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
So I think the next piece of code for helping with stabilization has to do with teaching cargo that the two URLs registry+https://github.com/rust-lang/crates.io-index and sparse+https://index.crates.io/ are equivalent. Specifically making sure that lock files only show one even if both are available, and that you can successfully publish with either configured.
[[package]]
name = "serde"
version = "1.0.137"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "61ea8d54c77f8315140a05f4c7237403bf38b72704d031543aa1d16abbf517d1"
So I don't think this is going to be super difficult. What I want is for cargo to act as if that override was set if -Z http-registry is passed (and no explicit override was set).
We may end up fixing the issue with publishing where there's a replacement as we work on this, but I think this will end up being implemented through a somewhat different code path. It will do the same thing as having a replacement, but it will not be implemented as literally a replacement. (Or if it is, cargo will know this one special and to treat it differently.)
So I think the next piece of code for helping with stabilization has to do with teaching cargo that the two URLs
registry+https://github.com/rust-lang/crates.io-index
andsparse+https://index.crates.io/
are equivalent. Specifically making sure that lock files only show one even if both are available, and that you can successfully publish with either configured.Testing with a
.cargo/config.toml
of:and it generates a lock file that has:
So I don't think this is going to be super difficult. What I want is for cargo to act as if that override was set if
-Z http-registry
is passed (and no explicit override was set).We may end up fixing the issue with publishing where there's a replacement as we work on this, but I think this will end up being implemented through a somewhat different code path. It will do the same thing as having a replacement, but it will not be implemented as literally a replacement. (Or if it is, cargo will know this one special and to treat it differently.)
cc: #9069
The text was updated successfully, but these errors were encountered: