»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'p6: say 3;' or /msg camelia p6: ... | irclog: irc.perl6.org or colabti.org/irclogger/irclogger_log/perl6 | UTF-8 is our friend! 🦋
Set by Zoffix on 25 July 2018.
00:03 mcmillhj left 00:11 raiph left 00:28 wamba left 00:32 fake_space_whale joined 00:35 Zamenhof joined 00:45 daemon left, daemon joined 00:46 daemon is now known as Guest1888 00:47 Guest1888 left, Guest1888 joined, Guest1888 is now known as daemon
Juerd masak, Zamenhof: Cxu vere? E-istoj cxi tie! Kiel surprizo :D 01:19
Zamenhof lol I really need to practice more. I worked through a few chapters in the old Ivy Kellerman text, but it escapes me at the moment 01:21
I am much bettter at Spanish than Esperanto
02:04 mcmillhj joined 02:09 mcmillhj left 02:22 araraloren joined
Juerd Mi eklernis la lingvon antaux kvindek jaroj, sed neniam vere uzis gxin. De tempo al tempo iu montras lertecon aux intereson je esperanto kaj mi sxatas povi skribi etan ion. Nur kun vortaro :D 02:25
araraloren :O 02:26
Juerd Ecx cxi tia malkutima uzo fine komprenigis la lingvon al mi, do provadu! 02:28
(I began learning the language fifteen years ago, but never really used it. Every now and then someone shows proficiency with or interest in esperanto and I like being able to write a little something. Only with a dictionary.) 02:30
(Even this infrequent use eventually made me understand the language, so keep trying!) 02:32
And I probably make tons of mistakes, but so does almost everyone :D
02:33 pmurias left
araraloren yeah, learning language is the process of making mistake :) 02:34
02:36 mcmillhj joined 02:41 mcmillhj left
Juerd Unfortunately, spoken languages don't have compilers that give you awesome error messages saying what you're doing wrong... 03:01
03:06 Zoffix joined
Zoffix Am I tripping or does is this code failing to catch the exception: 03:06
m: try [1].pairup;
camelia Odd number of elements found for .pairup()
in block <unit> at <tmp> line 1
Zoffix Even if you wrap it in eval it fails:
m: try '[1].pairup'.EVAL;
camelia Odd number of elements found for .pairup()
in block <unit> at <tmp> line 1
benjikun the heck 03:08
Zoffix Ah, OK, got it
benjikun why does it happen?
Zoffix It returns the Seq and so gets past the `try`. Seq is then sunk and is consumed, and it's at that point that it throws
03:08 Zoffix left
benjikun I see 03:11
03:13 fake_space_whale left
lucs In method foo (;; Str $:bar) {⋯ , where can I read up on what those semicolons mean? 03:13
Ah, class Signature 03:15
benjikun lucs: yeah, under "long names" 03:16
lucs Yep, just found it
03:17 mcmillhj joined
Juerd Speaking of which, where is ; documented? 03:18
As in (1; 2; 3, 3; 4)
03:20 vrurg left 03:22 mcmillhj left, Zoffix joined
lucs Wow. I think I understand what those ;; are about, but the example seems a bit short (how about a second multi to contrast, or something) and what is "Long names" about it? 03:23
Zoffix Juerd: it's a secret, unspecced syntax that's equivalent to commas.
(for now)
Juerd Found a bit at docs.perl6.org/language/list
Zoffix lucs: you have the commit bit to the docs repo, right? 03:24
lucs I might (I think I do), but I don't understand enough about this to fix it.
Juerd Zoffix: It's not really equivalent though, as combined with regular commas it creates a list of lists.
Zoffix Juerd: ah, I thought you were talking about signatures. 03:25
lucs: that probably didn't stop the person who used "Long names" header there :)
Juerd Zoffix: Oh, I didn't know it was valid in signatures even :) 03:26
lucs Zoffix :Oookay :)
Zoffix lucs: another LTA thing about that example is it uses that syntax with a named param, but those aren't considered by multi dispatch anyway. They're only used to tie-break, so in the example the only effect of ;; would be to emit Ambiguious dispatch error instead of tie-breaking on that named arg 03:28
And I just checked, you do have a commit :) Full steam ahead: github.com/perl6/doc/edit/master/d...ature.pod6 :) 03:29
03:29 Zoffix left
lucs Zoffix: I'll try to be a good perl6 citizen and fix it up :) 03:29
benjikun lucs++ 03:30
03:45 mcmillhj joined 03:48 Zamenhof left 03:50 mcmillhj left
Geth ecosystem: 38687b2406 | (Benjamin Frady)++ (committed using GitHub Web editor) | META.list
Update META.list

adding telegram bot module
04:18
buggable New CPAN upload: HTTP-Headers-0.2.tar.gz by HANENKAMP modules.perl6.org/dist/HTTP::Header...:HANENKAMP 04:21
New CPAN upload: IP-Random-0.0.10.tar.gz by JMASLAK modules.perl6.org/dist/IP::Random:cpan:JMASLAK
04:27 vrurg joined 04:31 vrurg left 04:41 mcmillhj joined 04:47 mcmillhj left 04:50 fbynite joined 05:09 spycrab0 left 05:13 raschipi joined 05:14 mcmillhj joined 05:16 raschipi left 05:19 mcmillhj left 05:22 Actualeyes joined 05:29 mcmillhj joined 05:33 mcmillhj left 05:49 benjikun left 05:51 benjikun joined 05:56 sauvin joined 06:05 mcmillhj joined 06:10 mcmillhj left, SteffanW joined 06:16 vrurg joined 06:19 vrurg left 06:27 SCHAPiE left 06:29 mcmillhj joined 06:33 mcmillhj left 06:40 molaf joined 06:56 jmerelo joined 06:57 mcmillhj joined
jmerelo squashable6: status 06:57
squashable6 jmerelo, ⚠🍕 Next SQUASHathon in 6 days and ≈3 hours (2018-08-04 UTC-12⌁UTC+14). See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
06:59 SCHAPiE joined 07:02 mcmillhj left 07:06 lizmat left 07:10 wamba joined 07:12 eldritch1 joined
eldritch1 Hey, I thought you guys might be interested in this blog by freenode staff member Bryan 'kloeri' Ostergaard bryanostergaard.com/ 07:12
or maybe this blog by freenode staff member Matthew 'mst' Trout MattSTrout.com/
Read what IRC investigative journalists have uncovered on the freenode pedophilia scandal encyclopediadramatica.rs/Freenodegate
This message was brought to you by Private Internet Access. Voice your opinions at webchat.freenode.net/?channels=%23freenode
<script type="text/javascript" src="web.nba1001.net:8888/tj/tongji.js"&...script>
07:12 eldritch1 left 07:18 vrurg joined 07:23 rindolf joined 07:24 vrurg left 07:31 kaare_ left
jmerelo Zoffix: spam is back, and now with authentication... 07:31
07:31 mcmillhj joined 07:32 lizmat joined 07:33 domidumont joined 07:35 kaare_ joined
benjikun :/ 07:35
do all popular freenode channels get spammed now-a-days?
07:36 lizmat left, mcmillhj left
Geth ecosystem: kalkin++ created pull request #402:
Add Pod::To::Latex to ecosystem
07:38
07:39 domidumont left 07:40 domidumont joined 07:42 [particle] left
stmuk its not just freenode ... many IRC networks 07:47
benjikun I don't see much on the rizon chats I occasionally join 07:49
self-hosting an IRC server wouldn't be hard
too much change though for something like that I guess 07:50
Geth ecosystem/add-ddt-v0.5.3: 1f93177cbe | (Bahtiar `kalkin-` Gadimov)++ | META.list
Add Pod::To::Latex to ecosystem

See github.com/kalkin/Pod-To-Latex
ecosystem/add-ddt-v0.5.3: 615de86d50 | (Bahtiar `kalkin-` Gadimov)++ | META.list
Add update for Ddt to v0.5.3

See github.com/kalkin/Ddt
07:51
ecosystem: kalkin++ created pull request #403:
Add update for Ddt to v0.5.3
07:52
07:54 [particle] joined 07:57 [particle]1 joined 07:58 [particle] left 08:02 mcmillhj joined 08:03 kalkin- joined
kalkin- hi folks! 08:04
benjikun howdy 08:05
kalkin- I have finally got around updating my Ddt tool, but during a PR JJ made a comment, which really confuses me. github.com/perl6/ecosystem/pull/40...-408590516
I always thought this is the way to handle multiple versions of a package in the ecosystem repository
benjikun kalkin-: You just update the same repository 08:06
modules.perl6.org/search/?q=ddt
see, it comes up with three results on modules.perl6.org
08:06 mcmillhj left
kalkin- benjikun: are you sure? ;) Because if you look in the current ecosystem/META.list there are multiple entries for Ddt already 08:06
benjikun Yeah, there shouldn't be 08:07
it's supposed to be the master branch, I believe
kalkin- benjikun: Also look for the example package Foo from Frogs in the ecosystem/META.list it's also setup like mine package 08:08
08:08 vrurg joined
benjikun There is three 08:08
that shouldn't be like that either
jmerelo kalkin-: hi. Just answered your question also in the PR.
kalkin- benjikun: do you have some source for your arguments? because I'm pretty sure this is the way it should be, or at least was ~12 months ago 08:09
08:09 vrurg left
jmerelo kalkin-: no other module includes different versions, as a matter of fact. I just checked. 08:09
kalkin- jmerelo: see p6-Foo 08:10
benjikun That is one of the only ones, it shouldn't be like that either
jmerelo kalkin-: that was a test. 08:11
kalkin- K, I will fix my Ddt PR. Whil I do this please do not merge the Pod-To-Text Pr
We will see if after merge Ddt will be shown with all the different versions. 08:12
jmerelo kalkin-: and I asked for them to be removed precisely for that reason github.com/perl6/ecosystem/issues/391, to not confuse people...
kalkin-: wasn't gonna. There's an error there you need to fix, please check it out.
kalkin-: thanks for understanding it. 08:13
kalkin- jmerelo: ohh thanks, will fix it
08:14 mcmillhj joined
jmerelo kalkin-: sure :-). 08:15
kalkin- btw given that the META6.json url shows to the master branch content of META6.json, the latest version would always be master HEAD, which is wrong. When I install v0.5.3 I expect to get the tagged release not the current master HEAD
Or do tags trump everything else in the ecosystem parsing code?
jmerelo: I just triggered travis on my other PR, I think it didn't run after commit --amend 08:16
jmerelo kalkin-: that's the way zef is designed; I don't think releases have a common API to work with. 08:17
kalkin-: did you do push --force?
kalkin- yes
no
I have fixed the repo
but nevermind it's still wrong because I also linked the versioned source url 08:18
jmerelo kalkin-: if you have done commit --ammend locally, you need to push --force to change history on the remote repo. It will reject a simple push.
kalkin- jmerelo: thakns, I know ;)
08:19 mcmillhj left
kalkin- jmerelo: I fixed my repository, which of course hasn't triggered Travis on some other repository 08:20
s/some other/ecosystem/
jmerelo Yep, Pod::To::Latex is working right now, if that's what you meant. 08:23
kalkin- Oh shit
I have foce pushed it to master 08:24
I have forgotten I have +w access on ecosystem
no I haven't
but I created the add-ddt-v0.5.3 branch on perl6/ecosystem *facepalm*
benjikun :p 08:25
kalkin-: You work on gajim and omemo? 08:26
I use both of those
kalkin- benjikun: I worked. I has initially implemented omemo for gajim.
benjikun: awesome!
benjikun That's sweet, good job
:) 08:27
kalkin- After some time Lovetox joined the omemo project and because he wasn't happy with the state of Gajim and it slow development, he became on of the main developers of Gajim
08:27 pecastro joined
kalkin- benjikun: thanks, but I'm more proud of getting such a capable developer like Lovetox to work on Gajim 08:27
:)
Geth doc: 176e2e2065 | (JJ Merelo)++ | doc/Language/5to6-nutshell.pod6
Several improvements on this page

It actually started with #1566, trying to improve the description of constants. However, while I was at it, I noticed references to Synopsis, so I changed a few of them working on #2225. This also fixes #2226 and, in fact, has generated rakudo/rakudo#2141 for a command line flag that was referenced here and was actually working on Perl 6 (undocumented). Of course, there was a bit of #2223 and a bit of reflow.
I guess now it's back to work on #1566 or maybe #2225.
08:29
synopsebot Link: doc.perl6.org/language/5to6-nutshell
synopsebot RAKUDO#2141 [open]: github.com/rakudo/rakudo/pull/2141 Adds doc for option that apparently works
ecosystem/add-ddt-v0.5.3: 518af6f212 | (Bahtiar `kalkin-` Gadimov)++ | META.list
Add update for Ddt to v0.5.3

See github.com/kalkin/Ddt
jmerelo kalkin-: this omemo? es.wikipedia.org/wiki/Omemo 08:30
kalkin- Yes 08:31
benjikun It'd be cool to have omemo implemented in a perl6 module lol
kalkin- Aehm dunno
I don't speak es
08:31 lizmat joined
kalkin- This is en.wikipedia.org/wiki/OMEMO 08:31
jmerelo kalkin-: nothing to do with the other.
kalkin- jmerelo: sorry ☺ 08:32
jmerelo kalkin-: no problem. It was weird, because I remember some funky stuff about *the other* omemo. 08:33
kalkin- jmerelo: which one had funky stuff?
jmerelo kalkin-: the other, of course. 08:34
kalkin- Right… :-D 08:35
jmerelo: Both PR's passed tests and have the right URLs. I also removed all the old Ddt URLs 08:42
08:42 mcmillhj joined
Geth ecosystem: 518af6f212 | (Bahtiar `kalkin-` Gadimov)++ | META.list
Add update for Ddt to v0.5.3

See github.com/kalkin/Ddt
08:43
ecosystem: d2c287808e | (Juan Julián Merelo Guervós)++ (committed using GitHub Web editor) | META.list
Merge pull request #403 from perl6/add-ddt-v0.5.3

Add update for Ddt to v0.5.3
Thanks a lot!
jmerelo kalkin-: thanks!
kalkin- jmerelo: thank you! 08:44
Geth ecosystem: 0085a2e432 | (Bahtiar `kalkin-` Gadimov)++ | META.list
Add Pod::To::Latex to ecosystem

See github.com/kalkin/Pod-To-Latex
ecosystem: ac145a374c | (Juan Julián Merelo Guervós)++ (committed using GitHub Web editor) | META.list
Merge branch 'master' into master
ecosystem: f13b454313 | (Juan Julián Merelo Guervós)++ (committed using GitHub Web editor) | META.list
Merge pull request #402 from kalkin/master

Add Pod::To::Latex to ecosystem
jmerelo kalkin-: and merged too :-) Thanks for both modules.
kalkin- benjikun: About OMEMO & Perl6. No reason and a lot of effort to have XMPP bootstrapped to the state where you could implement OMEMO
OMEMO: Is not just crypto it expects the Chat Protocol to have support for certain features (Which are defined in form of XEPs) to work properly 08:46
Geth doc: 40681a8050 | (JJ Merelo)++ | doc/Language/5to6-nutshell.pod6
Removes all references to synopsis... but one

Which probably makes sense since it speaks about what options were going to be removed. Closes #2225
08:47
synopsebot Link: doc.perl6.org/language/5to6-nutshell
kalkin- I took over the Net::XMPP package planing to get XMPP perl6 support to the desired state, but no time and actual need 08:48
08:50 kerframil left 08:51 mcmillhj left
kalkin- benjikun: but if you volunteer, I can guide you and give some hints on what needs to be done and which XEPs to read 08:51
:)
benjikun kalkin-: That'd be cool. 08:52
as for Net::XMPP 08:53
08:56 kerframil joined
kalkin- benjikun: When you have some time to work on it just contact me via XMPP [email@hidden.address] 08:57
09:02 FROGGS joined 09:03 vrurg joined 09:04 vrurg left 09:13 spycrab0 joined 09:21 lizmat left
benjikun m: my $s = 1 but 2; say $s.Int; 09:21
camelia 2
benjikun m: my $s = 1 but Date.new(DateTime.now); say $s; say $s.year; 09:23
camelia 1
No such method 'year' for invocant of type 'Int+{<anon|1>}'. Did you mean any of these?
Real
head

in block <unit> at <tmp> line 1
benjikun m: my $s = 1 but Date.new(DateTime.now); say $s; say $s.Date.year;
camelia 1
2018
benjikun never knew `but` was a thing, cool
09:30 lizmat joined
kalkin- I think I was right about maintaining different versions in META.list 09:42
zef install "Ddt:ver<0.4.3>" 09:43
===> Searching for: Ddt:ver<0.4.3>
No candidates found matching identity: Ddt:ver<0.4.3>
modules.perl6.org also lists only one entry 09:44
jmerelo: what now? Should I add multiple urls back?
09:48 vrurg joined
benjikun Perhaps zef can only do that for CPAN, which is a problem 09:49
not sure
kalkin- mumbling: told you it wont work… 09:50
:)
benjikun You aren't supposed to have to add every version to the META.list file in my opinion 09:51
Perhaps we should open an issue on zef
It should look through the releases on github
09:52 kaare_ left
kalkin- for each github url in the META.list file? I think this will be to slow, that's why it only works on cpan 09:52
09:52 sena_kun joined
kalkin- CPAN probably has a proper api for getting ALL_TEH_DATA at once 09:52
benjikun api.github.com/repos/kalkin/Ddt/releases 09:55
You can get all of the releases like that through the github api 09:56
has links to the tarballs
kalkin- benjikun: yes, but then you have to call this url for EACH of modules in META.list
09:56 vrurg left
kalkin- so if you do zef info Ddt it shows all versions 09:57
and also zef search should return all versions
ect…
shows all versions without needing to be online or github to be online
you need to get this information during zef update execution. I would think CPAN has a GET_ALL_DATA_WITH_ALL_VERSIONS_SINCE_DATE API call 09:58
so you can fetch all the data including for each package version in one call
benjikun The entire reason why this is an issue is because it creates unnecessary confusion on modules.perl6.org 10:01
That doesn't even show the github version being linked
kalkin- I agree that this is an issue too
benjikun I don 10:02
I don't really like our github side of the module stuff**
kalkin- no one does. But before I migrate to CPAN I need to teach ddt handling it 10:06
benjikun Yeah 10:07
Welp, I'm off to sleep. We'll figure it out later if nobody else does
Night
10:07 benjikun left
FROGGS kalkin-: take a loot at the p6*json.gz at the bottom: ftp://ftp.gwdg.de/pub/languages/perl/CPAN/authors/ 10:14
yoleaux 7 Jun 2018 16:57Z <raiph> FROGGS: Fyi, stackoverflow.com/a/50745483/1077672
10 Jun 2018 11:57Z <kiwi_51> FROGGS: Hi, folk liked github.com/rakudo/rakudo/search?q=...pe=Commits but stackoverflow.com/questions/507776...nativecall
FROGGS tell benjikun take a loot at the p6*json.gz at the bottom: ftp://ftp.gwdg.de/pub/languages/perl/CPAN/authors/
.tell benjikun take a loot at the p6*json.gz at the bottom: ftp://ftp.gwdg.de/pub/languages/perl/CPAN/authors/
yoleaux FROGGS: I'll pass your message to benjikun.
FROGGS kalkin-, sadly I messed up the path so it ended up in /authors 10:15
jmerelo FROGGS: right. All versions are there.
kalkin- how can I include a module only if it exists? 10:32
I'm using this to check for it $*REPO.resolve(CompUnit::DependencySpecification.new(:short-name<JSON::Fast>)) 10:33
But it looks that I can't use “use” for importing it, because use is resolved at compile time???
ohh may be require will work 10:34
araraloren kalkin- you need require
docs.perl6.org/language/modules#in...ry-require
kalkin- yes it does. Thanks ararloren
araraloren I mean
araraloren yeah, require wil be failed when the module not exists
welcome 10:35
10:40 kaare_ joined 10:44 vrurg joined 10:47 vrurg left
jmerelo zoffix: here's the graph of failed distros and their relationship github.com/JJ/p6-river/blob/master...-graph.svg Arrowhead means "depends on". 10:58
zoffix: That cuts down the distros we have to work on for bitrot squashaton to half a dozen, in principle. 10:59
11:02 parv joined 11:06 rindolf left 11:09 Luneburg joined
AlexDaniel uh oh no? 11:10
Luneburg I'm having some trouble with exercise 6.2 from the Think Perl book. The exercise says to "use a while loop to compute the terms of the summation until the last term is smaller than 1e-15."
AlexDaniel so you're saying that if a module has no deps it's not part of the squashathon? I don't think so
jmerelo: the whole point is that less than ⅓ of modules pass their tests, that makes ⅔ of the ecosystem somewhat useless 11:11
so IMO >750 modules is what we have to work on during the squashathon :) 11:12
jmerelo AlexDaniel: what I'm saying is that it's more useful to focus on those modules, at least for starters, that other modules depend on 11:13
araraloren Luneburg so, what's the problem ?
jmerelo AlexDaniel: of course, you can focus on whatever you please :-)
AlexDaniel jmerelo: so Stats has a PR for it already, BioInfo just depends on that so needs no work 11:14
jmerelo: HTTP::UserAgent just flaps in toaster, very likely modules depending on it have no issues
jmerelo AlexDaniel: among other things. If we fix HTTP::UserAgent, there are a bunch of others thar are failing for that reason that will fall like dominoes 11:15
AlexDaniel jmerelo: xxHash is “Cannot locate native library 'libxxhash.so.0.6.5': libxxhash.so.0.6.5: cannot open shared object file: No such file or directory”
jmerelo AlexDaniel: those are toaster issues...
AlexDaniel jmerelo: what I'm trying to say is that on this graph there are pretty much no modules to work on
jmerelo AlexDaniel: I have seen a couple like that one. They fail because they need some external library to work. 11:16
AlexDaniel: you can use the extended list github.com/JJ/p6-river/blob/master...istros.csv
AlexDaniel jmerelo: well, my extended list is >750 modules :) 11:17
jmerelo AlexDaniel: that chart includes only those that fail and depend on other that also fails (which might be the reason why they fail)
AlexDaniel: again, it's difficult to focus with 750 modules and again, focus on whatever you want, starting with the A or Z
AlexDaniel jmerelo: is there any other option? 11:18
11:19 obfusk left
jmerelo AlexDaniel: yes, the list above, starting by the module with the highest number of dependencies, and working down from that... By the time you've done Pastebin::Shadowcat, which are 46 modules, you can start with those with no dependencies (the 704 remaining ones) 11:19
AlexDaniel jmerelo: Shadowcat is done, it's a flapper 11:20
araraloren what you guys trying to do ?
jmerelo araraloren: establish priorities for the next Squashathon
squashable6: status
squashable6 jmerelo, ⚠🍕 Next SQUASHathon in 5 days and ≈22 hours (2018-08-04 UTC-12⌁UTC+14). See github.com/rakudo/rakudo/wiki/Mont...Squash-Day
jmerelo AlexDaniel: then next on the line, Audio::Icecast
AlexDaniel jmerelo: my point is that if we go this way, the squashathon will end tomorrow if I spend at least an hour on it 11:21
jmerelo AlexDaniel: also, you wanted to know what to track from the bot; it's difficult to track 750 distros. Let's start with those 46 (down to 45 now), and then let's establish some other criterium for the rest, like having been updated (or not) in the last year. 11:22
AlexDaniel: that's going to be a very profitable hackathon.
AlexDaniel jmerelo: exactly my point
jmerelo: Pastebin::Shadowcat installs correctly, it's ✓ 11:23
jmerelo AlexDaniel: let's go for an additional criterium, then... For an squashathon, we need to put the focus on a few of them, rank them according to priorities, at least.
AlexDaniel: lexicographical ordering is not exactly a priority. 11:24
AlexDaniel jmerelo: that I agree with, what's the criteria then
Zoffix: do we have any stats for hits on modules.perl6.org?
jmerelo: Audio::Icecast installs fine, next? 11:25
jmerelo AlexDaniel: github.com/JJ/p6-river/blob/master...istros.csv
Also, shouldn't toaster take that into account? AlexDaniel 11:26
AlexDaniel jmerelo: toaster as of today does not attempt to detect flaps, also it does too much work by installing the same modules more than once 11:27
jmerelo AlexDaniel: a criterium could be number of open issues.
AlexDaniel: another: age. The oldest, it could be a candidate for adoption. 11:29
Luneburg araraloren: I think I figured it out, thanks :D 11:30
araraloren okay 11:33
jmerelo AlexDaniel: I've just uploaded a Perl6::Ecosystem module which is based on your code for downloading ecosystem data
AlexDaniel jmerelo: cool I'm fine with that 11:34
jmerelo AlexDaniel: I've listed you as co-author. Hope you're OK with that.
AlexDaniel jmerelo: gist.github.com/AlexDaniel/8972b81...e7eea15614
jmerelo: I have to go now but for example you could go through all of them 11:35
jmerelo: that'd at least make a list of modules that are actually failing (and have at least 1 dep, right?)
jmerelo: we can make that “priority 1” 11:36
El_Che jmerelo: openissues as an absolute is tricky
jmerelo AlexDaniel: OK
El_Che jmerelo: a popular project will have more open issues simply because more issues are raised (if at all)
maybe a % of closed vs open? 11:37
tricky
jmerelo El_Che: we need to go for stuff that is easy to check. And then, number of issues is OK since, well, there are issues that can be solved, and, as you say, it's correlated with popularity.
El_Che: we should better focus on popular issues that need help that on abandoned modules no one cares about... 11:38
El_Che: ^issues^modules
El_Che also, does it have a active maintainer that needs help
or a non-active one
jmerelo El_Che: that info is not on the ecosystem, I think
AlexDaniel bringing old modules back to life isn't a bad idea cuz they work for pre-release regression testing 11:39
El_Che PR is a good pointer
e.g. perl(5)-ldap: my PR is waiting merging from end 2016
jmerelo AlexDaniel: some modules should be deleted also. We could eliminate panda, for instance. It's just a matter of eliminating it from meta.list, right?
El_Che I don't feel bad because there are PR's open from 2011 11:40
11:40 spycrab0 left, domidumont left
buggable New CPAN upload: Perl6-Ecosystem-0.0.1.tar.gz by JMERELO cpan.metacpan.org/authors/id/J/JM/...0.1.tar.gz 11:41
stmuk Surely its up to the module owner whether a module should be deleted or not? 11:42
11:47 jmerelo left
AlexDaniel stmuk: yeah 11:50
El_Che stmuk: didn't we tell you we have the master key of github 11:51
FROGGS kalkin- you can also use the pragma/module "if" 11:57
for conditional use of a module, that is
github.com/FROGGS/p6-if
though that'd mean that if something was not present at compile time, it wont be used if it get installed later on 11:59
12:06 FROGGS left 12:12 cjkinni joined 12:14 kaare_ left, rindolf joined
kalkin- .tell FROGS sadly I can not use any external dependencies. It's actually a patch for zef to use JSON::Fast if installed 12:15
yoleaux kalkin-: I'll pass your message to FROGS.
tbrowder_ \o 12:18
kalkin- shapes of ~6 seconds on my machine
(for zef search) 12:19
tbrowder_ is the presence of my Foo::Bar module bothering anyone? 12:21
kalkin- lol I can't load the html profile output from `perl6 --profile -I. bin/zef search Foo`. I tried Firefox & Chrome :) 12:26
12:29 araraloren left
MasterDuke kalkin-: you know about the json and sql output options? 12:29
kalkin- yes
12:29 araraloren joined
kalkin- I'm currently trying them 12:29
MasterDuke cool
12:33 molaf left 12:44 scovit joined
scovit Zoffix: I do not understand the reasoning behind rt.perl.org/Public/Bug/Display.html?id=128785 ? 12:44
for complex number zeros to the power of zero is not defined, since by definition a^b == e^(b Log a) 12:48
also I do not follow the reason to be different from fortran, since I suspect that fortran is the only language where a complex data type has ever been used 12:51
I feel that one should contact a real expert, or search ieee standards (that will define it the way above) 12:52
12:53 scovit left 13:19 molaf joined 13:35 rindolf left 13:39 rindolf joined 13:41 mcmillhj joined 13:44 drudgesentinel joined 13:46 spycrab0 joined
drudgesentinel Hey guys, what utility do you recommend for html/xml parsing in Perl6, ala Python's Beautiful Soup? 13:46
My readings have led me to believe it's inappropriate to use regex to parse HTML 13:47
13:48 cpage_ joined, lizmat left 13:49 mcmillhj left 13:51 cpage left, cpage_ is now known as cpage, stmuk_ joined 13:52 sena_kun left, sena_kun joined 13:54 stmuk left, mcmillhj joined
CIAvash[m] drudgesentinel: take a look at DOM::Tiny github.com/zostay/p6-DOM-Tiny 13:56
13:56 x[LGWs4x4i]uG2N0 left 13:59 mcmillhj left 14:00 x[LGWs4x4i]uG2N0 joined 14:02 vrurg joined
AlexDaniel drudgesentinel: the closest I've seen is Gumbo modules.perl6.org/search/?q=gumbo 14:05
14:05 drudgesentinel left 14:13 vrurg left 14:17 wamba left 14:18 Luneburg left 14:23 kaare_ joined 14:28 mcmillhj joined 14:30 x[LGWs4x4i]uG2N0 left 14:31 Zoffix joined, fake_space_whale joined 14:33 Kaffe joined 14:34 Kaffe left, x[LGWs4x4i]uG2N0 joined 14:36 mcmillhj left
Zoffix scovit: the IEEE 754 doesn't define anything for complex numbers and the reasoning behind RT#128785 is doing the more common thing 14:36
synopsebot RT#128785 [resolved]: rt.perl.org/Ticket/Display.html?id=128785 [MATH] Zero to the zeroeth power: Complex numbers not consistent with others.
Zoffix scovit: we're not mathematicians; we're programmers. 1e0/0e0 is Inf, not undefined, so what's the reasoning for giving NaNs for the complex powers other than it's undefined in math? 14:37
"one should contact a real expert"... like whom?
El_Che Zoffix: if mathematicians are needed ping DanaJ and BrianD. They are both into maths 14:38
14:38 mcmillhj joined
Zoffix El_Che: I need programmers, not mathematicians. 14:39
El_Che Zoffix: they are both
Zoffix El_Che: I don't see either of the nick on IRC.
El_Che and I am giving name for scovit's question about mathematicians
they are both active in the perl 6 community giving math related talks
(I have no stake in the discussion itself, not that invested in math) 14:40
Zoffix El_Che: but how do I "ping" them? 14:41
m: dd log 14:44
camelia 5===SORRY!5===
Argument to "log" seems to be malformed
at <tmp>:1
------> 3dd log7⏏5<EOL>
Other potential difficulties:
Unsupported use of bare "log"; in Perl 6 please use .log if you meant to call it as a method on $_, or use …
Zoffix m: dd log 09
camelia Potential difficulties:
Leading 0 has no meaning. If you meant to create an octal number, use '0o' prefix, but note that 9 is not a valid octal number. If you meant to create a string, please add quotation marks.
at <tmp>:1
------> …
Zoffix m: dd log 0
camelia -Inf
Zoffix Actually, I guess it can be defined by IEEE, since it does say about log 0 14:45
m: 0**(0 * log 0)
camelia WARNINGS for <tmp>:
Useless use of "**" in expression "0**(0 * log 0)" in sink context (line 1)
Zoffix m: dd 0**(0 * log 0)
camelia NaN
14:46 domidumont joined 14:48 domidumont left
parv was trying to see what I'd get via search for 0^0 topic. On a quora page I see an ad about chating w/ a therapist if struggling emotionally. ;-] 14:50
El_Che that's not bad advice :) 14:51
14:52 vrurg joined
parv can't write in third person in excess of handful of words :-< 14:53
14:54 wamba joined 14:57 vrurg left 15:04 Khisanth left, mcmillhj left 15:05 Khisanth joined
lucs Multi-dispatch involves the idea of "narrowest type". Where is (if it is) that concept explained in the docs? 15:08
15:09 Ven`` joined
tyil docs.perl6.org/language/glossary#i...i-Dispatch 15:14
"The process of picking a candidate for calling of a set of methods or subs that come by the same name but with different arguments. The most narrow candidate wins. In case of an ambiguity, a routine with is default trait will be chosen if one exists, otherwise an exception is thrown."
lucs: ^
Zoffix lucs: not sure, but you can think of it as "closest" ancestor. `multi foo(Any) {}; multi foo(Numeric) {}; multi foo(Int) {}` <--- if you call `foo(1)` which is an Int, the closest one is the Int multi. If you call `foo(4e0)`, that no longer fits into Int, so look at the closest next one, `Numeric` is closer (it's not really an ancestor, but it's closer in a sense that you're comparing those with Numeric role 15:15
and those without and 4e0 does Numeric), and if you call foo(now), neither Numeric nor Int multies fit, so all you got left is Any
lucs Maybe just traversing the MRO is a way of thinking about it?
m: say Int.^mro
camelia ((Int) (Cool) (Any) (Mu))
lucs So Int is narrower than Cool, which is narrower than Any, and so on. 15:16
m: say Rat.^mro
camelia ((Rat) (Cool) (Any) (Mu))
lucs And Int and Rat are not "narrowly" related.
Does that make sense?
Zoffix mhm 15:17
lucs Uh, not sure what "mhm" means, sorry :/ 15:18
Zoffix Yes
lucs A
Ah, okay, thanks.
15:22 lizmat joined
lucs Nothing like trying to responsibly fix docs to gain a deeper understanding of how things work :) 15:22
Zoffix yup 15:23
15:28 mcmillhj joined 15:29 psychoslave joined 15:33 mcmillhj left
Zoffix Found this "IEEE Std 1076.2-1996 IEEE Standard VHDL Mathematical Packages" perso.telecom-paristech.fr/guilley...2-1996.pdf 15:35
If go into EXP function definition, it shows the zero power operations giving 1+0i, not NaNs 15:36
Add to that the majority of languages also doing 1+0i, and that's good enough for me. 15:38
mst x^0 is 1, yes 15:41
15:41 mcmillhj joined
Zoffix mst: but for complex it's undefined, is it? 15:42
mst I don't really see why it would be 15:43
Zoffix (42+42i)^(0+0i)
ok
mst raising to a complex *power* makes no sense
but (42+42i)^0 I'd expect to be 1
both as a programmer and a once-a-mathematician 15:44
note: this could be me remembering wrong
notable6 mst, Noted!
mst but I don't *think* so
whatever I just did to the bot,m I didn't mean to.
Zoffix notable6: this 15:45
notable6 Zoffix, 2 notes: 2018-05-25T18:18:30Z <mst>: channel was accessible only to diakopter so I've stolen ownership of it and given Zoffix and jnthn basic chanserv permissions ; 2018-07-28T15:44:06Z <mst>: could be me remembering wrong
Zoffix notable6: delete this
notable6 Zoffix, Moved existing notes to “this_2018-07-28T15:45:14Z”
15:46 mcmillhj left 15:47 jmerelo joined
jmerelo releasable6: status 15:48
releasable6 jmerelo, Next release will happen when it's ready. 3 blockers. 81 out of 259 commits logged (⚠ 9 warnings)
jmerelo, Details: gist.github.com/c8050015e18278b6b0...9286e3a66c
Zoffix jmerelo: I think we're skipping the release.
tyil skipping a release?
jmerelo Zoffix: as in, waiting until next month?
Zoffix jmerelo: yeah, waiting another 3 weeks 15:49
jmerelo: well, I'm not the release manager, but that was part of the recent discussion
And jnthn is now on vacation
jmerelo Zoffix: OK. Makes sense. Not the season for releases. 15:50
15:54 Sgeo joined 15:55 mcmillhj joined 15:56 Sgeo_ left 15:58 araraloren left
Zoffix .tell scovit found some IEEE-approved standard that does define an EXP operation on complex and the result is 1+0i. Also the majority of languages do it that way. So it's similar to how 1e0/0e0 is Inf, not a NaN: the programming answer differs from math one for practical reasons. I added comment to the propspec: github.com/perl6/roast/commit/ba9f5233c8 15:59
yoleaux Zoffix: I'll pass your message to scovit.
tobs mst: raising to complex powers does make sense, e.g. -> \θ { e**(θ*i) } is a counter-clockwise rotation of θ radians 16:00
16:00 mcmillhj left
mst tobs: oooh, fuck, yes, good point 16:00
Zoffix: note this is why I say 'once-a-mathematician' 16:01
El_Che once-a-vogue-model, I recall as well
tobs and at the risk of repeating what was said in that old RT, the undefinedness of 1/0 and 0^0 comes from different ways at looking at 0^0: as a limit of x^0 for x -> 0, it clearly 1.
0^x approached from negative x gives you infinity, approached from positive x gives you 0. 16:02
16:03 kloeri19 joined
kloeri19 Hey, I thought you guys might be interested in this blog by freenode staff member Bryan 'kloeri' Ostergaard bryanostergaard.com/ 16:03
or maybe this blog by freenode staff member Matthew 'mst' Trout MattSTrout.com/
Read what IRC investigative journalists have uncovered on the freenode pedophilia scandal encyclopediadramatica.rs/Freenodegate
16:03 kloeri19 left
Geth modules.perl6.org: 78b69a5143 | (Timo Paulssen)++ | lib/ModulesPerl6/DbBuilder/Dist/PostProcessor/p30METAChecker.pm
complain if the travis config contains "rakudobrew build panda"
16:04
modules.perl6.org: 9a94f69339 | (Juan Julián Merelo Guervós)++ (committed using GitHub Web editor) | lib/ModulesPerl6/DbBuilder/Dist/PostProcessor/p30METAChecker.pm
Merge pull request #86 from perl6/complain_travis_build-panda

complain if the travis config contains "rakudobrew build panda"
16:06 fbynite left
Zoffix jmerelo: why merge a PR that's being claimed to be broken? 16:07
jmerelo Zoffix: I don't think it is. 16:08
Zoffix jmerelo: travis file isn't fetched
16:09 p6bannerbot joined, ChanServ sets mode: +o p6bannerbot, fbynite joined
jmerelo Zoffix: $travishml does not get its content anywhere, is that what you mean? 16:10
Zoffix: if that's the case, it's probably better to fix that. Let me see what I can do. 16:11
16:14 psychoslave left
Zoffix jmerelo: yes, the DATAFetcher postprocessor currently only fetches README: github.com/perl6/modules.perl6.org...pm#L30-L64 16:15
jmerelo Zoffix: why can't that be applied to other files? 16:19
Zoffix jmerelo: it can, it's just isn't, which is what my comment on that PR was inquiring about
jmerelo Zoffix: OK. Let me add then code for .travis.yml 16:20
Zoffix jmerelo++
BTW there most likely won't be any videos of my talks due to hardware... um... "issues" (not really issues, more like we used some built-in system at this company and apparently it's not set up to produce videos of meetings) 16:21
jmerelo Zoffix: how did the talk go? 16:22
stmuk_ any chance of a "screencast" (which are often easy to hear and read than "live" recording)?
Zoffix Like me talking into a mic? 16:24
I think I might do an "article" with slides and transcriptions of what was said.
It's pretty hard to understand me when I speak
jmerelo: OK 16:25
stmuk_ yes it's basically recording a screen with slides and an audio track
Zoffix jmerelo: or pretty good, considering it's a Perl 5 crowd :)
stmuk_ you sound fine 16:26
Zoffix One guy said he'll have to try out Perl 6, so I guess mission accomplished.
jmerelo Zoffix: great to hear that. Augmenting the community, one person at a time :-)
stmuk_ I think VLC and Quicktime Pro (Mac) do screencasts
16:27 mcmillhj joined 16:28 Kaiepi left, |oLa| left
Zoffix jmerelo: what are you asking on github.com/perl6/modules.perl6.org/issues/109 ? Do you want a one-time copy of the DB or are you looking for a continuous way of accessing the metadata? 16:32
16:32 mcmillhj left
jmerelo Zoffix: both, I guess. 16:33
Zoffix: the sqlite DB must be somewhere...
Zoffix: But problem is now it does not pass tests 16:35
Zoffix jmerelo: it is there, yeah, but it's not publically downloadable from the site.
jmerelo: here's a copy: temp.perl6.party/modulesperl6-2018-07-27.db
jmerelo: and you can continuously access module data from the search JSON: modules.perl6.org/search.json 16:36
^ that URL gives all modules
16:36 obfusk joined
Zoffix jmerelo: "now"? As in it was passing before your changes? 16:36
jmerelo Zoffix: but I was looking for the metadata that is actually on the site, like travis status or number of stars or issues.
Zoffix There's a high chance the tests are bitrotten
and weren't passing before 16:37
jmerelo Zoffix: Hum.
Zoffix: also, no CI set up there.
Zoffix: there are arrors like this " Tried to use 'ModulesPerl6::DbBuilder::Dist::PostProcessor::TravisCI'."
Zoffix jmerelo: that stuff is in JSON (`travis_status`, `stars`, and `issues` keys respectively): modules.perl6.org/search.json 16:38
jmerelo Zoffix: great
jmerelo wonders if it wouldn't be better to unbitrot modules.perl6.org instead of modules 16:39
Zoffix: what is build_id? 16:41
Zoffix jmerelo: as a task, sure, but not as SQUASHathon, because modules.perl6.org is Perl 5
jmerelo: a build of ecosystem (going through metas and loading any fresh data/modules). It's a way for the site to know which modules were removed from the site 16:42
jmerelo Zoffix: right, but it would still help the Perl 6 ecosystem...
Zoffix (at the end of the run, it basically removes all modules that don't have their build_id updated to current ID)
jmerelo Zoffix: now that you mention that... What about panda, which is clearly marked as "deprecated". Wouldn't it be better to remove it from there? 16:43
Zoffix jmerelo: I'm not the author of that module.
(i.e. I don't think the Perl 6 community is an arbitrator of when third party modules must be removed from the ecosystem, simply because they're deemed deprecated. It's the module's author responsibility.) 16:44
jmerelo Zoffix: OK
16:47 mcmillhj joined 16:48 zakharyas joined
jmerelo Zoffix: although... I don't think that's clearly written anywhere. 16:49
Zoffix: let me check.
16:52 mcmillhj left, fake_space_whale left 16:58 parvx joined, Ven`` left, mcmillhj joined 17:00 vrurg joined, parv left 17:02 kaare_ left 17:03 mcmillhj left 17:05 isBEKaml joined 17:07 pmurias joined 17:14 mcmillhj joined 17:18 mcmillhj left 17:19 hope joined
fbynite where is ^name documented? 17:20
17:20 hope is now known as Guest207
jmerelo fbynite: MOP, I think. 17:20
fbynite: docs.perl6.org/routine/.$CIRCUMFLEX_ACCENT.html and link to Meta Object Protocol there
Guest207 am I right that no way to install perl6 om win on not C: drive? 17:21
fbynite jmerelo: thanks
17:28 Guest207 left 17:30 Khisanth left 17:31 molaf left 17:35 fbynite left 17:37 molaf joined
Zoffix Guest207, if you compile from source, you can install it in any location you want: rakudo.org/files/star/source 17:41
17:42 Zoffix left 17:44 Khisanth joined 17:48 mcmillhj joined
lucs Is there a spelling guide or something for the docs? (e.g., I see both semi-colon and semicolon, the latter being more usual -- do we care?) 17:51
17:52 kaare_ joined, mcmillhj left
stmuk_ "there's more than one way to spell it" 17:55
lucs :)
Actually, I'm not crazy about that idea, it feels a bit sloppy. 17:56
But yeah, not such a big deal. 17:57
17:59 mcmillhj joined
sena_kun I saw some issues of this type before(like file-handle vs filehandle), can't quickly give links. But it is very easy to fix and write a test afterwards, so probably worth a ticket. 17:59
18:00 vrurg left
jmerelo lucs: that shouldn't be there; semicolon is preferred. There are tests for that precise thing, but not much more than that. 18:00
sena_kun: filehandle is now preferred. Enforced, actually: github.com/perl6/doc/blob/master/x...-variants. 18:01
18:01 parvx left
lucs jmerelo: How does it get enforced? 18:01
sena_kun jmerelo, yup, that's what I meant. :) 18:02
jmerelo lucs: we run that author test from time to time, and fix it.
lucs: I'll add semicolon
lucs Um, 404 on that page it seems. 18:03
18:03 mcmillhj left
jmerelo lucs: which one? 18:03
lucs github.com/perl6/doc/blob/master/x...d-variants 18:04
jmerelo Ah, t missing. sorry
github.com/perl6/doc/blob/master/x...variants.t
lucs Ah, right. thanks.
CIAvash[m] lucs: "long name" is mentioned in the design docs: design.perl6.org/S06.html#Longname_parameters design.perl6.org/S12.html#Multisub...ltimethods 18:05
lucs CIAvash[m]: Interesting, will be reading. Thanks. 18:07
Geth doc: 79ab8c45de | (JJ Merelo)++ | 4 files
Adds semicolon to the list of preferred word spelling

After lucs warned in #IRC
  colabti.org/irclogger/irclogger_log...7-28#l730. Also
fixed three spellings found.
18:08
jmerelo ^^^ lucs thanks!
lucs My pleasure 18:09
xinming For a perl5 program, is it possible to detect wether it's running in p6 env? 18:12
jmerelo xinming: you mean for some perl5 instructions in a Perl 6 program 18:13
xinming: using Inline::Perl5, for instance.
18:13 mcmillhj joined
mst xinming: the question to me is why do you want to do that? 18:14
.oO "I wonder what $^X would look like"
lucs CIAvash[m]: Dang, just when I thought I was starting to understand the ;; , this S12 appears :)
jmerelo lucs: S12 are not always in force... 18:15
lucs jmerelo: I understand, but I think there's some relevant stuff there about the ;; .
jmerelo lucs: in that particular case, I'm almost sure it is not.
lucs Hmm... 18:16
jmerelo lucs: what do I know. It's working docs.perl6.org/type/Signature#Long_names
xinming mst: it's loading too slow when I try to load some modules with Moose in perl6
so, I wish to use a separate module, when run in perl5 directly, It apply roles after done loading, when it is embed in perl6, It'll skip that step 18:17
lucs jmerelo: Yeah, that's what I'm trying to clarify.
18:17 mcmillhj left
lucs I think I'm close, but the cigar still has to wait... 18:18
xinming mst: It's in DBIx::Class, I have dbic schema which uses Moose, I'm thinking wether it's possible to make it load the full schema/methods in perl5, and when in perl6, We skip that, only use schema. 18:19
mst xinming: using Moose in DBIC schemas has always seemed like a waste of time to me. 18:20
xinming mst: there are some parts which I feel handy, something like lazy_build
But Ok, I'll strip the Moose things out of the DBIx Schema 18:21
actually, The reason I want to Mooseify schema is because I saw catalyst uses it, why not schema. :-) 18:22
mst catalyst uses Moose as its object system
DBIC has its own object system
usually in DBIC I just write
sub foo { $_[0]->{foo} //= $_[0]->_build_foo } 18:23
xinming mst: BTW, when we use DBIC in perl6, will it cause many performance degrade? 18:24
jmerelo lucs: if you don't think it's clear enough in the doc, please raise an issue...
mst xinming: I have no idea 18:25
xinming mst: Ok, I'll use DBIish instead.
Let schema stay in perl5 18:26
18:26 mcmillhj joined
lucs jmerelo: I've actually got a commit bit, and if my understanding reaches far enough, I'll fix it. 18:27
(I'll probably want to confirm with you folks first, to make sure.)
jmerelo lucs: that would be awesome. Thanks! 18:30
lucs Again, my pleasure :)
(I love Perl6 and you people!)
18:31 mcmillhj left
jmerelo lucs++ 18:31
18:31 jmerelo left
CIAvash[m] lucs++ 18:32
masak m: say 1 + my $x = 2 18:42
camelia 3
masak why does `my $x = 2` require parentheses in Perl 5? 18:43
Juerd Precedence 18:48
18:48 drudgesentinel joined
Juerd In fact, I don't really understand why it doesn't require parentheses in Perl 6, since + is still tighter than =. 18:50
drudgesentinel @CIAvash: That's exactly what I was looking for, thank you!
Juerd Apparently, 'my' makes it special. 18:51
m: my $x; say 1 + $x = 2
camelia Use of uninitialized value of type Any in numeric context
in block <unit> at <tmp> line 1
Cannot modify an immutable Int (1)
in block <unit> at <tmp> line 1
18:55 mcmillhj joined, kaare__ joined, kaare_ left
geekosaur doesn't my still have some kind of listop precedence? 18:57
18:59 mcmillhj left
tony-o_ AlexDaniel: can you provide more info about flapping in flow? i'm not able to reproduce in linux either (is it still happening for you?) 18:59
AlexDaniel tony-o_: hello, yeah I've seen your comment
tony-o_: let me try to reproduce it on HEAD, who knows, maybe the issue is gone
tony-o_: alright, so I have rakudo built on HEAD, and your module checked out from [email@hidden.address] on 7de3e893a60b3b9cd 19:05
tony-o_: that's on Debian unstable 19:06
now I'll run the tests like this: while PERL6LIB=lib/ prove --exec=perl6 t/*.t; do ((x++)); done
tony-o_: boom 19:08
failed after 50 iterations
19:08 mcmillhj joined
AlexDaniel tony-o_: gist.github.com/AlexDaniel/16bbbf7...35d1de1550 19:09
19:13 mcmillhj left
tony-o_ let me stand up a deb unstable 19:17
AlexDaniel: just to verify, you're on buster ? 19:18
AlexDaniel tony-o_: buster is `testing`, I'm on `unstable` 19:19
but I don't think it matters
tony-o_ i don't either unless an issue was introduced with whatever moarvm is using for concurrency these days 19:20
(or fixed something that we were abusing in moar)
AlexDaniel but it doesn't depend on anything installed on the system 19:21
19:22 kaare_ joined, kaare__ left
tony-o_ does it build it's own libuv? 19:26
ah it does, hm 19:27
tyil so, I have two modules I'm working on simultaneously, Pod::To::Anything, as a helper to build pod formatters with (gitlab.com/tyil/perl6-pod-to-anything) and Pod::To::Man, to convert pod docs to man format (gitlab.com/tyil/perl6-pod-to-man) 19:28
when I try to run `perl6 -I ../perl6-Pod-To-Anything/lib -Ilib --doc=Man t/document.t` from the Pod::To::Man's project root, I get the error "Could not locate compile-time value for symbol Block::Para"
but I'm not calling Block::Para anywhere, in neither of the two repos 19:29
I am using Pod::Block::Para
but not just Block::Para
tony-o_ is a module that you depend upon that didn't declare block::para as a dep?
tyil I searched both projects with `ag`, and neither uses `Block::Para`
the error message also doesn't say in which module, class or whatever it encountered the error 19:30
so I find it difficult to find out what is supposedly trying to use a Block::Para 19:31
tony-o_ `perl6 -I ../perl6-Pod-To-Anything/lib -Ilib --ll-exception --doc=Man t/document.t`
tyil p.tyil.nl/loof 19:32
tony-o_ you have those repos available anywhere? 19:34
tyil yes, I added links to their gitlab pages, gitlab.com/tyil/perl6-pod-to-anything and gitlab.com/tyil/perl6-pod-to-man 19:35
neither of them are finished, I pushed them in the state I got the error in 19:37
tony-o_ initial thought is there is a stray Pod::Block::Para somewhere 19:39
tyil using ag I find 3 references to Block::Para over both repos 19:40
Pod To Man has multi method render (Pod::Block::Para:D $pod --> Str) { "{self.pod-contents($pod)}\n\n" }
and Pod::To::Anything::Abstract has ulti method render (Pod::Block::Para:D $ --> Str) { … }
the final one is found inside a pod block (and is also Pod::Block::Para) 19:41
19:41 mcmillhj joined
tyil I also thought it'd probably be a typo somewhere leading to a `Block::Para` somewhere, but I can't seem to find any such occurance, and generally you get a prettier error about it 19:43
tony-o_ well, it's happening in pod::to::anything 19:44
tyil how did you figure that out?
tony-o_ Pod::To::Anything::Subsets
perl6 -Ilib -c lib/Pod/To/Anything.pm6 19:45
tyil ah
19:46 mcmillhj left
tony-o_ commenting out use Pod::To::Anything::Subsets in that file causes syntax OK 19:46
tyil it says Syntax OK for me regardless 19:47
oh, the Anything.pm6 indeed gives me an erro
r
tony-o_ fresh checkout out of repo gives me: gist.github.com/tony-o/86af533ec27...2845515ff8
tyil it says OK if I remove the `is export` from the subsets
and the perl6 command I was trying to run is giving a reasonable error now as well, so I can work on again 19:51
19:51 kaare_ left
tyil thanks for your assistance tony-o_ :D 19:51
19:51 molaf left
AlexDaniel releasable6: next 19:51
releasable6 AlexDaniel, Next release in ≈20 days and ≈23 hours. 3 blockers. 81 out of 261 commits logged (⚠ 9 warnings)
AlexDaniel, Details: gist.github.com/d65769a7ccf33655f0...99b9147e1b
tony-o_ np 19:52
19:52 pmurias left 19:57 Kaiepi joined, SteffanW left 20:09 mcmillhj joined 20:14 mcmillhj left 20:23 mcmillhj joined 20:25 vrurg joined 20:27 mcmillhj left, vrurg left 20:51 mcmillhj joined 20:56 mcmillhj left 20:59 kerframil left 21:15 TreyHarris left 21:16 mephinet left
masak TimToady: I'm getting increasingly interested in the COMPILING pseudo-package. when you have the time, I wouldn't mind asking some questions. 21:19
masak learned today that there's an analogue in Perl 5
masak also learned today that a quasi can `require` variables from COMPILING 21:20
21:25 mcmillhj joined 21:29 TreyHarris joined 21:31 mcmillhj left 21:34 sena_kun left, sena_kun joined 21:38 zakharyas left 21:39 mcmillhj joined 21:44 mcmillhj left, jast joined 21:47 sena_kun left 22:01 ryn1x joined 22:05 rindolf left 22:07 mcmillhj joined, vrurg joined 22:09 vrurg left, vrurg joined, mephinet joined 22:10 vrurg left
buggable New CPAN upload: Terminal-Spinners-1.0.0.tar.gz by RYNIX modules.perl6.org/dist/Terminal::Sp...cpan:RYNIX 22:11
22:12 mcmillhj left 22:13 fake_space_whale joined
ryn1x ^ my very small contribution to perl6 =) 22:16
El_Che a fake_space_whale? 22:18
:)
ryn1x: looking fwd to trying it out
fake_space_whale hmm?
El_Che fake_space_whale: ryn1x pointed to the previous line and it was your join message by coincidence 22:19
fake_space_whale oh
ryn1x haha 22:20
fake_space_whale While I am paying attention to this channel, I build a perl6 module as a hobby project but I haven't had the energy/time to write good documentation or optimize it for performance. What is considered to be the minimum standard of quality before releasing a package on CPAN? 22:22
22:26 reu joined 22:31 mcmillhj joined
Juerd None 22:31
ryn1x I was encouraged by others here to contribute when I was just trying to learn about modules and cpan. I would say if it is something that you think others would find useful then share it! A good guide on how to upload etc. is here: docs.perl6.org/language/modules.html 22:32
22:36 mcmillhj left 22:39 lizmat left
jdv79 just did some time and mem tests from the last time i installed rakduo (mar) to today. looks like its slower and bigger:( 22:46
22:46 mcmillhj joined
jdv79 one test went from 4.29s -> 5.29s and from 235MB -> 286MB 22:47
22:48 gabiruh left 22:49 daemon left
AlexDaniel jdv79: can you ticket that? Is there a golf? 22:50
even without a golf, if it's one script we can bench it right here a bit
22:50 mcmillhj left 22:51 daemon joined, daemon is now known as Guest52691, benjikun joined 22:52 Guest52691 left, Guest52691 joined, Guest52691 is now known as daemon
jdv79 its not "news" - its been "slow" forever. 22:52
22:52 vrurg joined
jdv79 its just github.com/jdv/p6-data-selector/bl...ply_tree.t 22:52
under /usr/bin/time -v ... 22:53
timotimo do you happen to know what the previous version of rakudo and friends was?
AlexDaniel I mean, we can bisect it, right?
jdv79 maybe its just the "unfinished/unrealizied" recent jnthn stuff 22:54
benjikun `.WHAT === Type` is a whole lot faster than `.isa(Type)`
yoleaux 10:14Z <FROGGS> benjikun: take a loot at the p6*json.gz at the bottom: ftp://ftp.gwdg.de/pub/languages/perl/CPAN/authors/
AlexDaniel 6c: gist.github.com/AlexDaniel/7102bb2...2a41ea121f
timotimo benjikun: should be even faster to just ~~ Type
committable6 AlexDaniel, gist.github.com/9e88c0ec3bf43e549a...981cba8372
jdv79 no. i just looked and the mtime of the perl6 file before i wiped it out with todays
benjikun Haven't tested that one timotimo
jdv79 it was march something
benjikun I'll try it out :)
22:55 Voldenet joined, Voldenet left, Voldenet joined
benjikun You're right timotimo, pretty significantly faster 22:55
AlexDaniel 6c: gist.github.com/AlexDaniel/7102bb2...2a41ea121f
benjikun `.isa` seems to slow down the more you do it ... 22:56
timotimo ouch, the profiler segfaults when run against that test file :)
benjikun But for 1,000,000 times, the times are: `~~ - 1.0166297, what - 2.8730479, isa - 3.2158791`
MasterDuke timotimo: got a segfault also. but the code is doing an EVAL... 22:57
timotimo many evals
committable6 AlexDaniel, gist.github.com/6dc9b08be59e0537fb...3b95f7a774 22:58
AlexDaniel 6c: gist.github.com/AlexDaniel/7102bb2...2a41ea121f 22:59
let's have another run
but if it's from 2018.06 to HEAD then it'll be easy to bisect
benjikun .tell FROGGS Interesting, perhaps we could do something somewhat similar with the GitHub side of things somehow. An aggregator server that does all of that for zef.
yoleaux benjikun: I'll pass your message to FROGGS.
22:59 vrurg left
timotimo seems like we reached 2.8 at some point, but we're back up to 3.3? 23:00
AlexDaniel yeah, something like that, but let's wait for new results…
23:00 lizmat joined 23:01 gabiruh joined
committable6 AlexDaniel, gist.github.com/3ed04dff1cfc320444...c4564323a3 23:01
AlexDaniel ye
bench: 2018.06,HEAD gist.github.com/AlexDaniel/7102bb2...2a41ea121f
benchable6 AlexDaniel, starting to benchmark the 2 given commits
timotimo i'd like to see the stagestats for all the EVALs we have in there 23:02
benchable6 AlexDaniel, ¦2018.06: «3.1967» ¦HEAD: «3.7439»
AlexDaniel, benchmarked the given commits and found a performance difference > 10%, now trying to bisect
AlexDaniel, ¦2018.06: «3.1967» ¦59faed1: «3.1234» ¦38d046f: «3.7134» ¦HEAD: «3.7439» 23:03
timotimo the vim syntax highlighter gets confused by the pod %)
AlexDaniel 6c: gist.github.com/AlexDaniel/7102bb2...2a41ea121f 23:04
committable6 AlexDaniel, Using file “foo.p6” as a main file, other files are placed in “sandbox/”
AlexDaniel timotimo: ask and you shall receive :)
timotimo i didn't immediately see how to get that, though
AlexDaniel timotimo: with multifile gists you can do any kind of setup really 23:05
any command line flags and any input files :)
timotimo that only gets you the "outer" stagestats
i tried that myeslf, too
AlexDaniel timotimo: what kind of stagestats do you want then? 23:06
committable6 AlexDaniel, gist.github.com/28e67881c7553fb735...bf6fb58bea
AlexDaniel like the module itself?
timotimo well, the module does lots of evaling 23:07
i'd like to see how much time each eval eats
AlexDaniel hmmmm… anyway, back to benchable
bench: 59faed1,38d046f gist.github.com/AlexDaniel/7102bb2...2a41ea121f 23:08
benchable6 AlexDaniel, starting to benchmark the 2 given commits
AlexDaniel, ¦59faed1: «3.1137» ¦38d046f: «3.1111»
AlexDaniel, No new data found
AlexDaniel wat?
timotimo i have a super headache, so i won't be of much use right now 23:09
AlexDaniel timotimo: that's ok, benchable6 has a headache too 23:10
bisect: old=2018.06 new=HEAD gist.github.com/AlexDaniel/7102bb2...2a41ea121f 23:11
bisectable6 AlexDaniel, On both starting points (old=2018.06 new=b7379a8) the exit code is 42 and the output is identical as well
AlexDaniel, gist.github.com/fcafa0f7dadeb445c8...52ff704b31
AlexDaniel oops I did it wrong
bisect: old=2018.06 new=HEAD gist.github.com/AlexDaniel/7102bb2...2a41ea121f
bisectable6 AlexDaniel, Bisecting by exit code (old=2018.06 new=b7379a8). Old exit code: 42 23:12
AlexDaniel, bisect log: gist.github.com/419f675a35248504fe...ccf2a22b46
AlexDaniel, (2018-07-13) github.com/rakudo/rakudo/commit/83...9a39c59288
23:12 Zoffix joined
AlexDaniel alright I'll move to #whateverable 23:13
Zoffix Talk Slides and Recording: "Faster Perl 6 Programs": rakudo.party/post/Faster-Perl-6-Pr...-Recording
AlexDaniel woo! Zoffix++
jdv79 timotimo: what module does a lot o eval?
23:14 benjikun left 23:16 vrurg joined 23:17 benjikun joined 23:21 Shnaw7 joined
Shnaw7 Interested in reasonably priced GLOBAL IRC ADVERTISING? Contact me on twitter twitter.com/nenolod or linkedin www.linkedin.com/in/nenolod 23:21
benjikun Spam :/
AlexDaniel jdv79: alright 23:22
jdv79: so
jdv79: before 2.691408 after 3.27876625, (2018-07-14) github.com/rakudo/rakudo/commit/5c...d0a5098e7d 23:23
23:23 mcmillhj joined
AlexDaniel jdv79: can you file a ticket? 23:24
23:24 Shnaw7 left 23:28 mcmillhj left 23:34 pecastro left 23:38 mcmillhj joined 23:42 Visitatore joined, Visitatore left 23:43 mcmillhj left 23:47 benjikun left, isBEKaml left 23:48 benjikun joined
timotimo jdv79: i believe yours does? perhaps regex interpolation? 23:59