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.
Kaiepi thoughts on adding jna bindings to libuv to the jvm for stuff like async udp sockets and IO::Socket::Unix ops if my grant gets accepted? 00:15
MasterDuke nqp: my num $s := nqp::time_n; my $a := 3; my $b := 1; while $b < 10_000_000 { $a := $a + $b; $b := $b + 1; }; say($a); say(nqp::sub_n(nqp::time_n, $s)) 00:23
camelia 49999995000003
0.8159172534942627
MasterDuke ^^^ doesn't run (noticeably) faster on my branch, but it does change the allocations from 19999998 BOOTNums to 19999981 BOOTInts 00:24
anyone have a good benchmark i can try? 01:10
japhb MasterDuke: github.com/japhb/perl6-bench has a few 01:33
MasterDuke japhb: i haven't run that in a little while. i don't remember, does it build rakudo/nqp/moarvm itself? and if so, easy to specify my branches? 01:42
such arbitrary limitations sometimes...`Unable to allocate an array of 4607182418800017409 elements` 02:19
fwiw, ^^^ happens when building NQP after attempting to add intify support to spesh's optimize_smart_coerce(), gist.github.com/MasterDuke17/a618b...d984af2d3e is the patch if anyone is interested 03:04
and here's the full command and error gist.github.com/MasterDuke17/2b956...d07facd838 03:06
Geth tap-harness6/dev: 4 commits pushed by (Leon Timmermans)++ 03:39
03:54 ufobat_ joined, ufobat left 03:55 p6bannerbot sets mode: +v ufobat_
Geth roast: 25ff1876e4 | (Nick Logan)++ (committed using GitHub Web editor) | S11-repository/cur-current-distribution.t
Use cmd.exe compatible hash lookup

Now passes on windows
06:29
06:42 jnthn left, moritz left, jdv79 left, krunen left, krunen joined, moritz joined, jnthn joined 06:43 p6bannerbot sets mode: +v krunen, p6bannerbot sets mode: +v moritz, p6bannerbot sets mode: +v jnthn 06:44 MasterDuke left 06:46 lizmat left 06:47 jdv79 joined 06:48 p6bannerbot sets mode: +v jdv79 12:11 lucasb joined, p6bannerbot sets mode: +v lucasb 12:20 llfourn left 12:22 masak left
lucasb AlexDaniel: Hey. Sorry for my statement, I was just trying to be funny. I dunno how to express myself in english :D 12:23
12:24 llfourn joined
lucasb with regard moving an issue from one repo to another, GH allows that, right? 12:24
12:25 p6bannerbot sets mode: +v llfourn
lucasb but well, I certainly don't have the rights to move :) 12:25
(unless you meant to open *a new issue* in user-experience, and then close the rakudo's one) 12:27
12:29 llfourn left, llfourn_ joined 12:30 p6bannerbot sets mode: +v llfourn_ 12:35 lizmat joined, p6bannerbot sets mode: +v lizmat
Geth synopsebot: c8a608e7ee | (Lucas Buchala)++ (committed using GitHub Web editor) | lib/SB/Plugin/GitHub.pm6
Add UE# prefix for user-experience issues
12:44
lucasb if any hack admin could pull synopsebot, I would appreciate :) github.com/perl6/synopsebot#pullin...ng-the-bot 12:50
13:08 masak_ joined, masak_ is now known as masak, p6bannerbot sets mode: +v masak 13:33 MasterDuke joined, p6bannerbot sets mode: +v MasterDuke 13:34 j3nnn1 joined, p6bannerbot sets mode: +v j3nnn1, MasterDuke left, MasterDuke joined, herbert.freenode.net sets mode: +v MasterDuke, p6bannerbot sets mode: +v MasterDuke
Geth nqp: bd7bf00b56 | (Jonathan Worthington)++ | src/vm/moar/QAST/QASTCompilerMAST.nqp
Fix loss of some line number annotations

The annotations live primarily on QAST::Stmts and QAST::Stmt nodes. However, if we had such a node directly as the chlid of a node that was not itself such a node, we'd drop the annotation. This, in combination with some tree simplification done by the Rakudo optimizer, meant that very simple routines like: ... (7 more lines)
13:39
rakudo: 5b231e0999 | (Jonathan Worthington)++ | tools/build/NQP_REVISION
Get an NQP with line number accuracy improvements
13:42
¦ rakudo: version bump brought these changes: github.com/perl6/nqp/compare/2018....1-gbd7bf00
roast: 1e2aa0e018 | (Jonathan Worthington)++ | integration/error-reporting.t
Unfudge now-passing line number test
13:43
lizmat confirms spectest clean 14:07
jnthn Was happy to find an existing spectest to unfudge after that fix, rather than needing to come up with one :) 14:10
timotimo free(): corrupted unsorted chunks 14:15
malloc(): memory corruption
getting a couple of these messages when precomping stuff for moarperf, probably including cro
MasterDuke timotimo: i got gist.github.com/MasterDuke17/d60ab...db09b9f088 when trying to chase one of those down a little while ago 14:17
bugs.kde.org/show_bug.cgi?id=290061 seemed like it had some relevant info 14:18
timotimo wtf
MasterDuke and yeah, it was during precompilation (why i had --trace-children=yes turned on). i tried getting something in gdb, but couldn't figure out how to correctly follow threads/forks/whatever 14:20
timotimo yeah, i haven't figured that out properly yet, either 14:21
MasterDuke timotimo: maybe something could be hardcoded here github.com/rakudo/rakudo/blob/mast...#L264-L273 to start them under gdb> 14:22
that's where i hardcoded a --profile-compile to try and figure out why tools/build/install-core-dist.p6 takes so long 14:24
timotimo mhm 14:25
14:37 titsuki joined, p6bannerbot sets mode: +v titsuki 15:09 Kaiepi left
AlexDaniel lucasb: no need to be sorry, you're actually right :) 16:16
16:18 synopsebot left 16:19 synopsebot joined, ChanServ sets mode: +v synopsebot
AlexDaniel UE#1 16:19
UE#36
synopsebot UE#36 [open]: github.com/perl6/user-experience/issues/36 Should tools support .t6 extension?
AlexDaniel lucasb: ↑ done! Thank you
16:19 p6bannerbot sets mode: +v synopsebot 16:22 ChanServ sets mode: +o AlexDaniel, p6bannerbot was kicked by AlexDaniel (Kicked by AlexDaniel)), ChanServ sets mode: -o AlexDaniel 16:27 Kaiepi joined
lucasb AlexDaniel: thanks :) 16:38
hehe forgot to add a #5 item to R#2627 16:56
synopsebot R#2627 [open]: github.com/rakudo/rakudo/issues/2627 [6.e][RFC][decluttering] Perl 6, 6.d, Diwali, Raku, Rakudo, Moar, NQP, Star, *
lucasb "(5) Revoke/retract the "Raku" mess. Say that it was just an elaborate prank." 16:57
:D
17:10 Kaiepi left 17:13 Kaiepi joined 17:33 Kaiepi left 17:37 Kaiepi joined 17:45 patrickb joined
patrickb Hi everyone! 19:08
yoleaux 20 Jan 2019 10:24Z <nine> patrickb: I wonder if we can salvage the RAKUDO_HOME/NQP_HOME idea. So openSUSE wants binary libs to be stored in lib/lib64. The same is true for virtually all other distros I guess. So why do we store stuff in /usr/share anyway? Perl 5 and Python libs are stored in /usr/lib already.
20 Jan 2019 10:47Z <nine> patrickb: The FHS states that /usr/lib is for "Libraries for programming and packages" while /usr/share is for "Architecture-independent *data*" (emphasis mine). Considering that I find plenty of *.gif, *.png, *.ogg, *.wav and whatever in my /usr/lib I'd guess "if in doubt, put it in /usr/lib" is the rule of thumb
patrickb mst: You there?
.tell Coke Can you give me a hint who's best to talk to about getting TPF involved in GSoC? I know near to nothing about the Perl 5 Community and the TPF people... 19:49
yoleaux patrickb: I'll pass your message to Coke.
AlexDaniel .tell [Coke] colabti.org/irclogger/irclogger_lo...01-21#l129 19:54
yoleaux AlexDaniel: I'll pass your message to [Coke].
AlexDaniel dogbert17: have you tried reverting it? 19:58
timotimo wow, URI::Encode is failing a whole lot of tests 20:07
m: my $RFC3986 = rx/<[0..9A..Za..z\-.~_]>/; my %escapes; for (0..255) { next if chr($_) ~~ /$RFC3986/; .say } 20:09
camelia 0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
47
58
59
60
61
62
63
6…
timotimo m: my $RFC3986 = rx/<[0..9A..Za..z\-.~_]>/; my %escapes; for (0..255) { next if chr($_) ~~ $RFC3986; .say }
camelia 0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
47
58
59
60
61
62
63
6…
timotimo m: my $RFC3986 = rx/<[0..9A..Za..z\-.~_]>/; my %escapes; for (0..255) { next if Bool.pick; .say }
camelia 0
4
5
6
8
12
15
16
17
18
19
20
21
23
29
30
32
33
34
35
36
37
38
39
41
42
43
44
46
51
52
53
56
58
60
62
63
64
65
66
70
73
74
75
76
77
79
80
81
85
86
9…
20:09 MasterDuke left
timotimo m: say "hi" ~~ rx/<[0..9A..Za..z\-.~_]>/; 20:09
camelia 「h」
timotimo m: my $rx = /<[0..9A..Za..z\-.~_]>/; for ^10 { say "hi" ~~ $rx } 20:10
camelia 「h」
「h」
「h」
「h」
「h」
「h」
「h」
「h」
「h」
「h」
timotimo m: my $RFC3986 = rx/<[0..9A..Za..z\-.~_]>/; my %escapes; for (0..255) { next if chr($_) ~~ $RFC3986; chr($_).say }
camelia











20:10 [TuxCM] joined
timotimo oh, maybe it is right 20:11
m: my $RFC3986 = rx/<[0..9A..Za..z\-.~_]>/; my %escapes; for (0..255) { next unless chr($_) ~~ $RFC3986; chr($_).say }
camelia -
.
0
1
2
3
4
5
6
7
8
9
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
Q
R
S
T
U
V
W
X
Y
Z
_
a
b
c
d
e
f
g
h
i
j
k
l
m
n
o
p
q
r
s
t
u
v
w
x…
timotimo m: my $RFC3986 = rx/<[0..9A..Za..z\-.~_]>/; my %escapes; for (0..255) { next unless chr($_) ~~ /$RFC3986/; chr($_).say }
camelia -
.
0
1
2
3
4
5
6
7
8
9
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
Q
R
S
T
U
V
W
X
Y
Z
_
a
b
c
d
e
f
g
h
i
j
k
l
m
n
o
p
q
r
s
t
u
v
w
x…
timotimo m: my $RFC3986_unreserved = rx/<[0..9A..Za..z\-.~_]>/; my %escapes; for (0..255) { next unless chr($_) ~~ /$RFC3986_unreserved/; chr($_).say }
camelia -
.
0
1
2
3
4
5
6
7
8
9
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
Q
R
S
T
U
V
W
X
Y
Z
_
a
b
c
d
e
f
g
h
i
j
k
l
m
n
o
p
q
r
s
t
u
v
w
x…
AlexDaniel timotimo: did you update the module? 20:12
timotimo oh
it's using the "boolifying a regex will use the $_ of its current scope" semantics 20:13
they were just recently broken
what do you mean update?
AlexDaniel aaaaaaaaaaaaaaaaah it was not merged…
github.com/perl6-community-modules...de/pull/11
timotimo oh, well :) 20:14
AlexDaniel ok, now it is…
timotimo better
AlexDaniel timotimo: while you're at it, please leave a comment on R#2608
synopsebot R#2608 [open]: github.com/rakudo/rakudo/issues/2608 [⚠ blocker ⚠] Scope flattening and regexes
timotimo "use dynamic '$_';" is perhaps the solution you want? 20:15
i haven't finished reading yet
20:15 [TuxCM] left
AlexDaniel timotimo: it looks like `use v6.c` is also a way to do it, but with regards to the upcoming release, what should be done? 20:18
timotimo you mean we should wait one more release before the $_-no-longer-dynamic-by-default behaviour is "turned on"?
AlexDaniel more like keep the behavior in v6.c/v6.d but change it for v6.e 20:20
timotimo i'd be interested to see if user programs have it in the wild 20:21
except for URI::Encode
i don't think it'd be easy to put in a deprecation warning
i'm not entirely sure if "with current rakudo" in jnthn's comment is refering to "after the optimization in question" or already before that 20:22
AlexDaniel pretty sure that yeah, there are programs in the wild. It looks like something I could write…
but at the same time it's probably not very common 20:23
timotimo i'd probably always be writing the ~~ out, tbh
making the presence of a regex literal in a scope poison all $_ to become dynamic wouldn't cover the case where a regex in a variable being imported by a module exists 20:25
AlexDaniel yeah… 20:26
timotimo if we could install a "fake" $_ is dynamic in those scopes that contains, for example, a Failure that would asplode when accessed ... but that would prevent cases where you're actually looking for an explicitly declared $_ is dynamic and that wouldn't fly 20:28
lizmat notable6: weekly 20:48
notable6 lizmat, 4 notes: gist.github.com/da3633027045946387...a0b3b7da6a
AlexDaniel reportable6: 2019-01-14T00:00:00Z 2019-01-21T00:00:00Z
reportable6 AlexDaniel, OK, working on it! This may take up to 40 seconds
AlexDaniel, gist.github.com/a820f7fd50ab9c76f6...683f6bc86a 20:49
AlexDaniel lizmat: ↑ 20:50
lizmat AlexDaniel++
notable6: reset 21:32
notable6 lizmat, No notes for “reset”
lizmat notable6: weekly reset 21:33
notable6 lizmat, Moved existing notes to “weekly_2019-01-21T21:33:04Z”
21:40 patrickb left
lizmat and another Perl 6 Weekly hits the Net: p6weekly.wordpress.com/2019/01/21/...-youtubed/ 22:11
moritz: ^^
timotimo lizmat++ 22:24
Geth nqp: 31a8b13852 | (Elizabeth Mattijsen)++ | docs/ops.markdown
Document cpucores/getrusage/RUSAGE constants
22:33
22:43 j3nnn1 left 22:46 Kaiepi left, Kaiepi joined
Geth tap-harness6/master: 5 commits pushed by (Leon Timmermans)++ 23:05
23:21 lucasb left