Files
hyungi_document_server/migrations/001_initial_schema.sql
Hyungi Ahn 131dbd7b7c feat: scaffold v2 project structure with Docker, FastAPI, and config
동작하는 최소 코드 수준의 v2 스캐폴딩:

- docker-compose.yml: postgres, fastapi, kordoc, frontend, caddy
- app/: FastAPI 백엔드 (main, core, models, ai, prompts)
- services/kordoc/: Node.js 문서 파싱 마이크로서비스
- gpu-server/: AI Gateway + GPU docker-compose
- frontend/: SvelteKit 기본 구조
- migrations/: PostgreSQL 초기 스키마 (documents, tasks, processing_queue)
- tests/: pytest conftest 기본 설정
- config.yaml, Caddyfile, credentials.env.example 갱신

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-04-02 10:20:15 +09:00

107 lines
3.5 KiB
SQL

-- hyungi_Document_Server 초기 스키마
-- PostgreSQL 16 + pgvector + pg_trgm
CREATE EXTENSION IF NOT EXISTS vector;
CREATE EXTENSION IF NOT EXISTS pg_trgm;
-- ENUM 타입
CREATE TYPE doc_type AS ENUM ('immutable', 'editable', 'note');
CREATE TYPE source_channel AS ENUM (
'law_monitor', 'devonagent', 'email', 'web_clip',
'tksafety', 'inbox_route', 'manual', 'drive_sync'
);
CREATE TYPE data_origin AS ENUM ('work', 'external');
CREATE TYPE process_stage AS ENUM ('extract', 'classify', 'embed');
CREATE TYPE process_status AS ENUM ('pending', 'processing', 'completed', 'failed');
-- documents 테이블
CREATE TABLE documents (
id BIGSERIAL PRIMARY KEY,
-- 1계층: 원본 파일 참조
file_path TEXT NOT NULL UNIQUE,
file_hash CHAR(64) NOT NULL,
file_format VARCHAR(20) NOT NULL,
file_size BIGINT,
file_type doc_type NOT NULL DEFAULT 'immutable',
import_source TEXT,
-- 2계층: 텍스트 추출
extracted_text TEXT,
extracted_at TIMESTAMPTZ,
extractor_version VARCHAR(50),
-- 2계층: AI 가공
ai_summary TEXT,
ai_tags JSONB DEFAULT '[]',
ai_domain VARCHAR(100),
ai_sub_group VARCHAR(100),
ai_model_version VARCHAR(50),
ai_processed_at TIMESTAMPTZ,
-- 3계층: 벡터 임베딩
embedding vector(768),
embed_model_version VARCHAR(50),
embedded_at TIMESTAMPTZ,
-- 메타데이터
source_channel source_channel,
data_origin data_origin,
title TEXT,
-- 타임스탬프
created_at TIMESTAMPTZ DEFAULT NOW(),
updated_at TIMESTAMPTZ DEFAULT NOW()
);
-- 전문검색 인덱스
CREATE INDEX idx_documents_fts ON documents
USING GIN (to_tsvector('simple', coalesce(title, '') || ' ' || coalesce(extracted_text, '')));
-- 트리그램 인덱스 (한국어 부분 매칭)
CREATE INDEX idx_documents_trgm ON documents
USING GIN ((coalesce(title, '') || ' ' || coalesce(extracted_text, '')) gin_trgm_ops);
-- 해시 기반 중복 검색
CREATE INDEX idx_documents_hash ON documents (file_hash);
-- 재가공 대상 필터링
CREATE INDEX idx_documents_ai_version ON documents (ai_model_version);
CREATE INDEX idx_documents_extractor_version ON documents (extractor_version);
CREATE INDEX idx_documents_embed_version ON documents (embed_model_version);
-- tasks 테이블 (CalDAV 캐시)
CREATE TABLE tasks (
id BIGSERIAL PRIMARY KEY,
caldav_uid TEXT UNIQUE,
title TEXT NOT NULL,
description TEXT,
due_date TIMESTAMPTZ,
priority SMALLINT DEFAULT 0,
completed BOOLEAN DEFAULT FALSE,
completed_at TIMESTAMPTZ,
document_id BIGINT REFERENCES documents(id),
source VARCHAR(50),
created_at TIMESTAMPTZ DEFAULT NOW(),
updated_at TIMESTAMPTZ DEFAULT NOW()
);
-- processing_queue 테이블 (비동기 가공 큐)
CREATE TABLE processing_queue (
id BIGSERIAL PRIMARY KEY,
document_id BIGINT REFERENCES documents(id) NOT NULL,
stage process_stage NOT NULL,
status process_status DEFAULT 'pending',
attempts SMALLINT DEFAULT 0,
max_attempts SMALLINT DEFAULT 3,
error_message TEXT,
created_at TIMESTAMPTZ DEFAULT NOW(),
started_at TIMESTAMPTZ,
completed_at TIMESTAMPTZ,
UNIQUE (document_id, stage, status)
);
CREATE INDEX idx_queue_pending ON processing_queue (stage, status)
WHERE status = 'pending';