Closed melezhik closed 2 years ago
Looks like Pretty::Table issue - http://sparrowhub.io:2222/report/388
@melezhik Thank you for bringing this issue to my attention!
It seems to me that Pretty::Table
is "abandoned." (I install it from GitLab, not from PAUSE / raku.land.)
What is the best way to address those issues?
Pretty::Table
in Data::Reshapers
with a more maintained package.Data::Reshapers
.Pretty::Table
.
@antononcube I'd ask in Raku IRC
I think fixed this issue by submitting a new version of the package "Pretty::Table".
Should bump a version of Pretty::Table as well? - it still fails - http://sparrowhub.io:2222/report/531
I bumped the version. But apparently I fixed another bug (about the slicing method) not the wrap-text
related one.
Currently, for some reason an old version of "Text::Wrap" is picked.
Here is the new version: https://raku.land/github:jkramer/Text::Wrap .
Here is what I get removing and re-installing "Text::Wrap":
> zef install Text::Wrap
===> Searching for: Text::Wrap
===> Updating fez mirror: https://360.zef.pm/
===> Updated fez mirror: https://360.zef.pm/
===> Testing: _:ver<0.0.1>:auth<zef:codesections>
===> Testing [OK] for _:ver<0.0.1>:auth<zef:codesections>
===> Installing: _:ver<0.0.1>:auth<zef:codesections>
So, it not yet fixed, yeah?
here is the report for your repo - http://sparrowhub.io:2222/report/532
I fixed the "Pretty::Table" issues.
Now I get failures in one of these test files: http://sparrowhub.io:2222/report/706 .
But I do not get these failures when I install that package locally (macOS) or on a certain remote Linux (Ubuntu) machine. What would be the reason? Is there something I can investigate / debug further? (I think I need your help... :) )
Please enable verbose mode in .sparkyci.yaml and run again - https://github.com/melezhik/sparkyci/blob/main/docs/dsl.md
I did that -- please take a look:
(I removed the redundant "use Hash::Ordered", but I have not seen a build after that...)
yeah, underlying sparkyci worker hung, so I restarted it and new build should come up
The last build shows a cause of the issue - sparrowhub.io:2222/report/763
Yeah, I saw the errors:
21:39:10 06/27/2022 [bash: run tests] :: # Failed test 'join across inner'
21:39:10 06/27/2022 [bash: run tests] :: # at t/08-join-across.rakutest line 61
21:39:10 06/27/2022 [bash: run tests] :: # expected: $(${:A("a"), :B("n"), :C(5), :D("i"), :E(13)}, ${:A("l"), :B("n"), :C(3), :D("i"), :E(13)})
21:39:10 06/27/2022 [bash: run tests] :: # got: $(${:A("a"), :B("n"), :C(5), :D("i"), :E(13)}, ${:A("a"), :B("n"), :C(5), :D("i"), :E(13)})
21:39:10 06/27/2022 [bash: run tests] :: Cannot modify an immutable Nil value
21:39:10 06/27/2022 [bash: run tests] :: in block at /home/raku/.raku/sources/F7993C3B2856EABC58CC7CF9EADCA6DDB4F03527 (Hash::Merge) line 53
21:39:10 06/27/2022 [bash: run tests] :: in sub merge-hash at /home/raku/.raku/sources/F7993C3B2856EABC58CC7CF9EADCA6DDB4F03527 (Hash::Merge) line 57
21:39:10 06/27/2022 [bash: run tests] :: in sub JoinAcross at /home/raku/.sparky/projects/raku/.sparrowdo/scm/lib/Data/Reshapers/JoinAcross.rakumod (Data::Reshapers::JoinAcross) line 93
21:39:10 06/27/2022 [bash: run tests] :: in sub JoinAcross at /home/raku/.sparky/projects/raku/.sparrowdo/scm/lib/Data/Reshapers/JoinAcross.rakumod (Data::Reshapers::JoinAcross) line 44
21:39:10 06/27/2022 [bash: run tests] :: in sub JoinAcross at /home/raku/.sparky/projects/raku/.sparrowdo/scm/lib/Data/Reshapers/JoinAcross.rakumod (Data::Reshapers::JoinAcross) line 25
21:39:10 06/27/2022 [bash: run tests] :: in sub join-across at /home/raku/.sparky/projects/raku/.sparrowdo/scm/lib/Data/Reshapers.rakumod (Data::Reshapers) line 94
21:39:10 06/27/2022 [bash: run tests] :: in block <unit> at t/08-join-across.rakutest line 62
21:39:10 06/27/2022 [bash: run tests] ::
Those errors do not happen on my machine (macOS). On what platform SparkyCI executes the tests?
On alpine Linux, also please see Rakudo version in a init tab, could be a rakudo specific bug 🤔
Well, it looks ok now: http://sparrowhub.io:2222/report/766 .
I made two changes:
Made the reduce statement to pass !:deep
to hash-merge
Put more detailed dependency on "Hash::Merge" (version and api)
I am not sure if 1 is "needed" to pass the tests, but it is more correct from a data wrangling perspective.
Good!
@melezhik Thank you for your efforts on SparkyCI, and thank you for bringing my attention to this issue!
Hi!
Installation fails, with such an error:
Full report - http://sparrowhub.io:2222/report/386
cc @uzluisf
HTH
Aleksei