wren
188 lines
8.8 KiB
PL/PgSQL
188 lines
8.8 KiB
PL/PgSQL
BEGIN;
|
|
|
|
CREATE TABLE IF NOT EXISTS rag_dataset (
|
|
id BIGSERIAL PRIMARY KEY,
|
|
name VARCHAR(255) NOT NULL,
|
|
description TEXT NOT NULL DEFAULT '',
|
|
area VARCHAR(50) NOT NULL DEFAULT '',
|
|
is_public BOOLEAN NOT NULL DEFAULT FALSE,
|
|
is_default BOOLEAN NOT NULL DEFAULT FALSE,
|
|
collection_name VARCHAR(100) NOT NULL UNIQUE,
|
|
embedding_model VARCHAR(100) NOT NULL DEFAULT 'text-embedding-v4',
|
|
embedding_dim INTEGER NOT NULL DEFAULT 1024,
|
|
chunk_max_size INTEGER NOT NULL DEFAULT 800,
|
|
chunk_min_size INTEGER NOT NULL DEFAULT 20,
|
|
document_count INTEGER NOT NULL DEFAULT 0,
|
|
total_chunks INTEGER NOT NULL DEFAULT 0,
|
|
retrieval_model JSONB NOT NULL DEFAULT '{}'::jsonb,
|
|
sort_order INTEGER NOT NULL DEFAULT 0,
|
|
status SMALLINT NOT NULL DEFAULT 1,
|
|
created_by BIGINT,
|
|
created_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
|
|
updated_by BIGINT,
|
|
updated_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
|
|
deleted_at TIMESTAMPTZ
|
|
);
|
|
|
|
CREATE TABLE IF NOT EXISTS rag_document (
|
|
id BIGSERIAL PRIMARY KEY,
|
|
dataset_id BIGINT NOT NULL REFERENCES rag_dataset(id),
|
|
filename VARCHAR(500) NOT NULL,
|
|
original_name VARCHAR(500) NOT NULL,
|
|
minio_path VARCHAR(1000) NOT NULL,
|
|
file_type VARCHAR(20) NOT NULL,
|
|
file_size BIGINT NOT NULL DEFAULT 0,
|
|
chunk_count INTEGER NOT NULL DEFAULT 0,
|
|
indexing_status VARCHAR(20) NOT NULL DEFAULT 'pending',
|
|
indexing_error TEXT,
|
|
indexing_started_at TIMESTAMPTZ,
|
|
indexing_completed_at TIMESTAMPTZ,
|
|
enabled BOOLEAN NOT NULL DEFAULT TRUE,
|
|
hit_count INTEGER NOT NULL DEFAULT 0,
|
|
created_by BIGINT,
|
|
created_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
|
|
updated_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
|
|
deleted_at TIMESTAMPTZ
|
|
);
|
|
|
|
CREATE TABLE IF NOT EXISTS rag_chat_app (
|
|
id BIGSERIAL PRIMARY KEY,
|
|
name VARCHAR(255) NOT NULL,
|
|
description TEXT NOT NULL DEFAULT '',
|
|
area VARCHAR(50) NOT NULL DEFAULT '',
|
|
dataset_id BIGINT REFERENCES rag_dataset(id),
|
|
system_prompt TEXT NOT NULL DEFAULT '',
|
|
llm_model VARCHAR(100) NOT NULL DEFAULT '',
|
|
temperature REAL NOT NULL DEFAULT 0.3,
|
|
max_tokens INTEGER NOT NULL DEFAULT 2048,
|
|
opening_statement TEXT NOT NULL DEFAULT '',
|
|
suggested_questions TEXT NOT NULL DEFAULT '[]',
|
|
is_default BOOLEAN NOT NULL DEFAULT FALSE,
|
|
sort_order INTEGER NOT NULL DEFAULT 0,
|
|
status SMALLINT NOT NULL DEFAULT 1,
|
|
created_by BIGINT,
|
|
created_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
|
|
updated_by BIGINT,
|
|
updated_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
|
|
deleted_at TIMESTAMPTZ
|
|
);
|
|
|
|
CREATE TABLE IF NOT EXISTS rag_conversation (
|
|
id BIGSERIAL PRIMARY KEY,
|
|
conversation_id VARCHAR(100) NOT NULL UNIQUE,
|
|
user_id BIGINT NOT NULL,
|
|
app_id BIGINT REFERENCES rag_chat_app(id),
|
|
name VARCHAR(500) NOT NULL DEFAULT '新对话',
|
|
introduction TEXT NOT NULL DEFAULT '',
|
|
created_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
|
|
updated_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
|
|
deleted_at TIMESTAMPTZ
|
|
);
|
|
|
|
CREATE TABLE IF NOT EXISTS rag_message (
|
|
id BIGSERIAL PRIMARY KEY,
|
|
message_id VARCHAR(100) NOT NULL UNIQUE,
|
|
conversation_id VARCHAR(100) NOT NULL REFERENCES rag_conversation(conversation_id),
|
|
role VARCHAR(20) NOT NULL,
|
|
content TEXT NOT NULL DEFAULT '',
|
|
sources JSONB NOT NULL DEFAULT '[]'::jsonb,
|
|
metadata JSONB NOT NULL DEFAULT '{}'::jsonb,
|
|
feedback VARCHAR(20),
|
|
created_at TIMESTAMPTZ NOT NULL DEFAULT NOW()
|
|
);
|
|
|
|
ALTER TABLE rag_dataset
|
|
ADD COLUMN IF NOT EXISTS retrieval_model JSONB NOT NULL DEFAULT '{}'::jsonb,
|
|
ADD COLUMN IF NOT EXISTS embedding_model VARCHAR(100) NOT NULL DEFAULT 'text-embedding-v4',
|
|
ADD COLUMN IF NOT EXISTS embedding_dim INTEGER NOT NULL DEFAULT 1024,
|
|
ADD COLUMN IF NOT EXISTS chunk_max_size INTEGER NOT NULL DEFAULT 800,
|
|
ADD COLUMN IF NOT EXISTS chunk_min_size INTEGER NOT NULL DEFAULT 20,
|
|
ADD COLUMN IF NOT EXISTS document_count INTEGER NOT NULL DEFAULT 0,
|
|
ADD COLUMN IF NOT EXISTS total_chunks INTEGER NOT NULL DEFAULT 0,
|
|
ADD COLUMN IF NOT EXISTS sort_order INTEGER NOT NULL DEFAULT 0,
|
|
ADD COLUMN IF NOT EXISTS status SMALLINT NOT NULL DEFAULT 1,
|
|
ADD COLUMN IF NOT EXISTS updated_by BIGINT,
|
|
ADD COLUMN IF NOT EXISTS updated_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
|
|
ADD COLUMN IF NOT EXISTS deleted_at TIMESTAMPTZ;
|
|
|
|
ALTER TABLE rag_document
|
|
ADD COLUMN IF NOT EXISTS enabled BOOLEAN NOT NULL DEFAULT TRUE,
|
|
ADD COLUMN IF NOT EXISTS hit_count INTEGER NOT NULL DEFAULT 0,
|
|
ADD COLUMN IF NOT EXISTS indexing_started_at TIMESTAMPTZ,
|
|
ADD COLUMN IF NOT EXISTS indexing_completed_at TIMESTAMPTZ,
|
|
ADD COLUMN IF NOT EXISTS updated_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
|
|
ADD COLUMN IF NOT EXISTS deleted_at TIMESTAMPTZ;
|
|
|
|
ALTER TABLE rag_chat_app
|
|
ADD COLUMN IF NOT EXISTS dataset_id BIGINT REFERENCES rag_dataset(id),
|
|
ADD COLUMN IF NOT EXISTS system_prompt TEXT NOT NULL DEFAULT '',
|
|
ADD COLUMN IF NOT EXISTS llm_model VARCHAR(100) NOT NULL DEFAULT '',
|
|
ADD COLUMN IF NOT EXISTS temperature REAL NOT NULL DEFAULT 0.3,
|
|
ADD COLUMN IF NOT EXISTS max_tokens INTEGER NOT NULL DEFAULT 2048,
|
|
ADD COLUMN IF NOT EXISTS opening_statement TEXT NOT NULL DEFAULT '',
|
|
ADD COLUMN IF NOT EXISTS suggested_questions TEXT NOT NULL DEFAULT '[]',
|
|
ADD COLUMN IF NOT EXISTS is_default BOOLEAN NOT NULL DEFAULT FALSE,
|
|
ADD COLUMN IF NOT EXISTS sort_order INTEGER NOT NULL DEFAULT 0,
|
|
ADD COLUMN IF NOT EXISTS status SMALLINT NOT NULL DEFAULT 1,
|
|
ADD COLUMN IF NOT EXISTS updated_by BIGINT,
|
|
ADD COLUMN IF NOT EXISTS updated_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
|
|
ADD COLUMN IF NOT EXISTS deleted_at TIMESTAMPTZ;
|
|
|
|
ALTER TABLE rag_conversation
|
|
ADD COLUMN IF NOT EXISTS app_id BIGINT REFERENCES rag_chat_app(id),
|
|
ADD COLUMN IF NOT EXISTS introduction TEXT NOT NULL DEFAULT '',
|
|
ADD COLUMN IF NOT EXISTS updated_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
|
|
ADD COLUMN IF NOT EXISTS deleted_at TIMESTAMPTZ;
|
|
|
|
ALTER TABLE rag_message
|
|
ADD COLUMN IF NOT EXISTS sources JSONB NOT NULL DEFAULT '[]'::jsonb,
|
|
ADD COLUMN IF NOT EXISTS metadata JSONB NOT NULL DEFAULT '{}'::jsonb,
|
|
ADD COLUMN IF NOT EXISTS feedback VARCHAR(20),
|
|
ADD COLUMN IF NOT EXISTS created_at TIMESTAMPTZ NOT NULL DEFAULT NOW();
|
|
|
|
UPDATE rag_dataset SET retrieval_model = '{}'::jsonb WHERE retrieval_model IS NULL;
|
|
UPDATE rag_document SET enabled = TRUE WHERE enabled IS NULL;
|
|
UPDATE rag_document SET hit_count = 0 WHERE hit_count IS NULL;
|
|
UPDATE rag_chat_app SET suggested_questions = '[]' WHERE suggested_questions IS NULL;
|
|
UPDATE rag_message SET sources = '[]'::jsonb WHERE sources IS NULL;
|
|
UPDATE rag_message SET metadata = '{}'::jsonb WHERE metadata IS NULL;
|
|
|
|
CREATE UNIQUE INDEX IF NOT EXISTS uq_rag_dataset_collection_name ON rag_dataset(collection_name);
|
|
CREATE INDEX IF NOT EXISTS idx_rag_dataset_area ON rag_dataset(area) WHERE deleted_at IS NULL;
|
|
CREATE INDEX IF NOT EXISTS idx_rag_dataset_status ON rag_dataset(status) WHERE deleted_at IS NULL;
|
|
CREATE INDEX IF NOT EXISTS idx_rag_document_dataset ON rag_document(dataset_id) WHERE deleted_at IS NULL;
|
|
CREATE INDEX IF NOT EXISTS idx_rag_document_status ON rag_document(indexing_status) WHERE deleted_at IS NULL;
|
|
CREATE INDEX IF NOT EXISTS idx_rag_document_original_name ON rag_document(dataset_id, original_name) WHERE deleted_at IS NULL;
|
|
CREATE INDEX IF NOT EXISTS idx_rag_chat_app_area ON rag_chat_app(area) WHERE deleted_at IS NULL;
|
|
CREATE INDEX IF NOT EXISTS idx_rag_chat_app_status ON rag_chat_app(status) WHERE deleted_at IS NULL;
|
|
CREATE INDEX IF NOT EXISTS idx_rag_conversation_user ON rag_conversation(user_id) WHERE deleted_at IS NULL;
|
|
CREATE INDEX IF NOT EXISTS idx_rag_conversation_app ON rag_conversation(app_id) WHERE deleted_at IS NULL;
|
|
CREATE INDEX IF NOT EXISTS idx_rag_message_conversation ON rag_message(conversation_id, created_at);
|
|
|
|
CREATE OR REPLACE FUNCTION update_rag_updated_at()
|
|
RETURNS TRIGGER AS $$
|
|
BEGIN
|
|
NEW.updated_at = NOW();
|
|
RETURN NEW;
|
|
END;
|
|
$$ LANGUAGE plpgsql;
|
|
|
|
DO $$
|
|
DECLARE
|
|
t TEXT;
|
|
BEGIN
|
|
FOREACH t IN ARRAY ARRAY['rag_dataset', 'rag_document', 'rag_chat_app', 'rag_conversation']
|
|
LOOP
|
|
EXECUTE format('DROP TRIGGER IF EXISTS trg_%s_updated_at ON %I', t, t);
|
|
EXECUTE format(
|
|
'CREATE TRIGGER trg_%s_updated_at
|
|
BEFORE UPDATE ON %I
|
|
FOR EACH ROW EXECUTE FUNCTION update_rag_updated_at()',
|
|
t, t
|
|
);
|
|
END LOOP;
|
|
END;
|
|
$$;
|
|
|
|
COMMIT;
|