For now, these emulate the support that `git archive` has for them (namely, that they present as empty directories, which are otherwise impossible to create/represent in Git).
(it verifies that the downloaded modules have not been tampered with since `go mod download`, which we do literally just prior so it is verifying nothing)
In Go 1.24, there's a new `go vet` rule that complains about `Example*` test functions that don't follow the documented naming convention. Combine that with `go test` running `go vet` by default, and you've got a perfect storm.
This renames our "example" tests to satisfy Go's naming convention.
This means slightly more typing in "zero-value" cases (`nil` vs `dockerfile.Metadata{}`), but the tradeoff is that it's simpler to use and reason about (and all the struct members are pointer-type map/slice values anyhow, so copying the struct is still pretty cheap).
This also swaps the scanner error handling to return the partially parsed Metadata object alongside the scanner error -- the error already tells us the object isn't fully complete data, so it's fair/fine to return and will likely just be ignored by the caller instead. This also allows us to get to 100% code coverage. 👀
This also updates our "treat `oci-import` just like `FROM scratch`" code to *actually* parse `FROM scratch` so we can't accidentally cause "missing data" bugs there in the future, and I implemented that using `sync.OnceValues` which requires upgrading to Go 1.21, but IMO that's a worthwhile tradeoff (because `sync.OnceValues` makes that code so clean/simple).
There were some very minor/subtle bugs in how I implemented continuation that wouldn't affect any real-world parsing we did, but still bothered me because I'm me. This fixes them (and further increases test coverage as a result).
This script needs/uses a custom `BASHBREW_LIBRARY` directory, but it stores that value in the exported environment slightly too soon such that `generate-stackbrew-library.sh` picks it up (and shouldn't be).
We often use "`BASHBREW_LIBRARY` is unset (or empty)" as a conditional for whether to fall back to using "https://github.com/docker-library/official-images/raw/HEAD/library/" as an explicit prefix for querying "source of truth" values for things like supported parent architectures.
These two things collided in cc2dc88e04 (and similar commits) because the script saw `BASHBREW_LIBRARY` set, trusted it, but then fails to find the parent image.
This is the cleanest place to fix this such that `generate-stackbrew-library.sh` can take `BASHBREW_LIBRARY` from the provided environment instead of using our generated value.
I thought this was already the behavior, but I guess it was relaxed because previous iterations of this validation had to apply to the older format where we'd been less meticulous about enforcing this. Since those are all gone now, we can safely update the validation to enforce that commit hashes *must* be fully qualified.
It's been failing to upload for a while, and I don't think the failures are *completely* their fault (GitHub's got to share some of the blame, I think), but the end result is that we don't really have any good options for continuing to use the service (bad/unacceptable options include install an app with way too many privileges on the org/repo or add a "secret" with a personal access token ... and make that available to Pull Requests too 🙃).
This should be more reliable/correct than `uname -m` (because we're almost always looking for the *userspace* architecture, not the kernel architecture).
In my refactoring to use `go-git`'s `Tree` objects, I missed this edge case (that symlinks get resolved to be relative to the Git root, but our `Tree` object is a subdirectory).
Tianon Gravi