AlexDaniel .tell jnthn if I tap on $procasync.stdout (without :bin), what kind of data am I getting? And what kind of assumptions can I have about it? 01:22
yoleaux AlexDaniel: I'll pass your message to jnthn.
AlexDaniel .tell jnthn I'm asking because I was expecting lines (stuff separated by \n), but apparently this is a dumb assumption given the limit of 65536
yoleaux AlexDaniel: I'll pass your message to jnthn.
AlexDaniel and if I want to read null-separated stuff, then what? 01:28
I'll get a bunch of Blobs and then I'll be doing some gluing business to get them the way I want? Hmm… 01:30
ugexe oh -without- bin... ignore everything ive said :) 01:53
AlexDaniel .tell Zoffix Try “treegrep: password”. It's undocumented, untested and unfinished, but may be helpful anyway 03:37
yoleaux AlexDaniel: I'll pass your message to Zoffix.
AlexDaniel treegrep: hello
greppable6 AlexDaniel, gist.github.com/a0f31a8418beb7c128...ea1030591d
AlexDaniel .tell Zoffix and also uncommitted :P 03:38
yoleaux AlexDaniel: I'll pass your message to Zoffix.
stmuk I assume I can approximate a crude equivalent of Toaster with "zef smoke --force"? 09:07
Zoffix: I don't see github.com/perl6-community-modules...META6.json being tested at toast.perl6.party/ (maybe I'm missing something?) 09:10
this passed under 2017.07 and fails currently
nine stmuk: there's also emmentaler 09:13
stmuk I'll probably eat some next weekend 09:21
Zoffix stmuk: look harder. It's there 11:58
yoleaux 03:37Z <AlexDaniel> Zoffix: Try “treegrep: password”. It's undocumented, untested and unfinished, but may be helpful anyway
03:38Z <AlexDaniel> Zoffix: and also uncommitted :P
Zoffix treegrep: password 12:06
greppable6 Zoffix, gist.github.com/6e3548eb604125b704...c6c030daea
Zoffix .tell AlexDaniel treegrep seems to work 12:07
yoleaux Zoffix: I'll pass your message to AlexDaniel.
Zoffix m: Supply.from-list("foo", "bar").split("x") 12:20
camelia Cannot resolve caller split(Supply, Str); Routine does not have any candidates. Is only the proto defined?? in block <unit> at <tmp> line 1??
Zoffix (last one was meant for #zofbot) 12:25
Geth rakudo/less-wrapping: 9a3bc77632 | (Jonathan Worthington)++ | src/Perl6/Actions.nqp
Try to produce less containers for $ parameters.

We wrap them up in a Scalar if we can't, at compile-time, see a type constraint that rules out Iterable. This patch attempts to do that only if a runtime check determines the parameter to be Iterable, and just deconts it if not. Unfortunately, this breaks a few tests, and I don't have time to look into that at the moment. But leaving this here ... (5 more lines)		 12:28
jnthn nine: ^^ is the patch I mentioned on #moarvm that didn't work out so well, if you feel like exploring further
nine jnthn: is the compile-time type check even enough? Nothing prevents me from passing an Int+Iterable to a sub foo(Int $a) {} 12:34
jnthn nine: That's a "DIHWIDT" :)
We've gotten away with it for years, so I suspect we'll continue to do so. 12:35
nine But is it worth it? If spesh will remove the check and the wrapping when it's not needed anyway, why bother with the compile time check at all? 12:38
jnthn Smaller code
But yeah, good point :)
mst DIHWIDT? 12:39
nine mst: design.perl6.org/S99.html#DIHWIDT
jnthn nine: Also there was no spesh back then :) 12:41
Geth rakudo/nom: 866405067a | (Jonathan Worthington)++ | src/core/atomicops.pm
Optimization to get better code for Scalar atomics
mst aha
lizmat m: my int $a; for ^100000 { ++$a }; say now - INIT now 12:52
camelia 0.03365874?
lizmat my int $a; use nqp; for ^100000 { $a = nqp::add_i($a,1) }; say now - INIT now 12:53
m: my int $a; use nqp; for ^100000 { $a = nqp::add_i($a,1) }; say now - INIT now
camelia 0.0335588?
lizmat hmmm... on my machine the prefix ++ version has become slight faster
nine maybe just noise?
lizmat m: my int $a; for ^100000 { ++$a }; say now - INIT now
camelia 0.0341122?
lizmat m: my int $a; use nqp; for ^100000 { $a = nqp::add_i($a,1) }; say now - INIT now
camelia 0.032682?
jnthn It may manage to only resolve the symbol once
But I'd expect it to be within noise 12:54
Also I'd imagine that is faster than the last release
nine hates noise
jnthn Since it probably gets the block inlined now
lizmat jnthn: which version?
jnthn Both
Every for blah() { ...tiny operation here... } 12:55
lizmat ok 12:56
jnthn That's a big part of why we can win on the million lines benchmark now :) 12:58
[Coke] Robert Lemmen++ 13:46
Zoffix [Coke]: RE #perl6 password on www.p6c.org .... I don't see your account in home. What's your username? 14:30
Also password? y u no key? :)
[Coke] I have changed laptops since then, and don't tend to keep keys 14:34
I'd create a new one if my documented password worked. It was 'coke', back in the day; no clue if i've been removed. was just going to check on the doc build failures reported in #perl6
I can give you the new key, that is also fine (I just realized this is probably what youw ere asking) 14:35
Zoffix [Coke]: the doc build is on hack. It's failing because npm stuff is missing: irclog.perlgeek.de/perl6/2017-08-18#i_15037611
[Coke] ok, if you're on it, that's great. I only saw the original report, not the followup. Thanks. 14:36
ugexe the password was pepsi 14:37
[Coke] That's just hurtful. :| 14:38
Zoffix "you must login to services to message this user" ffs
k 14:39
[Coke]: account added, login/pass /msg'ed you can set the key yourself, if yuou want :)
raschipi Why is it even saving clear text passwords? That's a no no.
Zoffix What? 14:40
raschipi ugexe was able to retrieve Coke's clear password. That's a problem. 14:41
Zoffix snickers
[Coke] raschipi: He's joking, in case it wasn't cleqar. 14:42
*clear
Coke and Pepsi have a brand rivalry.
ugexe crytal clear pepsi
raschipi Right, whoosh right over my head.
Zoffix :) 14:43
[Coke] btw, the new coke zero sugar is meh. I really prefer the "old" coke zero. 14:44
stmuk club mate! 14:46
lucasb :D
raschipi I'm from the region mate comes from... 14:47
[Coke] I tried that the SPW, I know many of you were super excited about it. I don't recall it being my cup of mate.
I'm at the point now where I'm considering stocking up on whatever's left of old Coke Zero so I have a stash. :)
raschipi Mate is sold everywhere around here... 14:48
jnthn I was one of the handful of kids that didn't like any <coke pepsi sprite fanta ...>. Pretty much anything with gas. 14:49
nine A very healthy trait :) 14:50
Zoffix has a carbonation machine and drinks carbonated tap water
jnthn I'm OK with, but don't care much for, sprite/fanta by now. I don't know if I ever drank one yet this year.
Carbonated water I really disliked.
Now I can tolerate it.
Beer is fine. :) 14:51
Zoffix eww beer
jnthn Though even then, I don't like the more heavily carbonated lagers
raschipi My city grew out of packaging mate for exporting. This created a paper industry and print shops. That attracted publishers and marketing shops. And they brought creatives with them. Which made the city an awesome place. 14:53
stmuk I was fairly shocked when my mother returned from Chile with Coca Tea (for altitude sickness she claimed) 14:55
ilmari we were serverd that when we arrived in cusco on our way to the world scout jamboree in 1998 14:56
it is (was?) apaprently the only known remedy for altitude sickness
raschipi Tea from Coca leaves is fine. Cocaine is actually hard to refine from it. 14:57
ilmari it's still illegal to import coca leaves to most countries
raschipi And the tree that gives out the leaves for Caca tea is different from the one used for extracting Cocaine. 14:58
ilmari one of the scout leaders had bought some bags of it when they were on a recce before the jaboree
he wasn't sure if it was legal to bring in, so he went on red at customs and asked, and got a _very_ stern talking to
"some bags" as in a box of tebags 14:59
*teabags
raschipi I don't know if that's usually the case around the world, but in the Mercosur borders it's forbidden to carry *any* plant matter or animals or even fresh food. 15:00
ilmari this was peru -> norway
raschipi Only highly-processed food is allowed. 15:01
ilmari I'd have thought teabags counted as highly-processed
raschipi They're pieces of leaves, they're not very processed at all.
ilmari they're dried and ground up 15:02
are you allowed to take tobacco across borders?
AlexDaniel Zoffix: any toaster news? 15:04
yoleaux 08:37Z <jnthn> AlexDaniel: You're getting what you'd get with :bin fed into a streaming decoder, then with everything that's "ready" taken out of it; bytes representing incomplete codepoints in the encoding, or codepoint(s) that might form a grapeheme with what comes next, will be held back.
08:38Z <jnthn> AlexDaniel: Note that you can use $procasync.stdout.lines to have the output broken up into lines as it arrives if that's what you want.
12:07Z <Zoffix> AlexDaniel: treegrep seems to work
Zoffix AlexDaniel: about to start another run 15:05
AlexDaniel jnthn: so how can I do line-by-line processing? 15:06
jnthn: ah, missed your second message 15:07
I see 15:09
jnthn: ok, but what if it's 0-separated? I see no .split, and newlines seem to be hardcoded into .lines 15:12
Skarsnik I wonder is NC and Test get a defined status concerning Perl6 release? like if it was discussed? I remember having NC dieing on bad type giveng and sub and it was reverted to keep 6.c compatible stuff not breaking 15:13
jnthn AlexDaniel: Then you're on your own for now
AlexDaniel: Though lines could be expanded to handle it better
Geth tap-harness6/preprocess: 16 commits pushed by (Leon Timmermans)++
review: github.com/perl6/tap-harness6/comp...c0c1dcaf67
jnthn Do you actually need this? :)
AlexDaniel jnthn: yes 15:15
I mean, how else would you do processing of stuff with newlines?
like file paths and stuff
jnthn I meant the zero separated stuff :) 15:16
Not .lines
raschipi He needs 0-separated because the data has newlines in it.
jnthn Anyway, it's pretty easy to write yourself 15:17
Soemthing like
AlexDaniel or *may* have newlines in it. Like, normally you don't put a newline into your file name, but it's legal
jnthn supply { my $buffer = ''; whenever $stdout { $buffer ~= $_; while $buffer.index("\0") -> $idx { emit $buffer.substr(0, $idx); $buffer .= substr($idx + 1); } LAST emit $buffer } }
AlexDaniel hmm 15:18
jnthn Which is approximately what .lines is doing
Just it's been micro-optimized
AlexDaniel I see! Ok
Zoffix AlexDaniel: well, bleed toast is failing to even start, if that says anything: gist.github.com/zoffixznet/8a5a4d8...2426ae9ab6 15:31
Zoffix tries using 2017.07 for toaster itself again, with updated zef
TimToady yeah, that's definitely an unchecked .parse error 15:33
Zoffix Yeah, no idea in what tho
oh damn, should've used --ll-exception >_<
gonna build it again after the toast run and do so
AlexDaniel alright, RT #131923 15:36
synopsebot6 Link: rt.perl.org/rt3/Public/Bug/Display...?id=131923
Geth tap-harness6/preprocess: 9a5c60d36d | (Leon Timmermans)++ | lib/TAP.pm
Explicitly devnull stderr output when asked
 15:45
TimToady so where is the source code for this Toaster? neither zef nor cpanm seem to know about it 16:27
nine TimToady: toast.perl6.party/ contains a link to the Github repo
TimToady google also found it
Zoffix github.com/zoffixznet/perl6-Toaster 16:29
TimToady so looks like the failure is out of JSON::Fast probably 16:30
Zoffix oh 16:31
stmuk damm you Log::Minimal 16:46
Geth tap-harness6/preprocess: 0a7be0bb67 | (Leon Timmermans)++ | lib/TAP.pm
Add a loose mode to the Harness
 16:47
Zoffix stmuk: ?
stmuk I've reproduced github.com/moznion/p6-Log-Minimal/issues/8 16:48
Zoffix heh
Geth tap-harness6/preprocess: 9e385534c4 | (Leon Timmermans)++ | lib/TAP.pm
Add a loose mode to the Harness
 16:49
rakudo/nom: 7154638f48 | (Aleks-Daniel Jakimenko-Aleksejev)++ | docs/ChangeLog
Log all changes

Some less important internal changes were removed from the log.
Deliberately not logged: 01553585 7ab6eb6a 9d5e9295 cbfda906 9501edae 591f69af 435f7e36 63cdc426 a30ce6bc 3e70d44d 01b6d3f5 40853b4f de312452 5ea85965 79c62c0b 6eb8f7e7 413b5331 46b1a7f2 22e25da2 cdbb0858 027e25fb 86640506 e7e33173 a45ede9c
e86bdfca 5f68c775 e4802f4a		 16:53
AlexDaniel [Coke]: do we want some special mention of '.' and 'blib' issue? 16:54
[TuxCM] This is Rakudo version 2017.07-149-g93a5eff7c built on MoarVM version 2017.07-382-g348b4262 16:57
csv-ip5xs 1.384 - 1.400
test 12.257
test-t 4.416 - 4.400
csv-parser 14.316
Skarsnik where are my comparaison and defined operator! I am tired for writing if $foo.defined and $foo eq $bar ! 16:59
Zoffix $foo andthen $_ eq $bar
AlexDaniel Skarsnik: By the way, do you know about “with”?
Skarsnik hm 17:00
Zoffix ($foo//'') eq $bar 17:01
Skarsnik but will be with $foo { if $foo eq $bar} ?
Zoffix hm, unless $bar can be empty :)
AlexDaniel Skarsnik: or just use andthen. with and andthen are very similar
Skarsnik dunno if $foo:D eq $bar could be fun? x) 17:02
Zoffix nope
AlexDaniel m: my $foo:D = ‘pizza’; say $foo:D 17:03
camelia pizza?
AlexDaniel :D
Zoffix Skarsnik: see #perl :)
Skarsnik: see #perl6 :)
TimToady or I could just dup it here 17:05
m: say "foo" when "bar" given Any
camelia ( no output )
TimToady m: say "foo" when "bar" given "bar"
camelia foo?
TimToady (sorry 'bout the mischan)
Skarsnik well, there is still a request/duplicate test for bar 17:06
m: sub infix<def-eq>($a, $b) { $a.defined and $b.defined and $a eq $b} ; my $a; my $b = "b"; say $b def-eq $a; 17:09
camelia 5===SORRY!5=== Error while compiling <tmp>?Missing block?at <tmp>:1?------> 3sub infix7?5<def-eq>($a, $b) { $a.defined and $b.def? expecting any of:? new name to be defined?
Skarsnik m: sub infix:<def-eq>($a, $b) { $a.defined and $b.defined and $a eq $b} ; my $a; my $b = "b"; say $b def-eq $a;
camelia False?
Geth nqp: f34562f5fb | TimToady++ | 2 files
uncollide nqp's starter/stopper from rakudo's

Turns out rakudo's starter/stopper use dynvars, while rakudo's use mixins, so accidentally calling rakudo's starter/stopper greatly confuse nqp. Here we rename nqp's starter/stopper to dynstarter/dynstopper to reflect their use of dynamic vars, since there are likely to be fewer grammars derived from nqp grammars than from rakudo grammars.		 17:12
Skarsnik love to wait 1mn for his script to run 17:13
TimToady that should be completely benign, and doesn't really need changelogging
Skarsnik unrelated. when trying to track/fix #127345 I noticed that moar get the line of the start of statement. So it was probably what build the ast that does not mark corectly the line. Should I search in nqp code? 17:15
synopsebot6 Link: rt.perl.org/rt3/Public/Bug/Display...?id=127345
timotimo Skarsnik: i believe we deliberately leave out line numbers for anything but QAST::Stmt or Stmts or something 17:19
Skarsnik I mean if I understand correctly a Statement is the whole instruction up to the ; and each part is a callframe? 17:21
You have the right to tell me that I know nothing about writing compiler and stuff, since it's true xD 17:24
Zoffix buggable: toast
buggable Zoffix, Between 2017.07-195-g8664050 and 2017.07: 35 (4.08%) modules got burnt; 36 (4.20%) got unsucced; 267 (31.16%) out of 857 modules appear unusable. See toast.perl6.party/ for details.
Zoffix AlexDaniel: ^ new run
timotimo is JSON::Fast in trouble? 17:25
Skarsnik: callframes are related to what subroutine invokes what other subroutine
AlexDaniel this does not look too bad 17:26
Zoffix Really? 17:35
Last toast was only like 4 or 8 modules burnt
AlexDaniel I didn't say that it wasn't bad :)
it's just that I expected worse
a couple of things have common problem with URI module it seems 17:36
… even p6doc 17:37
Geth rakudo/nom: 9e61827bf0 | TimToady++ | 2 files
make sure expected list is flattened

  (This ensures that each entry prints on a separate line.)		 17:38
TimToady that one is also very minor
both pass all spectests too
Skarsnik hm everycall to a // will update $0,$1.. ? 17:40
TimToady it will update $/, from which $0, etc are derived
timotimo did you already respond to "is X::Comp really the right exception for user-provided grammars?"? 17:41
Skarsnik Ok make sense. Should have named the capture in the first regex x)
timotimo i find it a bit surprising that i'd get a ===SORRY!=== for MyGrammar.parse
TimToady timotimo: nope, haven't seen that
we could split it out, but I was lazy :)
timotimo my role X::HighExplain 17:43
Zoffix ZOFFLOP: t/spec/S17-lowlevel/cas.t 17:50
failed during stresstest, passed when run by itself
[Coke] (special mention) since we don't have our act together, perhaps if we still have changelogs broken out by section, we can have a new "Security" section.
TimToady just updated #131922 to note that "slurpy" and "variadic" do not mean the same thing 17:56
Geth rakudo/nom: 97f9eb7f3e | (Aleks-Daniel Jakimenko-Aleksejev)++ | docs/ChangeLog
“Security:” section

  [Coke]++		 17:59
AlexDaniel [Coke]: like this? Anything else?
[Coke] AlexDaniel: +1 from me. Thanks for doing the release. 18:00
Geth rakudo/nom: 9303a6e422 | (Zoffix Znet)++ | lib/Test.pm6
Correctly handle test descriptions with new lines in them
 18:01
Zoffix mc: use Test; plan 1; skip "FOo", 1; ok 0 18:03
committable6 Zoffix, ¦2015.12: «1..1?ok 1 - # SKIP FOo?not ok 2 - ??# Failed test at /tmp/0HpzdfHO_c line 1?# Looks like you planned 1 test, but ran 2?# Looks like you failed 1 test of 2 «exit code = 1»»
Geth roast: 85edce2344 | (Zoffix Znet)++ | S24-testing/3-output.t
Test newlines in descriptions get properly `#`-prefixed

Rakudo fix: github.com/rakudo/rakudo/commit/9303a6e422		 18:06
Zoffix Another bunch of floppers, including lowlevel/cas.t one: gist.github.com/zoffixznet/d549471...bc546b35de 18:17
AlexDaniel huh… why am I never getting floppers 18:18
or is it from stress test?
Geth rakudo/nom: eb529f1495 | (Zoffix Znet)++ | lib/Test.pm6
Fix wrongful escaping of `# SKIP` TAP instruction
Zoffix AlexDaniel: stresstest on 24-core box with 30 jobs
Geth roast: 423cc68bf9 | (Zoffix Znet)++ | S24-testing/3-output.t
Test skip() correctly prints skip TAP instruction

Rakudo fix: github.com/rakudo/rakudo/commit/eb529f1495		 18:19
AlexDaniel right
ugexe: can you take a look at this github.com/perl6-community-modules/uri/pull/36 ? 18:28
Geth rakudo/nom: 1599873d14 | TimToady++ | tools/build/NQP_REVISION
bump nqp to get dynstarter/dynstopper tweak
 18:31
¦ rakudo/nom: version bump brought these changes: github.com/perl6/nqp/compare/2017....4-gf34562f
AlexDaniel # expected: "Subject: test\n\rTo: foo\n\r\n\rfoo\n\r" 18:33
# got: "Subject: test\n\rTo: foo\n\r\n\rfoo\n\r"
TimToady o_O 18:35
is it not an "is"? 18:36
AlexDaniel it has some “odd newlines”
TimToady so could be something odd in the NFGness 18:37
AlexDaniel if somebody is interested: github.com/retupmoca/p6-Email-Simp....t#L42-L48 18:38
“Round trip nasty” is failing
TimToady sounds like something for samcv++ or jnthn++
jnthn Does MVM_SPESH_DISABLE=1 help? :) 18:41
(When yes, it tends to be me that's on the hook :P)
geekosaur when I saw that I tried to point samcv at it and then got confused as to what triggered it 18:42
but yes, I suspect that NFG is going weird on the backwards newlines
AlexDaniel jnthn: does not help 18:43
I just dd-ed .ords of both strings, they are identical
yet say ~$mail eq $nasty; # false ???
or am I blind 18:44
geekosaur if you are then so am I... over two different days
AlexDaniel say (~$mail).ords.list ~~ ($nasty).ords.list; # true
evalable6 (exit code 1) 04===SORRY!04=== Error while compiling /tmp/fcAr3EER0a?Variabl…
AlexDaniel, Full output: gist.github.com/47e14ad8c23f73371e...05f4b7bebe
AlexDaniel slaps evalable6
dd (~$mail).substr(23, 1).ords; # (13,).Seq 18:47
dd (~$nasty).substr(23, 1).ords; # (13, 10).Seq
I'll have to go in ?10 minutes (food!). It won't hurt to have more eyes on this. There are a couple of modules depending on Email::Simple 18:50
hm, maybe something is not being renormalized? 18:52
I really don't know
yoleaux Zoffix: fix xtest on doc 19:27
timotimo AlexDaniel: there used to be a bug in eq where we were able to concat two strings and \n and \r didn't fuse into one grapheme 19:36
AlexDaniel back
timotimo should also give you two different amounts for nqp::graphs
also, i saw you already found the exact place where it happened
AlexDaniel timotimo: “used to be” means that it was fixed? 19:37
timotimo i thought it was, but it could easily have resurfaced from another change being not-perfect 19:38
AlexDaniel well… let's see
commit: 2017.07 gist.githubusercontent.com/AlexDan...15/basic.t 19:40
committable6 AlexDaniel, Successfully fetched the code from the provided URL.
AlexDaniel, gist.github.com/503541f1e0e6f3791e...b2b46ceed8
AlexDaniel argh
commit: 2017.07 gist.githubusercontent.com/AlexDan...82/basic.t 19:41
committable6 AlexDaniel, Successfully fetched the code from the provided URL.
AlexDaniel, ¦2017.07: «ok 1 - got correct line terminator?ok 2 - got correct body?ok 3 - Round trip nasty»
AlexDaniel commit: HEAD gist.githubusercontent.com/AlexDan...82/basic.t
committable6 AlexDaniel, Successfully fetched the code from the provided URL.
AlexDaniel, gist.github.com/820d2ad816882da36e...866e312ab9
AlexDaniel bisect: old=2017.07 gist.githubusercontent.com/AlexDan...82/basic.t
bisectable6 AlexDaniel, Successfully fetched the code from the provided URL.
AlexDaniel, Bisecting by exit code (old=2017.07 new=1599873). Old exit code: 0
AlexDaniel, bisect log: gist.github.com/d16b49e55a1102e25b...5df7bbf902
AlexDaniel, (2017-07-20) github.com/rakudo/rakudo/commit/b8...a13ede808c
AlexDaniel “String concatenation improvements” OK, seems relevant 19:42
timotimo good to find this before the release 19:43
AlexDaniel Zoffix++ for his toaster
timotimo is it wrong to expect "zef test" to also run Build.pm for me? 19:46
i'm looking at a module that relies on LibraryMake, but it doesn't work properly
AlexDaniel .tell samcv I'm marking RT #131924 as a blocker. Would be great if you could take a look 19:49
synopsebot6 Link: rt.perl.org/rt3/Public/Bug/Display...?id=131924
yoleaux AlexDaniel: I'll pass your message to samcv.
samcv hey AlexDaniel
yoleaux 13:29Z <Zoffix> samcv: would you be able to unbust the doc build. After hack's upgrade it's missing some sort of npm bits and I can't figure out which: github.com/perl6/doc/issues/1457#i...-251240775
19:49Z <AlexDaniel> samcv: I'm marking RT #131924 as a blocker. Would be great if you could take a look
synopsebot6 Link: rt.perl.org/rt3/Public/Bug/Display...?id=131924
AlexDaniel samcv: oh, hi! :)
samcv ok thanks for letting me know. will look at the blocker first
just got my tea so ready to go 19:50
AlexDaniel, have we run the toaster yet
AlexDaniel samcv: yes, this bug is exactly from toaster :)
samcv yay
AlexDaniel toast.perl6.party/
a lot of it is because of URI module which is now broken after Grammar.parse changes 19:51
I've submitted a pull request here: github.com/perl6-community-modules/uri/pull/36
then a bunch of stuff is not installable because Email::Simple is failing 19:52
and that's when I had to take a break for some food :) 19:53
timotimo the appveyor failure is because of a not-windows-compatible change i just fixed recently 19:54
when samcv fixes the concatenation/synthetic bug there'll be a bump anyway
AlexDaniel okay
damn, everyone on this channel is so helpful. Thank you 19:56
Zoffix m: cas $i, -> int $current { $current * 2 } 19:59
camelia 5===SORRY!5=== Error while compiling <tmp>?Variable '$i' is not declared?at <tmp>:1?------> 3cas 7?5$i, -> int $current { $current * 2 }?
Zoffix [Coke]: you around? xtest complaining about that ^ code sample with "Something went wrong in (Undeclared)". What do I do?
timotimo aha! it fails because Panda::Common can't be use-d 20:00
(the Build.pm i'm looking at)
Zoffix m: say atomicsize 20:01
camelia 5===SORRY!5=== Error while compiling <tmp>?Undeclared routine:? atomicsize used at line 1. Did you mean 'atomic-inc'???
Zoffix What's atomicsize ?
In docs "If C<$value> otherwise overflows C<atomicsize> then it will be silently truncated before the addition is performed." 20:02
AlexDaniel c: 2017.07 say [$*DISTRO.Str|$*KERNEL.Str] ~~ m:i/linux/;
committable6 AlexDaniel, ¦2017.07: «?linux?»
AlexDaniel bisect: old=2017.07 say [$*DISTRO.Str|$*KERNEL.Str] ~~ m:i/linux/;
bisectable6 AlexDaniel, Bisecting by exit code (old=2017.07 new=1599873). Old exit code: 0
AlexDaniel, bisect log: gist.github.com/aa04676317f2df828e...4bafad6954
AlexDaniel, (2017-08-09) github.com/rakudo/rakudo/commit/17...dab082750c
AlexDaniel …
oh…
samcv my current \r\n test doesn't catch it S15-nfg/concat-stable.t so once i install Mail::Simple imma test this out
AlexDaniel c: 2017.07,HEAD say [$*DISTRO.Str|$*KERNEL.Str].Str 20:03
committable6 AlexDaniel, ¦2017.07: «any("debian", "linux")» ¦HEAD(1599873): «This type cannot unbox to a native string: P6opaque, Junction? in block <unit> at /tmp/2a4rDBAlUa line 1? «exit code = 1»»
timotimo perhaps zef should put a custom CUR into the path that detects and maybe even handles Panda::Common being missing
samcv ok i found the issue 20:04
Zoffix \o/
AlexDaniel \o/ I didn't expect it to happen so fast
samcv m: "\r\n".chars.say; "\x0a\x0d".chars.say
camelia 1?2?
AlexDaniel Zoffix: any chance you can rerun toastor a little bit later?
samcv hmm. why would it...
AlexDaniel toastor, heh… 20:05
Zoffix ok
AlexDaniel Zoffix: ?
samcv bisectable6, "\x0a\x0d".chars.say
bisectable6 samcv, On both starting points (old=2015.12 new=1599873) the exit code is 0 and the output is identical as well
samcv, Output on both points: «2»
samcv ok so it's not a new bug?
still a bug though 20:06
AlexDaniel samcv: maybe not, yes. Perhaps previously things were renormalized more than they should have, so the bug was hidden by this?
lizmat: here?
samcv i mean. the thing that introduces it is "\x0a\x0d" 20:07
but i guess that's possible it renormalized \r\n unneededly. in that case probably a speedup when i changed when we trigger
to not trigger for \r or \n unless it sees \r\n (has a short circuit case for \r or \n characters now) 20:08
which is within unicode spec
timotimo so the upgrade path for Build.pm is probably just throwing out the Panda bits and make the Builder class just a Plain Old Class?
samcv what i want to know is how it constructs the string "\x0a\x0d" that it ends up not being normalized though
AlexDaniel c: 2017.07 say [$*DISTRO.Str|$*KERNEL.Str] ~~ m:i/‘any’/ 20:10
committable6 AlexDaniel, ¦2017.07: «?any?»
AlexDaniel c: HEAD say [$*DISTRO.Str|$*KERNEL.Str] ~~ m:i/‘any’/
committable6 AlexDaniel, ¦HEAD(1599873): «This type cannot unbox to a native string: P6opaque, Junction? in block <unit> at /tmp/8HQpdDUTkr line 1? «exit code = 1»»
Zoffix m: say [<a b>.any].Str 20:11
camelia This type cannot unbox to a native string: P6opaque, Junction? in block <unit> at <tmp> line 1??
Zoffix s: [], 'Str', \()
SourceBaby Zoffix, Sauce is at github.com/rakudo/rakudo/blob/1599...st.pm#L369
Zoffix m: say [<a b>.any].join: "meows" 20:12
camelia This type cannot unbox to a native string: P6opaque, Junction? in block <unit> at <tmp> line 1??
Zoffix s: [], 'join', \()
SourceBaby Zoffix, Sauce is at github.com/rakudo/rakudo/blob/1599...t.pm#L1149
samcv AlexDaniel, the problem happens during compilation
AlexDaniel Zoffix: the code is wrong, obviously, but I'm seeing something weird anyway
samcv so it compiles in the wrong string
AlexDaniel FWIW it's github.com/dnmfarrell/System-Passw...wd.pm6#L12 20:13
Zoffix AlexDaniel: FWIW: irclog.perlgeek.de/perl6-dev/2017-...i_14992203
AlexDaniel Zoffix: I've seen that, but the idea of reverting it makes me cry
Zoffix Why? 20:14
The native-string-Junction will keep popping up.
AlexDaniel m: say $*DISTRO.Str|$*KERNEL.Str ~~ /‘debian’/
camelia any(Nil, Nil)?
AlexDaniel m: say $*DISTRO.Str|$*KERNEL.Str ~~ /‘linux’/
camelia any(Nil, ?linux?)?
AlexDaniel m: say so $*DISTRO.Str|$*KERNEL.Str ~~ /‘linux’/
camelia True?
AlexDaniel m: say $*DISTRO.Str|$*KERNEL.Str ~~ m/‘linux’/
camelia False?
AlexDaniel is it correct?
samcv bisectable6, "\x[0a,0d]".chars.say 20:15
bisectable6 samcv, On both starting points (old=2015.12 new=1599873) the exit code is 0 and the output is identical as well
samcv, Output on both points: «2»
samcv ok so it's always been a problem
it compiles in the wrong string...
goddamit
ok so how does it construct that string though?
should be in nqp i think
AlexDaniel Zoffix: it's a very promising feature, I've needed this personally a couple of times 20:16
Zoffix: but I'm not saying that we shouldn't revert it, just that it's sad
Zoffix: do you have any idea why smartmatching a junction agains m// gives false in this case? 20:17
Zoffix m: say $*DISTRO.Str|$*KERNEL.Str
camelia any(opensuse, linux)?
Zoffix m: say ($*DISTRO.Str|$*KERNEL.Str).perl 20:18
camelia any("opensuse", "linux")?
samcv hmm wait that's not the issue that's just a reversed \r\n. sorry was misreading things
Zoffix s: /^/, 'ACCEPTS', \(any("opensuse", "linux"))
SourceBaby Zoffix, Sauce is at github.com/rakudo/rakudo/blob/1599...Code.pm#L7
AlexDaniel m: $_ = $*DISTRO.Str|$*KERNEL.Str; say m/‘linux’/;
camelia any(Nil, ?linux?)?
moritz m: say so $*DISTRO.Str|$*KERNEL.Str ~~ /linux/
camelia True?
moritz m: say so $*DISTRO.Str|$*KERNEL.Str ~~ m/linux/
camelia False?
moritz aha
it's possibly related to the weird stuff that we do to make m/.../ matching work in general 20:19
m/.../ matches immediately, and returns a Match
and then smart-matching against a Match returns True (or the Match), always 20:20
AlexDaniel moritz: so it's a bug, right? 20:23
moritz AlexDaniel: yes, I'd say so 20:24
Geth roast: e9a2bda8c4 | (Christopher Bottoms)++ | S02-literals/heredocs.t
Added failing heredoc tests
 20:25
AlexDaniel moritz: RT #131925 20:26
synopsebot6 Link: rt.perl.org/rt3/Public/Bug/Display...?id=131925
AlexDaniel Pull request for System::Passwd github.com/dnmfarrell/System-Passwd/pull/12 20:29
Geth roast: 9be474c12c | (Christopher Bottoms)++ | S02-literals/heredocs.t
Revert "Added failing heredoc tests"

This reverts commit e9a2bda8c4596f04b4549e416bba7b191a181999.		 20:32
AlexDaniel alright, next module… 20:35
alright, DOM::Tiny suffers from the Grammar.parse change 20:39
my plan is to fix as many modules as I can today, so that it will be easier for us next time when we introduce Grammar.parse change again 20:41
samcv ok i found the issue. it's in `split` 20:42
well maybe. investigating 20:43
Geth roast: molecules++ created pull request #295:
additional failing heredoc tests		 20:44
timotimo AlexDaniel: what a coinkidink, i'm also fixing a few modules! 20:51
AlexDaniel Could not find META6:ver<0.0.4..*> 20:52
META6:ver('0.0.12'):auth('github:jonathanstowe')
so 0.0.12 is installed but it cannot find 0.0.4..*, why?
Zoffix AlexDaniel: are you sure you're on latest zef?
AlexDaniel Zoffix: totally unsure 20:53
Zoffix I thought I saw some mention of this stuff awhile back and zef getting a fix for it
AlexDaniel zef uninstall zef 20:56
timotimo i went on latest zef and it still complained 20:57
so i used zef nuke site home and reinstalled zef and some stuff
Zoffix oh :(
What module is the issue in? 20:58
AlexDaniel DOM::Tiny
there's another issue in it with Grammar.parse and for that I have a pull request ready
timotimo oh, do you know about the citations project?
AlexDaniel ? 20:59
timotimo github.com/finanalyst/ModuleCitation - this has the code and data files but there's also a web frontend somewhere
Zoffix knows about 21:00
it
timotimo finanalyst.github.io/ModuleCitation/
Zoffix It's linked to from the nav of modules.perl6.org afterall :)
timotimo if we could cross-reference this with the toaster we could concentrate on things most used
Zoffix true
AlexDaniel yea 21:01
ok, yeah, it works! 21:02
nuking the whole thing helped
Zoffix Well, I'm not having any META issues, so I'd say you guys have too old zef
nuke
nuke ~/.zef 21:03
timotimo if you're having meta problems i feel bad for you son
i got 99 problems but perl6 ain't one
~ hit me!
Zoffix resist the urge to ask huggable for nuclear option >:)
AlexDaniel boom github.com/zostay/p6-DOM-Tiny/pull/3 21:04
next module!
samcv AlexDaniel, i'm not able to golf the problem so far with the email module 21:07
i'm not quite sure what it's actually doing. i mean i've tried splitting the string up many ways and joining it, and even creating only the codepoints and concatting it at every index by codepoint (not grapheme) and it always turns out the same
AlexDaniel samcv: should I try golfing it? 21:08
samcv file was last edited dec 2015...
Fix things after unicode \r\n changes
i feel it may do something really weird
yeah if you colud golf it that would be awesome
AlexDaniel meanwhile, does anybody know if SSL module is supposed to work? 21:09
timotimo the async ssl one? 21:10
AlexDaniel I don't know, this one: github.com/grondilu/openssl
toaster says that it Succ-ed on 2017.07, but its test files don't even have “use experimental :pack;”
timotimo oh, that one is probably not used terribly much, since it only covers digests? 21:11
or am i looking wrong?
Zoffix Where are you people finding these wrong URLs to repos? 21:12
[Coke] did it. Now AlexDaniel did it
timotimo hah what :)
Zoffix buggable: eco OpenSSL
buggable Zoffix, OpenSSL 'OpenSSL bindings': github.com/sergot/openssl 4 other matching results: modules.perl6.org/s/OpenSSL
AlexDaniel modules.perl6.org/dist/SSL 21:13
buggable: eco SSL
buggable AlexDaniel, SSL 'Perl6 interface to a OpenSSL (so far only digests though)': github.com/grondilu/openssl 5 other matching results: modules.perl6.org/s/SSL
AlexDaniel pffffft :D
jnthn buggable: SSL async 21:14
Zoffix Never mind
jnthn buggable: eco SSL async
buggable jnthn, Nothing found
Skarsnik m: use NativeCall; sub foo(num $foo) is native(){};
camelia 5===SORRY!5=== Error while compiling <tmp>?Cannot resolve caller infix:<==>(NQPMu, Int); none of these signatures match:? ($?)? (\a, \b)? (Real \a, Real \b)? (Int:D \a, Int:D \b)? (int $a, int $b)? (Num:D \a, Num:D \b …
jnthn aww :)
Zoffix AlexDaniel: but Succ toaster seems to have installed sergot's openssl?
"===> Filtering: OpenSSL:ver('0.1.14'):auth('github:sergot')" 21:15
toast.perl6.party/module?module=SS...it=2017.07
Skarsnik I should fix this error
Zoffix buggable: eco SSL author:jnthn
buggable Zoffix, IO::Socket::Async::SSL 'Provides an API like IO::Socket::Async, but with SSL support.': github.com/jnthn/p6-io-socket-async-ssl
AlexDaniel boom another pull request github.com/stmuk/p6-app-p6tags/pull/3 21:19
AlexDaniel is tired of this, now will look into the unicode issue 21:20
Zoffix Yeah, no idea why that output in toaster shows the result for OpenSSL one :S
AlexDaniel releasable6: status 21:23
releasable6 AlexDaniel, Next release in ?21 hours. 2 blockers. 195 out of 201 commits logged
AlexDaniel, Details: gist.github.com/c6847c34071813e827...2f483076aa
stmuk AlexDaniel: thanks 21:25
AlexDaniel stmuk: by the way, I feel your pain regarding the security issue. I really want to figure something out for it. If you have any ideas I'd love to hear them (probably after the release) 21:28
stmuk yeah I'm thinking there should be some sort of star release 21:30
AlexDaniel samcv: hey 21:31
samcv hey
AlexDaniel samcv: ah sorry, I already see that you noticed it: “hmm wait that's not the issue ” 21:32
samcv: is there any way I can check if a string somewhere is weird? 21:36
samcv: like to force renormalization or something
Skarsnik is-weird 21:37
samcv $a.NFC.Str eq $a
if it's not equal then that means it got messed up somewhere
travis-ci Rakudo build failed. TimToady 'bump nqp to get dynstarter/dynstopper tweak'
travis-ci.org/rakudo/rakudo/builds/266076340 github.com/rakudo/rakudo/compare/e...99873d1420
buggable [travis build above] ? All failures are due to timeout (0), missing build log (0), GitHub connectivity (1), or failed make test (0).
Skarsnik good night people, good luck with your awesome work :)
Zoffix \p 21:40
samcv AlexDaniel, i have it down to this: gist.github.com/5b8d633188baba6297...2fd7bc8516 21:42
if $mail-str.NFC.Str ne $nasty then usually it means Email::Simple gave us back a string that's not the same as we originally had, changed it in some way 21:43
timotimo samcv: maybe we want a piece of code in moarvm that optionally gets run after every string op that ensures nfg-related issues don't exist in any of our strings
samcv hmm 21:44
like STRAND_CHECK but for NFG?
timotimo yeah 21:46
samcv writing one now 21:51
AlexDaniel alright, now that's *very* golfed down 22:12
I give it 10 more minutes, it can't run away for me for so long 22:16
from*
it goes away if you move it out of the module 22:21
samcv hmm 22:22
timotimo it could have to do with how we serialize and read the string heap
AlexDaniel samcv: can you try this? gist.github.com/AlexDaniel/3882bc4...949035a59c 22:24
samcv does that trigger AlexDaniel ?
timotimo oh, don't forget to check if it differs between concat and join 22:25
AlexDaniel samcv: I've added sample output to the gist
timotimo and whether it's strings procured from a substr or literal or whatever that you use there
AlexDaniel samcv: I'd say yes, definitely
samcv ok give me a sec. almost done with my NFG_CHECK function
AlexDaniel I'm really surprised that $z makes any difference there 22:27
you can't just replace it with ‘’ 22:28
but you can do this: “: {‘’}\n\r”
slightly different version here, if needed: gist.github.com/AlexDaniel/bdba740...33f760ae87 22:31
Zoffix leaves gaming, then sleeping
Gonna toast tomorrow moarning if needed. Morning being in ~16 hours
AlexDaniel Zoffix: OK, perfect 22:32
Zoffix: thank you for your effort
Geth roast/master: 4 commits pushed by (Christopher Bottoms)++ 22:46
synopsebot6 Link: rt.perl.org/rt3/Public/Bug/Display...?id=131927
AlexDaniel
.oO( since when flappers became floppers? )
 22:50
but in any case, boom: github.com/tony-o/perl6-event-emitter/pull/11
(so that we won't see this module again getting burned)
Zoffix flip-flopping is going from one position to another one 22:51
hence floppers
AlexDaniel I thought it was about bugs with wings 22:53
hence flappers
timotimo AlexDaniel: that looks like a good place to use a "start react" :) 22:54
AlexDaniel timotimo: I agree, yes.
timotimo: the world would have been different if I had time to refactor everything I have to fix :) 22:56
timotimo true
AlexDaniel
.oO( /different/better/ … but who knows… )
 22:59
timotimo i wish you the best of luck for the release! 23:04
AlexDaniel thanks 23:06
Hiker module is simply a flapper, forget about it
timotimo goes to bed 23:07
AlexDaniel boom: github.com/tony-o/perl6-hiker/issues/8 and time to move on 23:08
samcv working on that function right now AlexDaniel 23:10
AlexDaniel .tell TimToady Perhaps take a look at this failing test: github.com/tadzik/Grammar-BNF/blob...bnf.t#L400 And here's some sample output: gist.github.com/AlexDaniel/7e6ebb6...93fd004b48 23:19
yoleaux AlexDaniel: I'll pass your message to TimToady.
AlexDaniel maybe there's an easy backward-compatible way to fix it ¯\_(?)_/¯ 23:20
hm, and that's it I think. This should be enough to unburn all modules 23:22
unless there are other failures of course
samcv AlexDaniel, if you checkout the latest moarvm. you can edit src/strings/ops.c at the top and uncomment and set MVM_DEBUG_NFG to 1 23:42
then it will check all concat operations, and give nice debug info
also MVM_DEBUG_NFG_STRICT must be set to 0 or 1 (comment that one out too) 23:43
NFG failure. Got different grapheme count. Got 8 but after re_nfg got 7 23:44
First differing grapheme was at index 5
orig: 13 (\r) after re_nfg: -1 (\r\n)
that was with my previous thing. let me download that gist now and run it
yeah i get similar error but it got 6 instead of 5 graphemes, but the differing graphemes were the same 23:46
this one differing at index 3

Powered by Cro and the Raku® Programming Language.

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

Thank you!