Meilisearch – Open-source Search API for Developers

Overview

Meilisearch is an open-source, developer-focused search engine designed to embed fast and relevant search experiences into applications. It offers a RESTful API that allows developers to index data and perform real-time searches with built-in typo-tolerance and relevancy ranking. Unlike some general-purpose database-driven search solutions, Meilisearch is engineered specifically for search, prioritizing speed and user experience from the ground up.

The core philosophy behind Meilisearch revolves around ease of use and immediate utility. Developers can deploy Meilisearch either as a self-hosted instance or utilize its hosted cloud service, Meilisearch Cloud. The engine automatically handles common search challenges like misspellings, plurals, and synonyms without requiring extensive manual configuration, which can be a time-saver compared to building custom search logic or fine-tuning more complex search platforms like Elasticsearch's distributed search capabilities.

Meilisearch is suitable for a range of applications that require efficient and user-friendly search. Its capabilities shine in e-commerce platforms where quick product discovery is crucial, allowing customers to find items despite typos or partial queries. Similarly, it is effective for internal knowledge bases, documentation portals, and content-heavy websites where users need to rapidly navigate information. The platform provides comprehensive SDKs across multiple programming languages, including JavaScript, Python, PHP, and Ruby, simplifying integration into existing development workflows. This focus on developer experience extends to its API design, which is intended to be intuitive and straightforward to interact with, enabling quicker implementation cycles.

Key features

• Typo Tolerance: Automatically handles common misspellings and typos in search queries, providing relevant results even with imperfect input.
• Relevancy Ranking: Employs a customizable ranking algorithm that prioritizes results based on factors like word proximity, attribute importance, and exact matches.
• Fast Search: Optimized for low-latency search queries, designed to return results in milliseconds.
• Faceting and Filtering: Supports filtering and faceting capabilities, enabling users to refine search results based on categories, price ranges, or other attributes.
• Multi-language Support: Capable of indexing and searching content in various languages, with specific optimizations for tokenization and stemming.
• Customizable Search Experience: Developers can configure ranking rules, define searchable attributes, and manage synonyms to tailor the search experience.
• RESTful API: Provides a straightforward HTTP API for indexing documents and performing search queries, accessible from any programming language.
• Comprehensive SDKs: Offers client libraries for JavaScript, Python, PHP, Ruby, Go, Java, C#, Rust, Dart, and Swift to streamline integration.
• Open Source: The core engine is open-source, allowing for self-hosting and community contributions, with the option for commercial cloud services.
• Document Indexing: Supports indexing JSON documents, allowing for flexible data structures to be searched.

Pricing

Meilisearch offers a free tier for its cloud service, along with various paid plans structured by usage. Self-hosting the open-source version incurs no software cost, only infrastructure expenses.

For detailed and up-to-date pricing information, refer to the official Meilisearch pricing page.

Common integrations

• Frontend Frameworks: Easily integrates with popular JavaScript frameworks like React, Vue, and Angular using its JavaScript SDK for dynamic search UIs.
• E-commerce Platforms: Can be integrated with platforms like Shopify or custom e-commerce solutions to power product search and filtering.
• Content Management Systems (CMS): Used to enhance search capabilities within CMS platforms for articles, blog posts, and documentation.
• Database Systems: Works alongside databases like PostgreSQL or MongoDB, indexing data from these sources to provide search functionality.
• API Gateways: Can be deployed behind API gateways like Kong API Gateway for centralized API management and security.
• Serverless Functions: Integrates with serverless platforms (AWS Lambda, Google Cloud Functions) for indexing and searching data without managing servers.

Alternatives

• Algolia: A hosted search API service known for its speed and relevance, offering similar features to Meilisearch Cloud but typically with a higher price point for comparable scale.
• Elasticsearch: A distributed, RESTful search and analytics engine. It is highly scalable and versatile, suitable for complex data analytics and logging, but generally requires more operational overhead and configuration for simple search use cases.
• Typesense: Another open-source, typo-tolerant search engine designed for speed and developer experience, often compared to Meilisearch and Algolia.

Getting started

The following example demonstrates how to set up a basic Meilisearch instance, add documents, and perform a search using the Python SDK. This assumes you have Meilisearch running locally or accessible via a server.

# pip install meilisearch
from meilisearch import Client

# Initialize the client with your Meilisearch instance URL and API key (if set)
# Default URL for a local instance is http://localhost:7700
client = Client('http://localhost:7700', 'YOUR_MASTER_KEY') # Replace with your actual key if applicable

# Create an index if it doesn't exist
index = client.index('movies')

# Add documents to the index
movies = [
    {"id": 1, "title": "The Lord of the Rings: The Fellowship of the Ring", "genres": ["Fantasy", "Adventure"]},
    {"id": 2, "title": "The Hitchhiker's Guide to the Galaxy", "genres": ["Sci-Fi", "Comedy"]},
    {"id": 3, "title": "Lord of the Flies", "genres": ["Drama"]}
]

# Add documents. Meilisearch automatically handles the primary key if 'id' is present.
task = index.add_documents(movies)
# Wait for the task to be completed (optional, good for ensuring data is indexed)
client.wait_for_task(task.task_uid)

print("Documents added successfully.")

# Perform a search
search_results = index.search('lord')
print("\nSearch results for 'lord':")
for hit in search_results['hits']:
    print(f"- {hit['title']} (ID: {hit['id']})")

# Perform a typo-tolerant search
typo_search_results = index.search('lrd')
print("\nSearch results for 'lrd' (typo-tolerant):")
for hit in typo_search_results['hits']:
    print(f"- {hit['title']} (ID: {hit['id']})")

# Search with filters
filtered_search_results = index.search('the', {'filter': ['genres = "Sci-Fi"']})
print("\nSearch results for 'the' filtered by 'Sci-Fi' genre:")
for hit in filtered_search_results['hits']:
    print(f"- {hit['title']} (ID: {hit['id']})")

This Python example demonstrates indexing movie data and performing both basic and typo-tolerant searches. For more advanced features like custom ranking rules, synonym management, and specific language configurations, refer to the official Meilisearch API reference documentation.