nxsearch

nxsearch is a full-text search engine library which is also provided with a web server integration.

The engine is written in C11 and is distributed under the 2-clause BSD license.

Upstream at: https://github.com/rmind/nxsearch

Features

• Lightweight, simple and fast.
• BM25 and TF-IDF algorithms.
• Integrates with the Snowball stemmer.
• Supports fuzzy matching (using the BK-tree with Levenshtein distance).
• Supports query logic operators, grouping, nesting, etc.
• Supports filters in Lua for easy extendibility.
• Basic UTF-8 and internationalization support.

Usage

To try as a web service:

# git submodule update --init --recursive  # ensure you have submodules
docker-compose up app  # spin up the service
open http://127.0.0.1:8000/docs  # documentation page

The NXS_BASEDIR environment variable specifies the base directory where the indexed documents as well as the application data files are stored.

Shell

# Create the index:
curl -XPOST http://127.0.0.1:8000/test-idx

# Index some test documents:
curl -d "cat dog cow" http://127.0.0.1:8000/test-idx/add/1
curl -d "dog cow" http://127.0.0.1:8000/test-idx/add/2
curl -d "cat cat cat" http://127.0.0.1:8000/test-idx/add/3

# Run a query:
curl -s -d "cat" http://127.0.0.1:8000/test-idx/search | jq

Python

You can also use the nxsearch as a Python library.

import nxsearch
from nxsearch import NxsResourceExistsError

nxs = nxsearch.init("./nxs-data")
index = nxs.open("animal-articles", create=True)

index.add(1, "cat dog cow")
index.add(2, "dog cow")
index.add(3, "cat cat cat")
index.add(4, "cat's catnip")

with index.search("cat") as result:
    for item in result:
        print(item)

Documentation

• Swagger UI with the endpoint documentation is provided at /docs URL.

• The C API documentation can be found HERE.

• The Lua filters API can be found HERE.