Implemented non-zero exit code when directory initialization fails

[Mastermindaxe]

Changes

• Implemented non-zero exit code when directory initalization fails

Related Issues

• Closes Update build script error behavior #110

Sorry for being absent for long. Life happened :D Thought I'd do some cleaning up. Not sure how to write tests for this behaviour though. Have a good one!

[Mastermindaxe]

Not sure why the clippy hook fails as it doesn't highlight that error on my machine. The pr.linux job failing doesn't seem to be my doing. Ignore and proceed or should we investigate as it happens in other PRs as well?

[calebcartwright]

Thanks for this! I actually want to think through whether this is still a change we really want to make, though if we do, we'll need to restrict it to running during the lib builds (we wouldn't want to do it for the bin).

Not sure why the clippy hook fails as it doesn't highlight that error on my machine. The pr.linux job failing doesn't seem to be my doing

Don't worry about it for now, not related to your changes. On the CI side we do some caching of certain dev tools that aren't preloaded on the build machines so that the builds can execute more quickly (e.g. don't have to recompile the code coverage tool on every run). However, the cache gets wiped after a while, as has happened here. I need to go in and refresh it.

Ignore and proceed or should we investigate as it happens in other PRs as well?

Don't want to ignore it, but also wouldn't be able to merge this just yet anyway

[Mastermindaxe]

Thanks for this! I actually want to think through whether this is still a change we really want to make, though if we do, we'll need to restrict it to running during the lib builds (we wouldn't want to do it for the bin).

I see! Do we set an env variable or something to control it? If so I can implement the change and just put it on ready until you decide what to do with it.

[calebcartwright]

I see! Do we set an env variable or something to control it? If so I can implement the change and just put it on ready until you decide what to do with it.

cargo sets various env variables which can be inspected, and it inherently has this knowledge so I'm sure there's a cargo var we can just check programmatically. Actually, now that I think about it, there's no reason to attempt to init during a bin install regardless of what we may or may not do on lib builds. Will open a separate issue for that as it can, and should, be addressed separately (refs #144)

(edit - added issue link)

[codecov]

Codecov Report

Merging #142 (fdfbf2e) into master (e284e15) will not change coverage.
The diff coverage is n/a.

@@            Coverage Diff            @@
##            master      #142   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files            4         4           
  Lines          144       142    -2     
=========================================
- Hits           144       142    -2     

Impacted Files	Coverage Δ
src/git.rs	100.00% <0.00%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update e284e15...fdfbf2e. Read the comment docs.

[calebcartwright]

Side note, have fixed the cache issue so now the only failing job is clippy. That's a legitimate failure, presumably caused by some updated lints, #145 opened to track resolving that issue