meilisearch-sdk使用指南：快速集成全文搜索功能到Rust应用

介绍

meilisearch-sdk是一个Rust客户端库，用于与MeiliSearch搜索引擎交互。MeiliSearch是一个开源的、超快速的搜索引擎，可以轻松地为你的应用添加强大的搜索功能。

主要特性

• 简单易用的API
• 即时搜索（输入时实时显示结果）
• 高度可定制（相关性排序、过滤、分面等）
• 支持多种数据类型
• 轻量级且快速

安装

在Cargo.toml中添加依赖：

[dependencies]
meilisearch-sdk = "0.20"
tokio = { version = "1.0", features = ["full"] }

基本使用方法

1. 连接到MeiliSearch服务器

use meilisearch_sdk::{client::*, indexes::*};

#[tokio::main]
async fn main() {
    // 连接到本地MeiliSearch实例
    let client = Client::new("http://localhost:7700", "masterKey");
    
    // 检查服务器健康状态
    if let Ok(health) = client.health().await {
        println!("Server is healthy: {:?}", health);
    }
}

2. 创建索引并添加文档

#[derive(Debug, serde::Serialize, serde::Deserialize)]
struct Movie {
    id: String,
    title: String,
    genres: Vec<String>,
    year: i32,
}

async fn create_index(client: &Client) {
    // 创建或获取索引
    let movies = client.index("movies");
    
    // 定义文档
    let documents = vec![
        Movie {
            id: "1".to_string(),
            title: "The Shawshank Redemption".to_string(),
            genres: vec!["Drama".to_string(), "Crime".to_string()],
            year: 1994,
        },
        Movie {
            id: "2".to_string(),
            title: "The Godfather".to_string(),
            genres: vec!["Crime".to_string(), "Drama"].to_string()],
            year: 1972,
        },
    ];
    
    // 添加文档到索引
    let task = movies.add_documents(&documents, Some("id")).await;
    match task {
        Ok(task) => println!("Documents added successfully: {:?}", task),
        Err(e) => eprintln!("Error adding documents: {}", e),
    }
}

3. 执行搜索

async fn search_movies(client: &Client) {
    let movies = client.index("movies");
    
    // 简单搜索
    let results = movies.search()
        .with_query("shawshank")
        .execute::<Movie>()
        .await
        .unwrap();
    
    println!("Search results for 'shawshank':");
    for hit in results.hits {
        println!("- {} ({})", hit.result.title, hit.result.year);
    }
    
    // 带过滤的搜索
    let filtered_results = movies.search()
        .with_query("crime")
        .with_filter("year > 1990")
        .execute::<Movie>()
        .await
        .unwrap();
    
    println!("\nFiltered results (crime movies after 1990):");
    for hit in filtered_results.hits {
        println!("- {} ({})", hit.result.title, hit.result.year);
    }
}

高级功能

1. 配置索引设置

async fn configure_index(client: &Client) {
    let movies = client.index("movies");
    
    // 设置搜索属性
    let settings = Settings::new()
        .with_searchable_attributes(["title", "genres"])
        .with_filterable_attributes(["year", "genres"]);
    
    movies.set_settings(&settings).await.unwrap();
}

2. 同义词和停用词

async fn configure_synonyms(client: &Client) {
    let movies = client.index("movies");
    
    // 设置同义词
    let synonyms = json!({
        "flick": ["movie", "film"],
        "movie": ["flick", "film"],
        "film": ["flick", "movie"]
    });
    
    movies.set_synonyms(&synonyms).await.unwrap();
}

3. 分面搜索

async fn facet_search(client: &Client) {
    let movies = client.index("movies");
    
    // 获取所有流派的分面统计
    let results = movies.search()
        .with_query("")
        .with_facets(["genres"])
        .execute::<Movie>()
        .await
        .unwrap();
    
    if let Some(facets) = results.facet_distribution {
        if let Some(genres) = facets.get("genres") {
            println!("Genre distribution:");
            for (genre, count) in genres {
                println!("- {}: {}", genre, count);
            }
        }
    }
}

完整示例代码

use meilisearch_sdk::{client::*, indexes::*, settings::Settings};
use serde_json::json;

#[derive(Debug, serde::Serialize, serde::Deserialize)]
struct Movie {
    id: String,
    title: String,
    genres: Vec<String>,
    year: i32,
}

#[tokio::main]
async fn main() {
    // 1. 连接到服务器
    let client = Client::new("http://localhost:7700", "masterKey");
    
    // 2. 创建索引并添加文档
    let movies = client.index("movies");
    
    let documents = vec![
        Movie {
            id: "1".to_string(),
            title: "The Shawshank Redemption".to_string(),
            genres: vec!["Drama".to_string(), "Crime".to_string()],
            year: 1994,
        },
        Movie {
            id: "2".to_string(),
            title: "The Godfather".to_string(),
            genres: vec!["Crime".to_string(), "Drama".to_string()],
            year: 1972,
        },
    ];
    
    movies.add_documents(&documents, Some("id")).await.unwrap();
    
    // 3. 配置索引
    let settings = Settings::new()
        .with_searchable_attributes(["title", "genres"])
        .with_filterable_attributes(["year", "genres"]);
    
    movies.set_settings(&settings).await.unwrap();
    
    // 4. 执行搜索
    let results = movies.search()
        .with_query("shawshank")
        .execute::<Movie>()
        .await
        .unwrap();
    
    println!("Search results:");
    for hit in results.hits {
        println!("- {} ({})", hit.result.title, hit.result.year);
    }
    
    // 5. 分面搜索
    let facet_results = movies.search()
        .with_query("")
        .with_facets(["genres"])
        .execute::<Movie>()
        .await
        .unwrap();
    
    if let Some(facets) = facet_results.facet_distribution {
        if let Some(genres) = facets.get("genres") {
            println!("\nGenre distribution:");
            for (genre, count) in genres {
                println!("- {}: {}", genre, count);
            }
        }
    }
}

总结

meilisearch-sdk为Rust应用提供了简单高效的方式来集成MeiliSearch的强大搜索功能。通过这个库，你可以快速实现全文搜索、过滤、分面搜索等高级功能，同时保持代码的简洁性和高性能。