00:06 lichtkind left 02:31 unclechu left 02:32 AlexDaniel` left 02:41 unclechu joined 02:55 AlexDaniel` joined
lizmat hmmm... raku.org/downloads still mentions 2020.01 Rakudo Star 09:31
as does rakudo.org/star/
09:33 sena_kun joined
sena_kun I did a blin check twice and both times there are more or less modules (a couple of dozens!) with "Could not obtain blocking, exclusive lock: Failed to lock filehandle: 35", anything rings the bell? 09:34
rba so far there is only a a src and linux binary release of rakudo star. 09:35
tellable6 2020-05-31T15:30:33Z #raku-dev <patrickb> rba colabti.org/irclogger/irclogger_lo...05-31#l191
lizmat ok, I'm going to pull the R* announcement from this weeks RWN 09:38
rba s/linux binary release/windows binary release/ 09:43
lizmat notable6: weekly 09:55
notable6 lizmat, 8 notes: gist.github.com/344af4d0c7ece502b4...29f21cb02f
lizmat notable6: weekly reset 10:02
notable6 lizmat, Moved existing notes to “weekly_2020-06-01T10:02:06Z”
10:09 Altai-man_ joined 10:12 sena_kun left
lizmat and yet another Rakudo Weekly News hits the Net: rakudoweekly.blog/2020/06/01/2020-...by-wenzel/ 11:44
Kaiepi m: proto sub succ(|) {*}; multi sub succ(Int:D $x) { $x + 1 }; multi sub succ(**@xs) { @xs.map(*.succ) }; say succ 1 11:45
camelia 2
Kaiepi does the Int:D candidate here get selected at compile-time?
11:56 leont joined
Kaiepi ah, forgot i could add stuff to the proto body to check 11:56
lizmat afk for a few hours& 12:01
12:10 sena_kun joined 12:12 Altai-man_ left
timotimo it randomly occurs to me that we may want to offer loading a module with a specific structure for stringification in the REPL, controllable via an environment variable (OTOH, security concerns) 12:31
(after seeing RAKU_REPL_OUTPUT_METHOD in the weekly)
12:46 camelCaser left 12:47 camelCaser joined 13:21 lichtkind joined
Geth rakudo/rakuast: 41 commits pushed by (Jonathan Worthington)++
review: github.com/rakudo/rakudo/compare/c...c0ae462af8
13:24
jnthn Just a rebase 13:25
Geth ¦ rakudo: tbrowder self-assigned docs/compiler_overview.pod needs updating github.com/rakudo/rakudo/issues/3734 13:35
13:52 bazzaar joined
[Coke] m: dd "SOMETHING".Date 13:57
camelia Invalid Date string 'SOMETHING'; use yyyy-mm-dd instead
in block <unit> at <tmp> line 1
[Coke] Can someone invoke a bot and tell me when that changed?
timotimo bisectable6: dd "SOMETHING".Date 13:58
bisectable6 timotimo, Bisecting by output (old=2015.12 new=62adc88) because on both starting points the exit code is 1
timotimo, bisect log: gist.github.com/8c087e89ff98de84bf...03e2edc9e0 13:59
timotimo, (2016-09-27) github.com/rakudo/rakudo/commit/22...0f14b9c05c
AlexDaniel` 6c: dd "SOMETHING".Date
committable6 AlexDaniel`, gist.github.com/c1422803ddb5cfb5b9...d3beb2d8cd
AlexDaniel` [Coke]: ↑ which change are you interested in? 14:00
if the most recent one, then 14:02
b: old=2020.02.1 dd "SOMETHING".Date
bisectable6 AlexDaniel`, Bisecting by output (old=2020.02.1 new=62adc88) because on both starting points the exit code is 1
AlexDaniel`, bisect log: gist.github.com/c6e13f4023a676e0b6...73969a48e8
AlexDaniel`, (2020-02-29) github.com/rakudo/rakudo/commit/d9...d198747f09
[Coke] we have a doc test that augments Str by adding Date; It is currently failing.
was trying to figure out if this was a change in Rakudo - doesn't look like it, and this test has been there since 2016, so something weird going on 14:03
github.com/Raku/doc/issues/3455
AlexDaniel` [Coke]: yes, it's caused by a change in rakudo 14:06
bisectable can reliably tell when something changed, that is as long as your question is correct
when you don't know if it changed at all it's the wrong tool to use 14:07
look at the output of committable, it gives the output on all releases, so it's much easier to see the picture
after that it's easier to ask bisectable for the thing you're interested in by providing a narrower range 14:08
so your answer is (2020-02-29) github.com/rakudo/rakudo/commit/d9...d198747f09
14:09 Altai-man_ joined
AlexDaniel` just always use committable first 14:09
arguably bisectable6 and committable6 should trigger at the same time, or maybe both bots should be combined into one 14:11
[Coke] ... this is why I asked someone else to do it. :)
14:11 sena_kun left
AlexDaniel` yes yes that's very nice, I'm just curious why timotimo triggered bisectable and then disappeared 😇 14:12
[Coke] AlexDaniel`++ 14:13
AlexDaniel` I should really fix the user interface of the bots one day, people always use bisectable first which nowadays gives misleading results because the range is so wide (from 2015 to 2020) 14:14
[Coke] AlexDaniel`: so if I had picked something that didn't *fail*, that would have been better, but since it just failed *differently*, that didn't help?
AlexDaniel` maybe it's time to change the default
[Coke]: generally, yes. Then it would've used the exit code, and bisecting by exit code is always more predictable than bisection by output (because any change in the error message will be considered different and as such “new”) 14:15
jnthn Could bissectable always run commitable too, and if it changed more than once in the range of commits being considered, then it can say "oh btw, this result might not be what you want" 14:16
AlexDaniel` jnthn: yeah, something like that, are we willing to wait though? :) 14:17
that is, currently committable is testing 43 revisions 14:18
and bisectable on average tests about 12 or 13 I think
unless it hits a lot of unbuildable commits
[Coke] (yay doc testing that found this was no longer a good example!) 14:19
AlexDaniel` of course I can parallelize it but then perf bisection is at risk, which I used to do relatively often
jnthn Ah, logarithms end up being too awesome once again... :) 14:20
.oO( just bissect on the commitable release commits too )
14:21
AlexDaniel` yeah, and committable is getting slower over time while bisectable is remaining the same
because even doubling the amount of commits in the range will only require one more test, while committable will get twice slower. Logarithms are nice indeed 14:22
hmmm 14:24
jnthn: I mean… let's see if I can do it in one evening…
I'll keep existing functionality the same but will add autobisect: feature 14:25
ok I'll also change what bisectable does by default when you address it… 14:32
or maybe what it does by default in general…
14:37 lucasb joined 15:57 Xliff joined
Xliff jnthn: Hi there! How can you access rows using Spreadsheet::XLSX? I can see columns and cells, but not rows. 15:59
m: my @a[3;2]; 16:03
camelia ( no output )
jnthn You mean row metadata, or the cells in a row? 16:04
Xliff The cells in the row
jnthn You need to index the cells
.cells[$row;$col]
There's not yet any sugar for pulling out a whole row at once 16:05
Xliff Ah. So that's a rakudo limitation?
Is that something that is just NYI?
jnthn Probably something the module should provide
Xliff m: my @a[3;2] = [ rand xx 2 ] xx 2; @a.gist.say 16:06
camelia [[0.6994475307350786 0.3987772143997744]
[0.445755831174833 0.8148259960965093]
[(Any) (Any)]]
Xliff m: my @a[3;2] = [ rand xx 3 ] xx 2; @a.gist.say
camelia Index 2 for dimension 2 out of range (must be 0..1)
in block <unit> at <tmp> line 1
Xliff m: my @a[3;2] = [ rand xx 2 ] xx 3; @a.gist.say
camelia [[0.6824559234434973 0.8812132385490395]
[0.37053142664719985 0.794236344412102]
[0.007650594138133071 0.7298466325525138]]
Xliff m: my @a[3;2] = [ rand xx 3 ] xx 2; @a[1].gist.say
camelia Index 2 for dimension 2 out of range (must be 0..1)
in block <unit> at <tmp> line 1
Xliff m: my @a[3;2] = [ rand xx 3 ] xx 2; @a[1;1,2].gist.say
camelia Index 2 for dimension 2 out of range (must be 0..1)
in block <unit> at <tmp> line 1
Xliff m: my @a[3;2] = [ rand xx 3 ] xx 2; @a[1;0,1].gist.say
camelia Index 2 for dimension 2 out of range (must be 0..1)
in block <unit> at <tmp> line 1
jnthn m: my @a[3;2] = [ rand xx 3 ] xx 2; say @a[3;*] 16:07
camelia Index 2 for dimension 2 out of range (must be 0..1)
in block <unit> at <tmp> line 1
Xliff m: my @a[3;2] = [ rand xx 3 ] xx 2; @a[1;0].gist.say
camelia Index 2 for dimension 2 out of range (must be 0..1)
in block <unit> at <tmp> line 1
jnthn m: my @a[3;2] = [ rand xx 2 ] xx 3; say @a[3;*]
camelia Partially dimensioned views of shaped arrays not yet implemented. Sorry.
in block <unit> at <tmp> line 1
jnthn That's what I was looking for.
Yeah, that isn't implemented yet
m: my @a[3;2] = [ rand xx 2 ] xx 3; say @a[3]
camelia Partially dimensioned views of shaped arrays not yet implemented. Sorry.
in block <unit> at <tmp> line 1
Xliff OK, thanks!
jnthn And that is a Rakudo-level limitation
Xliff jnthn: So what's the best way to get the actual size of the worksheet? 16:09
jnthn Implement support for dimensions, I guess :)
16:10 sena_kun joined
Xliff LOL. Just found out Cells doesn't implement .shape, so yeah. 16:11
jnthn Yeah, it's there in the XML, just nothing pulls it out yet
16:12 Altai-man_ left
Geth rakudo/rakuast: 308f398ae7 | (Jonathan Worthington)++ | 6 files
Add basic RakuAST support for calls

The new RakuAST::Name node will be used in a few further places that currently use strings also. For now, only non-flattening positional arguments are supported.
16:12
Xliff jnthn: Oh! I see that, now. 16:17
(Looking at the source as we talk)
jnthn: Your solution is rather elegant in a lazy-loading way. To get the dimensions might actually require a full document walk. 16:21
jnthn Xliff: But some document generators tell us up front by emitting a dimensions tag, which would save that :)
Xliff Ah. Need to find that tag in the specs. 16:22
jnthn grmbl, there's a really nice site with the spec and I can never find it if I don't have it open in a tab 16:23
Xliff jnthn: Getting this when attempting to read my spreadsheet: Missing attribute 'spans' on 'row' 16:27
jnthn: Full backtrace, here: pastebin.com/am0d2tZe 16:29
jnthn Hm, that ain't mandatory? :/
(Excel always emits it...) 16:30
Xliff Yeah, these were generated by an online tool
jnthn Ah...hmm. That could be a small nuisance
Xliff I can give you an example if you give me a second to sanitize the data. This is for $dayJob
Ah, damn. To save it, I'd have to go through OpenOffice and it would recreate the file...possibly eliminating the evidence. 16:33
:/
jnthn: Do you mean this one? download.microsoft.com/download/3/E...Basics.ppt 16:34
Geth rakudo/rakuast: ad93a0f7c4 | (Jonathan Worthington)++ | 2 files
Implement basic method calls in RakuAST
16:50
Xliff Hmmm... isn't there an easy way to take column "ABC" and turn it into a decimal index? 16:52
jnthn Xliff: No...that's just a high-level overview. 16:54
I think I've got the other one at home on a browser tab...will add a reference to it in the module
Can't believe that the most detailed reference I found for it is so hard to re-find...
Xliff my $c = 'ABC'; my $posV = 26 ** ($c.chars - 1); my $index = 0; for $c.comb -> { $index += (.ord - 64 * $posV); $posV /= 26 }; $index.say 16:56
m: my $c = 'ABC'; my $posV = 26 ** ($c.chars - 1); my $index = 0; for $c.comb -> { $index += (.ord - 64 * $posV); $posV /= 26 }; $index.say
camelia Too many positionals passed; expected 0 arguments but got 1
in block <unit> at <tmp> line 1
Xliff m: my $c = 'ABC'; my $posV = 26 ** ($c.chars - 1); my $index = 0; for $c.comb -> { $index += (ord($_) - 64 * $posV); $posV /= 26 }; $index.say
camelia Too many positionals passed; expected 0 arguments but got 1
in block <unit> at <tmp> line 1
Xliff m: my $c = 'ABC'; my $posV = 26 ** ($c.chars - 1); my $index = 0; for $c.comb { $index += (.ord - 64) * $posV); $posV /= 26 }; $index.say 16:57
camelia 5===SORRY!5=== Error while compiling <tmp>
Missing block
at <tmp>:1
------> 3 $c.comb { $index += (.ord - 64) * $posV7⏏5); $posV /= 26 }; $index.say
expecting any of:
statement end
statement modifier
Xliff m: my $c = 'ABC'; my $posV = 26 ** ($c.chars - 1); my $index = 0; for $c.comb { $index += (.ord - 64) * $posV; $posV /= 26 }; $index.say
camelia 731
Xliff m: my $c = 'Z'; my $posV = 26 ** ($c.chars - 1); my $index = 0; for $c.comb { $index += (.ord - 64) * $posV; $posV /= 26 }; $index.say
camelia 26
Xliff m: my $c = 'AA'; my $posV = 26 ** ($c.chars - 1); my $index = 0; for $c.comb { $index += (.ord - 64) * $posV; $posV /= 26 }; $index.say
camelia 27
Xliff m: my $c = 'AZ'; my $posV = 26 ** ($c.chars - 1); my $index = 0; for $c.comb { $index += (.ord - 64) * $posV; $posV /= 26 }; $index.say
camelia 52
Xliff m: my $c = 'AAA'; my $posV = 26 ** ($c.chars - 1); my $index = 0; for $c.comb { $index += (.ord - 64) * $posV; $posV /= 26 }; $index.say 16:58
camelia 703
Xliff m: my $c = 'BA'; my $posV = 26 ** ($c.chars - 1); my $index = 0; for $c.comb { $index += (.ord - 64) * $posV; $posV /= 26 }; $index.say
camelia 53
Xliff \o/
16:58 bazzaar left
jnthn home time & 17:05
18:09 Altai-man_ joined 18:12 sena_kun left 18:17 Xliff left 18:23 MasterDuke left 18:28 jjatria left 18:31 jjatria joined 18:35 AlexDaniel` left
tyil .tell hankache maybe you can help here stackoverflow.com/questions/621372...in-windows 18:36
tellable6 tyil, I'll pass your message to hankache
18:37 MasterDuke joined 18:39 MasterDuke left, MasterDuke joined 18:42 unclechu left 19:18 AlexDaniel` joined 19:33 unclechu joined 19:41 MasterDuke left
lizmat weekly: perlmonks.org/?node_id=11117564 19:56
notable6 lizmat, Noted! (weekly)
20:10 sena_kun joined 20:12 Altai-man_ left 20:14 lichtkind_ joined 20:15 lichtkind left 20:16 MasterDuke joined 20:22 sena_kun left
[Coke] <hagrid>I Shouldn't'a read that</hagrid> 20:22
21:33 AlexDaniel joined 21:34 AlexDaniel left, AlexDaniel joined 22:38 sena_kun joined 23:00 sena_kun left