Perl 6 language and compiler development | Logs at colabti.org/irclogger/irclogger_logs/perl6-dev | For toolchain/installation stuff see #perl6-toolchain | For MoarVM see #moarvm
Set by AlexDaniel on 12 June 2018.
00:15 BenGoldberg joined 00:28 MasterDuke joined 02:13 dogbert2 joined 02:16 dogbert17 left 03:01 stmuk_ joined 03:04 stmuk left
Geth nqp/truffle: 088087cbf0 | (Daniel Green)++ | 17 files
[truffle] Implement a bunch of _i and _n math ops

nqp::<abs add div gcd lcm mod mul neg sub>
03:32
04:56 lizmat left 05:20 BenGoldberg left
[TuxCM] Rakudo version 2018.06-52-g4bdb9787d - MoarVM version 2018.06-36-g6646038f5
csv-ip5xs0.900 - 0.912
csv-ip5xs-207.198 - 7.289
csv-parser26.083 - 26.161
csv-test-xs-200.449 - 0.483
test8.280 - 8.866
test-t2.303 - 2.376
test-t --race1.017 - 1.255
test-t-2042.563 - 43.910
test-t-20 --race16.451 - 16.640
06:24
06:28 |Tux| joined 06:29 robertle joined 06:57 brrt joined
Geth roast: JJ++ created pull request #443:
Adds tests for Distribution::Hash
07:06
07:38 pmurias joined
Geth rakudo: 2ff467647f | (Samantha McVey)++ | tools/build/makeUNIPROP.pl6
Update makeUNIPROP.pl6 to Unicode 10 property names

Update to the same version MoarVM uses.
07:44
rakudo: a9e844a8a7 | (Samantha McVey)++ | 2 files
Remove propname->propname mapping from Cool.pm6

We don't need this anymore since MoarVM/MoarVM#448 has been fixed.
Also update the propname preferences to Unicode 10.0
08:35 Ven` joined
Geth nqp/truffle: cb15b3fe2c | (Paweł Murias)++ | src/vm/jvm/bin/run_tests.pl
[truffle] Remove failing test from list of passing tests
08:36
nqp/truffle: adfa7aacba | (Paweł Murias)++ | nqp-truffle.nqp
[truffle] Fix wrongly specified type mistake
08:51
nqp/truffle: 6186280233 | (Paweł Murias)++ | 4 files
[truffle] Implement nqp::bitor_i, nqp::bitand_i, nqp::bitxor_i
nqp/truffle: 90a35870ac | (Paweł Murias)++ | nqp-truffle.nqp
[truffle] Use a human readable type name instead of a number in error messages
nqp/truffle: 4 commits pushed by (Paweł Murias)++ 09:01
nqp/truffle: 97dec8ec2f | (Paweł Murias)++ | src/vm/jvm/bin/run_tests.pl
[truffle] Add passing test 13 and 16
09:07
09:11 Ven` left 09:20 Ven` joined 09:27 lizmat joined
Geth nqp/truffle: fcdc348377 | (Paweł Murias)++ | nqp-truffle.nqp
[truffle] Use more whitespace for --target=tast
09:34
pmurias MasterDuke: I'm investigating how the other truffle implementations are doing closures efficently, once we have that figured out, we can tackle a bit more complex tests 09:36
Geth rakudo: lizmat self-assigned Regression with .comb on IntStr zero with leading zeros github.com/rakudo/rakudo/issues/2010
f8d6a3f423 | (Elizabeth Mattijsen)++ | src/core/Str.pm6

Fixes GH#2010.
In GH#1552 it was decided that the boolean value of allomorphs is decided by the numerical value, not by the string value. So from that, it *could* follow that the behaviour is correct. However, since `.comb` clearly works on the the string part of the allomorph, it seemed like the right thing to do to use the string semantics of Bool instead in this case.
09:51
synopsebot GH#2010 [open]: github.com/rakudo/rakudo/issues/2010 Regression with .comb on IntStr zero with leading zeros
synopsebot GH#1552 [closed]: github.com/rakudo/rakudo/issues/1552 [@LARRY][RFC] Booliness of Zero-Denominator Rationals
roast: 32157bd352 | (Elizabeth Mattijsen)++ | S02-literals/allomorphic.t
Add test for R#2010
10:34
synopsebot R#2010 [open]: github.com/rakudo/rakudo/issues/2010 Regression with .comb on IntStr zero with leading zeros
Geth rakudo/rescalar: 42 commits pushed by (Jonathan Worthington)++
review: github.com/rakudo/rakudo/compare/d...ac5defeb3d
10:37
jnthn (rebase) 10:38
Geth rakudo: 72ccd432a7 | (Elizabeth Mattijsen)++ | src/core/Str.pm6
Better fix for GH#1020

Let the empty string only suffer from having an iterator created that doesn't do anything, instead of *all* strings suffering from an extra check.
10:48
synopsebot GH#1020 [open]: github.com/rakudo/rakudo/pull/1020 [needs review] Revert "forbid to put variable in initializer of its declaration"
lizmat argh, that should have been #2010
yoleaux 08:44Z <brrt> lizmat: I feel (and git could probably tell me) that i'm no longer the major contributor of expression templates, and that seems like something worth mentioning in this weeks weekly
09:18Z <jnthn> lizmat: It was overall up to date, though a bit off in a few places. I've updated it.
lizmat GH#2010 10:49
synopsebot GH#2010 [closed]: github.com/rakudo/rakudo/issues/2010 Regression with .comb on IntStr zero with leading zeros
lizmat weekly: 08:44Z <brrt> lizmat: I feel (and git could probably tell me) that i'm no longer the major contributor of expression templates, and that seems like something worth mentioning in this weeks weekly 10:50
notable6 lizmat, Noted!
Geth rakudo/rescalar: 78cb739a78 | (Jonathan Worthington)++ | src/Perl6/Actions.nqp
Fix p6scalarwithvalue

We still need to do the type checking and Nil checking, so stick it through `p6assign`. Still very worth it, for the sake of saving the throwaway setup of `$!value` to `$!default`.
10:52
11:15 Ven` left 12:09 Ven` joined 12:10 MasterDuke left 13:09 j3nnn1 joined 13:21 Ven` left 13:22 Ven` joined 13:29 skids joined 14:21 Ven` left 14:30 Ven` joined 15:14 brrt left
Geth rakudo/rescalar: 63ecf2862c | (Jonathan Worthington)++ | src/vm/moar/spesh-plugins.nqp
Re-work how we guard Mu-typed assignments

Instead of building up a sizable guard set for the various variables that may be assigned at that site, we just guard against the Mu type itself. There may be ways we can be smarter later on, but this should mean we do rather better with all the code that is generic over many Scalar containers with different descriptors.
15:20
AlexDaniel reportable6: 2018-06-25T00:00:00Z 2018-07-02T00:00:00Z 15:41
reportable6 AlexDaniel, Couldn't find a snapshot for 2018-07-02T00:00:00Z (try 「list」 command to see what's available).
15:42 reportable6 joined
AlexDaniel :S 15:42
lizmat notable6: weekly 15:49
notable6 lizmat, 14 notes: gist.github.com/b6df803b619223078d...943600af5f 15:50
AlexDaniel reportable6: 2018-06-01T00:00:00Z 2018-06-30T06:00:00Z 15:51
reportable6 AlexDaniel, OK, working on it! This may take up to 40 seconds
AlexDaniel, gist.github.com/99cfc68e645cdbd9f3...26e7596c66 15:52
AlexDaniel weekly: monthly reportable: gist.github.com/99cfc68e645cdbd9f3...26e7596c66
notable6 AlexDaniel, Noted!
AlexDaniel and weekly one will be available in one hour… Gah 15:53
I don't know where this instability comes from lately 15:54
lizmat AlexDaniel: no worries, quite a few hours before I publishj :-) 16:02
16:03 robertle left
lizmat dinner& 16:43
16:46 Ven` left 18:21 Ven`` joined, lizmat left 18:46 ilogger2 joined 18:57 lizmat joined 19:02 lizmat left
dogbert2 .seen broken-robot 19:04
yoleaux I haven't seen broken-robot around.
AlexDaniel dogbert2: it's Zoffix
dogbert2 I know :) 19:05
I have fixed GH #1990 as per Zoffix++ instructions. Should I send a PR or is it better if someone else does this fix? 19:06
R #1990 19:07
hmm
AlexDaniel R#1990 19:09
synopsebot R#1990 [open]: github.com/rakudo/rakudo/issues/1990 [QAST][easy to resolve][good first issue][perf] Opt opportunity in METAOP_ASSIGN with % and @ vars
AlexDaniel dogbert2: send a pr
dogbert2 AlexDaniel: will do 19:10
AlexDaniel dogbert2: it also says “It's possible the tests covering the behaviour are sparse and it might be a good idea to write a few to cover this opt.” so maybe the issue is not that easy after all, if you write tests also :) 19:14
dogbert2 perhaps I should give up immediately :) 19:21
Geth rakudo: dogbert17++ created pull request #2012:
Fix R#1990
19:38
synopsebot R#1990 [open]: github.com/rakudo/rakudo/issues/1990 [QAST][easy to resolve][good first issue][perf] Opt opportunity in METAOP_ASSIGN with % and @ vars
19:38 lizmat joined
dogbert2 let's see how long this PR survives :) 19:39
lizmat dogbert2: why not elsif $sigil eq '@' || $sigil eq '%' ?? 19:40
yoleaux 19:08Z <jmerelo> lizmat: I have updated the front page of the survey repo with links to all charts perl6.github.io/p6survey/
dogbert2 lizmat: no good reason at all. Should I change it? 19:42
lizmat not especially, I guess the only difference is really fewer ops genned for the optimizer 19:43
hardly something to worry about
I was just wondering :-)
dogbert2 I'm a noob :) 19:45
AlexDaniel weekly: reportable weekly: gist.github.com/e92ec2d3e43c60eceb...3d974861a2 20:02
notable6 AlexDaniel, Noted!
AlexDaniel wrong link
weekly: oops. reportable weekly: gist.github.com/75578b2a18739d5fc8...5a785b34fd 20:03
notable6 AlexDaniel, Noted!
AlexDaniel weekly: 20:04
notable6 AlexDaniel, 17 notes: gist.github.com/d544a16daf061f00b9...106f22b9f1
AlexDaniel if only these links were clickable :)
Geth rakudo: 6668308097 | (Stefan Seifert)++ | src/core/ForeignCode.pm6
Port fix from 218f8c4600cc14f to EVAL :lang<Perl5>

Many thanks to ugexe++ for pointing this out!
20:10
dogbert2 lizmat: Zoffix suggested the same fix as you did. That's two against one, fix pushed :) 20:21
lizmat :-) 20:22
Geth rakudo/master: 4 commits pushed by (Jan-Olof Hendig)++, (Elizabeth Mattijsen)++ 20:29
synopsebot R#1990 [open]: github.com/rakudo/rakudo/issues/1990 [QAST][easy to resolve][good first issue][perf] Opt opportunity in METAOP_ASSIGN with % and @ vars
20:43 Ven`` joined 20:44 pmurias joined 20:45 dct joined
lizmat and another Perl 6 Weekly hits the Net: p6weekly.wordpress.com/2018/07/02/...-surveyed/ 20:51
20:56 Ven`` left
Geth ¦ star: stmuk self-assigned Win installer does not set up file associations github.com/rakudo/star/issues/115 20:59
¦ star: stmuk self-assigned WSL target modules-install-m failed for "make install" and "make modules-test" github.com/rakudo/star/issues/114 21:01
¦ star: stmuk self-assigned "make modules-test" failure on oo-monitors github.com/rakudo/star/issues/113 21:02
21:10 MasterDuke joined 21:33 pmurias left 21:56 dct left 22:51 skids joined
Geth nqp/truffle: 4c369044d6 | (Daniel Green)++ | 10 files
[truffle] Implement a bunch of _n numeric ops

nqp::<ceil exp floor inf log nan neginf pow sqrt>
23:43