auto merge of #8884 : blake2-ppc/rust/exact-size-hint, r=huonw

The message of the first commit explains (edited for changed trait name):

The trait `ExactSize` is introduced to solve a few small niggles:

* We can't reverse (`.invert()`) an enumeration iterator
* for a vector, we have `v.iter().position(f)` but `v.rposition(f)`.
* We can't reverse `Zip` even if both iterators are from vectors

`ExactSize` is an empty trait that is intended to indicate that an
iterator, for example `VecIterator`, knows its exact finite size and
reports it correctly using `.size_hint()`. Only adaptors that preserve
this at all times, can expose this trait further. (Where here we say
finite for fitting in uint).

---

It may seem complicated just to solve these small "niggles",
(It's really the reversible enumerate case that's the most interesting)
but only a few core iterators need to implement this trait.

While we gain more capabilities generically for some iterators,
it becomes a tad more complicated to figure out if a type has
the right trait impls for it.

Author: bors

src/libextra/bitv.rs

@@ -608,6 +608,8 @@ impl<'self> DoubleEndedIterator<bool> for BitvIterator<'self> {
     }
 }
 
+impl<'self> ExactSize<bool> for BitvIterator<'self> {}
+
 impl<'self> RandomAccessIterator<bool> for BitvIterator<'self> {
     #[inline]
     fn indexable(&self) -> uint {

src/libextra/dlist.rs

@@ -472,6 +472,8 @@ impl<'self, A> DoubleEndedIterator<&'self A> for DListIterator<'self, A> {
     }
 }
 
+impl<'self, A> ExactSize<&'self A> for DListIterator<'self, A> {}
+
 impl<'self, A> Iterator<&'self mut A> for MutDListIterator<'self, A> {
     #[inline]
     fn next(&mut self) -> Option<&'self mut A> {
@@ -508,6 +510,7 @@ impl<'self, A> DoubleEndedIterator<&'self mut A> for MutDListIterator<'self, A>
     }
 }
 
+impl<'self, A> ExactSize<&'self mut A> for MutDListIterator<'self, A> {}
 
 /// Allow mutating the DList while iterating
 pub trait ListInsertion<A> {

src/libextra/num/bigint.rs

@@ -525,7 +525,7 @@ impl BigUint {
     #[inline]
     pub fn new(v: ~[BigDigit]) -> BigUint {
         // omit trailing zeros
-        let new_len = v.rposition(|n| *n != 0).map_move_default(0, |p| p + 1);
+        let new_len = v.iter().rposition(|n| *n != 0).map_move_default(0, |p| p + 1);
 
         if new_len == v.len() { return BigUint { data: v }; }
         let mut v = v;

src/libextra/ringbuf.rs

@@ -243,6 +243,8 @@ pub struct RingBufIterator<'self, T> {
 iterator!{impl RingBufIterator -> &'self T, get_ref}
 iterator_rev!{impl RingBufIterator -> &'self T, get_ref}
 
+impl<'self, T> ExactSize<&'self T> for RingBufIterator<'self, T> {}
+
 impl<'self, T> RandomAccessIterator<&'self T> for RingBufIterator<'self, T> {
     #[inline]
     fn indexable(&self) -> uint { self.rindex - self.index }
@@ -268,6 +270,8 @@ pub struct RingBufMutIterator<'self, T> {
 iterator!{impl RingBufMutIterator -> &'self mut T, get_mut_ref}
 iterator_rev!{impl RingBufMutIterator -> &'self mut T, get_mut_ref}
 
+impl<'self, T> ExactSize<&'self mut T> for RingBufMutIterator<'self, T> {}
+
 /// Grow is only called on full elts, so nelts is also len(elts), unlike
 /// elsewhere.
 fn grow<T>(nelts: uint, loptr: &mut uint, elts: &mut ~[Option<T>]) {

src/libstd/iterator.rs

@@ -18,7 +18,7 @@ implementing the `Iterator` trait.
 */
 
 use cmp;
-use num::{Zero, One, Integer, CheckedAdd, Saturating};
+use num::{Zero, One, Integer, CheckedAdd, CheckedSub, Saturating};
 use option::{Option, Some, None};
 use ops::{Add, Mul, Sub};
 use cmp::Ord;
@@ -641,6 +641,7 @@ impl<'self, A, T: DoubleEndedIterator<&'self mut A>> MutableDoubleEndedIterator
     }
 }
 
+
 /// An object implementing random access indexing by `uint`
 ///
 /// A `RandomAccessIterator` should be either infinite or a `DoubleEndedIterator`.
@@ -653,6 +654,48 @@ pub trait RandomAccessIterator<A>: Iterator<A> {
     fn idx(&self, index: uint) -> Option<A>;
 }
 
+/// An iterator that knows its exact length
+///
+/// This trait is a helper for iterators like the vector iterator, so that
+/// it can support double-ended enumeration.
+///
+/// `Iterator::size_hint` *must* return the exact size of the iterator.
+/// Note that the size must fit in `uint`.
+pub trait ExactSize<A> : DoubleEndedIterator<A> {
+    /// Return the index of the last element satisfying the specified predicate
+    ///
+    /// If no element matches, None is returned.
+    #[inline]
+    fn rposition(&mut self, predicate: &fn(A) -> bool) -> Option<uint> {
+        let (lower, upper) = self.size_hint();
+        assert!(upper == Some(lower));
+        let mut i = lower;
+        loop {
+            match self.next_back() {
+                None => break,
+                Some(x) => {
+                    i = match i.checked_sub(&1) {
+                        Some(x) => x,
+                        None => fail!("rposition: incorrect ExactSize")
+                    };
+                    if predicate(x) {
+                        return Some(i)
+                    }
+                }
+            }
+        }
+        None
+    }
+}
+
+// All adaptors that preserve the size of the wrapped iterator are fine
+// Adaptors that may overflow in `size_hint` are not, i.e. `Chain`.
+impl<A, T: ExactSize<A>> ExactSize<(uint, A)> for Enumerate<T> {}
+impl<'self, A, T: ExactSize<A>> ExactSize<A> for Inspect<'self, A, T> {}
+impl<A, T: ExactSize<A>> ExactSize<A> for Invert<T> {}
+impl<'self, A, B, T: ExactSize<A>> ExactSize<B> for Map<'self, A, B, T> {}
+impl<A, B, T: ExactSize<A>, U: ExactSize<B>> ExactSize<(A, B)> for Zip<T, U> {}
+
 /// An double-ended iterator with the direction inverted
 #[deriving(Clone)]
 pub struct Invert<T> {
@@ -956,6 +999,29 @@ impl<A, B, T: Iterator<A>, U: Iterator<B>> Iterator<(A, B)> for Zip<T, U> {
     }
 }
 
+impl<A, B, T: ExactSize<A>, U: ExactSize<B>> DoubleEndedIterator<(A, B)>
+for Zip<T, U> {
+    #[inline]
+    fn next_back(&mut self) -> Option<(A, B)> {
+        let (a_sz, a_upper) = self.a.size_hint();
+        let (b_sz, b_upper) = self.b.size_hint();
+        assert!(a_upper == Some(a_sz));
+        assert!(b_upper == Some(b_sz));
+        if a_sz < b_sz {
+            for _ in range(0, b_sz - a_sz) { self.b.next_back(); }
+        } else if a_sz > b_sz {
+            for _ in range(0, a_sz - b_sz) { self.a.next_back(); }
+        }
+        let (a_sz, _) = self.a.size_hint();
+        let (b_sz, _) = self.b.size_hint();
+        assert!(a_sz == b_sz);
+        match (self.a.next_back(), self.b.next_back()) {
+            (Some(x), Some(y)) => Some((x, y)),
+            _ => None
+        }
+    }
+}
+
 impl<A, B, T: RandomAccessIterator<A>, U: RandomAccessIterator<B>>
 RandomAccessIterator<(A, B)> for Zip<T, U> {
     #[inline]
@@ -1137,6 +1203,20 @@ impl<A, T: Iterator<A>> Iterator<(uint, A)> for Enumerate<T> {
     }
 }
 
+impl<A, T: ExactSize<A>> DoubleEndedIterator<(uint, A)> for Enumerate<T> {
+    #[inline]
+    fn next_back(&mut self) -> Option<(uint, A)> {
+        match self.iter.next_back() {
+            Some(a) => {
+                let (lower, upper) = self.iter.size_hint();
+                assert!(upper == Some(lower));
+                Some((self.count + lower, a))
+            }
+            _ => None
+        }
+    }
+}
+
 impl<A, T: RandomAccessIterator<A>> RandomAccessIterator<(uint, A)> for Enumerate<T> {
     #[inline]
     fn indexable(&self) -> uint {
@@ -2331,6 +2411,33 @@ mod tests {
         assert_eq!(it.next(), None);
     }
 
+    #[test]
+    fn test_double_ended_enumerate() {
+        let xs = [1, 2, 3, 4, 5, 6];
+        let mut it = xs.iter().map(|&x| x).enumerate();
+        assert_eq!(it.next(), Some((0, 1)));
+        assert_eq!(it.next(), Some((1, 2)));
+        assert_eq!(it.next_back(), Some((5, 6)));
+        assert_eq!(it.next_back(), Some((4, 5)));
+        assert_eq!(it.next_back(), Some((3, 4)));
+        assert_eq!(it.next_back(), Some((2, 3)));
+        assert_eq!(it.next(), None);
+    }
+
+    #[test]
+    fn test_double_ended_zip() {
+        let xs = [1, 2, 3, 4, 5, 6];
+        let ys = [1, 2, 3, 7];
+        let a = xs.iter().map(|&x| x);
+        let b = ys.iter().map(|&x| x);
+        let mut it = a.zip(b);
+        assert_eq!(it.next(), Some((1, 1)));
+        assert_eq!(it.next(), Some((2, 2)));
+        assert_eq!(it.next_back(), Some((4, 7)));
+        assert_eq!(it.next_back(), Some((3, 3)));
+        assert_eq!(it.next(), None);
+    }
+
     #[test]
     fn test_double_ended_filter() {
         let xs = [1, 2, 3, 4, 5, 6];
@@ -2367,6 +2474,31 @@ mod tests {
         assert_eq!(it.next_back(), None)
     }
 
+    #[test]
+    fn test_rposition() {
+        fn f(xy: &(int, char)) -> bool { let (_x, y) = *xy; y == 'b' }
+        fn g(xy: &(int, char)) -> bool { let (_x, y) = *xy; y == 'd' }
+        let v = ~[(0, 'a'), (1, 'b'), (2, 'c'), (3, 'b')];
+
+        assert_eq!(v.iter().rposition(f), Some(3u));
+        assert!(v.iter().rposition(g).is_none());
+    }
+
+    #[test]
+    #[should_fail]
+    fn test_rposition_fail() {
+        let v = [(~0, @0), (~0, @0), (~0, @0), (~0, @0)];
+        let mut i = 0;
+        do v.iter().rposition |_elt| {
+            if i == 2 {
+                fail!()
+            }
+            i += 1;
+            false
+        };
+    }
+
+
     #[cfg(test)]
     fn check_randacc_iter<A: Eq, T: Clone + RandomAccessIterator<A>>(a: T, len: uint)
     {

src/libstd/option.rs

@@ -46,7 +46,7 @@ use cmp::{Eq,Ord};
 use util;
 use num::Zero;
 use iterator;
-use iterator::{Iterator, DoubleEndedIterator};
+use iterator::{Iterator, DoubleEndedIterator, ExactSize};
 use str::{StrSlice, OwnedStr};
 use to_str::ToStr;
 use clone::DeepClone;
@@ -402,6 +402,8 @@ impl<A> DoubleEndedIterator<A> for OptionIterator<A> {
     }
 }
 
+impl<A> ExactSize<A> for OptionIterator<A> {}
+
 #[cfg(test)]
 mod tests {
     use super::*;

src/libstd/prelude.rs

@@ -52,7 +52,7 @@ pub use hash::Hash;
 pub use num::Times;
 pub use iterator::{FromIterator, Extendable};
 pub use iterator::{Iterator, DoubleEndedIterator, RandomAccessIterator, ClonableIterator};
-pub use iterator::{OrdIterator, MutableDoubleEndedIterator};
+pub use iterator::{OrdIterator, MutableDoubleEndedIterator, ExactSize};
 pub use num::{Num, NumCast, CheckedAdd, CheckedSub, CheckedMul};
 pub use num::{Orderable, Signed, Unsigned, Round};
 pub use num::{Algebraic, Trigonometric, Exponential, Hyperbolic};

src/libstd/select.rs

@@ -11,7 +11,7 @@
 use cell::Cell;
 use comm;
 use container::Container;
-use iterator::Iterator;
+use iterator::{Iterator, DoubleEndedIterator};
 use option::*;
 // use either::{Either, Left, Right};
 // use rt::kill::BlockedTask;
@@ -87,7 +87,7 @@ pub fn select<A: Select>(ports: &mut [A]) -> uint {
     // Task resumes. Now unblock ourselves from all the ports we blocked on.
     // If the success index wasn't reset, 'take' will just take all of them.
     // Iterate in reverse so the 'earliest' index that's ready gets returned.
-    for (index, port) in ports.mut_slice(0, ready_index).mut_rev_iter().enumerate() {
+    for (index, port) in ports.mut_slice(0, ready_index).mut_iter().enumerate().invert() {
         if port.unblock_from() {
             ready_index = index;
         }

src/libstd/str.rs

@@ -24,7 +24,7 @@ use container::{Container, Mutable};
 use num::Times;
 use iterator::{Iterator, FromIterator, Extendable};
 use iterator::{Filter, AdditiveIterator, Map};
-use iterator::{Invert, DoubleEndedIterator};
+use iterator::{Invert, DoubleEndedIterator, ExactSize};
 use libc;
 use num::{Saturating};
 use option::{None, Option, Some};
@@ -484,9 +484,8 @@ for CharSplitIterator<'self, Sep> {
         let mut next_split = None;
 
         if self.only_ascii {
-            for (j, byte) in self.string.byte_rev_iter().enumerate() {
+            for (idx, byte) in self.string.byte_iter().enumerate().invert() {
                 if self.sep.matches(byte as char) && byte < 128u8 {
-                    let idx = len - j - 1;
                     next_split = Some((idx, idx + 1));
                     break;
                 }
@@ -2006,16 +2005,13 @@ impl<'self> StrSlice<'self> for &'self str {
     /// or `None` if there is no match
     fn find<C: CharEq>(&self, search: C) -> Option<uint> {
         if search.only_ascii() {
-            for (i, b) in self.byte_iter().enumerate() {
-                if search.matches(b as char) { return Some(i) }
-            }
+            self.byte_iter().position(|b| search.matches(b as char))
         } else {
             for (index, c) in self.char_offset_iter() {
                 if search.matches(c) { return Some(index); }
             }
+            None
         }
-
-        None
     }
 
     /// Returns the byte index of the last character of `self` that matches `search`
@@ -2026,18 +2022,13 @@ impl<'self> StrSlice<'self> for &'self str {
     /// or `None` if there is no match
     fn rfind<C: CharEq>(&self, search: C) -> Option<uint> {
         if search.only_ascii() {
-            let mut index = self.len();
-            for b in self.byte_rev_iter() {
-                index -= 1;
-                if search.matches(b as char) { return Some(index); }
-            }
+            self.byte_iter().rposition(|b| search.matches(b as char))
         } else {
             for (index, c) in self.char_offset_rev_iter() {
                 if search.matches(c) { return Some(index); }
             }
+            None
         }
-
-        None
     }
 
     /// Returns the byte index of the first matching substring