Version 1.1.5

Overview

This release reintroduces HikariCP for enhanced MySQL database performance and adds a public API for other plugins to interact with OllamaChat's AI chat functionality. It focuses on improving database efficiency, extensibility, and developer integration while maintaining compatibility with existing configurations.

New Features

• HikariCP Integration:
  • Added HikariCP connection pooling for MySQL databases to improve performance and resource management.
  • Configurable HikariCP settings in config.yml under database.mysql.hikari for tuning connection pool parameters (e.g., maximum-pool-size, connection-timeout).
• Public API:
  • Introduced OllamaChatAPI interface (com.ollamachat.api.OllamaChatAPI) for other plugins.
  • Supports sending AI queries, managing conversations, and accessing chat history programmatically with methods like sendAIQuery, createConversation, and listConversations.
  • Added API documentation in API.md for developer integration.

Enhancements

• Database Management:
  • Updated DatabaseManager to leverage HikariCP for MySQL connections, while preserving SQLite functionality.
• Dependency Management:
  • Updated DependencyLoader to include HikariCP (version 5.1.0) alongside MySQL and SQLite JDBC drivers, downloaded at runtime to plugins/OllamaChat/libs.
  • Maintained compileOnly in build.gradle for non-Paper dependencies to minimize JAR size.
• Configuration:
  • Added database.mysql.hikari section to config.yml with default settings for connection pooling.
  • Improved config.yml comments for clarity and ease of use.

Bug Fixes

• No specific bug fixes in this release, as the focus was on new features and performance improvements.

API Usage Example

Plugin ollamaPlugin = getServer().getPluginManager().getPlugin("OllamaChat");
if (ollamaPlugin != null && ollamaPlugin.isEnabled()) {
    OllamaChatAPI api = ((com.ollamachat.core.Ollamachat) ollamaPlugin).getAPI();
    api.sendAIQuery(player, "ollama", "Hello, how are you?")
        .thenAccept(response -> player.sendMessage("AI Response: " + response));
}

Configuration Example

database:
  type: mysql
  mysql:
    host: localhost
    port: 3306
    database: ollamachat
    username: root
    password: ""
    hikari:
      maximum-pool-size: 10
      connection-timeout: 30000

Notes for Upgrading

• Configuration:
  • Update config.yml to include the database.mysql.hikari section if using MySQL. Default values are added automatically.
  • Backup existing config.yml before updating to avoid data loss.
• Dependencies:
  • The plugin downloads HikariCP (version 5.1.0) automatically on first startup. Ensure server internet access or verify HikariCP-5.1.0.jar in plugins/OllamaChat/libs.
  • Delete the libs folder to force re-download if dependencies are corrupted.
• API Integration:
  • Developers can hook into OllamaChat using OllamaChatAPI. Add softdepend: [OllamaChat] to plugin.yml and refer to API.md.
• Compatibility:
  • This update is backward-compatible with existing configurations and databases.
• Testing:
  • Test database operations (e.g., /ollamachat conversation new) to verify MySQL/SQLite functionality.
  • Check server logs for dependency loading or database errors.

Known Issues

• HikariCP settings may require manual tuning for optimal performance based on server load and MySQL configuration.

版本 1.1.5

概述

本次更新重新引入HikariCP以提升MySQL数据库性能，并新增公共API供其他插件与OllamaChat的AI聊天功能交互。重点改进数据库效率、可扩展性和开发者集成，同时保持与现有配置的兼容性。

新特性

• HikariCP集成：
  • 为MySQL数据库添加HikariCP连接池，提升性能和资源管理
  • 在config.yml的database.mysql.hikari下可配置连接池参数（如maximum-pool-size、connection-timeout）
• 公共API：
  • 新增OllamaChatAPI接口（com.ollamachat.api.OllamaChatAPI）
  • 支持通过编程方式发送AI查询、管理会话和访问聊天记录（如sendAIQuery、createConversation、listConversations等方法）
  • 开发者文档见API.md

功能增强

• 数据库管理：
  • 更新DatabaseManager以使用HikariCP管理MySQL连接，同时保留SQLite功能
• 依赖管理：
  • 更新DependencyLoader包含HikariCP（5.1.0版）及MySQL/SQLite驱动，运行时下载至plugins/OllamaChat/libs
  • 保持build.gradle中非Paper依赖使用compileOnly以减小JAR体积
• 配置优化：
  • 在config.yml新增database.mysql.hikari配置段
  • 改进配置注释的清晰度

问题修复

• 本版本主要侧重新功能和性能提升，无特定问题修复

API使用示例

Plugin ollamaPlugin = getServer().getPluginManager().getPlugin("OllamaChat");
if (ollamaPlugin != null && ollamaPlugin.isEnabled()) {
    OllamaChatAPI api = ((com.ollamachat.core.Ollamachat) ollamaPlugin).getAPI();
    api.sendAIQuery(player, "ollama", "你好吗？")
        .thenAccept(response -> player.sendMessage("AI回复: " + response));
}

配置示例

database:
  type: mysql
  mysql:
    host: localhost
    port: 3306
    database: ollamachat
    username: root
    password: ""
    hikari:
      maximum-pool-size: 10
      connection-timeout: 30000

升级须知

• 配置：
  • 使用MySQL需更新config.yml添加hikari配置段（默认值会自动生成）
  • 升级前备份现有配置
• 依赖：
  • 首次启动自动下载HikariCP（5.1.0版），确保服务器联网或检查plugins/OllamaChat/libs目录
  • 依赖损坏时可删除libs文件夹强制重新下载
• API集成：
  • 开发者可通过OllamaChatAPI集成，在plugin.yml添加softdepend: [OllamaChat]并参考API.md
• 兼容性：
  • 保持与现有配置和数据库的向后兼容
• 测试建议：
  • 测试数据库操作（如/ollamachat conversation new命令）
  • 检查服务器日志中的依赖加载和数据库错误

已知问题

• HikariCP参数可能需要根据服务器负载手动调优