Zoffix m: class { has Str:D $.foo␤} 03:24
camelia rakudo-moar 5ca24c: OUTPUT«===SORRY!=== Error while compiling <tmp>␤Variable definition of type Str:D requires an initializer␤at <tmp>:2␤------> <BOL>⏏}␤»
Zoffix m: class { has Str:D $.foo;␤}
camelia rakudo-moar 5ca24c: OUTPUT«===SORRY!=== Error while compiling <tmp>␤Variable definition of type Str:D requires an initializer␤at <tmp>:1␤------> class { has Str:D $.foo⏏;␤ expecting any of:␤ constraint␤»
Zoffix LTA
m: try { class { has Str:D $.foo;␤}; CATCH { say .^name } 03:27
camelia rakudo-moar 5ca24c: OUTPUT«===SORRY!=== Error while compiling <tmp>␤Variable definition of type Str:D requires an initializer␤at <tmp>:1␤------> try { class { has Str:D $.foo⏏;␤ expecting any of:␤ constraint␤»
Zoffix m: try { EVAL 'class { has Str:D $.foo;␤};'; CATCH { say .^name }
camelia rakudo-moar 5ca24c: OUTPUT«===SORRY!=== Error while compiling <tmp>␤Missing block␤at <tmp>:2␤------> };'; CATCH { say .^name }⏏<EOL>␤»
Zoffix m: try { EVAL 'class { has Str:D $.foo;␤};'; CATCH { say .^name } }
camelia rakudo-moar 5ca24c: OUTPUT«X::Syntax::Variable::MissingInitializer␤===SORRY!=== Error while compiling /home/camelia/EVAL_0␤Variable definition of type Str:D requires an initializer␤at /home/camelia/EVAL_0:1␤------> class { has Str:D $.foo⏏;␤ expecting any…»
Zoffix m: try { EVAL 'class { has Str:D $.foo};'; CATCH { say .^name } }
camelia rakudo-moar 5ca24c: OUTPUT«X::Syntax::Variable::MissingInitializer␤===SORRY!=== Error while compiling /home/camelia/EVAL_0␤Variable definition of type Str:D requires an initializer␤at /home/camelia/EVAL_0:1␤------> class { has Str:D $.foo⏏};␤ expecting an…»
Zoffix *sigh* what an annoying bug I just experienced. 03:41
Was getting this error: gist.github.com/zoffixznet/105ee43...c8db3a029d while trying to `use` and `.new` this class: gist.github.com/zoffixznet/62cc1e2...429d6ffce9 03:42
And when I commented out everything but `unit` it went away and when I uncommented it all back no longer got the error :S What the f...
Method 'update' not found for invocant of class 'Ticket::Trakr::Tickets' 03:43
even though it's there!
removing lib/.precomp brought me back to the original error 03:44
*sigh*
Zoffix writes a raging blog post blaming nine for everything
Seems a combination of `use` Perl 6 module and `use` a Perl 5 module 03:46
Zoffix will file a bug report tomorrow
nine psch: when counting, don't forget possible multiple pre_deserialize and post_deserialize code refs that may accompany a comp unit. 06:13
psch nine: yeah, that's a point 06:24
nine Zoffix: does your Perl 5 using Perl 6 modules have an EXPORT routine? If yes, I may know what's going wrong. 06:33
[Tux] This is Rakudo version 2016.06-17-g5ca24cc built on MoarVM version 2016.06 06:46
test 16.137
test-t 9.756
csv-parser 21.924
jnthn morning, #perl6-dev :) 08:35
psch hm 09:28
so i build moar yesterday, and didn't have my jvm work in a branch
hence i'm on HEAD now
and there i get "Unhandled exception: getextype needs an object with VMException representation" 09:29
interestingly, adding --ll-exception gives "An exception occurred while evaluating a BEGIN" instead
i think this is a strong hint that everything's borked /o\ 09:30
dalek p: be9b89c | (Steve Mynott)++ | README.pod:
Tweet docs a little. Mostly toning down parrot refs
 10:28
p: 8f72fc9 | jnthn++ | README.pod:
Merge pull request #292 from stmuk/master

Tweet docs a little. Mostly toning down parrot refs
Zoffix nine, it's Perl 6 using Perl 5 modules. And I've just tried with modules none of which export anything and still same bug 10:43
But if the Perl 6 module is one included with Rakudo, the error is gone 10:45
It seems to stop updating the precomp too 10:57
nine, here's the ticket: rt.perl.org/Ticket/Display.html?id=128457
BrokenRobot And zef seems to break after installing a couple of modules: gist.github.com/zoffixznet/f954a82...e6f8baae3d 12:41
This is the oldest already-built Rakudo I got where I can still reproduce the bug. The newest being one built 18 hours ago: This is Rakudo version 2016.05-149-g66ebd86 built on MoarVM version 2016.05-17-g6075599 12:44
BrokenRobot tries older releases 12:46
bartolin psch: I had that problem on JVM after the merge of jnthn's branch as well: irclog.perlgeek.de/perl6-dev/2016-0...i_12688097 13:03
psch: another problem occured a bit later: irclog.perlgeek.de/perl6-dev/2016-0...i_12688097 13:04
psch bartolin: yeah, it's somewhat discouraging
bartolin unfortunately I didn't find the time to submit a rakudo bug, yet. Maybe I'll manage to do that on the weekend 13:05
psch although i obviously don't want to discourage development on moar either...
bartolin yes, I feel the same. so, jnthn++ :-)
psch but jvm breakage just keeps getting worse and, well, i am noticeably out of depth with the precomp/BEGINEVAL breakage already... vOv
bartolin I'm afraid, I'm totally unable to help there :-/ 13:06
psch well, it's really hard to pull apart what exactly is happening i think 13:08
bartolin psch: I think the problem you mentioned could be worked around with some '#?if jvm' statements (effictively reverting e5443765d4 for JVM ...
psch well, it's worth a try i suppose 13:09
bartolin *effectively
psch although we probably should rather just fix nqp-j sufficiently not to need that... :)
bartolin ++psch
*g*
bartolin goes back to $work 13:10
BrokenRobot Alright, 2016.04 release does *not* have my bug. 13:11
moritz BrokenRobot: a job for bisectable? 13:26
BrokenRobot AFAIK it can't do precomp stuff 13:31
I'm not gonna hunt down the exact commit, just trying to narrow it down a bit (and find a version I can use in the interim)
Somewhere between 2016.04 and 2016.05 releases 13:34
btyler_ timotimo: the comment about schedulers came from TimToady's comments during/after www.youtube.com/watch?v=73-XQEn6HfI (woo for YAPC organizers getting videos up so quickly!) 13:41
dalek kudo/nom: c3bb416 | lizmat++ | lib/Test.pm6:
Remove what doc we have in Test.pm6

And refer to what is in the doc repository, and what is available online on doc.perl6.org		 14:04
lizmat bikeshedding the text welcomed :-) 14:05
BrokenRobot Very well :) The URL should be https:// :) 14:15
dalek kudo/nom: 7b2710b | lizmat++ | lib/Test.pm6:
s/http/https/, Zoffix++
 14:19
BrokenRobot \o/
dalek kudo/nom: 053729c | timotimo++ | src/Perl6/Metamodel/BOOTSTRAP.nqp:
mention modifiable constraint in bind fail

for example, when you try to pass a constant int to a function with a "int $foo is rw" argument.		 17:18
kudo/nom: fd98fc3 | lizmat++ | src/core/ (9 files):
Remove support for count-only|bool-only

Finally agreed these were premature optimizations. This should make
  .elems always do the same as numerifying (so cache the iterator).		 19:37
lizmat as discussed with TimToady and pmichaud at TPC Orlando 2016 19:38
masak .oO( the three primary phases of a compiler/runtime are: compile time, runtime, and premature optimization time ) 19:39
nine likes to have his premature optimizations before breakfast 19:42
And no, I don't know what the hell this could mean either ;) 19:43
lizmat it was an afternoon snack for me :-) 19:44
more than 200lines less in the setting, and they were all mine afaik :-)
dalek kudo/ugexe-dist-interface: d513259 | ugexe++ | / (4 files):
Implement Distribution interface

  + Distribution is now a role enforcing an interface
  + Basic local file system based implementations
  + Socket based distribution file IO now possible
  + `resources` get prefixed with `resources/` before hashing
  + META6 provides install format changed from:
  `name => ext => { :file(), :time(), ... }` to
  `name => $file => { :time(), ... }`
which keeps the original format, and only extends upon it		 20:52
kudo/ugexe-dist-interface: 33f80a5 | ugexe++ | / (143 files):
Merge branch 'nom' of github.com/rakudo/rakudo into dist-interface
kudo/ugexe-dist-interface: f6514a7 | niner++ | / (4 files):
Fix backwards compatibility issues with repository versions < 2
kudo/ugexe-dist-interface: 1256808 | niner++ | / (5 files):
Fix backwards compatibility issues with repository versions < 2
 20:53
Zoffix :o 21:06
oh it's not nom
Zoffix goes back to bisecting 21:07
nine I wouldn't dare force pushing to nom :)
timotimo .tell lizmat you think you can figure out how to make comb's pull-one not allocate Int? i tried replacing < with islt_i, but that's not it, so maybe it's $!pos++? 21:52
yoleaux2 timotimo: I'll pass your message to lizmat.
dalek kudo/nom: 09cc2fb | timotimo++ | src/core/Mu.pm:
prevent a bunch of Scalar container allocs with dispatch meths
 22:20
Zoffix lizmat, you have a message 22:39
dalek ast/6.c-errata: 77efe8e | hoelzro++ | / (2 files):
Untodo two tests

They're currently passing		 23:10
Zoffix I'm done bisectioning the bug 23:13
.tell nine the bug I found last night where a P5 module loaded together with a P6 module had errors was introduced by this commit: github.com/rakudo/rakudo/commit/bf59085 23:14
yoleaux2 Zoffix: I'll pass your message to nine.
Zoffix And it's not just any P6 module that gives issues, but one installed by panda or zef

Powered by Cro and the Raku® Programming Language.

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

Thank you!