Don't trigger unsafe_op_in_unsafe_fn for deprecated safe fns
#125925
Conversation
|
r? @Nadrieril rustbot has assigned @Nadrieril. Use |
rustbot
added
S-waiting-on-review
This comment has been minimized.
This comment has been minimized.
tbu-
force-pushed
the
pr_unsafe_env_unsafe_op_in_unsafe_fn
branch
from
e605228
to
42b7153
Compare
| if !span.at_least_rust_2024() | ||
| && self.tcx.has_attr(id, sym::rustc_deprecated_safe_2024) => |
There was a problem hiding this comment.
While this PR should probably go through as-is to get the fix in as soon as possible, I feel like this system is somewhat brittle, relying on specifically the symbol rustc_deprecated_safe_2024 and the edition being hard coded to check at least 2024. I wonder if it would be a good idea for a follow up PR to somehow make this general in that the attribute could specify an edition in which a function starts becoming unsafe, and check that edition?
However, such a followup assumes that we will make more functions unsafe in the future due to oversight. Ideally we don't have to do that, and I think that the current state of things is such that just about everything has been audited? So perhaps it's not worth such a generalization, since it ideally doesn't happen again.
There was a problem hiding this comment.
While this PR should probably go through as-is to get the fix in as soon as possible, I feel like this system is somewhat brittle, relying on specifically the symbol
rustc_deprecated_safe_2024and the edition being hard coded to check at least2024. I wonder if it would be a good idea for a follow up PR to somehow make this general in that the attribute could specify an edition in which a function starts becoming unsafe, and check that edition?
Such a more general has been proposed and can probably be implemented. #[rustc_deprecated_safe_2024] was intentionally limited in scope in order to make std::env::{set_var, remove_var} unsafe in Rust 2024. See e.g. #94978.
|
I have one style comment, feel free to ignore it or fix it now or fix it in a later PR. @bors delegate+ |
|
✌️ @tbu-, you can now approve this pull request! If @Nadrieril told you to " |
tbu-
force-pushed
the
pr_unsafe_env_unsafe_op_in_unsafe_fn
branch
from
42b7153
to
bb901a1
Compare
|
@bors r=Nadrieril |
bors
added
S-waiting-on-bors
…safe_fn, r=Nadrieril Don't trigger `unsafe_op_in_unsafe_fn` for deprecated safe fns Fixes rust-lang#125875. Tracking: - rust-lang#124866
…iaskrgr Rollup of 7 pull requests Successful merges: - rust-lang#124731 (Add translation support by mdbook-i18n-helpers to bootstrap) - rust-lang#125168 (Match ergonomics 2024: align implementation with RFC) - rust-lang#125925 (Don't trigger `unsafe_op_in_unsafe_fn` for deprecated safe fns) - rust-lang#125966 (Implement `os_string_pathbuf_leak`) - rust-lang#125987 (When `derive`ing, account for HRTB on `BareFn` fields) - rust-lang#126045 (check_expr_struct_fields: taint context with errors if struct definit…) - rust-lang#126048 (Fix typos in cargo-specifics.md) r? `@ghost` `@rustbot` modify labels: rollup
…iaskrgr Rollup of 6 pull requests Successful merges: - rust-lang#124731 (Add translation support by mdbook-i18n-helpers to bootstrap) - rust-lang#125168 (Match ergonomics 2024: align implementation with RFC) - rust-lang#125925 (Don't trigger `unsafe_op_in_unsafe_fn` for deprecated safe fns) - rust-lang#125987 (When `derive`ing, account for HRTB on `BareFn` fields) - rust-lang#126045 (check_expr_struct_fields: taint context with errors if struct definit…) - rust-lang#126048 (Fix typos in cargo-specifics.md) r? `@ghost` `@rustbot` modify labels: rollup
Rollup merge of rust-lang#125925 - tbu-:pr_unsafe_env_unsafe_op_in_unsafe_fn, r=Nadrieril Don't trigger `unsafe_op_in_unsafe_fn` for deprecated safe fns Fixes rust-lang#125875. Tracking: - rust-lang#124866
Fixes #125875.
Tracking:
std::env::{set_var, remove_var}unsafe #124866