You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Describe the bug
This is dumb, but this actually happened with us. I won't disclose the name of the crate we have in our particular use case, but suppose it's named foo. We have it at the latest version x.y.z in our private crates registry. However, the same-named crate foo exists on crates.io and its version x.y.z was yanked.
To Reproduce
Steps to reproduce the behavior:
Publisn a crate with the same name and version as some existing crate on crates.io. For example, take syn v2.0.19, which is yanked.
Add this new crate from your private registry to another crate as a dependency. In our case, we still use the git-based index (see the offtopic section at the bottom of the issue). We have a config like this:
[registries]
private = { index = "https://dl.cloudsmith.io/basic/redacted-org/redacted-repo/cargo/index.git" }
Now run cargo deny check
You'll see an error saying that the crate from the private registry was yanked
Expected behavior
No "yanked crate" advisory should be reported, because the crate comes from our private registry, and it is different from the same-named crate on crates.io
Screenshots
Here is the redacted cargo-deny output:
Offtopic
I know Embark uses a private crates registry as well. I wonder if you still sit on a git-based registry index? We can't use the sparse crates registry index because it doesn't support auth tokens. Cloudsmith, for example still provides a way to use the sparse registry, but the auth token in this case must be part of the registry index URL, which in turn means this URL becomes part of Cargo.lock and therefore the secret auth token needs to be committed into source control. This is because the issue rust-lang/cargo#10474 is not done. So.. I just wanted to ask wether it also blocks your migration to spare registry index.
The text was updated successfully, but these errors were encountered:
Veetaha
changed the title
False-positively yank report if a crate is in a private registry, but same-named crate from public registry was yanked
False-positive yank report if a crate is in a private registry, but same-named crate from public registry was yanked
Jul 13, 2023
Describe the bug
This is dumb, but this actually happened with us. I won't disclose the name of the crate we have in our particular use case, but suppose it's named
foo
. We have it at the latest versionx.y.z
in our private crates registry. However, the same-named cratefoo
exists on crates.io and its versionx.y.z
was yanked.To Reproduce
Steps to reproduce the behavior:
cargo deny check
Expected behavior
No "yanked crate" advisory should be reported, because the crate comes from our private registry, and it is different from the same-named crate on crates.io
Screenshots
Here is the redacted cargo-deny output:
Offtopic
I know Embark uses a private crates registry as well. I wonder if you still sit on a git-based registry index? We can't use the sparse crates registry index because it doesn't support auth tokens. Cloudsmith, for example still provides a way to use the sparse registry, but the auth token in this case must be part of the registry index URL, which in turn means this URL becomes part of Cargo.lock and therefore the secret auth token needs to be committed into source control. This is because the issue rust-lang/cargo#10474 is not done. So.. I just wanted to ask wether it also blocks your migration to spare registry index.
The text was updated successfully, but these errors were encountered: