🎉 Initial commit: Document Server MVP

 Features implemented:
- FastAPI backend with JWT authentication
- PostgreSQL database with async SQLAlchemy
- HTML document viewer with smart highlighting
- Note system connected to highlights (1:1 relationship)
- Bookmark system for quick navigation
- Integrated search (documents + notes)
- Tag system for document organization
- Docker containerization with Nginx

🔧 Technical stack:
- Backend: FastAPI + PostgreSQL + Redis
- Frontend: Alpine.js + Tailwind CSS
- Authentication: JWT tokens
- File handling: HTML + PDF support
- Search: Full-text search with relevance scoring

📋 Core functionality:
- Text selection → Highlight creation
- Highlight → Note attachment
- Note management with search/filtering
- Bookmark creation at scroll positions
- Document upload with metadata
- User management (admin creates accounts)
This commit is contained in:
Hyungi Ahn
2025-08-21 16:09:17 +09:00
commit 3036b8f0fb
40 changed files with 6303 additions and 0 deletions

View File

@@ -0,0 +1,354 @@
"""
검색 API 라우터
"""
from fastapi import APIRouter, Depends, Query
from sqlalchemy.ext.asyncio import AsyncSession
from sqlalchemy import select, or_, and_, text
from sqlalchemy.orm import joinedload, selectinload
from typing import List, Optional, Dict, Any
from datetime import datetime
from src.core.database import get_db
from src.models.user import User
from src.models.document import Document, Tag
from src.models.highlight import Highlight
from src.models.note import Note
from src.api.dependencies import get_current_active_user
from pydantic import BaseModel
class SearchResult(BaseModel):
"""검색 결과"""
type: str # "document", "note", "highlight"
id: str
title: str
content: str
document_id: str
document_title: str
created_at: datetime
relevance_score: float = 0.0
highlight_info: Optional[Dict[str, Any]] = None
class Config:
from_attributes = True
class SearchResponse(BaseModel):
"""검색 응답"""
query: str
total_results: int
results: List[SearchResult]
facets: Dict[str, List[Dict[str, Any]]] = {}
router = APIRouter()
@router.get("/", response_model=SearchResponse)
async def search_all(
q: str = Query(..., description="검색어"),
type_filter: Optional[str] = Query(None, description="검색 타입 필터: document, note, highlight"),
document_id: Optional[str] = Query(None, description="특정 문서 내 검색"),
tag: Optional[str] = Query(None, description="태그 필터"),
skip: int = Query(0, ge=0),
limit: int = Query(50, ge=1, le=100),
current_user: User = Depends(get_current_active_user),
db: AsyncSession = Depends(get_db)
):
"""통합 검색 (문서 + 메모 + 하이라이트)"""
results = []
# 1. 문서 검색
if not type_filter or type_filter == "document":
document_results = await search_documents(q, document_id, tag, current_user, db)
results.extend(document_results)
# 2. 메모 검색
if not type_filter or type_filter == "note":
note_results = await search_notes(q, document_id, tag, current_user, db)
results.extend(note_results)
# 3. 하이라이트 검색
if not type_filter or type_filter == "highlight":
highlight_results = await search_highlights(q, document_id, current_user, db)
results.extend(highlight_results)
# 관련성 점수로 정렬
results.sort(key=lambda x: x.relevance_score, reverse=True)
# 페이지네이션
total_results = len(results)
paginated_results = results[skip:skip + limit]
# 패싯 정보 생성
facets = await generate_search_facets(results, current_user, db)
return SearchResponse(
query=q,
total_results=total_results,
results=paginated_results,
facets=facets
)
async def search_documents(
query: str,
document_id: Optional[str],
tag: Optional[str],
current_user: User,
db: AsyncSession
) -> List[SearchResult]:
"""문서 검색"""
query_obj = select(Document).options(
selectinload(Document.uploader),
selectinload(Document.tags)
)
# 권한 필터링
if not current_user.is_admin:
query_obj = query_obj.where(
or_(
Document.is_public == True,
Document.uploaded_by == current_user.id
)
)
# 특정 문서 필터
if document_id:
query_obj = query_obj.where(Document.id == document_id)
# 태그 필터
if tag:
query_obj = query_obj.join(Document.tags).where(Tag.name == tag)
# 텍스트 검색
search_condition = or_(
Document.title.ilike(f"%{query}%"),
Document.description.ilike(f"%{query}%")
)
query_obj = query_obj.where(search_condition)
result = await db.execute(query_obj)
documents = result.scalars().all()
search_results = []
for doc in documents:
# 관련성 점수 계산 (제목 매치가 더 높은 점수)
score = 0.0
if query.lower() in doc.title.lower():
score += 2.0
if doc.description and query.lower() in doc.description.lower():
score += 1.0
search_results.append(SearchResult(
type="document",
id=str(doc.id),
title=doc.title,
content=doc.description or "",
document_id=str(doc.id),
document_title=doc.title,
created_at=doc.created_at,
relevance_score=score
))
return search_results
async def search_notes(
query: str,
document_id: Optional[str],
tag: Optional[str],
current_user: User,
db: AsyncSession
) -> List[SearchResult]:
"""메모 검색"""
query_obj = (
select(Note)
.options(
joinedload(Note.highlight).joinedload(Highlight.document)
)
.join(Highlight)
.where(Highlight.user_id == current_user.id)
)
# 특정 문서 필터
if document_id:
query_obj = query_obj.where(Highlight.document_id == document_id)
# 태그 필터
if tag:
query_obj = query_obj.where(Note.tags.contains([tag]))
# 텍스트 검색 (메모 내용 + 하이라이트된 텍스트)
search_condition = or_(
Note.content.ilike(f"%{query}%"),
Highlight.selected_text.ilike(f"%{query}%")
)
query_obj = query_obj.where(search_condition)
result = await db.execute(query_obj)
notes = result.scalars().all()
search_results = []
for note in notes:
# 관련성 점수 계산
score = 0.0
if query.lower() in note.content.lower():
score += 2.0
if query.lower() in note.highlight.selected_text.lower():
score += 1.5
search_results.append(SearchResult(
type="note",
id=str(note.id),
title=f"메모: {note.highlight.selected_text[:50]}...",
content=note.content,
document_id=str(note.highlight.document.id),
document_title=note.highlight.document.title,
created_at=note.created_at,
relevance_score=score,
highlight_info={
"highlight_id": str(note.highlight.id),
"selected_text": note.highlight.selected_text,
"start_offset": note.highlight.start_offset,
"end_offset": note.highlight.end_offset
}
))
return search_results
async def search_highlights(
query: str,
document_id: Optional[str],
current_user: User,
db: AsyncSession
) -> List[SearchResult]:
"""하이라이트 검색"""
query_obj = (
select(Highlight)
.options(joinedload(Highlight.document))
.where(Highlight.user_id == current_user.id)
)
# 특정 문서 필터
if document_id:
query_obj = query_obj.where(Highlight.document_id == document_id)
# 텍스트 검색
query_obj = query_obj.where(Highlight.selected_text.ilike(f"%{query}%"))
result = await db.execute(query_obj)
highlights = result.scalars().all()
search_results = []
for highlight in highlights:
# 관련성 점수 계산
score = 1.0 if query.lower() in highlight.selected_text.lower() else 0.5
search_results.append(SearchResult(
type="highlight",
id=str(highlight.id),
title=f"하이라이트: {highlight.selected_text[:50]}...",
content=highlight.selected_text,
document_id=str(highlight.document.id),
document_title=highlight.document.title,
created_at=highlight.created_at,
relevance_score=score,
highlight_info={
"highlight_id": str(highlight.id),
"selected_text": highlight.selected_text,
"start_offset": highlight.start_offset,
"end_offset": highlight.end_offset,
"highlight_color": highlight.highlight_color
}
))
return search_results
async def generate_search_facets(
results: List[SearchResult],
current_user: User,
db: AsyncSession
) -> Dict[str, List[Dict[str, Any]]]:
"""검색 결과 패싯 생성"""
facets = {}
# 타입별 개수
type_counts = {}
for result in results:
type_counts[result.type] = type_counts.get(result.type, 0) + 1
facets["types"] = [
{"name": type_name, "count": count}
for type_name, count in type_counts.items()
]
# 문서별 개수
document_counts = {}
for result in results:
doc_title = result.document_title
document_counts[doc_title] = document_counts.get(doc_title, 0) + 1
facets["documents"] = [
{"name": doc_title, "count": count}
for doc_title, count in sorted(document_counts.items(), key=lambda x: x[1], reverse=True)[:10]
]
return facets
@router.get("/suggestions")
async def get_search_suggestions(
q: str = Query(..., min_length=2, description="검색어 (최소 2글자)"),
current_user: User = Depends(get_current_active_user),
db: AsyncSession = Depends(get_db)
):
"""검색어 자동완성 제안"""
suggestions = []
# 문서 제목에서 제안
doc_result = await db.execute(
select(Document.title)
.where(
and_(
Document.title.ilike(f"%{q}%"),
or_(
Document.is_public == True,
Document.uploaded_by == current_user.id
) if not current_user.is_admin else text("true")
)
)
.limit(5)
)
doc_titles = doc_result.scalars().all()
suggestions.extend([{"text": title, "type": "document"} for title in doc_titles])
# 태그에서 제안
tag_result = await db.execute(
select(Tag.name)
.where(Tag.name.ilike(f"%{q}%"))
.limit(5)
)
tag_names = tag_result.scalars().all()
suggestions.extend([{"text": name, "type": "tag"} for name in tag_names])
# 메모 태그에서 제안
note_result = await db.execute(
select(Note.tags)
.join(Highlight)
.where(Highlight.user_id == current_user.id)
)
notes = note_result.scalars().all()
note_tags = set()
for note in notes:
if note and isinstance(note, list):
for tag in note:
if q.lower() in tag.lower():
note_tags.add(tag)
suggestions.extend([{"text": tag, "type": "note_tag"} for tag in list(note_tags)[:5]])
return {"suggestions": suggestions[:10]}