lingyue-zhibao/database/init.sql

-- =====================================================
-- 灵越智报 v2.0 数据库初始化脚本（单文件）
-- PostgreSQL 15+
-- 包含所有表结构，不含 graph_nodes / graph_relations（已移除）
-- =====================================================

CREATE EXTENSION IF NOT EXISTS "uuid-ossp";
CREATE EXTENSION IF NOT EXISTS vector;

-- ============================================ 一、基础模块 ============================================

CREATE TABLE IF NOT EXISTS users (
    id VARCHAR(36) PRIMARY KEY,
    username VARCHAR(50) UNIQUE NOT NULL,
    email VARCHAR(100) UNIQUE NOT NULL,
    password_hash VARCHAR(255) NOT NULL,
    avatar_url VARCHAR(500),
    role VARCHAR(20) NOT NULL DEFAULT 'user',
    preferences TEXT DEFAULT '{}',
    last_login_at TIMESTAMP,
    create_by VARCHAR(36),
    create_by_name VARCHAR(100),
    create_time TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,
    update_by VARCHAR(36),
    update_by_name VARCHAR(100),
    update_time TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP
);
CREATE INDEX IF NOT EXISTS idx_users_username ON users(username);
CREATE INDEX IF NOT EXISTS idx_users_email ON users(email);

CREATE TABLE IF NOT EXISTS documents (
    id VARCHAR(36) PRIMARY KEY,
    user_id VARCHAR(36) NOT NULL REFERENCES users(id) ON DELETE CASCADE,
    name VARCHAR(255) NOT NULL,
    type VARCHAR(20) NOT NULL,
    status VARCHAR(20) NOT NULL DEFAULT 'pending',
    file_size BIGINT,
    file_url VARCHAR(500),
    thumbnail_url VARCHAR(500),
    parsed_text TEXT,
    parse_status VARCHAR(20),
    parse_progress INTEGER DEFAULT 0,
    parse_error TEXT,
    parse_started_at TIMESTAMP,
    parse_completed_at TIMESTAMP,
    metadata JSONB DEFAULT '{}',
    structured_status VARCHAR(20) DEFAULT 'pending',
    image_count INT DEFAULT 0,
    table_count INT DEFAULT 0,
    element_count INT DEFAULT 0,
    create_by VARCHAR(36),
    create_by_name VARCHAR(100),
    create_time TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,
    update_by VARCHAR(36),
    update_by_name VARCHAR(100),
    update_time TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP
);
CREATE INDEX IF NOT EXISTS idx_documents_user_id ON documents(user_id);
CREATE INDEX IF NOT EXISTS idx_documents_status ON documents(status);
CREATE INDEX IF NOT EXISTS idx_documents_type ON documents(type);
CREATE INDEX IF NOT EXISTS idx_documents_created_at ON documents(create_time DESC);
CREATE INDEX IF NOT EXISTS idx_documents_metadata ON documents USING GIN(metadata);

CREATE TABLE IF NOT EXISTS elements (
    id VARCHAR(36) PRIMARY KEY,
    document_id VARCHAR(36) NOT NULL REFERENCES documents(id) ON DELETE CASCADE,
    user_id VARCHAR(36) NOT NULL REFERENCES users(id) ON DELETE CASCADE,
    type VARCHAR(20) NOT NULL,
    label VARCHAR(100) NOT NULL,
    value TEXT NOT NULL,
    position JSONB,
    confidence DECIMAL(3,2),
    extraction_method VARCHAR(20),
    metadata JSONB DEFAULT '{}',
    create_by VARCHAR(36),
    create_by_name VARCHAR(100),
    create_time TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,
    update_by VARCHAR(36),
    update_by_name VARCHAR(100),
    update_time TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP
);
CREATE INDEX IF NOT EXISTS idx_elements_document_id ON elements(document_id);
CREATE INDEX IF NOT EXISTS idx_elements_user_id ON elements(user_id);
CREATE INDEX IF NOT EXISTS idx_elements_type ON elements(type);
CREATE INDEX IF NOT EXISTS idx_elements_position ON elements USING GIN(position);

CREATE TABLE IF NOT EXISTS annotations (
    id VARCHAR(36) PRIMARY KEY,
    document_id VARCHAR(36) NOT NULL REFERENCES documents(id) ON DELETE CASCADE,
    user_id VARCHAR(36) NOT NULL REFERENCES users(id) ON DELETE CASCADE,
    text TEXT NOT NULL,
    position JSONB NOT NULL,
    type VARCHAR(20) NOT NULL,
    suggestion TEXT,
    ai_generated BOOLEAN DEFAULT FALSE,
    confidence DECIMAL(3,2),
    status VARCHAR(20) DEFAULT 'pending',
    create_by VARCHAR(36),
    create_by_name VARCHAR(100),
    create_time TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,
    update_by VARCHAR(36),
    update_by_name VARCHAR(100),
    update_time TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP
);
CREATE INDEX IF NOT EXISTS idx_annotations_document_id ON annotations(document_id);
CREATE INDEX IF NOT EXISTS idx_annotations_user_id ON annotations(user_id);
CREATE INDEX IF NOT EXISTS idx_annotations_type ON annotations(type);
CREATE INDEX IF NOT EXISTS idx_annotations_status ON annotations(status);

CREATE TABLE IF NOT EXISTS graphs (
    id VARCHAR(36) PRIMARY KEY,
    document_id VARCHAR(36) NOT NULL REFERENCES documents(id) ON DELETE CASCADE,
    user_id VARCHAR(36) NOT NULL REFERENCES users(id) ON DELETE CASCADE,
    name VARCHAR(255) NOT NULL,
    nodes JSONB NOT NULL DEFAULT '[]',
    edges JSONB NOT NULL DEFAULT '[]',
    calculation_result JSONB,
    calculation_status VARCHAR(20),
    metadata JSONB DEFAULT '{}',
    create_by VARCHAR(36),
    create_by_name VARCHAR(100),
    create_time TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,
    update_by VARCHAR(36),
    update_by_name VARCHAR(100),
    update_time TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP
);
CREATE INDEX IF NOT EXISTS idx_graphs_document_id ON graphs(document_id);
CREATE INDEX IF NOT EXISTS idx_graphs_user_id ON graphs(user_id);
CREATE INDEX IF NOT EXISTS idx_graphs_nodes ON graphs USING GIN(nodes);
CREATE INDEX IF NOT EXISTS idx_graphs_edges ON graphs USING GIN(edges);

CREATE TABLE IF NOT EXISTS parse_tasks (
    id VARCHAR(36) PRIMARY KEY,
    document_id VARCHAR(36) NOT NULL REFERENCES documents(id) ON DELETE CASCADE,
    status VARCHAR(20) NOT NULL DEFAULT 'pending',
    progress INTEGER DEFAULT 0,
    current_step VARCHAR(100),
    error_message TEXT,
    options JSONB DEFAULT '{}',
    started_at TIMESTAMP,
    completed_at TIMESTAMP,
    parse_status VARCHAR(20) DEFAULT 'pending',
    parse_progress INTEGER DEFAULT 0,
    rag_status VARCHAR(20) DEFAULT 'pending',
    rag_progress INTEGER DEFAULT 0,
    structured_status VARCHAR(20) DEFAULT 'pending',
    structured_progress INTEGER DEFAULT 0,
    structured_element_count INTEGER,
    structured_image_count INTEGER,
    structured_table_count INTEGER,
    ner_status VARCHAR(20) DEFAULT 'pending',
    ner_progress INTEGER DEFAULT 0,
    ner_task_id VARCHAR(64),
    ner_entity_count INTEGER,
    ner_relation_count INTEGER,
    ner_message VARCHAR(255),
    graph_status VARCHAR(20) DEFAULT 'pending',
    graph_progress INTEGER DEFAULT 0,
    create_by VARCHAR(36),
    create_by_name VARCHAR(100),
    create_time TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,
    update_by VARCHAR(36),
    update_by_name VARCHAR(100),
    update_time TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP
);
CREATE INDEX IF NOT EXISTS idx_parse_tasks_document_id ON parse_tasks(document_id);
CREATE INDEX IF NOT EXISTS idx_parse_tasks_status ON parse_tasks(status);

CREATE TABLE IF NOT EXISTS sessions (
    id VARCHAR(36) PRIMARY KEY,
    user_id VARCHAR(36) NOT NULL REFERENCES users(id) ON DELETE CASCADE,
    token_hash VARCHAR(255) NOT NULL UNIQUE,
    refresh_token_hash VARCHAR(255) NOT NULL UNIQUE,
    expires_at TIMESTAMP NOT NULL,
    ip_address VARCHAR(45),
    user_agent TEXT,
    create_time TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,
    last_used_at TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP
);
CREATE INDEX IF NOT EXISTS idx_sessions_user_id ON sessions(user_id);
CREATE INDEX IF NOT EXISTS idx_sessions_token_hash ON sessions(token_hash);
CREATE INDEX IF NOT EXISTS idx_sessions_expires_at ON sessions(expires_at);

-- ============================================ 二、补充模块 ============================================

CREATE TABLE IF NOT EXISTS rules (
    id VARCHAR(36) PRIMARY KEY,
    user_id VARCHAR(36) NOT NULL REFERENCES users(id) ON DELETE CASCADE,
    name VARCHAR(255) NOT NULL,
    description TEXT,
    entry_node_id VARCHAR(36),
    exit_node_id VARCHAR(36),
    rule_chain JSONB NOT NULL DEFAULT '[]',
    status VARCHAR(20) DEFAULT 'active',
    metadata JSONB DEFAULT '{}',
    create_by VARCHAR(36),
    create_by_name VARCHAR(100),
    create_time TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,
    update_by VARCHAR(36),
    update_by_name VARCHAR(100),
    update_time TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP
);
CREATE INDEX IF NOT EXISTS idx_rules_user_id ON rules(user_id);
CREATE INDEX IF NOT EXISTS idx_rules_status ON rules(status);

CREATE TABLE IF NOT EXISTS data_sources (
    id VARCHAR(36) PRIMARY KEY,
    user_id VARCHAR(36) NOT NULL REFERENCES users(id) ON DELETE CASCADE,
    document_id VARCHAR(36) REFERENCES documents(id) ON DELETE SET NULL,
    name VARCHAR(255) NOT NULL,
    type VARCHAR(50) NOT NULL,
    source_type VARCHAR(50) NOT NULL DEFAULT 'manual',
    node_ids JSONB DEFAULT '{"refs": []}',
    config JSONB DEFAULT '{}',
    metadata JSONB DEFAULT '{}',
    value_type VARCHAR(20) DEFAULT 'text',
    aggregate_type VARCHAR(20) DEFAULT 'first',
    separator VARCHAR(50) DEFAULT '',
    create_by VARCHAR(36),
    create_by_name VARCHAR(100),
    create_time TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,
    update_by VARCHAR(36),
    update_by_name VARCHAR(100),
    update_time TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP
);
CREATE INDEX IF NOT EXISTS idx_data_sources_user_id ON data_sources(user_id);
CREATE INDEX IF NOT EXISTS idx_data_sources_document_id ON data_sources(document_id);
CREATE INDEX IF NOT EXISTS idx_data_sources_type ON data_sources(type);

CREATE TABLE IF NOT EXISTS text_storage (
    id VARCHAR(36) PRIMARY KEY,
    document_id VARCHAR(36) NOT NULL REFERENCES documents(id) ON DELETE CASCADE,
    file_path VARCHAR(500) NOT NULL,
    file_size BIGINT,
    checksum VARCHAR(64),
    create_by VARCHAR(36),
    create_by_name VARCHAR(100),
    create_time TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,
    update_by VARCHAR(36),
    update_by_name VARCHAR(100),
    update_time TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP
);
CREATE INDEX IF NOT EXISTS idx_text_storage_document_id ON text_storage(document_id);
CREATE UNIQUE INDEX IF NOT EXISTS idx_text_storage_document_unique ON text_storage(document_id);

-- ============================================ 三、RAG 模块 ============================================

CREATE TABLE IF NOT EXISTS text_chunks (
    id VARCHAR(36) PRIMARY KEY,
    document_id VARCHAR(36) NOT NULL,
    text_storage_id VARCHAR(36),
    chunk_index INTEGER NOT NULL,
    content TEXT NOT NULL,
    token_count INTEGER,
    metadata JSONB DEFAULT '{}',
    create_by VARCHAR(36),
    create_by_name VARCHAR(100),
    create_time TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,
    update_by VARCHAR(36),
    update_by_name VARCHAR(100),
    update_time TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP
);
CREATE INDEX IF NOT EXISTS idx_text_chunks_document_id ON text_chunks(document_id);
CREATE INDEX IF NOT EXISTS idx_text_chunks_text_storage_id ON text_chunks(text_storage_id);
CREATE UNIQUE INDEX IF NOT EXISTS idx_text_chunks_doc_index ON text_chunks(document_id, chunk_index);

CREATE TABLE IF NOT EXISTS vector_embeddings (
    id VARCHAR(36) PRIMARY KEY,
    chunk_id VARCHAR(36) NOT NULL REFERENCES text_chunks(id) ON DELETE CASCADE,
    embedding vector(768),
    model_name VARCHAR(100) DEFAULT 'nomic-embed-text',
    create_time TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP
);
CREATE INDEX IF NOT EXISTS idx_vector_embeddings_chunk_id ON vector_embeddings(chunk_id);
CREATE INDEX IF NOT EXISTS idx_vector_embeddings_model ON vector_embeddings(model_name);
CREATE INDEX IF NOT EXISTS idx_vector_embeddings_hnsw ON vector_embeddings USING hnsw (embedding vector_cosine_ops);

-- ============================================ 四、文档结构化模块 ============================================

CREATE TABLE IF NOT EXISTS document_blocks (
    id VARCHAR(64) PRIMARY KEY,
    document_id VARCHAR(64) NOT NULL,
    parent_id VARCHAR(64),
    children JSONB,
    block_index INTEGER NOT NULL,
    block_type VARCHAR(32) NOT NULL,
    elements JSONB,
    style JSONB,
    metadata JSONB,
    create_by VARCHAR(64),
    create_by_name VARCHAR(128),
    create_time TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
    update_by VARCHAR(64),
    update_by_name VARCHAR(128),
    update_time TIMESTAMP DEFAULT CURRENT_TIMESTAMP
);
CREATE INDEX IF NOT EXISTS idx_document_blocks_document_id ON document_blocks(document_id);
CREATE INDEX IF NOT EXISTS idx_document_blocks_parent_id ON document_blocks(parent_id);
CREATE INDEX IF NOT EXISTS idx_document_blocks_block_type ON document_blocks(block_type);
CREATE INDEX IF NOT EXISTS idx_document_blocks_elements_gin ON document_blocks USING GIN (elements jsonb_path_ops);

CREATE TABLE IF NOT EXISTS document_entities (
    id VARCHAR(64) PRIMARY KEY,
    document_id VARCHAR(64) NOT NULL,
    block_id VARCHAR(64),
    name VARCHAR(512) NOT NULL,
    entity_type VARCHAR(32) NOT NULL,
    value TEXT,
    block_char_start INTEGER,
    block_char_end INTEGER,
    global_char_start INTEGER,
    global_char_end INTEGER,
    anchor_before VARCHAR(100),
    anchor_after VARCHAR(100),
    source VARCHAR(16) DEFAULT 'auto',
    confidence DECIMAL(5,4),
    confirmed BOOLEAN DEFAULT FALSE,
    metadata JSONB,
    create_by VARCHAR(64),
    create_by_name VARCHAR(128),
    create_time TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
    update_by VARCHAR(64),
    update_by_name VARCHAR(128),
    update_time TIMESTAMP DEFAULT CURRENT_TIMESTAMP
);
CREATE INDEX IF NOT EXISTS idx_document_entities_document_id ON document_entities(document_id);
CREATE INDEX IF NOT EXISTS idx_document_entities_block_id ON document_entities(block_id);
CREATE INDEX IF NOT EXISTS idx_document_entities_type ON document_entities(entity_type);
CREATE INDEX IF NOT EXISTS idx_document_entities_name ON document_entities(name);
CREATE INDEX IF NOT EXISTS idx_document_entities_global_char ON document_entities(document_id, global_char_start, global_char_end);

CREATE TABLE IF NOT EXISTS document_elements (
    id VARCHAR(64) PRIMARY KEY,
    document_id VARCHAR(64) NOT NULL,
    element_index INT NOT NULL,
    element_type VARCHAR(32) NOT NULL,
    content TEXT,
    style JSONB,
    runs JSONB,
    image_url VARCHAR(500),
    image_path VARCHAR(500),
    image_alt VARCHAR(255),
    image_width INT,
    image_height INT,
    image_format VARCHAR(16),
    table_index INT,
    table_data JSONB,
    table_row_count INT,
    table_col_count INT,
    table_text TEXT,
    create_by VARCHAR(64),
    create_by_name VARCHAR(128),
    create_time TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
    update_by VARCHAR(64),
    update_by_name VARCHAR(128),
    update_time TIMESTAMP DEFAULT CURRENT_TIMESTAMP
);
CREATE INDEX IF NOT EXISTS idx_document_elements_document_id ON document_elements(document_id);
CREATE INDEX IF NOT EXISTS idx_document_elements_type ON document_elements(element_type);
CREATE INDEX IF NOT EXISTS idx_document_elements_order ON document_elements(document_id, element_index);

-- ============================================ 五、模板系统模块 ============================================

CREATE TABLE IF NOT EXISTS templates (
    id VARCHAR(36) PRIMARY KEY,
    user_id VARCHAR(36) NOT NULL REFERENCES users(id) ON DELETE CASCADE,
    name VARCHAR(255) NOT NULL,
    description TEXT,
    base_document_id VARCHAR(36) REFERENCES documents(id) ON DELETE SET NULL,
    status VARCHAR(32) DEFAULT 'draft',
    config JSONB DEFAULT '{}',
    is_public BOOLEAN DEFAULT FALSE,
    use_count INT DEFAULT 0,
    rating DECIMAL(2,1) DEFAULT 0.0,
    create_time TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
    update_time TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
    create_by VARCHAR(36),
    create_by_name VARCHAR(100),
    update_by VARCHAR(36),
    update_by_name VARCHAR(100)
);
CREATE INDEX IF NOT EXISTS idx_templates_user_id ON templates(user_id);
CREATE INDEX IF NOT EXISTS idx_templates_status ON templates(status);
CREATE INDEX IF NOT EXISTS idx_templates_is_public ON templates(is_public);
CREATE INDEX IF NOT EXISTS idx_templates_base_document ON templates(base_document_id);

CREATE TABLE IF NOT EXISTS source_files (
    id VARCHAR(36) PRIMARY KEY,
    template_id VARCHAR(36) NOT NULL REFERENCES templates(id) ON DELETE CASCADE,
    alias VARCHAR(100) NOT NULL,
    description TEXT,
    file_types JSONB DEFAULT '["pdf", "docx"]',
    required BOOLEAN DEFAULT TRUE,
    example_document_id VARCHAR(36) REFERENCES documents(id) ON DELETE SET NULL,
    display_order INT DEFAULT 0,
    create_time TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
    update_time TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
    CONSTRAINT uk_source_files_alias UNIQUE (template_id, alias)
);
CREATE INDEX IF NOT EXISTS idx_source_files_template ON source_files(template_id);

CREATE TABLE IF NOT EXISTS variables (
    id VARCHAR(36) PRIMARY KEY,
    template_id VARCHAR(36) NOT NULL REFERENCES templates(id) ON DELETE CASCADE,
    name VARCHAR(100) NOT NULL,
    display_name VARCHAR(200) NOT NULL,
    variable_group VARCHAR(100),
    category VARCHAR(32),
    location JSONB NOT NULL,
    example_value TEXT,
    value_type VARCHAR(32) DEFAULT 'text',
    source_file_alias VARCHAR(100),
    source_type VARCHAR(32) NOT NULL,
    source_config JSONB,
    extract_type VARCHAR(32),
    extract_config JSONB,
    display_order INT DEFAULT 0,
    create_time TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
    update_time TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
    CONSTRAINT uk_variables_name UNIQUE (template_id, name)
);
CREATE INDEX IF NOT EXISTS idx_variables_template ON variables(template_id);
CREATE INDEX IF NOT EXISTS idx_variables_source_alias ON variables(source_file_alias);
CREATE INDEX IF NOT EXISTS idx_variables_source_type ON variables(source_type);
CREATE INDEX IF NOT EXISTS idx_variables_category ON variables(category);

CREATE TABLE IF NOT EXISTS generations (
    id VARCHAR(36) PRIMARY KEY,
    template_id VARCHAR(36) NOT NULL REFERENCES templates(id) ON DELETE RESTRICT,
    user_id VARCHAR(36) NOT NULL REFERENCES users(id) ON DELETE CASCADE,
    name VARCHAR(255),
    source_file_map JSONB NOT NULL,
    variable_values JSONB,
    output_document_id VARCHAR(36) REFERENCES documents(id) ON DELETE SET NULL,
    output_file_path VARCHAR(500),
    status VARCHAR(32) DEFAULT 'pending',
    error_message TEXT,
    progress INT DEFAULT 0,
    create_time TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
    completed_at TIMESTAMP
);
CREATE INDEX IF NOT EXISTS idx_generations_template ON generations(template_id);
CREATE INDEX IF NOT EXISTS idx_generations_user ON generations(user_id);
CREATE INDEX IF NOT EXISTS idx_generations_status ON generations(status);
CREATE INDEX IF NOT EXISTS idx_generations_create_time ON generations(create_time DESC);

-- ============================================ 触发器 ============================================

CREATE OR REPLACE FUNCTION update_update_time_column()
RETURNS TRIGGER AS $$
BEGIN
    NEW.update_time = CURRENT_TIMESTAMP;
    RETURN NEW;
END;
$$ LANGUAGE plpgsql;

DO $$
DECLARE
    tbl TEXT;
    tables TEXT[] := ARRAY[
        'users', 'documents', 'elements', 'annotations', 'graphs',
        'parse_tasks', 'sessions', 'rules', 'data_sources', 'text_storage',
        'text_chunks', 'document_blocks', 'document_entities', 'document_elements',
        'templates', 'source_files', 'variables', 'generations'
    ];
BEGIN
    FOREACH tbl IN ARRAY tables LOOP
        EXECUTE format('DROP TRIGGER IF EXISTS trigger_%s_update_time ON %I', tbl, tbl);
        EXECUTE format('CREATE TRIGGER trigger_%s_update_time BEFORE UPDATE ON %I FOR EACH ROW EXECUTE FUNCTION update_update_time_column()', tbl, tbl);
    END LOOP;
END $$;

-- ============================================ 向量检索函数 ============================================

CREATE OR REPLACE FUNCTION search_similar_chunks(
    query_embedding vector(768),
    target_document_id VARCHAR(36),
    result_limit INTEGER DEFAULT 3
)
RETURNS TABLE (chunk_id VARCHAR(36), document_id VARCHAR(36), content TEXT, chunk_index INTEGER, similarity FLOAT) AS $$
BEGIN
    RETURN QUERY
    SELECT tc.id, tc.document_id, tc.content, tc.chunk_index,
           1 - (ve.embedding <=> query_embedding) AS similarity
    FROM text_chunks tc
    JOIN vector_embeddings ve ON tc.id = ve.chunk_id
    WHERE tc.document_id = target_document_id
    ORDER BY ve.embedding <=> query_embedding
    LIMIT result_limit;
END;
$$ LANGUAGE plpgsql;

CREATE OR REPLACE FUNCTION search_similar_chunks_global(
    query_embedding vector(768),
    result_limit INTEGER DEFAULT 5
)
RETURNS TABLE (chunk_id VARCHAR(36), document_id VARCHAR(36), content TEXT, chunk_index INTEGER, similarity FLOAT) AS $$
BEGIN
    RETURN QUERY
    SELECT tc.id, tc.document_id, tc.content, tc.chunk_index,
           1 - (ve.embedding <=> query_embedding) AS similarity
    FROM text_chunks tc
    JOIN vector_embeddings ve ON tc.id = ve.chunk_id
    ORDER BY ve.embedding <=> query_embedding
    LIMIT result_limit;
END;
$$ LANGUAGE plpgsql;

SELECT '灵越智报 v2.0 数据库初始化完成' AS result;