José Julián Espina

Thank you for the quick response! > Usually for bigger changes it is nice to have an issue up before a PR emerges to allow for discussion beforehand. Noted for...

> I think it is best if you just run the benchmarks on your own computer to get a feeling. Just checkout `master` branch, run `cargo bench`, then checkout your...

On another note, I had to adjust the allocation benchmarks for the `BucketBackend`, because calling `shrink_to_fit` results in undefined behaviour. The cause was the definition of `shrink_to_fit`: https://github.com/Robbepop/string-interner/blob/d552e760dae3fa1a2c3cb9a00c9232c218d0f58a/src/backend/bucket/mod.rs#L119-L123 Calling `self.head.shrink_to_fit`...

> Can you elaborate on this? I don't see how this results in undefined behavior atm. The memory consumption regressions for the `BucketBackend` are extreme. Sure :) The definition of...

> I'd like for `FixedContainer` not to be exposed. I don't think that's possible. You cannot require that a generic type ought to be an implementor of a certain trait...

@Robbepop Did you have the chance to read my explanation and comments? Also, if you're still unsure about this PR, I could reduce its complexity by completely removing the ability...

> Hey, sorry for not responding for a long time. Thanks a lot for the issue and the PR. I actually was already thinking for quite a while to remove...

> Are we talking about >(10^3) static strings or just like Because I guess it won't matter then. Have you got benchmarks? No benchmarks, unfortunately. However, that backend has the...

> > The team kindly recommended us to treat the input locale as a simple Language Identifier instead, which is a bit of a "hack" but it should work. >...

Just want to report that, with the new [workspace inherited dependencies](https://doc.rust-lang.org/cargo/reference/workspaces.html#the-dependencies-table), this syntax will have a lot more usage, so it would be good to prioritize this.