iter: use Option::map() in struct Iterater::map() #22958
Conversation
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
|
r? @aturon (rust_highfive has picked a reviewer for you, use r? to override) |
| #[stable(feature = "rust1", since = "1.0.0")] | ||
| impl<B, I: Iterator, F> Iterator for Map<I, F> where F: FnMut(I::Item) -> B { | ||
| type Item = B; | ||
|
|
||
| #[inline] | ||
| fn next(&mut self) -> Option<B> { | ||
| let next = self.iter.next(); | ||
| self.do_map(next) | ||
| self.iter.next().map(|a| (self.f)(a)) |
There was a problem hiding this comment.
Can this just be replaced with self.iter.next().map(self.f)?
There was a problem hiding this comment.
self.f can not be moved.
if we have "impl FnMut for &FnMut", then we can use &self.iter.next().map(&self.f)
I'm curious that it is not implemented yet.
I'm also curious that fn(Arg) is not trait Fn(Arg). If it is implemented, it also helps (in other code).
There was a problem hiding this comment.
I believe this should be possible with self.iter.next().map(&mut self.f). The fn types are distinct from the Fn traits because fn is a specific type for a function pointer (but implements the Fn trait).
We cannot implement FnMut for &FnMut, but it is implemented for &mut FnMut (as the mutable borrow is required).
There was a problem hiding this comment.
Due to the auto implmeting among the Fn, FnMut, FnOnce, we have some constrains.
Among these 3 types: &mut T where T: Fn, &mut T where T:FnMut, &mut T where T:FnOnce, we can only impl one Fn* trait to them, otherwise they will conflict. The only choice is impl FnMut for &mut T where T:FnMut, other choices make no sense.
The same, Among these 3 types: &T where T: Fn, &T where T:FnMut, &T where T:FnOnce, we can only impl one Fn* trait to them, otherwise they will conflict. The best choice is impl Fn for &T where T:Fn, other choices will cover less situation.
Could I implement the above?
There was a problem hiding this comment.
Ah sorry I was under the impression that this was possible when it in fact isn't! I've opened #23015 on the topic.
There was a problem hiding this comment.
Ok, I will leave this and look forward to your implementing.
|
⌛ Testing commit cd65156 with merge ffe3976... |
|
⌛ Testing commit cd65156 with merge 7003120... |
|
💔 Test failed - auto-linux-64-x-android-t |
|
@bors: retry |
|
⚡ Previous build results are reusable. Rebuilding only auto-linux-32-nopt-t, auto-linux-32-opt, auto-linux-64-nopt-t, auto-linux-64-opt, auto-linux-64-x-android-t, auto-mac-32-opt, auto-mac-64-nopt-t, auto-mac-64-opt, auto-win-32-nopt-t, auto-win-32-opt, auto-win-64-nopt-t, auto-win-64-opt... |
|
💔 Test failed - auto-win-64-opt |
|
@bors: retry |
|
@bors: clean |
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
…excrichton Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
|
Ah no the "hack" was doing the match manually, which that PR didn't change. My git-blame-foo is weak, and I gotta go. Hopefully someone else can figure it out. |
|
(The code has been that way since it was introduced, in 8bf9fc5.) |
|
I'm pretty sure I did that |
Signed-off-by: Lai Jiangshan laijs@cn.fujitsu.com