mirror of
https://github.com/rust-lang/rust.git
synced 2025-04-14 04:56:49 +00:00
Auto merge of #8644 - yoav-lavi:squashed-master, r=flip1995
update unnecessary_join documentation changelog: none Updates the description of `unnecessary_join` in accordance with https://github.com/rust-lang/rust-clippy/pull/8579#issuecomment-1089969859. I've also added a line regarding differences in assembly output, please let me know if it should also make it in.
This commit is contained in:
commit
30019d1d05
@ -2055,7 +2055,7 @@ declare_clippy_lint! {
|
||||
/// Checks for use of `.collect::<Vec<String>>().join("")` on iterators.
|
||||
///
|
||||
/// ### Why is this bad?
|
||||
/// `.collect::<String>()` is more concise and usually more performant
|
||||
/// `.collect::<String>()` is more concise and might be more performant
|
||||
///
|
||||
/// ### Example
|
||||
/// ```rust
|
||||
@ -2070,9 +2070,12 @@ declare_clippy_lint! {
|
||||
/// println!("{}", output);
|
||||
/// ```
|
||||
/// ### Known problems
|
||||
/// While `.collect::<String>()` is more performant in most cases, there are cases where
|
||||
/// While `.collect::<String>()` is sometimes more performant, there are cases where
|
||||
/// using `.collect::<String>()` over `.collect::<Vec<String>>().join("")`
|
||||
/// will prevent loop unrolling and will result in a negative performance impact.
|
||||
///
|
||||
/// Additionlly, differences have been observed between aarch64 and x86_64 assembly output,
|
||||
/// with aarch64 tending to producing faster assembly in more cases when using `.collect::<String>()`
|
||||
#[clippy::version = "1.61.0"]
|
||||
pub UNNECESSARY_JOIN,
|
||||
pedantic,
|
||||
|
Loading…
Reference in New Issue
Block a user