WeKnora deployment source with the reviewed wiki identity, summary and recovery fixes

migrations/versioned/000000_init.up.sql

@@ -0,0 +1,205 @@
+-- Migration: 000000_init
+-- Description: Initialize database schema
+DO $$ BEGIN RAISE NOTICE '[Migration 000000] Starting initial database setup...'; END $$;
+
+-- Create extensions
+DO $$ BEGIN RAISE NOTICE '[Migration 000000] Creating extensions...'; END $$;
+CREATE EXTENSION IF NOT EXISTS "uuid-ossp";
+
+-- Create tenant table
+DO $$ BEGIN RAISE NOTICE '[Migration 000000] Creating table: tenants'; END $$;
+CREATE TABLE IF NOT EXISTS tenants (
+    id SERIAL PRIMARY KEY,
+    name VARCHAR(255) NOT NULL,
+    description TEXT,
+    api_key VARCHAR(64) NOT NULL,
+    retriever_engines JSONB NOT NULL DEFAULT '[]',
+    status VARCHAR(50) DEFAULT 'active',
+    business VARCHAR(255) NOT NULL,
+    storage_quota BIGINT NOT NULL DEFAULT 10737418240, -- 默认10GB配额(Bytes)
+    storage_used BIGINT NOT NULL DEFAULT 0, -- 已使用的存储空间(Bytes)
+    agent_config JSONB DEFAULT NULL,
+    created_at TIMESTAMP WITH TIME ZONE DEFAULT CURRENT_TIMESTAMP,
+    updated_at TIMESTAMP WITH TIME ZONE DEFAULT CURRENT_TIMESTAMP,
+    deleted_at TIMESTAMP WITH TIME ZONE
+);
+
+-- Set the starting value for tenants id sequence (only if current value is less than 10000)
+DO $$
+DECLARE
+    current_val BIGINT;
+BEGIN
+    SELECT last_value INTO current_val FROM tenants_id_seq;
+    IF current_val < 10000 THEN
+        ALTER SEQUENCE tenants_id_seq RESTART WITH 10000;
+        RAISE NOTICE '[Migration 000000] Set tenants_id_seq to start at 10000';
+    ELSE
+        RAISE NOTICE '[Migration 000000] tenants_id_seq already at % (>= 10000), skipping', current_val;
+    END IF;
+EXCEPTION
+    WHEN undefined_table THEN
+        -- Sequence doesn't exist yet, will be created with table
+        RAISE NOTICE '[Migration 000000] tenants_id_seq not found, will be created with table';
+END $$;
+
+-- Add indexes
+CREATE INDEX IF NOT EXISTS idx_tenants_api_key ON tenants(api_key);
+CREATE INDEX IF NOT EXISTS idx_tenants_status ON tenants(status);
+
+-- Create model table
+DO $$ BEGIN RAISE NOTICE '[Migration 000000] Creating table: models'; END $$;
+CREATE TABLE IF NOT EXISTS models (
+    id VARCHAR(64) PRIMARY KEY DEFAULT uuid_generate_v4(),
+    tenant_id INTEGER NOT NULL,
+    name VARCHAR(255) NOT NULL,
+    display_name VARCHAR(255) NOT NULL DEFAULT '',
+    type VARCHAR(50) NOT NULL,
+    source VARCHAR(50) NOT NULL,
+    description TEXT,
+    parameters JSONB NOT NULL,
+    is_default BOOLEAN NOT NULL DEFAULT false,
+    status VARCHAR(50) NOT NULL DEFAULT 'active',
+    created_at TIMESTAMP WITH TIME ZONE DEFAULT CURRENT_TIMESTAMP,
+    updated_at TIMESTAMP WITH TIME ZONE DEFAULT CURRENT_TIMESTAMP,
+    deleted_at TIMESTAMP WITH TIME ZONE
+);  
+
+-- Add indexes for models
+CREATE INDEX IF NOT EXISTS idx_models_type ON models(type);
+CREATE INDEX IF NOT EXISTS idx_models_source ON models(source);
+
+-- Create knowledge_base table
+DO $$ BEGIN RAISE NOTICE '[Migration 000000] Creating table: knowledge_bases'; END $$;
+CREATE TABLE IF NOT EXISTS knowledge_bases (
+    id VARCHAR(36) PRIMARY KEY DEFAULT uuid_generate_v4(),
+    name VARCHAR(255) NOT NULL,
+    description TEXT,
+    tenant_id INTEGER NOT NULL,
+    chunking_config JSONB NOT NULL DEFAULT '{"chunk_size": 512, "chunk_overlap": 50, "split_markers": ["\n\n", "\n", "。"], "keep_separator": true}',
+    image_processing_config JSONB NOT NULL DEFAULT '{"enable_multimodal": false, "model_id": ""}',
+    embedding_model_id VARCHAR(64) NOT NULL,
+    summary_model_id VARCHAR(64) NOT NULL,
+    rerank_model_id VARCHAR(64) NOT NULL,
+    cos_config JSONB NOT NULL DEFAULT '{}',
+    vlm_config JSONB NOT NULL DEFAULT '{}',
+    extract_config JSONB NULL DEFAULT NULL,
+    created_at TIMESTAMP WITH TIME ZONE DEFAULT CURRENT_TIMESTAMP,
+    updated_at TIMESTAMP WITH TIME ZONE DEFAULT CURRENT_TIMESTAMP,
+    deleted_at TIMESTAMP WITH TIME ZONE
+);
+
+-- Add indexes for knowledge_bases
+CREATE INDEX IF NOT EXISTS idx_knowledge_bases_tenant_id ON knowledge_bases(tenant_id);
+
+-- Create knowledge table
+DO $$ BEGIN RAISE NOTICE '[Migration 000000] Creating table: knowledges'; END $$;
+CREATE TABLE IF NOT EXISTS knowledges (
+    id VARCHAR(36) PRIMARY KEY DEFAULT uuid_generate_v4(),
+    tenant_id INTEGER NOT NULL,
+    knowledge_base_id VARCHAR(36) NOT NULL,
+    type VARCHAR(50) NOT NULL,
+    title VARCHAR(255) NOT NULL,
+    description TEXT,
+    source VARCHAR(2048) NOT NULL,
+    parse_status VARCHAR(50) NOT NULL DEFAULT 'unprocessed',
+    enable_status VARCHAR(50) NOT NULL DEFAULT 'enabled',
+    embedding_model_id VARCHAR(64),
+    file_name VARCHAR(255),
+    file_type VARCHAR(50),
+    file_size BIGINT,
+    file_path TEXT,
+    file_hash VARCHAR(64),
+    storage_size BIGINT NOT NULL DEFAULT 0, -- 存储大小(Byte)
+    metadata JSONB,
+    created_at TIMESTAMP WITH TIME ZONE DEFAULT CURRENT_TIMESTAMP,
+    updated_at TIMESTAMP WITH TIME ZONE DEFAULT CURRENT_TIMESTAMP,
+    processed_at TIMESTAMP WITH TIME ZONE,
+    error_message TEXT,
+    deleted_at TIMESTAMP WITH TIME ZONE
+);
+
+-- Add indexes for knowledge
+CREATE INDEX IF NOT EXISTS idx_knowledges_tenant_id ON knowledges(tenant_id);
+CREATE INDEX IF NOT EXISTS idx_knowledges_base_id ON knowledges(knowledge_base_id);
+CREATE INDEX IF NOT EXISTS idx_knowledges_parse_status ON knowledges(parse_status);
+CREATE INDEX IF NOT EXISTS idx_knowledges_enable_status ON knowledges(enable_status);
+
+-- Create session table
+DO $$ BEGIN RAISE NOTICE '[Migration 000000] Creating table: sessions'; END $$;
+CREATE TABLE IF NOT EXISTS sessions (
+    id VARCHAR(36) PRIMARY KEY DEFAULT uuid_generate_v4(),
+    tenant_id INTEGER NOT NULL,
+    title VARCHAR(255),
+    description TEXT,
+    knowledge_base_id VARCHAR(36),
+    max_rounds INTEGER NOT NULL DEFAULT 5,
+    enable_rewrite BOOLEAN NOT NULL DEFAULT true,
+    fallback_strategy VARCHAR(255) NOT NULL DEFAULT 'fixed',
+    fallback_response TEXT NOT NULL DEFAULT '很抱歉，我暂时无法回答这个问题。',
+    keyword_threshold FLOAT NOT NULL DEFAULT 0.5,
+    vector_threshold FLOAT NOT NULL DEFAULT 0.5,
+    rerank_model_id VARCHAR(64),
+    embedding_top_k INTEGER NOT NULL DEFAULT 10,
+    rerank_top_k INTEGER NOT NULL DEFAULT 10,
+    rerank_threshold FLOAT NOT NULL DEFAULT 0.65,
+    summary_model_id VARCHAR(64),
+    summary_parameters JSONB NOT NULL DEFAULT '{}',
+    agent_config JSONB DEFAULT NULL,
+    context_config JSONB DEFAULT NULL,
+    created_at TIMESTAMP WITH TIME ZONE DEFAULT CURRENT_TIMESTAMP,
+    updated_at TIMESTAMP WITH TIME ZONE DEFAULT CURRENT_TIMESTAMP,
+    deleted_at TIMESTAMP WITH TIME ZONE
+);
+
+-- Create Index for sessions
+CREATE INDEX IF NOT EXISTS idx_sessions_tenant_id ON sessions(tenant_id);
+
+
+-- Create message table
+DO $$ BEGIN RAISE NOTICE '[Migration 000000] Creating table: messages'; END $$;
+CREATE TABLE IF NOT EXISTS messages (
+    id VARCHAR(36) PRIMARY KEY DEFAULT uuid_generate_v4(),
+    request_id VARCHAR(36) NOT NULL,
+    session_id VARCHAR(36) NOT NULL,
+    role VARCHAR(50) NOT NULL,
+    content TEXT NOT NULL,
+    knowledge_references JSONB NOT NULL DEFAULT '[]',
+    agent_steps JSONB DEFAULT NULL,
+    is_completed BOOLEAN NOT NULL DEFAULT false,
+    created_at TIMESTAMP WITH TIME ZONE DEFAULT CURRENT_TIMESTAMP,
+    updated_at TIMESTAMP WITH TIME ZONE DEFAULT CURRENT_TIMESTAMP,
+    deleted_at TIMESTAMP WITH TIME ZONE
+);
+
+-- Create Index for messages
+CREATE INDEX IF NOT EXISTS idx_messages_session_id ON messages(session_id); 
+
+-- Create chunks table
+DO $$ BEGIN RAISE NOTICE '[Migration 000000] Creating table: chunks'; END $$;
+CREATE TABLE IF NOT EXISTS chunks (
+    id VARCHAR(36) PRIMARY KEY DEFAULT uuid_generate_v4(),
+    tenant_id INTEGER NOT NULL,
+    knowledge_base_id VARCHAR(36) NOT NULL,
+    knowledge_id VARCHAR(36) NOT NULL,
+    content TEXT NOT NULL,
+    chunk_index INTEGER NOT NULL,
+    is_enabled BOOLEAN NOT NULL DEFAULT true,
+    start_at INTEGER NOT NULL,
+    end_at INTEGER NOT NULL,
+    pre_chunk_id VARCHAR(36),
+    next_chunk_id VARCHAR(36),
+    chunk_type VARCHAR(20) NOT NULL DEFAULT 'text',
+    parent_chunk_id VARCHAR(36),
+    image_info TEXT,
+    relation_chunks JSONB,
+    indirect_relation_chunks JSONB,
+    created_at TIMESTAMP WITH TIME ZONE DEFAULT CURRENT_TIMESTAMP,
+    updated_at TIMESTAMP WITH TIME ZONE DEFAULT CURRENT_TIMESTAMP,
+    deleted_at TIMESTAMP WITH TIME ZONE
+);
+
+CREATE INDEX IF NOT EXISTS idx_chunks_tenant_kg ON chunks(tenant_id, knowledge_id);
+CREATE INDEX IF NOT EXISTS idx_chunks_parent_id ON chunks(parent_chunk_id);
+CREATE INDEX IF NOT EXISTS idx_chunks_chunk_type ON chunks(chunk_type);
+
+DO $$ BEGIN RAISE NOTICE '[Migration 000000] Initial database setup completed successfully!'; END $$;