Path improvements
#546
Path improvements
#546
Conversation
|
Draft for now: everything should work but I'd like to know which of these changes would be welcome, if any. Thanks for maintaining this by the way 🙂 |
|
One alternative I'm considering is a |
src/path/mod.rs
Outdated
| /// # Performance | ||
| /// | ||
| /// This operation is `O(n)`, similar to calling `.parts().count()` manually. | ||
| pub fn len(&self) -> usize { |
There was a problem hiding this comment.
This collides with the existing len provided by the deref to str
Edit: actually we don't implement deref but still callings this parts_count or something might be less ambiguous
There was a problem hiding this comment.
Would renaming empty and is_empty into root and is_root make sense?
Kinrany
force-pushed
the
push-qnqozslwvlop
branch
from
44f4967 to
b49dd85
Compare
Kinrany
force-pushed
the
push-qnqozslwvlop
branch
from
b49dd85 to
c6ccb69
Compare
|
Could I have one more CI run please? |
Done |
|
|
||
| /// Returns a copy of this [`Path`] with the last path segment removed | ||
| /// | ||
| /// Returns `None` if this path has zero segments. |
There was a problem hiding this comment.
this seems pretty similar to https://doc.rust-lang.org/std/path/struct.Path.html#method.parent
Should we call this method parent as well to follow existing convention?
| assert_eq!(Path::ROOT.parts().count(), Path::ROOT.parts_count()); | ||
|
|
||
| let path = path("foo/bar/baz"); | ||
| assert_eq!(path.parts().count(), path.parts_count()); |
There was a problem hiding this comment.
could you also please add an assertion that the count is 3?
|
|
||
| Self { raw } | ||
| impl<'a, I: Into<PathPart<'a>>> Extend<I> for Path { | ||
| fn extend<T: IntoIterator<Item = I>>(&mut self, iter: T) { |
There was a problem hiding this comment.
could you please add tests (perhaps doctests) using this feature explicitly to help others find it?
| if s.raw.is_empty() { | ||
| continue; | ||
| } | ||
| self.raw.push(DELIMITER_CHAR); |
There was a problem hiding this comment.
Can you add some tests showing how this works for:
- Empty iterators
- Iterator with a single element
- Iterator with multiple elements
- Extending an existing (non empty) Path? (it looks like maybe this will add an extra
/🤔 )
|
CI appears to show some regressions |
|
Thank you, will address! |
|
marking as draft while CI issues are resolved |
|
please mark it ready for review when it is ready for another look |
3 participants
Which issue does this PR close?
Closes #545
Rationale for this change
fn prefix(&self)was the first thing I really wanted to have onPath, as opposed to copying from project to project.It also benefits from direct access to
raw. In userland we have to usepathsand collect, and omitting the last item is not trivial.What changes are included in this PR?
Implementations of everything listed with minimal docs, without tests.
Are there any user-facing changes?
Most of these changes are user-facing.
Internal changes:
impl FromIteratordelegates most of the logic toimpl ExpandPath::partsnow returns a custom iterator typeDELIMITER_CHARThere should be no breaking changes.