💡 Fastest O(n) diffing algorithm optimized for Swift collection
💡 Calculate diffs for batch updates of list UI in UIKit, AppKit and Texture
💡 Supports both linear and sectioned collection even if contains duplicates
💡 Supports all kind of diffs for animated UI batch updates
Algorithm
This is a diffing algorithm developed for Carbon, works stand alone. The algorithm optimized based on the Paul Heckel’s algorithm. See also his paper “A technique for isolating differences between files” released in 1978. It allows all kind of diffs to be calculated in linear time O(n). RxDataSources and IGListKit are also implemented based on his algorithm.
However, in performBatchUpdates of UITableView, UICollectionView, etc, there are combinations of diffs that cause crash when applied simultaneously. To solve this problem, DifferenceKit takes an approach of split the set of diffs at the minimal stages that can be perform batch updates with no crashes.
The type of the element that to take diffs must be conform to the Differentiable protocol. The differenceIdentifier‘s type is generic associated type:
struct User: Differentiable {
let id: Int
let name: String
var differenceIdentifier: Int {
return id
}
func isContentEqual(to source: User) -> Bool {
return name == source.name
}
}
In the case of definition above, id uniquely identifies the element and get to know the user updated by comparing equality of name of the elements in source and target.
There are default implementations of Differentiable for the types that conforming to Equatable or Hashable:
// If `Self` conforming to `Hashable`.
var differenceIdentifier: Self {
return self
}
// If `Self` conforming to `Equatable`.
func isContentEqual(to source: Self) -> Bool {
return self == source
}
Therefore, you can simply:
extension String: Differentiable {}
Calculate the diffs by creating StagedChangeset from two collections of elements conforming to Differentiable:
If you want to include multiple types conforming to Differentiable in the collection, use AnyDifferentiable:
let source = [
AnyDifferentiable("A"),
AnyDifferentiable(User(id: 0, name: "Vincent"))
]
In the case of sectioned collection, the section itself must have a unique identifier and be able to compare whether there is an update. So each section must conforming to DifferentiableSection protocol, but in most cases you can use ArraySection that general type conforming to it. ArraySection requires a model conforming to Differentiable for diffing from other sections:
enum Model: Differentiable {
case a, b, c
}
let source: [ArraySection<Model, String>] = [
ArraySection(model: .a, elements: ["A", "B"]),
ArraySection(model: .b, elements: ["C"])
]
let target: [ArraySection<Model, String>] = [
ArraySection(model: .c, elements: ["D", "E"]),
ArraySection(model: .a, elements: ["A"]),
ArraySection(model: .b, elements: ["B", "C"])
]
let changeset = StagedChangeset(source: source, target: target)
You can perform diffing batch updates of UITableView and UICollectionView using the created StagedChangeset.
⚠️ Don’t forget to synchronously update the data referenced by the data-source, with the data passed in the setData closure. The diffs are applied in stages, and failing to do so is bound to create a crash:
tableView.reload(using: changeset, with: .fade) { data in
dataSource.data = data
}
Batch updates using too large amount of diffs may adversely affect to performance. Returning true with interrupt closure then falls back to reloadData:
collectionView.reload(using: changeset, interrupt: { $0.changeCount > 100 }) { data in
dataSource.data = data
}
Made a fair comparison as much as possible in performance and features with other popular and awesome frameworks. This does NOT determine superiority or inferiority of the frameworks. I know that each framework has different benefits. The frameworks and its version that compared is below.
Benchmark project is here. Performance was mesured by code compiled using Xcode11.1 and Swift 5.1 with -O optimization and run on iPhone11 Pro simulator. Use Foundation.UUID as an element of collections.
- From 5,000 elements to 1,000 deleted, 1,000 inserted and 200 shuffled
Time(sec)
DifferenceKit
0.0019
RxDataSources
0.0074
IGListKit
0.0346
FlexibleDiff
0.0161
DeepDiff
0.0373
Differ
1.0581
Dwifft
0.4732
Swift.CollectionDifference
0.0620
- From 100,000 elements to 10,000 deleted, 10,000 inserted and 2,000 shuffled
A fast and flexible O(n) difference algorithm framework for Swift collection. The algorithm is optimized based on the Paul Heckel's algorithm.
Made with ❤️ by Ryo Aoyama and Contributors
Features
💡 Fastest O(n) diffing algorithm optimized for Swift collection
💡 Calculate diffs for batch updates of list UI in
UIKit
,AppKit
and Texture💡 Supports both linear and sectioned collection even if contains duplicates
💡 Supports all kind of diffs for animated UI batch updates
Algorithm
This is a diffing algorithm developed for Carbon, works stand alone.
The algorithm optimized based on the Paul Heckel’s algorithm.
See also his paper “A technique for isolating differences between files” released in 1978.
It allows all kind of diffs to be calculated in linear time O(n).
RxDataSources and IGListKit are also implemented based on his algorithm.
However, in
performBatchUpdates
ofUITableView
,UICollectionView
, etc, there are combinations of diffs that cause crash when applied simultaneously.To solve this problem,
DifferenceKit
takes an approach of split the set of diffs at the minimal stages that can be perform batch updates with no crashes.Implementation is here.
Getting Started
Basic Usage
The type of the element that to take diffs must be conform to the
Differentiable
protocol.The
differenceIdentifier
‘s type is generic associated type:In the case of definition above,
id
uniquely identifies the element and get to know the user updated by comparing equality ofname
of the elements in source and target.There are default implementations of
Differentiable
for the types that conforming toEquatable
orHashable
:Therefore, you can simply:
Calculate the diffs by creating
StagedChangeset
from two collections of elements conforming toDifferentiable
:If you want to include multiple types conforming to
Differentiable
in the collection, useAnyDifferentiable
:In the case of sectioned collection, the section itself must have a unique identifier and be able to compare whether there is an update.
So each section must conforming to
DifferentiableSection
protocol, but in most cases you can useArraySection
that general type conforming to it.ArraySection
requires a model conforming toDifferentiable
for diffing from other sections:You can perform diffing batch updates of
UITableView
andUICollectionView
using the createdStagedChangeset
.⚠️ Don’t forget to synchronously update the data referenced by the data-source, with the data passed in the
setData
closure. The diffs are applied in stages, and failing to do so is bound to create a crash:Batch updates using too large amount of diffs may adversely affect to performance.
Returning
true
withinterrupt
closure then falls back toreloadData
:[See More Usage]
Comparison with Other Frameworks
Made a fair comparison as much as possible in performance and features with other popular and awesome frameworks.
This does NOT determine superiority or inferiority of the frameworks.
I know that each framework has different benefits.
The frameworks and its version that compared is below.
Performance Comparison
Benchmark project is here.
Performance was mesured by code compiled using
Xcode11.1
andSwift 5.1
with-O
optimization and run oniPhone11 Pro simulator
.Use
Foundation.UUID
as an element of collections.- From 5,000 elements to 1,000 deleted, 1,000 inserted and 200 shuffled
0.0019
0.0074
0.0346
0.0161
0.0373
1.0581
0.4732
0.0620
- From 100,000 elements to 10,000 deleted, 10,000 inserted and 2,000 shuffled
0.0348
0.1024
0.7002
0.2189
0.5537
153.8007
187.1341
5.0281
Features Comparison
- Algorithm
* Heckel algorithm
* Myers algorithm
- Supported Collection
* Linear means 1-dimensional collection
* Sectioned means 2-dimensional collection
- Supported Element Diff
- Supported Section Diff
Requirements
Installation
CocoaPods
To use only algorithm without extensions for UI, add the following to your
Podfile
:iOS / tvOS
To use DifferenceKit with UIKit extension, add the following to your
Podfile
:or
macOS
To use DifferenceKit with AppKit extension, add the following to your
Podfile
:watchOS
There is no UI extension for watchOS.
To use only algorithm without extensions for UI, add the following to your
Podfile
:Carthage
Add the following to your
Cartfile
:Swift Package Manager for Apple platforms
Select Xcode menu
File > Swift Packages > Add Package Dependency
and enter repository URL with GUI.Swift Package Manager
Add the following to the dependencies of your
Package.swift
:Contribution
Pull requests, bug reports and feature requests are welcome 🚀
Please see the CONTRIBUTING file for learn how to contribute to DifferenceKit.
Credit
Bibliography
DifferenceKit was developed with reference to the following excellent materials and framework.
OSS using DifferenceKit
The list of the awesome OSS which uses this library. They also help to understanding how to use DifferenceKit.
Other diffing libraries
I respect and ️❤️ all libraries involved in diffing.
License
DifferenceKit is released under the Apache 2.0 License.