🦋 Welcome to Raku! raku.org/ | evalbot usage: 'p6: say 3;' or /msg camelia p6: ... | irclog: colabti.org/irclogger/irclogger_log/raku
Set by ChanServ on 14 October 2019.
SmokeMachine Could some one give me some opinion about this? github.com/FCO/Red/issues/141 02:22
abraxxa I've just updated my raku from the new cloudsmith repo after wondering why I don't get any updates. Now zef weils because it can't write to the raku directory: Failed to create directory '/opt/rakudo-pkg/share/perl6/site/short' with mode '0o777': Failed to mkdir: Permission denied 09:29
El_Che mmm 09:30
can you rm -rf .raku .perl and run /opt/rakudo-pkg/bin/install-zef-as-user 09:31
abraxxa in my home dir?
I don't have a ~/.perl dir
same error messages when running /opt/rakudo-pkg/bin/install-zef-as-user 09:32
El_Che sorry
.perl6 09:33
abraxxa don't have that either
I guess I have cleaned that up already
El_Che .raku is the most recent one used by zef
which zef
abraxxa /opt/rakudo-pkg/bin/zef
El_Che ok
that only works are root
maybe I need to add a note 09:34
abraxxa it did work as user before
El_Che /opt/rakudo-pkg/bin/install-zef-as-user installs a local zef
in .raku/bin
abraxxa that fails with the same error message
El_Che that one has rw access to your home\
/opt/rakudo-pkg/bin/install-zef-as-user then
.raku/bin/zef
you can add it t oyour path later 09:35
abraxxa when I run /opt/rakudo-pkg/bin/install-zef-as-user it fails with the same error message
El_Che ah the script failes you mean?
abraxxa yes
El_Che oh, that's annoying 09:36
abraxxa [ahartmai@ahartmai-nb:~$ ]$ /opt/rakudo-pkg/bin/install-zef-as-user
WARNING: unhandled Failure detected in DESTROY. If you meant to ignore it, you can mark it as handled by calling .Bool, .so, .not, or .defined methods. The Failure was:
Failed to create directory '/opt/rakudo-pkg/share/perl6/site/short' with mode '0o777': Failed to mkdir: Permission denied
in any statement_control at /opt/rakudo-pkg/bin/../share/perl6/lib/Perl6/Grammar.moarvm line 1
in sub MAIN at /opt/rakudo-pkg/bin/zef line 3
in block <unit> at /opt/rakudo-pkg/bin/zef line 1
the first error in a list of many
python precomplies all included py files on installation 09:37
El_Che zef gives terrible advise, chmod 777 09:38
abraxxa maybe the raku package can do the same
El_Che I am looking if something changed in zef
abraxxa thanks!
zef --version gives v0.11.4 after all those errors
El_Che on my install the erros are cosmetic, but zef does work 09:39
is it also in your case?
El_Che abraxxa: what distro are you using? 09:40
abraxxa after the list of errors it says 09:41
All candidates are currently installed
No reason to proceed. Use --force-install to continue anyway
Don't forget to add '~/.raku/bin' to your PATH,
e.g. by adding this to .profile:
export PATH=~/.raku/bin:$PATH
but .raku/bin is empty 09:42
Ubuntu 20.10
El_Che find .raku -name "zef"
abraxxa nothing
El_Che of system zef then
looking at it now 09:43
abraxxa I've tried purging and reinstalling rakudo-pkg, didn't change anything 09:53
El_Che I have a radical fix, but it's the correct one
zef is confused about root and user install
I'll remove root zef and make the install-zef-as-user script smarter 09:54
El_Che now install-zef-as-user uses zef to install the user's zef 09:54
abraxxa I'd let zef always save the precomp files in a per user directory instead of rakudo directory
El_Che sonmething changed in upstream rakudo (or zef)
but it was a source of confusion anyways 09:55
El_Che new packages will be up very soon 09:59
testing to making sure nothing breaks 10:00
should have done it ages ago, but it's the first time zef breaks (for some people) 10:02
not really break, just complains :)
sena_kun El_Che, details?
El_Che sena_kun: github.com/nxadm/rakudo-pkg/issues/78 10:03
by adapting how zef is installing by the user, we won't see that problem any more 10:04
sena_kun Oh, this is not cool.
But I think it was not introduced by a point, was it?
El_Che and it makes more sense anyway, because people are confused by root zef and user zef all the time
sena_kun Ah, it dates back to 2018.
El_Che fixing it once and for all
sena_kun El_Che++
El_Che sena_kun: I only saw it with 2020.12 10:05
and not always
sena_kun El_Che, still not something that calls for another point.
El_Che I thought it was fixed once the cached was cleaned up
sena_kun: no, not created by the latest release
sena_kun OTOH I won't be able to make it anyway, not sure how many times I can do it again at all.
El_Che and the package workaround will fix the confusion anyway 10:06
so sidestepping the bug
El_Che abraxxa: testing the changes as we speak on github actions 10:21
if ok, packages should be up after tagging the build 10:22
abraxxa El_Che++
abraxxa Github Actions really queues the job quite long! 10:57
PimDaniel \o 11:02
I struggle to capture raw keyboard events into THIS code: pastebin.com/GaLQnau2
The problem is that nothing happens when i hit a key for the first time. :( 11:03
and the next time : all is fine. 11:04
To use my class : use Events :DEFAULT, :EVENTS; 11:05
tyil abraxxa: good to know it wasn't just me, then 12:20
El_Che yeah, I cancelled it (seems not to work) and started a new build 12:24
tyil El_Che: I can't even seem to cancel this workflow, it just gives me an error that even that failed :p 12:26
El_Che tyil: yeaj, I also have one stuck, but others runs after work fine 12:48
tyil perhaps I'll take another look tonights
El_Che abraxxa: released 14:04
PimDaniel \o 14:07
PimDaniel Does Term::Readkey works really? modules.raku.org/dist/Term::ReadKe...eadKey.pm6 14:08
Running the Pod example i get : Use of uninitialized value of type Str in string context.\nMethods .^name, .raku, .gist, or .say can be used to stringify it to something meaningful.\nNil. Displayed on stderr by an infinite loop. :/ 14:10
abraxxa /opt/rakudo-pkg/bin/install-zef ? 14:11
El_Che: same error 14:13
El_Che: Unpacking rakudo-pkg (2021.2.1-01) over (2021.2.0-01) ...
El_Che: is this the correct, new package?
PimDaniel How to uninstall a module and all its dependencies with zef? 14:18
abraxxa i can't run zef any more 14:25
El_Che abraxxa: did you run /opt/rakudo-pkg/bin/install-zef as a user? 14:33
abraxxa El_Che: yes, same errors as before 14:34
El_Che: [ahartmai@ahartmai-nb:~$ ]$ /opt/rakudo-pkg/bin/install-zef
WARNING: unhandled Failure detected in DESTROY. If you meant to ignore it, you can mark it as handled by calling .Bool, .so, .not, or .defined methods. The Failure was:
Failed to create directory '/opt/rakudo-pkg/share/perl6/vendor/short' with mode '0o777': Failed to mkdir: Permission denied
in method probe at bin/zef line 71
in block at bin/zef line 135
in code at bin/zef line 121
in submethod TWEAK at bin/zef line 70
in submethod TWEAK at bin/zef line 326
in sub get-client at bin/zef line 1173
in sub MAIN at bin/zef line 420
in sub MAIN at bin/zef line 314
in block <unit> at bin/zef line 3
El_Che I wonder why it didn't happen in github actions 14:35
perry Careful with your copy-pasting, you'll be booted if you send too many messages.
(Well, in most channels anyway—I assume this one as well.)
PimDaniel I 14:36
El_Che of, I'll add the dir myself with sane rights
PimDaniel hi
What is the module helper command in raku?
El_Che sudo mkdir /opt/rakudo-pkg/share/perl6/site/short before running zef does the trick here 14:37
so zef does not need to write there, just make sure the dir is there
a rakudo bug
abraxxa: thx for testing with me, I'll add the dir and release a revision
building revision here: github.com/nxadm/rakudo-pkg/action.../610818709 14:42
abraxxa El_Che: should I test anything before you cut a release?
El_Che sudo mkdir -p /opt/rakudo-pkg/share/perl6/site/short 14:43
rm -rf ~/.raku
/opt/rakudo-pkg/bin/install-zef
that should do it
abraxxa Failed to create directory '/opt/rakudo-pkg/share/perl6/vendor/short' 14:53
but just a single error message
so this dir might be needed too 14:54
El_Che I'll add it 14:56
good catch 14:58
abraxxa now zef works 15:19
thanks for the quick fix!
El_Che \o/ 15:23
thx for the help abraxxa 15:35
abraxxa++
lizmat and yet another Rakudo Weekly News hits the Net: rakudoweekly.blog/2021/03/01/2021-...t-of-raku/ 18:12
El_Che already? 18:13
:)
codesections lizmat++ 18:14
eseyman seems like only 7 days since the last one... 18:22
lizmat that *is* the goal :-) 18:24
leont was too late for his new module 18:39
lizmat leont: there will be another weekly :-) 19:02
guifa hey leont, that’s MY line to use when the weekly hits 19:37
also, lizmat++
lizmat: I’m not on the twitter, but you might point out that my dissertation (in digital humanities) was done half/half in RAku and JS (the latter for a UI in a broswer) 19:43
El_Che guifa: do you have a link? 19:44
guifa El_Che: trace.tennessee.edu/utk_graddiss/5840/
NO JUDGING FOR MY CODE
lol 19:45
Pandemic and all made me have to rush a lot of things to get them to a “just work” state
El_Che no, the humanities part is the one that triggered me
guifa I’ll be feeding off of the dissertation for a while (I only did an initial complex diff/analysis for a few chapters, an dI’ll eventually do all of them once I have time to transcribe. Which will then feed into an ML project to automate diplomatic transcriptions of older texts), so by the end of summer there will probably be a few modules released relating to it 19:47
El_Che Downloaded to have a look, it looks promising 19:50
In a past life I did research on original documents around 1600 19:51
and used programming to help me with discourse analysis
El_Che Because I didn't attend classes, I was badly surprised that paleography was a subject that needed attendance :) 19:52
codesections vrurg: wait, I'm confused. Did your PR/work on the new protocol add the TypeName($val) syntax for coercion? Or was that already there? 20:53
I thought the new part was just the COERCE method? 20:54
vrurg codesections: It was always there. Try class Foo { }; say Foo(42); on anything before the merge. 20:56
codesections Ok, that's what I thought. So the issue you linked _isn't_ the same thing 20:56
that's about the new feature not being documented 20:57
but there's some old syntax that also isn't in the docs
codesections (admittedly newly useful syntax) 20:57
vrurg codesections: BTW, as I'm stuck in my car for a while having fun playing "commit police". Could you, please, use more informative commit message in R#4231? Referencing to an issue is not really helpful for 'git log' readers. 20:58
vrurg I don't think it makes sense to have separate issues on very related topics. Coercions just need updated docs, what else? 20:59
codesections vrurg: ??? R#4231 has a 4 paragraph commit msg 21:00
vrurg codesections: Oops, I meant the roast PR. Let me re-check. 21:01
codesections Do you mean roast 719?
vrurg Yeap, that one. Sorry.
codesections oh, ok, will do
vrurg Just copied wrong thing
vrurg is leaving. Will be back in few hours. 21:02
codesections (I'd seen some Roast PRs that just referenced the feature under test, so I thought that was the practice) 21:02
Thanks for letting me know!
sortiz \o #raku 23:40