WikiGo - ElasticSearch search engine

Wiki Go is a search engine that enables users to find articles and other information using various filters and advanced search functionalities. Developed as part of the Introduction to ElasticSearch at Universidade Federal de Alfenas, this project demonstrates the power of ElasticSearch in combination with modern web development frameworks.

Table of Contents

1. Overview
2. Technologies Used
3. User Interface
4. Usage
5. Project Structure
6. API Endpoints
7. Examples of Queries
8. Contributors

Overview

Wiki Go provides an intuitive interface for searching and retrieving content efficiently. Leveraging ElasticSearch, the project highlights:

• Fast and scalable search capabilities.
• Integration of keyword highlighting and advanced filtering.
• A seamless experience between the backend (Java) and the frontend (React).

This project was created to demonstrate proficiency in ElasticSearch and full-stack development.

Technologies Used

• ElasticSearch: For storing and indexing data.
• Java: Backend development with Spring Boot.
• React: Frontend development with a responsive design.
• Axios: For managing HTTP requests between the frontend and backend.
• HTML/CSS/JavaScript: Core web technologies for UI implementation.

User Interface

WikiGo home page

Usage

The Wiki Go search engine provides a user-friendly interface to search and filter articles. Users can:

• Input keywords in the search bar.
• Apply filters such as publication date, category, and tags.
• View highlighted keywords in search results for better readability.

Project Structure

The project is organized as follows:

• backend/: Contains the Java Spring Boot application that interacts with ElasticSearch.

  • Controllers: Handle incoming HTTP requests.
  • Services: Contain business logic.
  • Repositories: Define ElasticSearch queries.

• frontend_wikigo-main/: Contains the React application for the user interface.

  • Components: Modular UI elements.
  • Pages: Define the layout of key pages like the search and result pages.

API Endpoints

Base URL

The backend runs on /api by default.

Endpoints

• GET /search
  • Description: Retrieves search results based on query parameters.
  • Query Parameters:
    • query: The search keyword(s).
    • page: Page number to be returned.
    • filters: Optional filters (e.g., category, date).
  • Example:

    curl -X GET "http://localhost:8080/api/search?query=exampleQuery&page=1&filter=filter1&filter=filter2"

Examples of Queries

Example 1: Filtered Search

GET "http://localhost/api/search?query=machine+learning&page=2" 

Response:

[
  {
    "abs": "Machine Learning Article 1 example.",
    "title": "Machine Learning introduction",
    "url": "https://example.com/machine-learning-introduction"
  },
  {
    "abs": "Machine Learning Article 2 example.",
    "title": "Machine is Learning",
    "url": "https://example.com/machine-is-learning"
  }
]

Supported Filters

1. No Filters (Basic Search)

Executes a basic search query without applying any filters.
Endpoint: /search
Description: Retrieves results based on the provided query string.

• Example:

  GET /api/search?query=example&page=1

---

2. Search with AND Operator

Searches with all terms required in the content.
Endpoint: /search
Description: Uses the AND operator to ensure all terms match.

• Example:

  GET /api/search?query=example&filter=and&page=1

---

3. Exclude Terms (Must Not)

Filters out results containing specific terms.
Endpoint: /search
Description: Excludes content that matches the specified terms.

• Example:

  GET /api/search?query=example&filter=must_not:excluded_term&page=1

---

4. Filter by Creation Date (dt_creation)

Filters results based on a range of creation dates.
Endpoint: /search
Description: Use comparison operators (lt, gte) with a date.

• Parameters:

  • lt: Less than
  • gte: Greater than or equal to

• Example:

  GET /api/search?query=example&filter=dt_creation:lt:2023-01-01&page=1
  GET /api/search?query=example&filter=dt_creation:gte:2023-01-01&page=1

---

5. Filter by Reading Time (reading_time)

Filters results based on the reading time of the content.
Endpoint: /search
Description: Filters based on a range of reading times using lt or gte.

• Parameters:

  • lt: Less than
  • gte: Greater than or equal to

• Example:

  GET /api/search?query=example&filter=reading_time:lt:10&page=1
  GET /api/search?query=example&filter=reading_time:gte:5&page=1

---

6. Fuzzy Search

Executes a fuzzy search to find matches with slight variations.
Endpoint: /search
Description: Allows minor differences in the query, such as typos.

• Example:

  GET /api/search?query=exmple&filter=fuzziness:2&page=1

---

Highlighting Results

All searches support highlighting of matched content. By default, results include highlighted terms enclosed in <strong> tags.

Example Response:

[
  {
    "abs": "This is a <strong>highlighted</strong> result.",
    "title": "Highlighted Title",
    "url": "https://example.com/result"
  }
]

Contributors

Giovana Nogueira backend

Lucas Dogo frontend