00:02 reportable6 left 00:03 reportable6 joined 00:33 leont left 00:47 elcaro joined
elcaro FYI, I rebuilt rakudo, probably last update was on Thursday/Friday. Now running REPL spits out error when it starts 00:48
Use of uninitialized value element[0] of type Any in string context in any protect at gen/moar/stage2/NQPCORE.setting line 1216
so most likely caused by some update in last 2 days 00:49
AlexDaniel m: run ‘perl6’ 00:50
camelia 5===SORRY!5=== Error while compiling -
Missing infix inside hyper
at -:1
------> 3»7⏏5Wann treffen wir drei wieder zusamm?«
expecting any of:
infix stopper
The spawned command 'perl6' exited unsucces…
AlexDaniel m: run <perl6 -e>, ‘run ‘perl6’’ 00:51
camelia 5===SORRY!5=== Error while compiling -
Missing infix inside hyper
at -:1
------> 3»7⏏5Wann treffen wir drei wieder zusamm?«
expecting any of:
infix stopper
The spawned command 'perl6' exited unsucces…
AlexDaniel m: slurp; run <perl6 -e>, ‘run ‘perl6’’
camelia ( no output )
AlexDaniel e: slurp; run <perl6 -e>, ‘run ‘perl6’’
AlexDaniel okay that doesn't work of course 00:52
elcaro rolled back to aed658d and the error's gone... so it's one of the commits in the past 12 hours 00:57
AlexDaniel elcaro: can you bisect it? :) 01:00
01:06 squashable6 left 01:08 squashable6 joined
elcaro will do 01:08
just listening to my company's COVID-19 updates right now :/ 01:09
01:16 Altai-man_ joined 01:18 sena_kun left
elcaro AlexDaniel: 86bc3b3 is the culprit 01:30
AlexDaniel 86bc3b3
linkable6 (2020-03-15) github.com/rakudo/rakudo/commit/86bc3b3d29 Remove unnecessary implementation detail
AlexDaniel yeah, too short for the bot to pick up :)
elcaro: can you file a ticket? 01:31
elcaro sure
#3552 01:47
AlexDaniel R#3552 02:54
linkable6 R#3552 [open]: github.com/rakudo/rakudo/issues/3552 Errors on REPL launch
03:09 b2gills joined 03:16 sena_kun joined 03:18 Altai-man_ left 05:16 Altai-man_ joined 05:18 sena_kun left 07:16 sena_kun joined 07:18 Altai-man_ left 07:57 leont joined
nine Looks like repl.t is consistently failing now: build.opensuse.org/package/live_bu...5.2/x86_64 08:03
sena_kun how can all tests pass when using `prove --jobs=4 -e "raku -Ilib" t/` but with `zef test .` some (not all) tests die with a proper object being treated as, eh, a Stash? going afk before $dayjob, maybe anyone can take a look at github.com/Altai-man/perl6-Perl6-Parser ? This is just bizarre and looks like something is wrong with either rakudo or zef (I'd bet on rakudo). 08:10
gist.github.com/Altai-man/f7931a05...f28a070a45 is the log 08:12
08:19 stoned75 left
lizmat Files=1305, Tests=111216, 207 wallclock secs (28.20 usr 8.38 sys + 2892.74 cusr 275.34 csys = 3204.66 CPU) 08:28
leont Could be a number of things 08:47
I think zef sets PERL6LIB, which means its also passed to child processes 08:48
What does «PERL6LIB=lib prove --jobs=4 -e raku t/» do?
08:52 stoned75 joined 09:16 Altai-man_ joined
nine lizmat: first repl.t failure was yesterday 20:05 UTC+1. So must be a change shortly before that 09:17
09:18 sena_kun left
lizmat I think we have an issue and an offending commit: github.com/rakudo/rakudo/issues/3552 09:18
nine lizmat: ah, very well. What an odd issue 09:20
Geth rakudo: d3385e73c4 | (Elizabeth Mattijsen)++ | 7 files
Revert "Remove unnecessary implementation detail"

This reverts commit 86bc3b3d296fde7330b9ab17ce74f101ae07b9b9.
This appears to be breaking REPL startup on linuxen
nine FWIW I can't reproduce it locally on Linux either, but it fails all the time on the OBS
lizmat on MacOS it's fine
anyways, it was part of a cleanup: we've been living with that "dirt" for many years... so a little longer won't hurt 09:21
09:24 hungrydonkey joined
lizmat will be busy with the RWN first 09:25
Altai-man_ `PERL6LIB=lib prove --jobs=4 -e raku t/` passes fine. :S 09:28
lizmat afk for a few hours& 09:48
rba sena_kun: I have installed libperl-dev Ubuntu/Debian package on blin server. May you check if this resolved the dependency for Inline::Perl5. (Sorry for the delay...) 09:53
tellable6 rba, I'll pass your message to sena_kun
rba Altai-man_: I have installed libperl-dev Ubuntu/Debian package on blin server. May you check if this resolved the dependency for Inline::Perl5. (Sorry for the delay...) 09:56
tellable6 2020-03-15T11:52:54Z #raku-dev <sena_kun> rba hi! Can you please install perl-dev (or how is it called) to the blin server, so -lperl flag was supported to build Inline::Perl5 there?
2020-03-15T23:42:43Z #raku-dev <patrickb> rba I uploaded rakubrew v6 (patszim.volans.uberspace.de/patclo...LQDZcjbf). Upload at your convenience. :-)
2020-03-15T23:43:08Z #raku-dev <patrickb> rba Also I have a new version of the rakubrew.org website. Can you bump that?
Altai-man_ let's see...
09:58 hungrydonkey left
Altai-man_ rba, installs fine now, thanks! 09:58
rba Altai-man_: cool.
10:10 hungrydonkey joined
rba patrickb: rakubrew.org is on v6, running on the newest container build from rakubrew.org:master github repo. 10:13
tellable6 rba, I'll pass your message to patrickb
10:16 hungrydonkey left 10:20 ufobat joined 10:55 ufobat_ joined 10:56 ufobat left 10:58 ufobat_ left
AlexDaniel Altai-man_: what a fruitful run! 11:10
leont Also, IMHO people should be using prove6 instead of prove in general. It's silly to depend on perl for raku testing. 11:11
AlexDaniel why 11:13
we still depend on it to compile rakudo, don't we? 11:14
lizmat AlexDaniel: yeah, and that will have to change 11:15
11:16 ufobat joined
lizmat as the main reason: perl being available on all OSes, is becoming less and less true 11:16
AlexDaniel okay then something changed while I was away :D
last time we talked about this everyone was like “meh, it works, no need to change it”
11:17 sena_kun joined
leont Also, most users of raku shouldn't be compiling their own rakudo 11:17
Needing it for testing means the dependency affect many more people 11:18
11:18 Altai-man_ left
sena_kun I don't really mind using prove or prove6 or just running files by hands, the problem is that zef test fails. 11:20
tellable6 2020-03-16T09:53:59Z #raku-dev <rba> sena_kun: I have installed libperl-dev Ubuntu/Debian package on blin server. May you check if this resolved the dependency for Inline::Perl5. (Sorry for the delay...)
sena_kun While running files on their own is successul, which is... odd. 11:21
lizmat notable6: weekly 11:49
notable6 lizmat, 2 notes: 2020-03-13T17:10:30Z <lizmat>: grants.perlfoundation.org ; 2020-03-14T22:18:04Z <lizmat>: www.facebook.com/groups/raku.perl6...287605181/
12:02 reportable6 left 12:03 reportable6 joined 12:04 MasterDuke left 13:18 sena_kun left 13:19 sena_kun joined 14:08 hungrydonkey joined
[Tux] Rakudo version 2020.02.1-193-gd3385e73c - MoarVM version 2020.02.1-45-g2252a95df
csv-ip5xs0.720 - 0.786
csv-ip5xs-206.238 - 6.268
csv-parser26.769 - 27.542
csv-test-xs-200.369 - 0.372
test8.463 - 8.948
test-t1.950 - 1.982
test-t --race0.930 - 0.958
test-t-2033.041 - 33.405
test-t-20 --race9.939 - 10.003
14:38 leont left 14:42 squashable6 left 14:45 squashable6 joined 15:18 Altai-man_ joined 15:21 sena_kun left 16:00 lucasb joined 16:16 stoned75 left 16:21 hungrydonkey left
lizmat notable6: weekly reset 16:43
notable6 lizmat, Moved existing notes to “weekly_2020-03-16T16:43:48Z”
17:03 stoned75 joined 17:19 sena_kun joined 17:21 Altai-man_ left 17:36 stoned75 left 17:57 Xliff joined
Xliff Can someone help me with a simple Cro app? 18:00
sena_kun Xliff, maybe I can? also, #cro
Xliff I have the routes right, but so far I am not getting any output from any of the controllers.
sena_kun: github.com/Xliff/p6-RouterProject
Why are my routes not showing up. I can point to the controllers but I just get a blank page. 18:01
sena_kun Xliff, I don't think that's how `include` work, can you try `include routes1` then `include routes2` etc? 18:02
here, instead of passing them as an array 18:03
jnthn I think the array of pairs probably should work, fwiw
The key is the prefix to add to the routes
sena_kun jnthn, even if it should, we can narrow the problem scope if know separate includes work
jnthn Sure :)
Xliff: What URLs are you trying? Blank page suggests 404 (try it with curl -v or some such to see the status code) 18:04
18:04 stoned75 joined
Xliff When I run the app, I can see all pages return 200, but no output. 18:05
The slip is working.
sena_kun github.com/Xliff/p6-RouterProject/...mon.pm6#L6 looks suspicious, if you do just `content 'text/plain', 'foo'` it works?
Xliff Wow. loading takes for ever. 18:06
OK. Just 'foo' works. Odd.
sena_kun then `warn $your-code-here` and see what it ends up there 18:07
Xliff :S 18:08
Does &?ROUTINE work in a pointy block?
jnthn Well, it means the enclosing routine, and a pointy block is not a routine 18:09
Did you mean &?BLOCK?
Xliff my $a = -> 'two', 'three' { &?BLOCK.signature.params.map( *.gist.Str ).join(', ') }; $a('two', 'three')
m: my $a = -> 'two', 'three' { &?BLOCK.signature.params.map( *.gist.Str ).join(', ') }; $a('two', 'three')
camelia ( no output )
Xliff m: my $a = -> 'two', 'three' { &?BLOCK.signature.params.map( *.gist.Str ).join(', ').say }; $a('two', 'three')
camelia "two", "three"
Xliff Excellent! 18:10
jnthn++: Yep! It was the &?BLOCK instead of &?ROUTINE, and the array is fine! \o/ 18:11
Now... jnthn and sena_kun... what's the best way to remove a route from Cro? 18:12
18:15 stoned75 left
sena_kun based on what I saw in Router.pm6, this isn't really possible 18:16
Xliff, what do you want to do? 18:17
maybe just a very generic route with a smart `where` clause will do
jnthn You can't really remove one, but you can probably hide it by writing a `route` block that matches the route you want to hide, and then delegates everything else 18:19
Then just `not-found` or so in the body of the handler
18:29 ilogger2 joined
jnthn Xliff: If you want to do this across a bunch of routes, maybe `before { not-found if $hidden }` is better than putting something in all of them 18:37
18:37 pmurias joined
pmurias bartolin_: ping 18:37
bartolin_ pmurias: hi 18:41
bartolin_ will be back later 18:55
19:18 Altai-man_ joined
Xliff jnthn: Ooh! That's a big help, thanks! 19:26
jnthn: Does before need to be in the route block? 19:27
19:34 MasterDuke joined
jnthn Yes 19:35
Xliff When I add that, I get "Can only use 'include' with 'route' block, not a Cro::CompositeTransform::WithConnectionState" 19:36
github.com/Xliff/p6-RouterProject/...outer1.pm6 19:38
Should before use a pointy? 19:39
lizmat A very sad Rakudo Weekly News hits the Net: rakudoweekly.blog/2020/03/16/2020-...od-friend/ 19:43
Xliff OMG!
jnthn Xliff: You should use delegate, not include, when doing this 19:55
dinner, bbl
19:55 pmurias left 20:01 stoned75 joined
Xliff Does delegate have the ability to accept a block, or does it require a clasS? 20:51
Altai-man_ delegate works with Cro::Transform. You can write one yourself if something specific or just use `route` routine and it'll create one for you 20:53
Xliff OK... again, the documents are not very helpful when trying to figure out the details as to how things work. 20:55
Heres what I want to do: I want to be able to create routes that may NOT be the same between server invocations. And I want the various routes to be able to have the ability to be hidden by a specific group. Each group is has a sub that defines all of the routes. 20:56
Given the need to use before() for the simple hidden handlers implies the use of delegate. 20:57
I have to step away for a bit. Altai-man: if you could give me the pointers on how to do above, it would really be appreciated!
Altai-man_ Xliff, I think you want to have a single generic route handler for all possible urls and do this work yourself, because cro router is not so flexible to add/remove routes on the fly. 21:00
so you have a single route which takes all URL pieces, say `*@path-pieces` in signature, and then in this route you expect headers, cookies, patch pieces to decide if you're returning something to the user or not. 21:01
21:06 stoned75 left 21:14 stoned75 joined 21:19 sena_kun joined 21:21 Altai-man_ left 21:56 stoned75 left
Geth nqp: f8740ded1f | (Elizabeth Mattijsen)++ | tools/templates/MOAR_REVISION
Bump Moar to get latest JIT fixes
rakudo: 9ebbad3062 | (Elizabeth Mattijsen)++ | tools/templates/NQP_REVISION
Bump NQP to get latest MoarVM improvements
lizmat sleep&
23:18 Altai-man_ joined 23:21 sena_kun left