red-future/rag
4
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

refactor: 重构文档向量相关代码结构

Browse Source
This commit is contained in:
曲皓东
2026-04-10 13:12:19 +08:00
parent a7b8713e26
commit 94df015aa9
30 changed files with 335 additions and 506 deletions
Download Patch File Download Diff File Expand all files Collapse all files

64
update.sql
View File

@@ -211,9 +211,9 @@ COMMENT ON COLUMN rag_knowledge_task.remark IS '备注';
-- 向量数据集索引表
CREATE TABLE IF NOT EXISTS rag_vector_dataset_index (
-- 基础字段
id BIGINT PRIMARY KEY, -- 主键ID（非自增）
tenant_id BIGINT NOT NULL DEFAULT 0, -- 租户ID int8
creator VARCHAR(64) NOT NULL,
id BIGINT PRIMARY KEY, -- 主键ID（非自增）
tenant_id BIGINT NOT NULL DEFAULT 0, -- 租户ID int8
creator VARCHAR(64) NOT NULL,
created_at TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,
updater VARCHAR(64) NOT NULL,
updated_at TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,
@@ -260,16 +260,16 @@ COMMENT ON COLUMN rag_vector_dataset_index.description IS '描述';
--------------------pgsql创建rag_vector_dataset_index表语句---------------------------
--------------------pgsql创建rag_vector_document_chunk表语句---------------------------
--------------------pgsql创建rag_vector_document_vector表语句---------------------------
CREATE EXTENSION IF NOT EXISTS vector;
-- 文档分块向量表
CREATE TABLE IF NOT EXISTS rag_vector_document_chunk (
CREATE TABLE IF NOT EXISTS rag_vector_document_vector (
-- 基础字段
id BIGINT PRIMARY KEY, -- 主键ID（非自增）
tenant_id BIGINT NOT NULL DEFAULT 0, -- 租户ID int8
creator VARCHAR(64) NOT NULL,
id BIGINT PRIMARY KEY, -- 主键ID（非自增）
tenant_id BIGINT NOT NULL DEFAULT 0, -- 租户ID int8
creator VARCHAR(64) NOT NULL,
created_at TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,
updater VARCHAR(64) NOT NULL,
updated_at TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,
@@ -292,30 +292,30 @@ CREATE TABLE IF NOT EXISTS rag_vector_document_chunk (
);
-- 索引
CREATE INDEX idx_chunk_tenant_id ON rag_vector_document_chunk(tenant_id);
CREATE INDEX idx_chunk_dataset_id ON rag_vector_document_chunk(dataset_id);
CREATE INDEX idx_chunk_document_id ON rag_vector_document_chunk(document_id);
CREATE INDEX idx_chunk_content_hash ON rag_vector_document_chunk(content_hash);
CREATE INDEX idx_chunk_status ON rag_vector_document_chunk(status);
CREATE INDEX idx_chunk_vector_status ON rag_vector_document_chunk(vector_status);
CREATE INDEX idx_vector_tenant_id ON rag_vector_document_vector(tenant_id);
CREATE INDEX idx_vector_dataset_id ON rag_vector_document_vector(dataset_id);
CREATE INDEX idx_vector_document_id ON rag_vector_document_vector(document_id);
CREATE INDEX idx_vector_content_hash ON rag_vector_document_vector(content_hash);
CREATE INDEX idx_vector_status ON rag_vector_document_vector(status);
CREATE INDEX idx_vector_vector_status ON rag_vector_document_vector(vector_status);
-- 注释
COMMENT ON TABLE rag_vector_document_chunk IS '文档分块向量表';
COMMENT ON COLUMN rag_vector_document_chunk.id IS '主键ID（非自增）';
COMMENT ON COLUMN rag_vector_document_chunk.tenant_id IS '租户ID';
COMMENT ON COLUMN rag_vector_document_chunk.creator IS '创建人';
COMMENT ON COLUMN rag_vector_document_chunk.created_at IS '创建时间';
COMMENT ON COLUMN rag_vector_document_chunk.updater IS '更新人';
COMMENT ON COLUMN rag_vector_document_chunk.updated_at IS '更新时间';
COMMENT ON COLUMN rag_vector_document_chunk.deleted_at IS '删除时间（软删）';
COMMENT ON COLUMN rag_vector_document_chunk.status IS '状态';
COMMENT ON COLUMN rag_vector_document_chunk.vector_status IS '向量生成状态';
COMMENT ON COLUMN rag_vector_document_chunk.dataset_id IS '数据集ID';
COMMENT ON COLUMN rag_vector_document_chunk.document_id IS '文档ID';
COMMENT ON COLUMN rag_vector_document_chunk.content IS '分块内容';
COMMENT ON COLUMN rag_vector_document_chunk.content_hash IS '内容哈希';
COMMENT ON COLUMN rag_vector_document_chunk.chunk_index IS '分块序号';
COMMENT ON COLUMN rag_vector_document_chunk.vector IS '向量数据';
COMMENT ON COLUMN rag_vector_document_chunk.metadata IS '扩展元数据';
COMMENT ON TABLE rag_vector_document_vector IS '文档分块向量表';
COMMENT ON COLUMN rag_vector_document_vector.id IS '主键ID（非自增）';
COMMENT ON COLUMN rag_vector_document_vector.tenant_id IS '租户ID';
COMMENT ON COLUMN rag_vector_document_vector.creator IS '创建人';
COMMENT ON COLUMN rag_vector_document_vector.created_at IS '创建时间';
COMMENT ON COLUMN rag_vector_document_vector.updater IS '更新人';
COMMENT ON COLUMN rag_vector_document_vector.updated_at IS '更新时间';
COMMENT ON COLUMN rag_vector_document_vector.deleted_at IS '删除时间（软删）';
COMMENT ON COLUMN rag_vector_document_vector.status IS '状态';
COMMENT ON COLUMN rag_vector_document_vector.vector_status IS '向量生成状态';
COMMENT ON COLUMN rag_vector_document_vector.dataset_id IS '数据集ID';
COMMENT ON COLUMN rag_vector_document_vector.document_id IS '文档ID';
COMMENT ON COLUMN rag_vector_document_vector.content IS '分块内容';
COMMENT ON COLUMN rag_vector_document_vector.content_hash IS '内容哈希';
COMMENT ON COLUMN rag_vector_document_vector.chunk_index IS '分块序号';
COMMENT ON COLUMN rag_vector_document_vector.vector IS '向量数据';
COMMENT ON COLUMN rag_vector_document_vector.metadata IS '扩展元数据';
--------------------pgsql创建rag_vector_document_chunk表语句---------------------------
--------------------pgsql创建rag_vector_document_vector表语句---------------------------