Perl 6 language and compiler development | Logs at colabti.org/irclogger/irclogger_log/perl6-dev | For toolchain/installation stuff see #perl6-toolchain | For MoarVM see #moarvm
Set by Zoffix on 27 July 2018.
00:01 lizmat joined, p6bannerbot sets mode: +v lizmat 00:08 stmuk joined 00:09 p6bannerbot sets mode: +v stmuk 00:10 stmuk_ left 00:11 epony left
geekosaur that might mean it's a syscall failing , in which case you don't get much help aside from maybe strace and syslog 00:22
00:36 epony joined, p6bannerbot sets mode: +v epony, stmuk_ joined 00:37 p6bannerbot sets mode: +v stmuk_ 00:39 stmuk left 01:20 peterrus joined 01:24 peterrus left 02:05 cognominal-p6 joined 02:06 Kaiepi left, epony left, p6bannerbot sets mode: +v cognominal-p6 02:07 Kaiepi joined, p6bannerbot sets mode: +v Kaiepi 02:09 b2gills left, cognominal-p6 left, cognominal-p6 joined 02:10 p6bannerbot sets mode: +v cognominal-p6 02:11 stmuk joined, epony joined, p6bannerbot sets mode: +v epony 02:12 p6bannerbot sets mode: +v stmuk 02:14 stmuk_ left 02:29 cognominal-p6 left 02:41 b2gills joined 02:42 p6bannerbot sets mode: +v b2gills 02:47 ufobat_ joined 02:48 p6bannerbot sets mode: +v ufobat_ 02:51 fake_space_whale left, ufobat___ left 02:53 epony left 03:00 epony joined, p6bannerbot sets mode: +v epony 03:04 epony left, epony joined, p6bannerbot sets mode: +v epony 03:11 epony left 03:18 epony joined, p6bannerbot sets mode: +v epony 03:20 epony left 03:26 EmzyFq joined 03:27 EmzyFq left 03:32 epony joined, p6bannerbot sets mode: +v epony 04:03 Zoffix joined, p6bannerbot sets mode: +v Zoffix
Zoffix ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ 04:32
.tell TimToady As promised, I compiled all the info on the creation of the language alias and I'm now ready to "formally" request you to consider the creation of the name alias: perl6.party/post/A-Request-to-Larr...for-Perl-6
yoleaux Zoffix: I'll pass your message to TimToady.
Zoffix ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ
weekly: colabti.org/irclogger/irclogger_log...-10-07#l54 04:45
notable6 Zoffix, Noted!
04:46 Zoffix left 05:12 osx joined 05:16 osx left 05:25 cognominal-p6 joined 05:26 p6bannerbot sets mode: +v cognominal-p6 05:47 TimToady left 05:49 cognominal-p6 left 06:32 TimToady joined, p6bannerbot sets mode: +v TimToady 07:35 AthasxR joined, AthasxR left 08:45 cognominal-p6 joined 08:46 p6bannerbot sets mode: +v cognominal-p6 08:59 cognominal-p6 left 09:00 cognominal-p6 joined, p6bannerbot sets mode: +v cognominal-p6
lizmat Files=1255, Tests=75814, 338 wallclock secs (15.66 usr 5.23 sys + 2379.59 cusr 245.12 csys = 2645.60 CPU) 09:08
.tell pmurias js branch spectest times the same as master, so it *was* a fluke that it didn't before 09:09
yoleaux lizmat: I'll pass your message to pmurias.
lizmat Zoffix: s/ proponets/proponents/ 09:13
09:21 dct joined, p6bannerbot sets mode: +v dct 09:32 cognominal-p6 left 09:40 pmurias joined, p6bannerbot sets mode: +v pmurias
pmurias jnthn: is it ok if I merge the js branch in? (I fixed the stuff people found during reviews) 09:45
yoleaux 09:09Z <lizmat> pmurias: js branch spectest times the same as master, so it *was* a fluke that it didn't before
09:47 cognominal-p6 joined 09:48 p6bannerbot sets mode: +v cognominal-p6 09:57 brrt joined 09:58 p6bannerbot sets mode: +v brrt
brrt \o 09:58
09:59 cognominal-p6 left
brrt .tell masak re macros, it might be interesting for you to learn that since I implemented hygienic macros in the the expression template compiler, I can actually abstract things that I couldn't in C 09:59
yoleaux brrt: I'll pass your message to masak.
09:59 cognominal-p6 joined 10:00 p6bannerbot sets mode: +v cognominal-p6
pmurias lizmat: good, was afraid it was some JIT issue or some weird VM inefficieny getting accidently triggered ;) 10:02
[Tux] Rakudo version 2018.09-75-g2d7724f8b - MoarVM version 2018.09-124-g588873f7a
csv-ip5xs0.888 - 0.953
csv-ip5xs-207.271 - 7.378
csv-parser22.369 - 22.672
csv-test-xs-200.445 - 0.446
test8.045 - 8.136
test-t1.773 - 1.865
test-t --race0.800 - 0.842
test-t-2030.984 - 31.397
test-t-20 --race11.132 - 12.239
 10:10
lizmat pmurias: that was my fear as well, but no more
brrt .tell Zoffix: Summation -> Summary, Numerious -> Numerous 10:13
yoleaux brrt: What kind of a name is "Zoffix:"?!
brrt .tell Zoffix Summation -> Summary, Numerious -> Numerous
yoleaux brrt: I'll pass your message to Zoffix.
brrt (not an english language native, but still)
lizmat pmurias: I wouldn't mind you going ahead and merge the js branch: you worked long enough on it do to the honours! 10:20
10:22 ExtraCrispy joined, p6bannerbot sets mode: +v ExtraCrispy 10:28 cognominal-p6 left 10:32 cognominal-p6 joined 10:33 p6bannerbot sets mode: +v cognominal-p6 10:39 cognominal-p6 left 10:40 cognominal-p6 joined, p6bannerbot sets mode: +v cognominal-p6 10:44 brrt left 10:53 ExtraCrispy left 10:57 ExtraCrispy joined, p6bannerbot sets mode: +v ExtraCrispy 11:03 cognominal-p6 left, cognominal-p6 joined 11:04 p6bannerbot sets mode: +v cognominal-p6 11:10 robertle joined, p6bannerbot sets mode: +v robertle
jnthn pmurias: No objections from me to merge. :) 11:23
yoleaux 09:29Z <lizmat> jnthn: s/construts/constructs/
jnthn Spell checkers are useless...
lizmat: Fixed, thanks 11:24
Geth rakudo/master: 295 commits pushed by (Pawel Murias)++, pmurias++, (Garrett Goebel)++, (Paweล‚ Murias)++
review: github.com/rakudo/rakudo/compare/2...9536fd942d		 11:28
ยฆ rakudo/master: version bump brought these changes: github.com/perl6/nqp/compare/2017....gcb169e00b
AlexDaniel โ€œSecond, even after you make that clarification, people still continue to shorten the name to just Perl ("You're the Perl guy, right?").โ€ 11:44
hah, hehโ€ฆ :'(
so relatable
pmurias I feel what we rename to is super important before we decide if a rename is a good idea 11:51
lizmat pmurias: did you see jnthn's ok ? 11:53
aww... Geth is down again :-( nobody saw pmurias' merge of the JS branch :-( 11:58
pmurias: congratulations!
nine lizmat: it was reported! But Geth is smart enough to not list all the 295 commits :)
lizmat ah, /me is blind :-) 11:59
jnthn Yay, le merge :)
nine What a fantastic day :)
jnthn pmurias++ \o/
This makes pmurias++ the second person to develop/merge a new Rakudo backend. :) $bus-number++ :D 12:00
12:00 [TuxCM] left
nine Btw. the first NQP build that assembles bytecode from QAST directly just finished successfully :) 12:01
lizmat
.oO( more great news ahead, she thought while working on the P6W )
 12:04
12:06 cognominal-p6 left
timotimo weekly: mastodon.social/@timotimo/100848930972850155 12:14
notable6 timotimo, Noted!
12:16 lizmat left 12:18 lizmat joined, p6bannerbot sets mode: +v lizmat
Geth roast: 42801d96ac | (Zoffix Znet)++ | S32-hash/map.t
[v6.d REVIEW] Clarify Map.gist sorts

Orig: github.com/perl6/roast/commit/8a3908c3e		 12:23
12:26 cognominal-p6 joined, p6bannerbot sets mode: +v cognominal-p6
nine Aaaand nqp's test suite passes :) 12:35
Geth nqp: 8b681e9669 | (Elizabeth Mattijsen)++ | tools/build/MOAR_REVISION
Bump Moar to get bit shift fix
 12:47
ยฆ nqp: version bump brought these changes: github.com/MoarVM/MoarVM/compare/2...4-g588873f
12:48 ExtraCrispy left 12:58 cognominal-p6 left 13:07 cognominal-p6 joined, p6bannerbot sets mode: +v cognominal-p6
Geth rakudo: 8bb34ff07c | (Elizabeth Mattijsen)++ | tools/build/NQP_REVISION
Bump NQP to get the bit shift JIT removal fix
 13:08
ยฆ rakudo: version bump brought these changes: github.com/perl6/nqp/compare/2018....7-g8b681e9
cognominal-p6 Eventually Perl 6 will take over Perl 5. So renaming is much ado about nothing. As far as renaming goes, I would like 6lang, but so far Perl 6 fails to deliver about macros. 13:09
13:16 cognominal-p6 left, cognominal-p6 joined 13:17 p6bannerbot sets mode: +v cognominal-p6
stmuk I can't see how an alias could be harmful. At worst it makes no difference and at best it helps 13:31
yoleaux 9 Sep 2018 18:46Z <AlexDaniel> stmuk: can you fix that please? It's Test::META not Test::Meta gitlab.com/stmuk/p6-task-galaxy/bl...6.json#L52
lizmat stmuk: we already have a de-facto alias: Rakudo 13:32
some people call Perl 6 "Camelia" on PerlMonks
people will call it names whatever the "official" name or alias is
stmuk lizmat: I agree but the "its an implementation not the language people" won't 13:34
anyway if larry makes a decision hopefully the whole name issue will go away 13:37
13:38 cognominal-p6 left 13:49 cognominal-p6 joined 13:50 p6bannerbot sets mode: +v cognominal-p6 14:05 cognominal-p6 left 14:06 cognominal-p6 joined 14:07 p6bannerbot sets mode: +v cognominal-p6 14:28 lizmat left 14:39 Kaiepi left 14:40 Kaiepi joined, PufferBot joined 14:41 p6bannerbot sets mode: +v Kaiepi, p6bannerbot sets mode: +v PufferBot
nine And rakudo build finishes, too :) 15:19
15:20 buggable joined, ChanServ sets mode: +v buggable 15:21 p6bannerbot sets mode: +v buggable
Geth nqp/nqp-mbc: ef282200eb | (Stefan Seifert)++ | 6 files
Assemble byte code directly instead of assembling a MAST
 15:21
nqp/nqp-mbc: 8aa963db2a | (Stefan Seifert)++ | src/vm/moar/QAST/QASTOperationsMAST.nqp
Work around order issue when compiling repeat loops
nqp/nqp-mbc: 50ed735ac6 | (Stefan Seifert)++ | src/vm/moar/QAST/QASTOperationsMAST.nqp
Fix mbc stage of nqp::if when the then branch doesn't return a result
rakudo/nqp-mbc: 4de6b8596b | (Stefan Seifert)++ | src/vm/moar/Perl6/Ops.nqp
Get rid of MAST
 15:22
roast: ef369a45ea | (Zoffix Znet)++ | S05-modifier/ignorecase.t
[v6.d REVIEW] Fix ignorecase ligature test descriptions

Orig: github.com/perl6/roast/commit/946d2ff79		 15:26
roast: 4c3778a4db | (Zoffix Znet)++ | S06-signature/slurpy-and-interpolation.t
Remove trailing whitespace
 15:28
roast: 6e52734322 | (Zoffix Znet)++ | S12-introspection/attributes.t
[v6.d REVIEW] Fix typo in attr test desc
 15:30
15:32 fake_space_whale joined
Geth roast: 3bef51542a | (Zoffix Znet)++ | S32-exceptions/misc.t
[v6.d REVIEW] Use more unique name in type suggestion error

In case we ever introduce some similar-named type and it'd end up being in the list of suggestions.
Orig: github.com/perl6/roast/commit/3014999fc		 15:32
15:33 p6bannerbot sets mode: +v fake_space_whale 15:51 lizmat joined, p6bannerbot sets mode: +v lizmat
Geth nqp/nqp-mbc: 534d81b817 | (Stefan Seifert)++ | 2 files
Get rid of push_ilist

Since we don't use the assembled instruction lists any longer, there's no need to append lists anymore.		 16:09
nqp/nqp-mbc: 131c6d29ef | (Stefan Seifert)++ | 2 files
Remove @il argument of push_op

It's now just a wrapper for MAST::Op.new_with_operand_array
AlexDaniel releasable6: next 16:36
yoleaux 16:15Z <jmerelo> AlexDaniel: thanks!
releasable6 AlexDaniel, Next release in โ‰ˆ13 days and โ‰ˆ2 hours. 1 blocker. 0 out of 371 commits logged
AlexDaniel, Details: gist.github.com/99933c2fb138a502ae...fff4fbcb8a
AlexDaniel holy moly 371 commits
lizmat AlexDaniel: js branch :-) 17:14
18:03 MasterDuke joined, p6bannerbot sets mode: +v MasterDuke, MasterDuke left, MasterDuke joined, herbert.freenode.net sets mode: +v MasterDuke, p6bannerbot sets mode: +v MasterDuke
Geth rakudo/nqp-mbc: b9fc202cac | (Stefan Seifert)++ | src/vm/moar/Perl6/Ops.nqp
Take the list ouf of MAST::InstructionList
 18:09
rakudo/nqp-mbc: 3f9eca850d | (Stefan Seifert)++ | src/vm/moar/Perl6/Ops.nqp
Get rid of push_ilist
18:21 ggoebel left 18:36 ggoebel joined 18:37 p6bannerbot sets mode: +v ggoebel 18:41 brrt joined 18:42 p6bannerbot sets mode: +v brrt
brrt \o 18:44
since the whole language name debate has come up, again, I will say the only thing I have to say about it here 18:51
Perl is a scripting language. All perls have always been 18:55
The potential of perl 6 is to be both suitable for adhoc as well as systematically developed programs 18:56
To have concurrency and scalability and oneliners in one language 18:57
Also to have good performance, for scripting notions of this anyway, and a language you can extend to do unexpected things 18:59
If we, as a community, can realize this, then nobody in the world is going to care what we call it
If perl6 is the language it could be, then people *will* use it, under *any* name
... or in other words, what we need is people to think 'I was going to do it in ruby/php/python/javascript, but I decided to use perl6 instead because it was better that way' 19:01
aiming for anything less than that is pointless
now I'm happy to admit that this will take a long, long time
lizmat brrt: +1
brrt as a meta topic, I'd say that the whole 'scripting' community is facing headwinds these days; and frankly, I see little potential from any other language than perl6 in turning that arround 19:03
pmurias brrt: static typing/performance and tools support seems to be the things that the 'scripting' community is loosing on and they aren't our strong points :( 19:07
brrt pmurias: agreed. There's actually two things going on, and one of them is easily understood, and the other is a bit weird 19:08
the easy one is performance - single threaded performance has hit a plateau. But there are more users (devices) online then ever. So either we need more boxes, or we need to do more work per box. 19:09
It used to be the case, not so long ago, that the notion 'programmer time is cheapr than machine time' was beyond doubt
These days people are writing rust, so that kind of disproves that 19:10
The static typing thing is less easy to explain 19:11
I may be cynical, but I think I *can* explain it 19:12
Bootcamps
So the just-so story is this
- first, we have an increase in the economic potential of computer technology, basically because more people (and devices) are connected for longer times, and they can do more 19:13
- and because technologically intermediated transactions are typically better (faster, more reliable, more transparent, cheaper) than traditional methods, a lot of economic activity gradually moves online 19:15
- because that happens, there is a huge demand for people who have skills to manage/enable this transition, i.e. programmers
- because of this huge demand, programming computers is suddenly a Good Carreer 19:16
It wasn't that way not so long ago
- because of this new status of a Good Carreer, there is an influx of people that wouldn't otherwise be programming, and many of them not via traditional paths, i.e. bootcamps 19:17
How's that related to static typing, you might ask. 19:18
well, static typing is a defense mechanism; it locks down the set of behaviours of a program 19:19
This is going to sound elitist but isn't really; if there's a lot of influx of new programmers, a lot of them are going to be inexperienced 19:21
geekosaur note how many people start looking at haskell because they think static types are a magic bullet, expressed as "if it typechecks, it'll work first time" (whichโ€ฆ no)
brrt programmers that are less experienced are more likely to make the kind of mistakes that a type system defends against 19:22
geekosaur it helps, but it's no magic bullet
brrt programmers that are more experienceed will mentally apply those type constraints and are more likely to encounter situations in which the type system prevents them from doing things that they want and know are valid 19:23
(for instance, I'm working a bunch in java lately, and if there's anybody anywhere that relies on a class hierarchy being met, good luck in slipping in your modification in) 19:24
geekosaur: indeed. I see the same with rust
Kaiepi the only experience i have with java is stack overflow and nqp 19:25
and their documentation
brrt so the system is this: more economic value of programs <- More demand for programmers <- more influx of inexperienced programs <- higher value of defense against 'beginners mistakes'
pmurias brrt: seems a bit elitist, a lot of better coders like static typing too 19:27
brrt ultimately, that influx will be balanced, the average experience of a programmer will increase, the value of a static type system will decrease and its cost will increase
pmurias and looking at a random list of 'top bootcamps' a lot of them are training scripting programmers 19:28
brrt proving my point exactly
Well, the other coin I can flip is that static type systems are more valuable in large codebases, and large codebases are a beginners mistake :-P 19:29
geekosaur pmurias, for different reasosn though
brrt I can give you a more pessimistic story, still 19:30
geekosaur static types are a tool, helpful in some cases, wrong tool in others
brrt the advantage of scriptng languages is that they allow you to whip up something without a lot of fuss
it is generally harder to make a 'robust' system out of scripts 19:31
geekosaur and tring to extend tat tool too far leads to something like tim sheards omega; it has some neat concepts, but those concepts are also why you've never heard of it
brrt
.oO( What is omega? )
geekosaur it's a functional programming language where the type level language is identical to the value level. and has its own "types of types" level language, which is again the same, and on up indefinitely 19:32
brrt now making something quickly without a lot of fuss is *very* valuable in a dynamic market where you (and everybody else) is trying to do everything as quickly as possible
right.... 19:33
doing something robust and well executed and perhaps a bit slower than otherwise would be the case, that is typical of well-established, entrenched markets
geekosaur so, once you start mucking with movig stuff to type level, you discover that it's less expressive than value level, but checking some things at type level requires more expressiveness. which leads to dependent types
omega takes that to its irrational limit, every level is fully expressive and there's an infinite hierarchy of levels. which makes it hard to keep track of what level you're on introducing whole new kinds of weird bugs 19:34
and incomprehensibility 19:35
it's nciely symmetric, but symmetry turns out to equate to confusion for people
brrt :-) 19:36
pmurias brrt: I think a bigger factor is that the newer static languages are better nowadays then the old ones
brrt that may well be part of it
but, I've been learning rust, and ...
well, hmm
have you seen lifetime annotations? 19:37
pmurias brrt: so it's python vs go not python vs (plainly disgusting) C++ or python vs old school java without closers
* closures
brrt hehe, go is also on the receiving end of a lot of hate
Kaiepi i'm surprised no one's brought up js
geekosaur and you do always have that confusion between ease of use vs. stringency 19:38
Kaiepi guess there's no point in beating a dead horse though
geekosaur (I'd say strictness btu I've already mentioned haskell, where that means something else)
pmurias brrt: rust being complex is a benefit in some ways
geekosaur ehhhh, that's potentially getting back to elitism
brrt (js is nice enough)
hehehehe 19:39
pmurias brrt: people have something to learn and a such I guess are more invested in it
brrt if by invested you mean 'ready to join the evangelism strike force' :-P
jdv79_ the java over perl thing is not new 19:40
brrt also, if you want to see the elitism here, then you can also say that static types are a defense of traditinally trained developers against bootcampers
19:40 MasterDuke left
jdv79_ i remember experiencing and hearing about migration projects like that since the early 2000s 19:40
brrt well, the early 2000s were less dynamical than the early 2010s 19:41
jdv79_ so the bootcamper thing may have added to it but it existed before them
pmurias brrt: you can do all sorts of wizardy with macros and stuff in rust and you can actually feel you are learning something at least interesting 19:42
brrt I'm not saying that rust isn't an interesting language. It certainly is. And super ambitious
What I will say is that it is a usability nightmare. And I think that this is part of the attraction for many. 19:43
geekosaur go look at folks who were doing clever things with shell scripts in the late 70s/early 80s
โ€ฆthen at folks doing the same with ibm jcl in the 60s
brrt :-)
what hopefully will happen is that there emerges a domain of work that isn't neatly expressed by static languages 19:44
pmurias brrt: I don't imagine having fun playing with go (as opposed to building stuff in it)
s/don't/can't
brrt I'm saying hopeful because I think that this situation is more interesting than reengineering the same things to be faster in a static language 19:45
pmurias Perl 6 is more in Rust like in that regard
in regards to speed graalvm trying to capture the scripting market is interesting 19:46
19:54 AlexDaniel left, AlexDaniel joined 19:55 p6bannerbot sets mode: +v AlexDaniel
brrt Certainly 19:56
(well, not that perl6 is like rust. rust is the anti-adhoc engineering language) 19:57
Kaiepi still mad rust doesn't have openbsd as a very high priority on their list of languages to support 19:58
pmurias brrt: I view Rust as a very perlish design as it tries to solve problems by sticking a bunch of stuff in 20:01
brrt I view it as 'C++, but with more theory' 20:02
I guess you can make the point that perl is kind of like that as well
20:16 brrt left, gfldex_ is now known as gfldex, gfldex left, gfldex joined 20:17 p6bannerbot sets mode: +v gfldex 21:10 pmurias left 21:12 pmurias joined, p6bannerbot sets mode: +v pmurias
Geth 6.d-prep: a46078fa92 | (Zoffix Znet)++ (committed using GitHub Web editor) | TODO/README.md
Naming TODO is now completed
 21:13
21:19 AlexDaniel left, robertle left, AlexDaniel joined 21:20 p6bannerbot sets mode: +v AlexDaniel 22:05 dogbert21 joined, dogbert21 left, dogbert2_ left 22:08 pmurias left, pmurias joined, p6bannerbot sets mode: +v pmurias 22:13 lizmat left 22:14 ufobat_ left
japhb New module zef install failure at Rakudo HEAD (my full rebuild script from last night failed earlier than this, so I'm not sure what commit broke it): 22:18
===> Testing: IO::Path::ChildSecure:ver<1.001011>
Type check failed for return value; expected Match but got IO::Path (IO::Path.new("/tmp/perl6_roas...)
in sub make-rand-path at t/01-operation.t line 16
in sub make-temp-dir at t/01-operation.t line 33
in block <unit> at t/01-operation.t line 58
AlexDaniel o nice 22:58
23:01 lizmat joined, p6bannerbot sets mode: +v lizmat
AlexDaniel ๐Ÿฅž Bisecting IO::Path::ChildSecure 23:03
:)
it doesn't have any dependencies, unfortunately, so not that fancy
23:12 dct left
AlexDaniel japhb: I think it is this commit: github.com/rakudo/rakudo/commit/3b...59ddccc3cd 23:13
yoleaux 23:12Z <uzl> AlexDaniel: Not directly related with a Russian tutorial but I've been working in a Spanish translation of
23:12Z <uzl> AlexDaniel: 'Think Perl 6' and a while back I asked for a (community?) review of the translation. Lately
AlexDaniel japhb: at least my new fancy tool says that it is that commit, it'd be nice if you could verify it 23:15
I didn't look at the code, and the commit does not look like it is related 23:16
but โ€œbisectable is always rightโ€ and this tool is built upon bisectableโ€ฆ who knows! 23:17
japhb AlexDaniel: OK, will check 23:22
AlexDaniel: Running a build at that commit, and after that will try the commit right before it. 23:23
AlexDaniel japhb: awesome! Can't wait to know the result
23:24 MasterDuke joined, p6bannerbot sets mode: +v MasterDuke 23:28 MasterDuke left, MasterDuke joined, herbert.freenode.net sets mode: +v MasterDuke, p6bannerbot sets mode: +v MasterDuke
japhb Wow, that rev blows up spectacularly -- it won't even install zef itself. `cd zef; .../perl6/rakudo-moar-3b453c575de14efd3abfa4f49c82a059ddccc3cd/install/bin/perl6-m -Ilib bin/zef install .` gives: When invoking 33 '', provided outer frame 0x55562adf72e8 (62 '') does not match expected static frame 0x55562adf73e0 (34 '') 23:44
Trying previous
AlexDaniel damn, I guess we got a bit unlucky with that bisect 23:45
timotimo turn spesh off? :) 23:48
japhb timotimo: I'm trying to minimize variables changed. 23:49
Previous rev is making it farther.

Powered by Cro and the Rakuยฎ Programming Language.

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

Thank you!