NovaCraft.FullTextSearch 1.0.0

NovaCraft Full-Text Search

NovaCraft Full-Text Search is a lightweight .NET package that enables full-text search functionality for your applications, similar to Azure AI Search, using Lucene.Net as the underlying engine. It allows indexing of database tables and provides fast search capabilities with support for AND, OR, wildcard, and fuzzy searches.

Features

• Full-text search over multiple columns and tables.
• AND / OR query support.
• Wildcard search (prefix, suffix, and contains).
• Fuzzy search for approximate matching.
• Incremental and asynchronous indexing.
• Track indexed row counts.
• Configurable columns and tables.
• Cross-platform .NET support.
• Simple API, no external dependencies except Lucene.Net.

Table of Contents

• Features
• Installation
• Usage
• Configuration
• Best Practices
• Security Considerations
• Next Version / Roadmap
• License

Installation

Install the NuGet package:

dotnet add package NovaCraft.FullTextSearch

Usage

1. Configure Search

using NovaCraft.FullTextSearch;
using NovaCraft.FullTextSearch.Models;

var engine = new SearchEngine();

// Add search configuration
engine.AddSearchConfig(new SearchConfig
{
    Name = "Products",
    Columns = new List<ColumnMapping>
    {
        new ColumnMapping { Table = "highValueProducts", Column = "Name" },
        new ColumnMapping { Table = "highValueProducts", Column = "Value" },
        new ColumnMapping { Table = "highValueProducts", Column = "Skid" }
    }
});

2. Index Data

// Synchronous
engine.IndexData("highValueProducts", new Dictionary<string, object>
{
    { "Name", "Product1" },
    { "Value", 100 },
    { "Skid", "SK123" }
});

// Asynchronous batch indexing
engine.IndexDataAsync("highValueProducts", listOfRows);

Track indexed row counts:

int indexedCount = engine.GetIndexedCount("Products");
Console.WriteLine($"Indexed rows: {indexedCount}");

3. Index Data

Supports AND / OR operators, wildcards, and fuzzy searches:

var results = engine.Find("Products", "Product AND 100");

foreach (var row in results)
{
    foreach (var kvp in row)
        Console.WriteLine($"{kvp.Key}: {kvp.Value}");
}

Specify maximum number of results:

var results = engine.Find("Products", "Product", maxResults: 500);

Configuration

1. SearchConfig

public class SearchConfig
{
    public string Name { get; set; }
    public List<ColumnMapping> Columns { get; set; }
}

2. ColumnMapping

public class ColumnMapping
{
    public string Table { get; set; }
    public string Column { get; set; }
}

Best Practices

• Use asynchronous indexing for large datasets.
• Limit search results with maxResults to improve performance.
• Avoid indexing extremely large columns with unbounded text.
• Persist and reuse the index directory to avoid re-indexing.
• Monitor GetIndexedCount() to track indexing progress.

Security Considerations

• Sanitize input from external sources.
• Use parameterized queries when fetching data.
• Avoid exposing the raw Lucene index path.
• Run indexing with a restricted service account.

Next Version / Roadmap

• Fuzzy search improvements.
• Highlighting matched terms in results.
• Multi-language support.
• Search scoring and ranking improvements.

License

MIT License. Free for personal and commercial use.