Closed Stranger6667 closed 5 months ago
Hmmm... this seems to be the relevant part:
C:/hostedtoolcache/windows/Ruby/3.2.3/x64/lib/ruby/3.2.0/pathname.rb:532:in `relative_path_from': different prefix: "D:/" and "C:/Users/runneradmin/.cargo/registry/src/index.crates.io-6f17d22bba15001f/rb-sys-0.9.90" (ArgumentError)
from D:/a/css-inline/css-inline/bindings/ruby/vendor/bundle/ruby/3.2.0/gems/bundler-2.4.10/lib/bundler/definition.rb:361:in `ensure_equivalent_gemfile_and_lockfile'
from D:/a/css-inline/css-inline/bindings/ruby/vendor/bundle/ruby/3.2.0/gems/bundler-2.4.10/lib/bundler/runtime.rb:13:in `setup'
from D:/a/css-inline/css-inline/bindings/ruby/vendor/bundle/ruby/3.2.0/gems/bundler-2.4.10/lib/bundler.rb:171:in `setup'
from D:/a/css-inline/css-inline/bindings/ruby/vendor/bundle/ruby/3.2.0/gems/bundler-2.4.10/lib/bundler/setup.rb:23:in `block in <top (required)>'
from D:/a/css-inline/css-inline/bindings/ruby/vendor/bundle/ruby/3.2.0/gems/bundler-2.4.10/lib/bundler/ui/shell.rb:159:in `with_level'
from D:/a/css-inline/css-inline/bindings/ruby/vendor/bundle/ruby/3.2.0/gems/bundler-2.4.10/lib/bundler/ui/shell.rb:111:in `silence'
from D:/a/css-inline/css-inline/bindings/ruby/vendor/bundle/ruby/3.2.0/gems/bundler-2.4.10/lib/bundler/setup.rb:23:in `<top (required)>'
from -e:in `require'
It seems like when trying to get the rbconfig info, bundler is choking. Something about the C:/ and D:/ windows volumes. I must admit, I'm pretty stumped on this one.
It seems like bundler cache is getting restored to D:/
, but Ruby install is in C:/
, causing Pathname#relative_path_from
to choke. At first glance, it seems like a bundler issue. Can you try to update bundler and see if that helps?
bundler update --bundler
I'm hoping there is an upstream issue that was resolved since v2.4.10
Hi @ianks
Indeed, updating bundler resolves the issue! Thank you very much!
As https://github.com/oxidize-rb/rb-sys/commit/04e0634ccec33751420abe6c3cac4f6b1cc5dd36 landed, I noticed that builds of
css-inline
started failing. As far as I understand, it might have exposed some underlying problems there, however withv0.9.87
, the build was successful, and the tests passed. Even though I userb-sys
viamagnus
, it is not 100% clear whether it should be addressed incss-inline
,magnus
, orrb-sys
. Posting it here to start the discussion, let me know if I can provide any additional information: