https://github.com/ukushu/ifrit
Swfit library for fuzzy search. No dependencies lib.
https://github.com/ukushu/ifrit
app-search full-text-search fuzzy-matching fuzzy-search hybrid-search levenstain search search-algorithm search-as-you-type search-engine swift typo-tolerance typo-tolerant
Last synced: 3 months ago
JSON representation
Swfit library for fuzzy search. No dependencies lib.
- Host: GitHub
- URL: https://github.com/ukushu/ifrit
- Owner: ukushu
- License: mit
- Created: 2024-05-11T15:52:00.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2025-04-22T07:27:27.000Z (6 months ago)
- Last Synced: 2025-06-08T08:23:45.534Z (4 months ago)
- Topics: app-search, full-text-search, fuzzy-matching, fuzzy-search, hybrid-search, levenstain, search, search-algorithm, search-as-you-type, search-engine, swift, typo-tolerance, typo-tolerant
- Language: Swift
- Homepage:
- Size: 3.05 MB
- Stars: 60
- Watchers: 2
- Forks: 8
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- Changelog: CHANGELOG.md
- License: LICENSE.md
Awesome Lists containing this project
README
# Ifrit
[](https://swiftpackageindex.com/ukushu/Ifrit)
[](https://swiftpackageindex.com/ukushu/Ifrit)
```
https://github.com/ukushu/Ifrit.git
git@github.com:ukushu/Ifrit.git
```
## What is Ifrit?
Ifrit is fuzzy search library written in pure swift.
### What is fuzzy Search?
A fuzzy search algorithm is designed to find approximate matches for a given search query.
Let's say you have a list of products, and a user is looking for a "Czy Swetr,". Exact match will return nothing, but the fuzzy search will find "Cozy Sweater.".
If you need exact match - you need to use standard function `.filter { }` instead of this library.
Demo of Fuzzy Search:
[![Ifrit demo][1]][1]
### Advanteges / Disadvantages of Ifrit
```diff
Advanteges:
+ Super lightweight
+ No dependencies
+ OS supported: MacOS(>=v10_15), iOS(>=v13), visionOS, watchOS, twOS, Linux
+ Async ( async/await + callbacks both )
+ Text highlight support (AttributedString)
+ Code covered with automated tests
+ MIT Licensed
+ Can be linked as both Static and Dynamic package — choose what works best for you
Disadvantages:
- Not tested in production
- Requires Swift 5.9 or higher (older versions not supported)
```
This repository is built upon the archived Fuse-Swift project by KRisk: https://github.com/krisk/fuse-swift
Swift Package Index information: https://swiftpackageindex.com/ukushu/Ifrit
## Difference: Ifrit VS Fuze-Swift?
```diff
+ Fuze-Swift support Pods and Packages :)
- Ifrit supports only Packages :(
- Fuze-Swift is dead :(
+ 🔥🔥🔥 Ifrit is Fuse-Swift, reborn from the hells ashes! But more powerful and alive than ever! 🔥🔥🔥 :)
- Fuse-Swift supports asynchronous functions with callbacks but lacks native async/await support. :(
+ Ifrit supports callbacks and async/await both :)
- Fuze-Swift have async/multithreading issues. Crashes is possible :(
+ Ifrit is stable and crash-free!
- Fuse-Swift's async functions return incorrect results - wrong indexes. :((((
+ Ifrit's async functions return correct indexes and are fully covered with tests :)
- Fuze-Swift was written for Xcode 11 and now generates numerous warnings :(
+ Ifrit's code is updated to the latest Swift standards and fully supports Swift 6 :)
- But Ifrit have a few build warnings :(
+ Ifrit - extended coverage with automated tests :)
+ Ifrit features an additional search system using the Levenshtein distance algorithm. :)
+ Ifrit allows searching by multiple search templates
+ Ifrit's syntax has been slightly adjusted for improved flexibility
+ Ifrit's sync search is 2% faster than Fuse's (using M1 processor and standard settings).
+ Async search performance remains the same as Fuse's (standard settings).
```
## Installation
1. XCode -> Menu Line -> Add Package Dependencies -> `https://github.com/ukushu/Ifrit.git`
2. `import Ifrit` to your source files.
## Documentation / Instructions / Example Project
Ifrit repository have no example project. Use instructions below:
[How to use Ifrit's Fuse](https://github.com/ukushu/Ifrit/blob/main/Documentation/FuseInstructions.md)
[How to use Ifrit's Levenstain](https://github.com/ukushu/Ifrit/blob/main/Documentation/LevenstainInstructions.md)
[How to display search results in UI](https://github.com/ukushu/Ifrit/blob/main/Documentation/HowToUseSrchResult.md)
## Contributing
Ifrit welcomes contributions in the form of GitHub issues and pull-requests.
[1]: https://i.sstatic.net/8MwFeAHT.gif