1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174

//! Iterator adapter to flatten an `Iterator<Item> where: Item = Result<IntoIterator,_>`.
//! See documentation for [FlattenResults] for details.
//! 
//! Based on solution proposed by redditor [earthengine](https://www.reddit.com/user/earthengine/) in [this post](https://www.reddit.com/r/rust/comments/9u6846/rust_puzzle_flatten_a_nested_iterator_of_results/) and on the [rust playground](https://play.rust-lang.org/?version=stable&mode=debug&edition=2015&gist=30e0fb57a8ba63777052a344054b22c0).
// 
// Implementation is non-trivial because trait methods cannot return impl trait, see the relevant RFC:
// https://github.com/rust-lang/rfcs/blob/master/text/1522-conservative-impl-trait.md
// This necessitates the rather complex and manually-defined return type of FlattenResults::flatten_result().
// To further complicated matters, rust rlosures are anonymous types and therefore cannot be expressed in a manually-defined return type.  Instead we declare module-private functions flatten() and wrap_result() below, and we use pointers to these functions (which do have a type) instead of closures.

// Module-private function to flatten an iterator of results, see earthengine's solution for an explanation.
fn flatten<InnerIter, T, Error>(res: Result<InnerIter, Error>) -> std::iter::Chain<std::iter::Map<std::iter::Flatten<std::option::IntoIter<InnerIter>>,fn(T)->Result<T, Error>>, std::option::IntoIter<std::result::Result<T, Error>>>
where
	InnerIter: IntoIterator<Item = T>
{
	let (v,r) = match res {
		Ok(v) => (Some(v), None),
		Err(e) => (None, Some(Err(e)))
	};
	v.into_iter().flatten().map(wrap_result as fn(T)->Result<T, Error>).chain(r)
}

// Function pointer to pass to map() adapter in flatten() above.
// Wraps a type T in a Result<T,_>.
fn wrap_result<T,E>(t: T) -> Result<T, E> {
	Ok(t)
}

/// Module-level free-standing function analog to [flatten_results()](FlattenResults::flatten_results()) that can be called without using the [FlattenResults] trait.  It is more idiomatic to use the iterator adapter and call [flatten_results()](FlattenResults::flatten_results()) instead.
///
/// For example:
///
/// ```
/// # struct MyError;
/// # impl std::fmt::Display for MyError {
/// # 	fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
/// # 		write!(f, "MyError")
/// # 	}
/// # }
/// # impl std::fmt::Debug for MyError {
/// # 	fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
/// # 		<MyError as std::fmt::Display>::fmt(self, f)
/// # 	}
/// # }
/// # impl std::error::Error for MyError { }
/// // Vector of results, each of which contains its own vector of numbers.
/// let v: Vec<Result<Vec<i32>, MyError>> = vec![
/// 	Ok(vec![1, 2]),
/// 	Ok(vec![3, 4]),
/// 	Err(MyError{}),
/// 	Ok(vec![5, 6])
/// ];
///
/// // Flatten the vector of results. 
/// // Propagates the error instead of panicking.
/// let v: Vec<Result<i32, MyError>> =
/// 	resultit::flatten_results::transform(v.into_iter())
/// 	.collect();
/// 
/// println!("{:?}", v);
/// // [Ok(1), Ok(2), Ok(3), Ok(4), Err(MyError), Ok(5), Ok(6)]
/// # assert_eq!(*v[0].as_ref().unwrap(), 1);
/// # assert_eq!(*v[1].as_ref().unwrap(), 2);
/// # assert_eq!(*v[2].as_ref().unwrap(), 3);
/// # assert_eq!(*v[3].as_ref().unwrap(), 4);
/// # assert_eq!(v[4].is_err(), true);
/// # assert_eq!(*v[5].as_ref().unwrap(), 5);
/// # assert_eq!(*v[6].as_ref().unwrap(), 6);
/// ```
pub fn transform<OuterIter, InnerIter, T, Error>(outer_iter: OuterIter) -> std::iter::FlatMap<OuterIter, std::iter::Chain<std::iter::Map<std::iter::Flatten<std::option::IntoIter<InnerIter>>,fn(T)->Result<T, Error>>, std::option::IntoIter<std::result::Result<T, Error>>>, /* type we are mapping to */ fn(Result<InnerIter, Error>)->std::iter::Chain<std::iter::Map<std::iter::Flatten<std::option::IntoIter<InnerIter>>,fn(T)->Result<T, Error>>, std::option::IntoIter<std::result::Result<T, Error>>>>
	where
		OuterIter: Iterator<Item = Result<InnerIter, Error>> + Sized,
		InnerIter: IntoIterator<Item = T>
	{
		outer_iter.flat_map(flatten)
	}

/// Iterator adapter to flatten an `Iterator<Item> where: Item = Result<IntoIterator,_>`.  This is needed because the standard [flatten()](std::iter::Iterator::flatten()) adapter only works on an iterator of iterators and does not work on an iterator of results.
/// 
/// Example of what not to do:
///
/// ```should_panic
/// # struct MyError;
/// # impl std::fmt::Display for MyError {
/// # 	fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
/// # 		write!(f, "MyError")
/// # 	}
/// # }
/// # impl std::fmt::Debug for MyError {
/// # 	fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
/// # 		<MyError as std::fmt::Display>::fmt(self, f)
/// # 	}
/// # }
/// # impl std::error::Error for MyError { }
/// // Vector of results, each of which contains its own vector of numbers.
/// let v: Vec<Result<Vec<i32>, MyError>> = vec![
/// 	Ok(vec![1, 2]),
/// 	Ok(vec![3, 4]),
/// 	Err(MyError{}),
/// 	Ok(vec![5, 6])
/// ];
/// 
/// // Panics after printing the number 4.  Probably not what you want!
/// v.into_iter()
/// 	.map(|res| res.unwrap())
/// 	.flatten()
/// 	.for_each(|i| println!("{}", i));
/// ```
/// 
/// Instead of [flatten()](std::iter::Iterator::flatten()), use [flatten_results()](FlattenResults::flatten_results()) to flatten each `Ok(IntoInterator)` and pass through the errors.
/// 
/// ```
/// # struct MyError;
/// # impl std::fmt::Display for MyError {
/// # 	fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
/// # 		write!(f, "MyError")
/// # 	}
/// # }
/// # impl std::fmt::Debug for MyError {
/// # 	fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
/// # 		<MyError as std::fmt::Display>::fmt(self, f)
/// # 	}
/// # }
/// # impl std::error::Error for MyError { }
/// // Use the FlattenResults trait to enable flatten_results() on iterators.
/// use resultit::FlattenResults;
///
/// // Vector of results, each of which contains its own vector of numbers.
/// let v: Vec<Result<Vec<i32>, MyError>> = vec![
/// 	Ok(vec![1, 2]),
/// 	Ok(vec![3, 4]),
/// 	Err(MyError{}),
/// 	Ok(vec![5, 6])
/// ];
/// 
/// // Flatten the vector of results.
/// // Propagates the error instead of panicking.
/// let v: Vec<Result<i32, MyError>> = v.into_iter()
/// 	.flatten_results()
///		.collect();
/// 
/// println!("{:?}", v);
/// // [Ok(1), Ok(2), Ok(3), Ok(4), Err(MyError), Ok(5), Ok(6)]
/// # assert_eq!(*v[0].as_ref().unwrap(), 1);
/// # assert_eq!(*v[1].as_ref().unwrap(), 2);
/// # assert_eq!(*v[2].as_ref().unwrap(), 3);
/// # assert_eq!(*v[3].as_ref().unwrap(), 4);
/// # assert_eq!(v[4].is_err(), true);
/// # assert_eq!(*v[5].as_ref().unwrap(), 5);
/// # assert_eq!(*v[6].as_ref().unwrap(), 6);
/// ```
pub trait FlattenResults {
	/// Iterator adapter to flatten an `Iterator<Item> where: Item = Result<IntoIterator,_>`.  See the documentation of [FlattenResults].
	fn flatten_results<InnerIter, Error, T>(self) -> std::iter::FlatMap<Self, std::iter::Chain<std::iter::Map<std::iter::Flatten<std::option::IntoIter<InnerIter>>,fn(T)->Result<T, Error>>, std::option::IntoIter<std::result::Result<T, Error>>>, /* type we are mapping to */ fn(Result<InnerIter, Error>)->std::iter::Chain<std::iter::Map<std::iter::Flatten<std::option::IntoIter<InnerIter>>,fn(T)->Result<T, Error>>, std::option::IntoIter<std::result::Result<T, Error>>>>
	where
		Self: Iterator<Item = Result<InnerIter, Error>> + Sized,
		InnerIter: IntoIterator<Item = T>;
}

// Blanket implementation of the FlattenREsults trait for all iterators.
// This is what enables us to call flatten_results() on any iterator.
impl<It> FlattenResults for It
where
	It: Iterator + Sized
{
	fn flatten_results<InnerIter, Error, T>(self) -> std::iter::FlatMap<Self, std::iter::Chain<std::iter::Map<std::iter::Flatten<std::option::IntoIter<InnerIter>>,fn(T)->Result<T, Error>>, std::option::IntoIter<std::result::Result<T, Error>>>, /* type we are mapping to */ fn(Result<InnerIter, Error>)->std::iter::Chain<std::iter::Map<std::iter::Flatten<std::option::IntoIter<InnerIter>>,fn(T)->Result<T, Error>>, std::option::IntoIter<std::result::Result<T, Error>>>>
	where
		Self: Iterator<Item = Result<InnerIter, Error>> + Sized,
		InnerIter: IntoIterator<Item = T>
	{
		// Delegate to flatten_results::transform() to avoid code duplication.
		transform(self)
	}
}