Closed Veetaha closed 2 years ago
Interesting. What was the output from cargo deny?
@cholcombe973 ah, it's just the way we use cargo-deny
.
First of all we track the duplicates of different versions of the same crate in the dependencies tree. Since now we gradually transition to 0.23
of strum we see crates from our tree still using 0.22
making that version of the library compiled in together with 0.23
.
Here is the example output of that (cargo deny check
):
error[B004]: found 2 duplicate entries for crate 'strum'
┌─ /home/veetaha/work/{redacted}
│
739 │ ╭ strum 0.22.0 registry+https://github.com/rust-lang/crates.io-index
740 │ │ strum 0.23.0 registry+https://github.com/rust-lang/crates.io-index
│ ╰──────────────────────────────────────────────────────────────────^ lock entries
Also, we have the following config in deny.toml
[bans]
deny = [
# `strum_macros` direct usage is deprecated. Use `strum = { features = ["derive"] }` instead.
{ name = "strum_macros", wrappers = ["strum"] },
]
This lets us track if anyone still uses strum_macros
crate directly instead of using strum
's "derive"
feature
Yeah that makes sense. Sure I'll cut a patch release for it. Thanks!
This appeared on our
cargo-deny
radar. Would be glad if you could do a patch release.Even though "technically"
strum::ParseError
is part of publicBlockUtilsError
and0.22 -> 0.23
is a breaking change, I doubt that anyone would break as a result of this, and if it would break, the resolution of the problem would not be too dificult