This tracking issue is to create a more ergonomic method to join on Iterator without collect. Hopefully, we can do some optimizations on top of it but I haven't think of any case to improve it.
The feature gate for the issue is #![feature(iterator_join)].
Before
["hello", "world"].iter().collect::<Vec<_>>().join(" ")
After
["hello", "world"].iter().join(" ")
It makes use of Join under slice_concat_trait feature #27747 and have a similar behavior to Iterator::collect.
It is similar to join from itertools except that it makes use of the stuff mentioned for feature parity between Iterator and join and ergonomics to not need to do .collect::<Vec<_>>().
Technical implementation proof of concept.
pub fn join<T, Separator>(iter: impl Iterator<Item = T>, sep: Separator) -> <[T] as Join<Separator>>::Output
where
[T]: Join<Separator>,
{
// <[S] as std::slice::Join<&str>>
// <[V] as std::slice::Join<&T>>
// <[V] as std::slice::Join<&[T]>>
Join::join(iter.collect::<Vec<T>>().as_slice(), sep)
}
Existing discussion on zulip https://rust-lang.zulipchat.com/#narrow/stream/219381-t-libs/topic/join.20on.20Iterator
Related issue #22754
About tracking issues
Tracking issues are used to record the overall progress of implementation.
They are also uses as hubs connecting to other relevant issues, e.g., bugs or open design questions.
A tracking issue is however not meant for large scale discussion, questions, or bug reports about a feature.
Instead, open a dedicated issue for the specific matter and add the relevant feature gate label.
Steps
Unresolved Questions
Implementation history
- Initial Implementation: WIP (please let me do this)
This tracking issue is to create a more ergonomic method to
joinonIteratorwithout collect. Hopefully, we can do some optimizations on top of it but I haven't think of any case to improve it.The feature gate for the issue is
#![feature(iterator_join)].Before
After
It makes use of
Joinunderslice_concat_traitfeature #27747 and have a similar behavior toIterator::collect.It is similar to
joinfromitertoolsexcept that it makes use of the stuff mentioned for feature parity betweenIteratorandjoinand ergonomics to not need to do.collect::<Vec<_>>().Technical implementation proof of concept.
Existing discussion on zulip https://rust-lang.zulipchat.com/#narrow/stream/219381-t-libs/topic/join.20on.20Iterator
Related issue #22754
About tracking issues
Tracking issues are used to record the overall progress of implementation.
They are also uses as hubs connecting to other relevant issues, e.g., bugs or open design questions.
A tracking issue is however not meant for large scale discussion, questions, or bug reports about a feature.
Instead, open a dedicated issue for the specific matter and add the relevant feature gate label.
Steps
Unresolved Questions
.join(sep)rather than.collect::<Vec<_>>().join()after stabilization? Is it even possible when it goes across lines?Implementation history