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.
Geth roast: tbrowder++ created pull request #465:
Rename pm 2 pm6		 01:36
roast: 4d6918ab32 | (Tom Browder)++ (committed using GitHub Web editor) | 222 files
Rename pm 2 pm6 (#465)

  * add tests for =defn implementation
  * renamed modules *.pm to *.pm6
  * add another two tests to ensure .pm modules are tested
  * make internal changes due to renaming .pm to .pm6		 01:40
roast: 5c93ee3168 | (Nick Logan)++ (committed using GitHub Web editor) | S10-packages/basic.t
Remove incorrect-ish test

I would argue the file extension test is not appropriate for the roast. More importantly it uses relative file paths for its `use lib` ( see: github.com/perl6/roast#test-utilities and github.com/perl6/roast/commit/dfe7...c1555ed8c4 )		 02:09
roast: bcc988ecbe | (Nick Logan)++ (committed using GitHub Web editor) | 2 files
Remove extra .pm in filename
 02:13
Geth roast: 741d473506 | 陈梓立++ (committed using GitHub Web editor) | S10-packages/basic.t
correct plan number
 09:34
nqp/truffle: 91a5f6c775 | (Paweł Murias)++ | 4 files
[truffle] Rename scRef to sc
 09:41
pmurias jnthn: any significant reasons why the jvm backend keeps the REPR and REPRData as separate objects? 10:58
yoleaux 28 Aug 2018 22:56Z <AlexDaniel> pmurias: this test does not pass on jvm github.com/perl6/nqp/commit/f0f384...28755fa1f6
timotimo pmurias: probably just taken over from moarvm; REPR_data is variable-sized, that's why it's behind a pointer in moarvm
pmurias wasn't the JVM backend done before MoarVM? 10:59
timotimo er
pmurias (it could be taken from Parrot)
timotimo parrot probably also had REPR_data :)
jnthn The initial 6model implementation was in C; various things were adapted for the JVM, but where it wasn't deemed performance sensitive I stuck with keeping things similar. 11:00
But in C, a REPR is a static thing, whereas REPR data is per type, subject to GC, etc. 11:01
timotimo oh, oops, of course
jnthn And yes, that C was a Parrot extension :)
timotimo well, shows that i still need to finish waking up
jnthn Though the design of the C code survived largely intact in MoarVM 11:02
pmurias I think I'll just merge that for truffle, seem most types actually have some data and this will get rid of casting everywhere 11:12
Geth nqp/truffle: a203dc07e9 | (Paweł Murias)++ | 7 files
[truffle] Stub STables while deserializing
 12:34
roast: tbrowder++ created pull request #466:
Add test for loading module .pm files		 13:07
roast: 5c7df0d522 | (Tom Browder)++ (committed using GitHub Web editor) | 2 files
Add test for loading module .pm files (#466)

  * add tests for =defn implementation
  * add new file and new tests for loading .pm modules		 13:08
rakudo: 656b0730f9 | (Elizabeth Mattijsen)++ | docs/release_guide.pod
Remove chromatic as release manager of 2009.12 release
 13:21
Geth roast: e8864cee30 | (Nick Logan)++ | 2 files
Revert "Add test for loading module .pm files (#466)"

This reverts commit 5c7df0d522a24d352ae229ea75f277fa8310b7a1.
As mentioned in my previous commit I do not think this is appropriate to test in the roast. i.e. the tests we replaced were not for explicitly testing what file extension works, so its not a single persons place to decide that now the spec incorporates specific file extensions in this instance.		 13:34
lizmat is surprised by the revert in roast 13:56
lizmat is also going to be afk for a bit&
|Tux| Rakudo version 2018.06-500-g656b0730f - MoarVM version 2018.06-442-g633604d30
csv-ip5xs0.931 - 0.962
csv-ip5xs-207.885 - 7.979
csv-parser22.905 - 22.970
csv-test-xs-200.454 - 0.468
test8.682 - 9.230
test-t2.015 - 2.059
test-t --race0.875 - 0.890
test-t-2035.560 - 36.366
test-t-20 --race11.544 - 11.912
 15:48
Geth nqp/truffle: 321c790756 | (Daniel Green)++ | 5 files
[truffle] Implement nqp::getlex and _<i n s>
 16:26
lizmat timewasteable6? 17:27
Geth nqp/truffle: ed653e198a | (Daniel Green)++ | src/vm/jvm/runtime/org/perl6/nqp/truffle/nodes/expression/NQPGetlexNode.java
[truffle] Fix wrong class name in getlex
 17:44
nqp/truffle: c893f05b8c | (Daniel Green)++ | 6 files
[truffle] Implement nqp::iscont and _<i n s>
timotimo yeah because if you're a legitimate visitor, your time is wasted by not getting voice immediately 17:48
[Coke] I don't think we need to optimize for "visitors". 18:03
TimToady m: my $x => 42; # interesting little buglet 18:53
camelia 5===SORRY!5=== Error while compiling <tmp>
Preceding context expects a term, but found infix > instead.
at <tmp>:1
------> 3my $x =>7⏏5 42; # interesting little buglet
AlexDaniel TimToady: what would be the right behavior? 19:07
m: my $x = > 42;
camelia 5===SORRY!5=== Error while compiling <tmp>
Preceding context expects a term, but found infix > instead.
at <tmp>:1
------> 3my $x = >7⏏5 42;
AlexDaniel timotimo: also it wastes our time to maintain that stuff :(
timotimo right 19:09
AlexDaniel also I was talking to myself again because I had no voice 19:11
AlexDaniel forgot to voice myself… let's pretend that was intentional :P 19:12
jdv79 what visitors do you mean? 19:28
timotimo anyone coming to the irc channel, perhaps for their first time 19:31
brrt spam makes irc a bit less effective and useful this way 19:32
Zoffix AlexDaniel: I can kill my bot permanently now right? 20:20
tje bannerbot
pretty insane Freenode is still being hammered. 20:22
AlexDaniel: right behaviour in TimToady's example is to create a Pair object 20:23
Zoffix m: (my $x => 42) 20:23
camelia 5===SORRY!5=== Error while compiling <tmp>
Preceding context expects a term, but found infix > instead.
at <tmp>:1
------> 3(my $x =>7⏏5 42)		 20:24
Zoffix m: (my $x) => 42
camelia WARNINGS for <tmp>:
Useless use of "=>" in expression "(my $x) => 42" in sink context (line 1)
Zoffix well, I'm sirmising here
sur
Geth roast: jstuder-gh++ created pull request #467:
Test STOREing / splicing lazy Seq into native array throws		 20:26
Geth roast: 1713da0720 | (Jeremy Studer)++ | S09-typed-arrays/native.t
Test STOREing lazy Seq into native array throws

splicing as well.
Corresponds to the following
  [Rakudo development](github.com/rakudo/rakudo/pull/2240)		 22:01

Powered by Cro and the Raku® Programming Language.

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

Thank you!