variable? 22:46
i could not get that to work
jnthn <...> is literal syntax 22:47
Would have to be {$name-here} or so
samcv jnthn, that creates a block object 22:48
ugexe m: BEGIN my $foo = q|&bar|; require ::("XXX") << $foo >>; ::("&bar")() 22:51
camelia Could not find XXX at line 0 in:
/home/camelia/.perl6
/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/perl6
CompUnit::Repository::AbsolutePa…
geekosaur samcv, afaik that's just a list of expressions at that point 22:52
although it's entirely possibl nobody implemented or tested the general case; I know various other documented things with respect to importing are missing 22:54
ugexe m: BEGIN my $foo = q|&source-file|; BEGIN "lib/XXX.pm6".IO.spurt(q|unit module XXX; sub source-file is export { $?FILE }|); require ::("XXX") << $foo >>; say ::($foo)() 22:55
camelia Could not find XXX at line 0 in:
/home/camelia/.perl6
/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/perl6
CompUnit::Repository::AbsolutePa…
ugexe m: BEGIN use lib 'lib'; BEGIN my $foo = q|&source-file|; BEGIN "lib/XXX.pm6".IO.spurt(q|unit module XXX; sub source-file is export { $?FILE }|); require ::("XXX") << $foo >>; say ::($foo)()
camelia /home/camelia/lib/XXX.pm6 (XXX)
jnthn samcv: I meant after the ::('module-name').WHO or so 22:56
samcv: But what ugexe is suggested is a better way :) 22:57
*suggesting
samcv ok cool. so use << >>
ugexe a list of strings 22:58
"$foo", "baz"; also works 22:59
samcv that doesn't seem to work 23:00
hm
it says
Use of uninitialized value $subname of type Any in string context.
Methods .^name, .perl, .gist, or .say can be used to stringify it to something meaningful.
in any statement_control:sym<require> at /home/samantha/perl6/share/nqp/lib/Perl6/Grammar.moarvm line 1
ugexe m: BEGIN use lib 'lib'; BEGIN my $foo = q|&source-file|; BEGIN "lib/XXX.pm6".IO.spurt(q|unit module XXX; sub source-file is export { $?FILE }|); require ::("XXX") "$foo"; say ::($foo)()
camelia /home/camelia/lib/XXX.pm6 (XXX)
samcv the valua isn't known at compile time ugexe 23:01
like this github.com/samcv/test/blob/master/lib/main.pm6 23:02
if i do <<$modulename>> then i get a different error during compilation about uninitialized value used in a string context 23:03
but with the comma it seems to be doing the correct thing but it can't handle it not being defined yet 23:04
jnthn Sleep here; 'night 23:05
timotimo night jnthn 23:06
ugexe yeah kinda weird it can delay knowing the ::("XXX") but not the << $subname >> 23:09
timotimo but the sub is supposed to be installed in the lexical scope and that's compile-time? 23:10
or am i misunderstanding?
ugexe well if `require ::($foo)` can install that name at runtime, why can't subs be imported from it at that time? 23:12
timotimo when you require ::($foo) you can only access its contents indirectly, though? 23:14
the part after that, the names you want to import, is supposed to give you these thigs as lexical symbols
and then how is the code supposed to use it :) 23:15
require ::($mymodule) (prompt "which sub to take?"); i-guess-this-sub-maybe-exists-now(); # how is this supposed to work?
ugexe ah yeah 23:17
m: my $foo = q|&source-file|; BEGIN "lib/XXX.pm6".IO.spurt(q|unit module XXX; our sub source-file is export { $?FILE }|); require ::("XXX"); say ::("XXX::$foo")() 23:21
camelia Could not find XXX at line 0 in:
/home/camelia/.perl6
/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/perl6
CompUnit::Repository::AbsolutePa…
ugexe m: use lib 'lib'; my $foo = q|&source-file|; BEGIN "lib/XXX.pm6".IO.spurt(q|unit module XXX; our sub source-file is export { $?FILE }|); require ::("XXX"); say ::("XXX::$foo")() 23:22
camelia /home/camelia/lib/XXX.pm6 (XXX)
timotimo heh. 23:23
ugexe m: my $foo = q|&source-file|; BEGIN "lib/XXX.pm6".IO.spurt(q|unit module XXX; our sub source-file is export { $?FILE }|); require "/home/camelia/lib/XXX.pm6"; say ::("XXX::$foo")() 23:24
camelia /home/camelia/lib/XXX.pm6
timotimo i'm a little surprised you are able to use IO at begin time even though restricted setting :) 23:27
geekosaur restricted setting isn't very 23:31
timotimo quite true 23:32
Geth roast: skids++ created pull request #311:
Add tests for RT#129008		 23:50
synopsebot6 Link: rt.perl.org/rt3/Public/Bug/Display...?id=129008
Geth roast: 9a09b4ee15 | skids++ | S02-literals/pairs.t
Add tests for RT#129008
roast: d762a6f017 | skids++ (committed using GitHub Web editor) | S02-literals/pairs.t
Merge pull request #311 from skids/rt129008

Add tests for RT#129008
synopsebot6 Link: rt.perl.org/rt3/Public/Bug/Display...?id=129008
synopsebot6 Link: rt.perl.org/rt3/Public/Bug/Display...?id=129008

Powered by Cro and the Raku® Programming Language.

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

Thank you!