Perl 6 language and compiler development 2.0 | Logs at irclog.perlgeek.de/perl6-dev/today | For toolchain/installation stuff see #perl6-toolchain | For MoarVM see #moarvm
Set by moderator on 16 May 2018.
00:10 Kaiepi joined
Zoffix How would that be differentiated from a string '<null>'? 00:18
There's also a PR to address the inconsistency: github.com/MoarVM/MoarVM/pull/767 00:19
.tell lucasb irclog.perlgeek.de/perl6-dev/2018-...i_16173824
yoleaux Zoffix: I'll pass your message to lucasb.
AlexDaniel samcv: “This devolves to O(n️²) time for the hash lookup”
samcv: the description looks O(n) to me 00:20
00:31 shareable6 joined 01:58 ilbot3 joined
moderator Perl 6 language and compiler development 2.0 | Logs at irclog.perlgeek.de/perl6-dev/today | For toolchain/installation stuff see #perl6-toolchain | For MoarVM see #moarvm
AlexDaniel ah, length of strings matters, I see 02:22
right
02:48 ufobat joined
samcv i just fixed an issue that would happen if the hash generated for a string is 0. We check if cached_hash_code is 0 and if it's 0 we generate a hash code from the string 03:23
so if the string happens to have a hash of 0 then we end up generating a hash from the string every single time. now what it does is if the generated hash code is zero, it adds the string's length to the hash (at the very end)
we already mixed in the strings length into the hash generation, but it's still possible for the output to be 0 03:24
(by the end of the hashing function)
AlexDaniel what if the string length is 0?
samcv then it'll have to generate it every time. but that's pretty unlikely
and will be very fast
i could fix that for string of length 0, but then there's a possible overflow 03:25
i'll probably fix that when we move to 64 bit hashes after the release
also did some preliminary work on getting it so the hash value has its own MVM data type. before cached_hash_code was MVMint32, while in our hashing function we used MVMuint32. while uthash.h used `unsigned` 03:26
which kind of perturbed me. but that's fixed now 03:27
though at least x86 seems to set 32 bit ints for 'unsigned' but it's not guarenteed by C
skids hrm that last MoarVM commit gives a gcc error here. 03:40
samcv yep 03:48
fixed now
skids :-)
MasterDuke m: my $a; $a = PseudoStash.new for ^4_747; say $a; say now - INIT now 03:50
camelia PseudoStash.new(($_ => 4746))
0.016728
MasterDuke m: my $a; $a = PseudoStash.new for ^4_747; say $a; say now - INIT now
camelia PseudoStash.new(($_ => 4746))
0.01873318
MasterDuke m: my $a; $a = PseudoStash.new for ^4_747; say $a; say now - INIT now
camelia PseudoStash.new(($_ => 4746))
0.018344
MasterDuke m: my $a; $a = PseudoStash.new for ^4_747; say $a; say now - INIT now
camelia PseudoStash.new(($_ => 4746))
0.0164851
MasterDuke m: my $a; $a = PseudoStash.new for ^4_747; say $a; say now - INIT now
camelia PseudoStash.new(($_ => 4746))
0.01803696
MasterDuke hm. locally it pegs a core and doesn't finish after 10+m every once in a while 03:52
AlexDaniel c: HEAD,HEAD,HEAD,HEAD,HEAD,HEAD,HEAD,HEAD,HEAD,HEAD,HEAD,HEAD,HEAD,HEAD,HEAD,HEAD my $a; $a = PseudoStash.new for ^4_747; say $a 03:53
committable6 AlexDaniel, gist.github.com/fe38a4ac33c60c1ca3...167f1d9663
AlexDaniel c: b8318b8,b8318b8,b8318b8,b8318b8,b8318b8,b8318b8,b8318b8,b8318b8,b8318b8,b8318b8,b8318b8,b8318b8,b8318b8,b8318b8 my $a; $a = PseudoStash.new for ^4_747; say $a
MasterDuke might need a bigger number if camelia is running on a beefier machine than mine 03:54
committable6 AlexDaniel, ¦b8318b8,b8318b8,b8318b8,b8318b8,b8318b8,b8318b8,b8318b8,b8318b8,b8318b8,b8318b8,b8318b8,b8318b8,b8318b8: «PseudoStash.new(($_ => 4746))␤» ¦b8318b8: ««timed out after 10 seconds» «exit signal = SIGHUP (1)»»
AlexDaniel there you have it?
MasterDuke ah
AlexDaniel is it a regression?
c: 2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1 my $a; $a = PseudoStash.new for ^4_747; say $a 03:55
committable6 AlexDaniel, ¦2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1: «PseudoStash.new(($_ => 4746))␤»
AlexDaniel c: 2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1 my $a; $a = PseudoStash.new for ^4_747; say $a
committable6 AlexDaniel, ¦2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1: «PseudoStash.new(($_ => 4746))␤»
AlexDaniel c: 2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1 my $a; $a = PseudoStash.new for ^4_747; say $a
MasterDuke no idea
committable6 AlexDaniel, ¦2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1: «PseudoStash.new(($_ => 4746))␤»
AlexDaniel c: 2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1 my $a; $a = PseudoStash.new for ^4_747; say $a
committable6 AlexDaniel, ¦2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1: «PseudoStash.new(($_ => 4746))␤»
AlexDaniel c: 2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1 my $a; $a = PseudoStash.new for ^4_747; say $a
committable6 AlexDaniel, ¦2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1: «PseudoStash.new(($_ => 4746))␤»
AlexDaniel c: 2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1 my $a; $a = PseudoStash.new for ^4_747; say $a
committable6 AlexDaniel, ¦2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1,2018.04.1: «PseudoStash.new(($_ => 4746))␤»
AlexDaniel I guess it is?
ah no 03:56
I just managed to repro it on 2018.04.1 on #whateverable
we were just a bit unlucky on this channel :)
03:59 Zoffix left 04:20 squashable6 joined
samcv AlexDaniel: so are things all ok then? 04:37
AlexDaniel samcv: if mean the bug mentioned above, then it's still a bug, but I guess not a blocker 04:38
MasterDuke: btw will you submit a ticket?
samcv what is the bug though? or is it just slower than it should be? 04:39
AlexDaniel samcv: btw as for other blockers, R#1832 will be “resolved” for the release by reverting the commit (will do it later today if nobody beats me to it), and I don't know what's up with R#1810 04:40
synopsebot R#1832 [open]: github.com/rakudo/rakudo/issues/1832 [regression][⚠ blocker ⚠] Regression exposed in Crane and other modules
R#1810 [open]: github.com/rakudo/rakudo/issues/1810 [regression][⚠ blocker ⚠] Recent regression "ctxlexpad needs an MVMContext"
AlexDaniel samcv: I'm not sure, it just hangs
samcv oh
AlexDaniel sometimes
I can reproduce it locally fairly easy with a loop 04:41
while perl6 -e 'my $a; $a = PseudoStash.new for ^4_747; say $a'; do :; done
oh, actually 04:43
c: HEAD my $a; $a = PseudoStash.new for ^9999; say $a
committable6 AlexDaniel, ¦HEAD(b8318b8): ««timed out after 10 seconds» «exit signal = SIGHUP (1)»» 04:44
AlexDaniel c: MVM_SPESH_DISABLE=1 HEAD my $a; $a = PseudoStash.new for ^9999; say $a
committable6 AlexDaniel, ¦HEAD(b8318b8): «PseudoStash.new(($_ => 9998))␤»
AlexDaniel MasterDuke: ↑
oh, I guess this way we can even bisect it?
c: 2015.12 my $a; $a = PseudoStash.new for ^9999; say $a
committable6 AlexDaniel, ¦2015.12: «PseudoStash.new(("\\$_" => 9998))␤»
AlexDaniel bisect: my $a; $a = PseudoStash.new for ^9999; say $a
bisectable6 AlexDaniel, Bisecting by exit signal (old=2015.12 new=b8318b8). Old exit signal: 0 (None) 04:45
AlexDaniel will take a lot of time because of timeouts
bisectable6 AlexDaniel, bisect log: gist.github.com/e3cb6a0d63e4d77af2...87d75f5ff1 04:46
AlexDaniel, (2017-08-11) github.com/rakudo/rakudo/commit/76...ed2e2fe011
AlexDaniel c: 76af17a8b1d^ my $a; $a = PseudoStash.new for ^99999; say $a
committable6 AlexDaniel, ¦76af17a8b1d^: «PseudoStash.new(("\\$_" => 99998))␤»
AlexDaniel c: 76af17a8b1d^ my $a; $a = PseudoStash.new for ^999999; say $a
committable6 AlexDaniel, ¦76af17a8b1d^: «PseudoStash.new(("\\$_" => 999998))␤» 04:47
AlexDaniel yep
c: 76af17a8b1d my $a; $a = PseudoStash.new for ^9999; say $a 04:48
committable6 AlexDaniel, ¦76af17a: ««timed out after 10 seconds» «exit signal = SIGHUP (1)»»
samcv AlexDaniel: know how to bind IRC::Client to a specific ip? 04:53
or to force ipv4?
AlexDaniel well, “yes”… 04:54
whateverable does this: github.com/perl6/whateverable/blob...e.pm6#L586
so if you connect to ipv4 address, then it's ipv4…
samcv oh..
how does it validate the ssl then? 04:55
if you give it an ip
AlexDaniel oh whoops
btw there's :$family in IO::Socket::INET, not sure about the Async version 04:56
samcv: as for binding to specific IPs, we have these set of PRs to make it possible: github.com/MoarVM/MoarVM/pull/659 04:59
well, one set of PRs obviously :) 05:00
samcv ah nice
AlexDaniel but again it does not seem to change IO::Socket::Async ? 05:01
either way we should work on merging it at some point
05:03 |Tux| joined 05:04 Tux__ joined
MasterDuke github.com/rakudo/rakudo/issues/1835 05:11
AlexDaniel MasterDuke++ 05:16
|Tux| Rakudo version 2018.04.1-91-gb8318b8ed - MoarVM version 2018.04.1-119-g6bb647d74
csv-ip5xs0.869 - 0.870
csv-ip5xs-208.561 - 8.773
csv-parser24.035 - 24.342
csv-test-xs-200.456 - 0.464
test8.727 - 8.746
test-t2.395 - 2.600
test-t --race0.958 - 0.982
test-t-2042.011 - 42.522
test-t-20 --race14.436 - 14.819
 05:57
Geth nqp: d775c09b1b | usev6++ | 5 files
[JVM] Implement output buffering for file handles
 06:16
06:34 robertle joined 06:55 Ven`` joined 07:01 shareable6 joined
lizmat Files=1239, Tests=76357, 319 wallclock secs (15.65 usr 5.62 sys + 2205.51 cusr 225.38 csys = 2452.16 CPU) 07:15
AlexDaniel`: I did not know that 07:16
07:17 Ven`` joined 08:20 Ven`` joined 08:31 Ven`` joined 08:50 shareable6 joined 09:29 FROGGS joined 11:03 Ven`` joined
Geth rakudo: 8ae82a558b | (Elizabeth Mattijsen)++ | 2 files
Introduce "use p5isms" pragma

When this pragma is activated in a scope, then the errors about using Perl 5 constructs will *not* be thrown. This does *not* do anything else. So "defined" without any parameter will still not check $_: instead it will just complain that calling defined without any parameters will never work. And *that* allows a module to export a "defined" sub that will check
  $_ when called without any parameters.
I realize this is close to the 2018.05 release. If it's too controversial then please remove this before the release.		 11:38
11:52 softmoth joined 12:14 ExtraCrispy joined
lizmat .ask TimToady opinions about github.com/rakudo/rakudo/commit/8ae82a558b ? 12:22
yoleaux lizmat: I'll pass your message to TimToady.
lizmat m: use p5isms; say defined 12:24
camelia 5===SORRY!5=== Error while compiling <tmp>
Argument to "defined" seems to be malformed
at <tmp>:1
------> 3use p5isms; say defined7⏏5<EOL>
lizmat m: use p5isms; say defined;
camelia 5===SORRY!5=== Error while compiling <tmp>
Calling defined() will never work with signature of the proto (Mu, *%)
at <tmp>:1
------> 3use p5isms; say 7⏏5defined;
lizmat m: no p5isms; say defined; 12:25
12:25 Zoffix joined
camelia 5===SORRY!5=== Error while compiling <tmp>
Unsupported use of bare "defined"; in Perl 6 please use .defined if you meant to call it as a method on $_, or use an explicit invocant or argument, or use &defined to refer to the function as a noun
…		 12:25
Zoffix lizmat: the name of the pragma doesn't quite fit. Some X::Obsolete errors are for languages other than p5
m: try ｢new class {}:｣.EVAL; say $!.^name
camelia X::Obsolete
Zoffix m: new class {}:
camelia 5===SORRY!5=== Error while compiling <tmp>
Unsupported use of C++ constructor syntax; in Perl 6 please use method call syntax
at <tmp>:1
------> 3new class 7⏏5{}:
lizmat m: use p5isms; new class {}: # ok, that's LTA 12:27
camelia ===SORRY!===
This representation (P6int) does not support attribute storage (for type BOOTInt)
lizmat Zoffix: apart from the name, do you see other issues ? 12:28
Zoffix Nope
lizmat use alienisms;
feels a bit artsy fartsy to me
Zoffix: do you know of any other cases apart from the C++ "new" syntax one? 12:32
FROGGS .oO( no isms; ) 12:33
ahh, 'no' isnt there (yet) 12:35
Zoffix It's there
`no worries`
FROGGS well, it is parsed right? but not more
m: use Test; no Test
camelia ===SORRY!===
Don't know how to 'no Test' just yet
Zoffix `no worries` works
FROGGS ohh 12:36
lizmat FROGGS: depends on the pragma
Zoffix Test isn't a pragma
timotimo m: no shit
camelia ===SORRY!===
Don't know how to 'no shit' just yet
FROGGS no obsoletes
lizmat m: use p5isms; { no p5isms; say defined }
camelia 5===SORRY!5=== Error while compiling <tmp>
Unsupported use of bare "defined"; in Perl 6 please use .defined if you meant to call it as a method on $_, or use an explicit invocant or argument, or use &defined to refer to the function as a noun
…
geekosaur no nonsense
lizmat m: use p5isms; { say defined }
camelia 5===SORRY!5=== Error while compiling <tmp>
Calling defined() will never work with signature of the proto (Mu, *%)
at <tmp>:1
------> 3use p5isms; { say 7⏏5defined }
FROGGS :P
Zoffix lizmat: no, C++ is the only one I know. Other than more generic stuff like octal literals that exist in other languages 12:37
m: 04
camelia Potential difficulties:
Leading 0 has no meaning. If you meant to create an octal number, use '0o' prefix; like, '0o4'. If you meant to create a string, please add quotation marks.
at <tmp>:1
------> 03047⏏5<EOL>
WARNINGS for…
Zoffix m: say 04
camelia Potential difficulties:
Leading 0 has no meaning. If you meant to create an octal number, use '0o' prefix; like, '0o4'. If you meant to create a string, please add quotation marks.
at <tmp>:1
------> 3say 047⏏5<EOL>
4
Zoffix m: no worries; say 04
camelia 4
FROGGS and extending 'no worries' is not an option? 12:38
I mean, it is similar, no? maybe it should sound similar if it is
lizmat well, the P5 things are no worries... 12:39
Zoffix "worries" are warnings tho
FROGGS ahh, well
12:39 Zoffix left
lizmat FWIW, I've tried to do this change in a slang as a module, but that went nowhere :-( 12:40
FROGGS :o(
lizmat commute to NRpm meeting& 12:42
13:02 lizmat joined 13:15 lizmat_ joined
Geth rakudo: 8ccda1890a | (Elizabeth Mattijsen)++ | src/Perl6/Grammar.nqp
Make sure p5isms doesn't hide C++ errors
 13:18
13:32 skids joined 13:36 lucasb joined 13:40 Ven`` joined 13:42 softmoth joined 13:45 softmoth joined
lucasb . 13:54
yoleaux 00:19Z <Zoffix> lucasb: irclog.perlgeek.de/perl6-dev/2018-...i_16173824
lucasb I suggested "<null>" (instead of the empty string) as a *temporary* placeholder to be able to ".perl" print native arrays of strings, as a visual indication that that value isn't an ordinary string. 13:55
I see this null thing as an accident, not as a valid first class value in the P6 universe. Am I wrong?
m: my str @a; @a[2] = ''; say @a.perl # Will the real string please stand up? 13:57
camelia array[str].new("", "", "")
lucasb m: use nqp; my str @a; @a[2] = ''; say @a.map({nqp::isnull_s($_) ?? '<null>' !! $_}).perl
camelia ("<null>", "<null>", "").Seq
14:04 Ven`` joined, Zoffix joined
Zoffix lucasb: well yeah you're wrong. What .perl looks like is largely irrelevant. It's purpose is to serialize data to be EVALable back to the original. What your suggestion does is populate the array with strings containing data that wasn't there before 14:05
m: use nqp; my str @a = "<null>"; @a[2] = ''; say @a.map({nqp::isnull_s($_) ?? '<null>' !! $_}).perl 14:06
camelia ("<null>", "<null>", "").Seq
Zoffix "Will the real "<null>" please stand up?"
And the "temporary" fixes are a terrible idea. They quickly become permanent because there's no longer enough hurt to fix the issue properly, but because they were meant to be temporary they're shit. 14:07
I'd be +1 on banning "temporary" fixes entirely in our codebase. Fix it right or don't fix it at all.
lucasb I think the problem is not with .perl stringfication, but rather with the existence of this null value itself in P6 user side. 14:08
Zoffix Right.
lucasb ah, thanks
Zoffix $ G 'hack' src | wc -l
7
lucasb I like the commit that made it possible to be able to .perl print null values. I just see the empty string "" *as* temporary as "<null>" 14:10
Zoffix No, it's meant to be empty
same as "uninitialized" `int` is a zero
m: my int $x; my str $y; dd $x; dd $y 14:11
camelia 0
""
Zoffix m: my int $x; my str $y; dd $x; dd $y ~ "x"
camelia 0
"x"
lucasb ok, I get that
Zoffix R#1566 : It was severe enough to be a blocker. A temporary fix went in on Mar 6 and I recall plans to look into proper fix a week later. Nothing happened still 14:12
synopsebot R#1566 [open]: github.com/rakudo/rakudo/issues/1566 [regression] multi built-ins are not assignable to Callable
lucasb IOW, I think that .perl stringfication shouldn't even have to care about NULL strings 14:15
14:15 Ven`` joined
lucasb It's signature is method PERLIFY-STR(Str \\string), right? 14:15
"Str", so, it's either a Str type object or a defined Str object, right? 14:16
If Zoffix's moar's patch is the correct fix, then why can't it be merged?
Zoffix No one with enough knowledge available to review it.
lucasb hmm 14:17
Zoffix OOC, why the urgency to fix this?
lucasb no urgency. I just saw this specific commit in the changelog for this comming release 14:18
Zoffix m: -> Str $x { use nqp; dd $x }((my str @)[1])
camelia ""
lucasb c: 2018.04.1 -> Str $x { use nqp; dd $x }((my str @)[1])
committable6 lucasb, ¦2018.04.1: «chars requires a concrete string, but got null␤ in block <unit> at /tmp/Jwir375dgv line 1␤␤ «exit code = 1»»
Zoffix m: -> Str $x { use nqp; dd $x ~ "" }((my str @)[1]) 14:19
camelia concatenate requires a concrete string, but got null
in block <unit> at <tmp> line 1
lucasb my worry was exactly as yours: "there's no longer enough hurt to fix the issue properly"
Zoffix Well yeah it sweeps the problem under the rug there, but I wouldn't say adding "<null>" solves anything.
bisect: -> Str $x { use nqp; dd $x }((my str @)[1]) 14:20
bisectable6 Zoffix, Bisecting by exit code (old=2015.12 new=8ccda18). Old exit code: 1
Zoffix, bisect log: gist.github.com/0d00d9b42157eaaa37...31f7cf0973
Zoffix, (2018-01-07) github.com/rakudo/rakudo/commit/99...018eb065df
Zoffix m: -> Str $x { use nqp; $x.perl.say }((my str @)[1])
camelia ""
Zoffix bisect: -> Str $x { use nqp; $x.perl.say }((my str @)[1])
bisectable6 Zoffix, Bisecting by exit code (old=2015.12 new=8ccda18). Old exit code: 1
Zoffix, bisect log: gist.github.com/aeee7089eb67ca67d8...489a619092
Zoffix, (2018-05-08) github.com/rakudo/rakudo/commit/b1...f87c488703
Zoffix R#1837 14:23
synopsebot R#1837 [open]: github.com/rakudo/rakudo/issues/1837 TODO: remove kludge in Str.perl after null str bug is resolved
lucasb thanks! 14:24
14:41 brrt joined 15:01 Zoffix left 15:25 shareable6 joined 15:39 softmoth joined 15:51 skids joined 16:04 AlexDaniel joined, softmoth joined 16:26 skids joined 16:57 softmoth joined 17:00 softmoth joined 18:34 FROGGS joined 18:41 lizmat joined
releasable6 Next release in ≈1 day and ≈23 hours. 2 blockers. Please log your changes in the ChangeLog 19:00
19:14 brrt joined 19:27 shareable6 joined
bartolin r: my $s = Supply.from-list(1..10).rotor(3 => 1, 2 => -1,:partial); $s.tap( { say $_ } ) 20:03
camelia [1 2 3]
[5 6]
[5 6 7]
[10]
[1 2 3]
[5 6]
[6 7 8]
[10]
bartolin r-j started to give the wrong result after github.com/rakudo/rakudo/commit/a5d46b4273
lizmat: do you have an idea what could go wrong there, maybe? ^^ 20:04
bartolin has stared at the code of nqp::slice for a while but didn't spot something wrong
lizmat bartolin: am at the NR.pm meeting now...please make a ticket, will look at it tomorrow 20:06
bartolin: perhaps also clue Jeremy Studer in somehow
bartolin lizmat: will do, thanks. no need to rush, anyway
lizmat well, I guess we want to fix that before the release, right ? 20:07
so at least we would have to revert that for the JVM before the release
bartolin has no strong opinion on that. There are other failing tests, too *g* 20:09
MasterDuke btw, i emailed a CLA a couple days ago, don't remember if there's something else i'm supposed to do 20:17
AlexDaniel I only make sure that r-j compiles, there are probably many regressions like this that I simply don't see
MasterDuke: no, that's it
MasterDuke: \\o/
I mean, from your side that's it
[Coke]: ping ↑ 20:18
bartolin lizmat: github.com/rakudo/rakudo/issues/1838 20:20
lizmat bartolin++ 20:21
bartolin AlexDaniel: I run a daily spectest with r-j and try to report (or even fix) regressions
Geth rakudo: lucasbuchala++ created pull request #1839:
Test.pm: Tweaks to pass() and flunk()		 20:25
bartolin .seen jstuder 20:28
yoleaux I saw jstuder 1 Apr 2018 15:51Z in #perl6-dev: <jstuder> yeah, don't understand why it would be performing that op on the iterator itself though
20:37 lucasb joined
lucasb most Test.pm functions use $desc as the parameter name, but some use $reason 20:43
normally I wouldn't mind about var names, but in the case of Test.pm, I think some consitent style would be nice 20:44
would you accept the patch s/$reason/$desc/g ? 20:45
*consistent
Geth roast: 09b14c055c | usev6++ | S29-context/eval.t
[JVM] Another case of equal Bufs not recognized
 20:48
roast: d07a9cd464 | usev6++ | S29-context/evalfile.t
[JVM] Another case of equal Bufs not recognized
 20:52
lucasb Please, allow me a few more comments about Test.pm 21:03
If you do 'grep die-on-fail Test.pm6' you will see 21 occurrences of this line:
$ok or ($die_on_fail and die-on-fail) or $ok;
$ok is always either True or False 21:04
and die-on-fail() either dies or returns False
oh, nevermind. $die_on_fail is an integer 21:05
I was going to argue that the trailing " or $ok" was not need at all, but I guess it must be there to always return False, and not integer zero 21:06
21:16 yoleaux joined, nativecallable6 joined, unicodable6 joined, statisfiable6 joined, reportable6 joined, squashable6 joined
tbrowder_ [Coke]: ping, still no commit bit.., 21:25
22:12 dct joined 22:13 committable6 joined, undersightable6 joined 22:55 softmoth joined 23:09 MasterDuke joined

Powered by Cro and the Raku® Programming Language.

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

Thank you!