https://github.com/icelk/elipdotter

Embeddable full text search engine.
https://github.com/icelk/elipdotter

Last synced: about 1 year ago
JSON representation

Embeddable full text search engine.

• Host: GitHub
• URL: https://github.com/icelk/elipdotter
• Owner: Icelk
• License: lgpl-3.0
• Created: 2021-12-22T21:01:33.000Z (over 4 years ago)
• Default Branch: main
• Last Pushed: 2024-05-25T21:29:34.000Z (about 2 years ago)
• Last Synced: 2024-10-29T21:46:18.204Z (over 1 year ago)
• Language: Rust
• Size: 169 KB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: COPYING

Awesome Lists containing this project

README

          
[![crates.io version](https://img.shields.io/crates/v/elipdotter)](https://crates.io/crates/elipdotter)

![lines of code](https://tokei.rs/b1/github/Icelk/elipdotter?style=flat)

![license](https://img.shields.io/github/license/Icelk/elipdotter)

# Elipdotter

> The forgotten daughter of Elip, inheriting it's minimalism.

Elipdotter is an embeddable, [rusty](https://rust-lang.org) full-text search

engine, with fuzzy search and rating of results. It supports complex queries

using `AND`, `OR`, and `NOT` operators with parentheses.

## Kvarn integration

Using the [`kvarn-search`](https://github.com/Icelk/kvarn-search), you can

easily use this search engine in your [Kvarn web server](https://kvarn.org).

# Usage example

See the

[source of `kvarn-search`](https://github.com/Icelk/kvarn-search/tree/main/src/)

for a comprehensive example.

The [tests of this crate](https://github.com/Icelk/elipdotter/tree/main/tests/)

contain minimal examples.

# Changelog

## v0.3.5

-   Fix panic in some circumstances when non-ASCII characters were indexed.

-   Fix "starts with" detection when content contains non-alphanumeric or

    non-ASCII characters.

## v0.3.4

-   Update `simstr`

-   Fix some clippy lints.

## v0.3.3

-   Fix another issue with AND NOT, where some results would disappear when

    adding the NOT part.

## v0.3.2

-   Fix an issue with AND NOT, where some results would disappear when adding

    the NOT part.

## v0.3.1

-   Removed unwanted debugging.

## v0.3.0

-   Added lossless index for faster query resolution times.

    -   10x better performance at the cost of having all the documents in

        memory.

-   Fixed parsing issue where `for me` would be parsed as `OR(f, me)`.

-   Fixed issue with AND NOT, where AND didn't find the closest NOT occurrence.

-   Added size method to indices to estimate the memory usage.

-   Improvements to [docs](https://doc.icelk.dev/elipdotter/elipdotter/).

## v0.2.0

-   [Fixed issue](https://github.com/Icelk/elipdotter/commit/7ab071c) where AND

    NOT queries got erroneous results.

-   [Fixed issue](https://github.com/Icelk/elipdotter/commit/51370f7) with OR

    queries. Now all occurrences in either of the documents are returned.

-   Major improvements to relevancy of results, by checking more combinations of

    occurrences within a document. Small performance impact.

-   Text `next-gen` is now matched by the query `nextgen` - words with hyphens

    are registered as both separate words and one single.

-   Better [docs](https://doc.icelk.dev/elipdotter/elipdotter/).

-   Fewer allocations - less memory usage.

# License

Elipdotter is licensed under the [GNU LGPLv3](COPYING). All contributions must

also be.