Skip to content

Commit

Permalink
Merge #623
Browse files Browse the repository at this point in the history
623: Explain why Hashmap is used instead of Hashset in unique_impl r=jswrenn a=hellow554

The `Hashmap` was chosen because of the `Entry` API, but one can work around
that easiliy.

Co-authored-by: Marcel Hellwig <github@cookiesoft.de>
  • Loading branch information
bors[bot] and hellow554 committed Jun 7, 2022
2 parents 1f0a8c1 + 2c74555 commit 28cff76
Showing 1 changed file with 4 additions and 3 deletions.
7 changes: 4 additions & 3 deletions src/unique_impl.rs
@@ -1,6 +1,5 @@

use std::collections::HashMap;
use std::collections::hash_map::{Entry};
use std::collections::hash_map::Entry;
use std::hash::Hash;
use std::fmt;
use std::iter::FusedIterator;
Expand All @@ -12,7 +11,9 @@ use std::iter::FusedIterator;
#[must_use = "iterator adaptors are lazy and do nothing unless consumed"]
pub struct UniqueBy<I: Iterator, V, F> {
iter: I,
// Use a hashmap for the entry API
// Use a Hashmap for the Entry API in order to prevent hashing twice.
// This can maybe be replaced with a HashSet once `get_or_insert_with`
// or a proper Entry API for Hashset is stable and meets this msrv
used: HashMap<V, ()>,
f: F,
}
Expand Down

0 comments on commit 28cff76

Please sign in to comment.