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
Following up from the rapidhash PR #633, this bumps foldhash to 0.2.0 for a large improvement in foldhash's string hashing.
Both rapidhash and foldhash are equal in the hashbrown benchmark suite as it only tests the hashing of usize objects.
The reason foldhash 0.2.0 required a major version bump is because FoldHasher now has a lifetime parameter. I don't know how users are expected to use DefaultHashBuilder, but noting that it likely makes this a breaking change for hashbrown?
I'm not sure if it's a breaking release, the DefaultHashBuilder still implements the HashBuilder contract all the same, and that lifetime is just 'static on an associated type.
I believe it technically is breaking because you could pass DefaultHashBuilder to a function expecting foldhash-0.1::RandomState whereas that breaks with this. DefaultHashBuilder probably should be using the newtype pattern.
FYI... I also got this compilation error when I attempted to use this PR on [patch.crates-io] of my project's Cargo.toml
error[E0204]: the trait `core::marker::Copy` cannot be implemented for this type
--> /Users/jdoe/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/hashlink-0.10.0/src/lib.rs:19:17
|
19 | #[derive(Clone, Copy, Default, Debug)]
| ^^^^
20 | pub struct DefaultHashBuilder(hashbrown::DefaultHashBuilder);
| ----------------------------- this field does not implement `core::marker::Copy`
FYI... I also got this compilation error when I attempted to use this PR on [patch.crates-io] of my project's Cargo.toml
error[E0204]: the trait `core::marker::Copy` cannot be implemented for this type
--> /Users/jdoe/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/hashlink-0.10.0/src/lib.rs:19:17
|
19 | #[derive(Clone, Copy, Default, Debug)]
| ^^^^
20 | pub struct DefaultHashBuilder(hashbrown::DefaultHashBuilder);
| ----------------------------- this field does not implement `core::marker::Copy`
But yes, the lack of a newtype pattern meant that this breaking change would leak into hashbrown as it directly passes through any traits that were implemented.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Following up from the rapidhash PR #633, this bumps foldhash to
0.2.0for a large improvement in foldhash's string hashing.Both rapidhash and foldhash are equal in the hashbrown benchmark suite as it only tests the hashing of
usizeobjects.The reason foldhash 0.2.0 required a major version bump is because
FoldHashernow has a lifetime parameter. I don't know how users are expected to useDefaultHashBuilder, but noting that it likely makes this a breaking change for hashbrown?cc. @orlp @Amanieu from #633