mirror of
https://github.com/handsomezhuzhu/QQuiz.git
synced 2026-02-20 20:10:14 +00:00
🎉 Initial commit: QQuiz - 智能刷题与题库管理平台
## 功能特性 ✅ **核心功能** - 多文件上传与智能去重（基于 content_hash） - 异步文档解析（支持 TXT/PDF/DOCX/XLSX） - AI 智能题目提取与评分（OpenAI/Anthropic/Qwen） - 断点续做与进度管理 - 自动错题本收集 ✅ **技术栈** - Backend: FastAPI + SQLAlchemy 2.0 + PostgreSQL - Frontend: React 18 + Vite + Tailwind CSS - Deployment: Docker Compose ✅ **项目结构** - 53 个文件 - 完整的前后端分离架构 - Docker/源码双模部署支持 🚀 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
This commit is contained in:
13
backend/services/__init__.py
Normal file
13
backend/services/__init__.py
Normal file
@@ -0,0 +1,13 @@
|
||||
"""
|
||||
Services package
|
||||
"""
|
||||
from .auth_service import get_current_user, get_current_admin_user
|
||||
from .llm_service import llm_service
|
||||
from .document_parser import document_parser
|
||||
|
||||
__all__ = [
|
||||
"get_current_user",
|
||||
"get_current_admin_user",
|
||||
"llm_service",
|
||||
"document_parser"
|
||||
]
|
||||
78
backend/services/auth_service.py
Normal file
78
backend/services/auth_service.py
Normal file
@@ -0,0 +1,78 @@
|
||||
"""
|
||||
Authentication Service
|
||||
"""
|
||||
from fastapi import Depends, HTTPException, status
|
||||
from fastapi.security import HTTPBearer, HTTPAuthorizationCredentials
|
||||
from sqlalchemy.ext.asyncio import AsyncSession
|
||||
from sqlalchemy import select
|
||||
from typing import Optional
|
||||
|
||||
from models import User
|
||||
from database import get_db
|
||||
from utils import decode_access_token
|
||||
|
||||
security = HTTPBearer()
|
||||
|
||||
|
||||
async def get_current_user(
|
||||
credentials: HTTPAuthorizationCredentials = Depends(security),
|
||||
db: AsyncSession = Depends(get_db)
|
||||
) -> User:
|
||||
"""
|
||||
Get current authenticated user from JWT token.
|
||||
"""
|
||||
token = credentials.credentials
|
||||
credentials_exception = HTTPException(
|
||||
status_code=status.HTTP_401_UNAUTHORIZED,
|
||||
detail="Could not validate credentials",
|
||||
headers={"WWW-Authenticate": "Bearer"},
|
||||
)
|
||||
|
||||
# Decode token
|
||||
payload = decode_access_token(token)
|
||||
if payload is None:
|
||||
raise credentials_exception
|
||||
|
||||
user_id: int = payload.get("sub")
|
||||
if user_id is None:
|
||||
raise credentials_exception
|
||||
|
||||
# Get user from database
|
||||
result = await db.execute(select(User).where(User.id == user_id))
|
||||
user = result.scalar_one_or_none()
|
||||
|
||||
if user is None:
|
||||
raise credentials_exception
|
||||
|
||||
return user
|
||||
|
||||
|
||||
async def get_current_admin_user(
|
||||
current_user: User = Depends(get_current_user)
|
||||
) -> User:
|
||||
"""
|
||||
Get current user and verify admin permissions.
|
||||
"""
|
||||
if not current_user.is_admin:
|
||||
raise HTTPException(
|
||||
status_code=status.HTTP_403_FORBIDDEN,
|
||||
detail="Not enough permissions"
|
||||
)
|
||||
return current_user
|
||||
|
||||
|
||||
async def get_optional_user(
|
||||
credentials: Optional[HTTPAuthorizationCredentials] = Depends(security),
|
||||
db: AsyncSession = Depends(get_db)
|
||||
) -> Optional[User]:
|
||||
"""
|
||||
Get current user if token is provided, otherwise return None.
|
||||
Useful for endpoints that work for both authenticated and anonymous users.
|
||||
"""
|
||||
if credentials is None:
|
||||
return None
|
||||
|
||||
try:
|
||||
return await get_current_user(credentials, db)
|
||||
except HTTPException:
|
||||
return None
|
||||
121
backend/services/document_parser.py
Normal file
121
backend/services/document_parser.py
Normal file
@@ -0,0 +1,121 @@
|
||||
"""
|
||||
Document Parser Service
|
||||
Supports: TXT, PDF, DOCX, XLSX
|
||||
"""
|
||||
import io
|
||||
from typing import Optional
|
||||
import PyPDF2
|
||||
from docx import Document
|
||||
import openpyxl
|
||||
|
||||
|
||||
class DocumentParser:
|
||||
"""Parse various document formats to extract text content"""
|
||||
|
||||
@staticmethod
|
||||
async def parse_txt(file_content: bytes) -> str:
|
||||
"""Parse TXT file"""
|
||||
try:
|
||||
return file_content.decode('utf-8')
|
||||
except UnicodeDecodeError:
|
||||
try:
|
||||
return file_content.decode('gbk')
|
||||
except:
|
||||
return file_content.decode('utf-8', errors='ignore')
|
||||
|
||||
@staticmethod
|
||||
async def parse_pdf(file_content: bytes) -> str:
|
||||
"""Parse PDF file"""
|
||||
try:
|
||||
pdf_file = io.BytesIO(file_content)
|
||||
pdf_reader = PyPDF2.PdfReader(pdf_file)
|
||||
|
||||
text_content = []
|
||||
for page in pdf_reader.pages:
|
||||
text = page.extract_text()
|
||||
if text:
|
||||
text_content.append(text)
|
||||
|
||||
return '\n\n'.join(text_content)
|
||||
except Exception as e:
|
||||
raise Exception(f"Failed to parse PDF: {str(e)}")
|
||||
|
||||
@staticmethod
|
||||
async def parse_docx(file_content: bytes) -> str:
|
||||
"""Parse DOCX file"""
|
||||
try:
|
||||
docx_file = io.BytesIO(file_content)
|
||||
doc = Document(docx_file)
|
||||
|
||||
text_content = []
|
||||
for paragraph in doc.paragraphs:
|
||||
if paragraph.text.strip():
|
||||
text_content.append(paragraph.text)
|
||||
|
||||
# Also extract text from tables
|
||||
for table in doc.tables:
|
||||
for row in table.rows:
|
||||
row_text = ' | '.join(cell.text.strip() for cell in row.cells)
|
||||
if row_text.strip():
|
||||
text_content.append(row_text)
|
||||
|
||||
return '\n\n'.join(text_content)
|
||||
except Exception as e:
|
||||
raise Exception(f"Failed to parse DOCX: {str(e)}")
|
||||
|
||||
@staticmethod
|
||||
async def parse_xlsx(file_content: bytes) -> str:
|
||||
"""Parse XLSX file"""
|
||||
try:
|
||||
xlsx_file = io.BytesIO(file_content)
|
||||
workbook = openpyxl.load_workbook(xlsx_file, data_only=True)
|
||||
|
||||
text_content = []
|
||||
for sheet_name in workbook.sheetnames:
|
||||
sheet = workbook[sheet_name]
|
||||
text_content.append(f"=== Sheet: {sheet_name} ===")
|
||||
|
||||
for row in sheet.iter_rows(values_only=True):
|
||||
row_text = ' | '.join(str(cell) if cell is not None else '' for cell in row)
|
||||
if row_text.strip(' |'):
|
||||
text_content.append(row_text)
|
||||
|
||||
return '\n\n'.join(text_content)
|
||||
except Exception as e:
|
||||
raise Exception(f"Failed to parse XLSX: {str(e)}")
|
||||
|
||||
@staticmethod
|
||||
async def parse_file(file_content: bytes, filename: str) -> str:
|
||||
"""
|
||||
Parse file based on extension.
|
||||
|
||||
Args:
|
||||
file_content: File content as bytes
|
||||
filename: Original filename
|
||||
|
||||
Returns:
|
||||
Extracted text content
|
||||
|
||||
Raises:
|
||||
Exception: If file format is unsupported or parsing fails
|
||||
"""
|
||||
extension = filename.rsplit('.', 1)[-1].lower() if '.' in filename else ''
|
||||
|
||||
parsers = {
|
||||
'txt': DocumentParser.parse_txt,
|
||||
'pdf': DocumentParser.parse_pdf,
|
||||
'docx': DocumentParser.parse_docx,
|
||||
'doc': DocumentParser.parse_docx, # Try to parse DOC as DOCX
|
||||
'xlsx': DocumentParser.parse_xlsx,
|
||||
'xls': DocumentParser.parse_xlsx, # Try to parse XLS as XLSX
|
||||
}
|
||||
|
||||
parser = parsers.get(extension)
|
||||
if not parser:
|
||||
raise Exception(f"Unsupported file format: {extension}")
|
||||
|
||||
return await parser(file_content)
|
||||
|
||||
|
||||
# Singleton instance
|
||||
document_parser = DocumentParser()
|
||||
216
backend/services/llm_service.py
Normal file
216
backend/services/llm_service.py
Normal file
@@ -0,0 +1,216 @@
|
||||
"""
|
||||
LLM Service for AI-powered question parsing and grading
|
||||
"""
|
||||
import os
|
||||
import json
|
||||
from typing import List, Dict, Any, Optional
|
||||
from openai import AsyncOpenAI
|
||||
from anthropic import AsyncAnthropic
|
||||
import httpx
|
||||
|
||||
from models import QuestionType
|
||||
from utils import calculate_content_hash
|
||||
|
||||
|
||||
class LLMService:
|
||||
"""Service for interacting with various LLM providers"""
|
||||
|
||||
def __init__(self):
|
||||
self.provider = os.getenv("AI_PROVIDER", "openai")
|
||||
|
||||
if self.provider == "openai":
|
||||
self.client = AsyncOpenAI(
|
||||
api_key=os.getenv("OPENAI_API_KEY"),
|
||||
base_url=os.getenv("OPENAI_BASE_URL", "https://api.openai.com/v1")
|
||||
)
|
||||
self.model = os.getenv("OPENAI_MODEL", "gpt-4o-mini")
|
||||
|
||||
elif self.provider == "anthropic":
|
||||
self.client = AsyncAnthropic(
|
||||
api_key=os.getenv("ANTHROPIC_API_KEY")
|
||||
)
|
||||
self.model = os.getenv("ANTHROPIC_MODEL", "claude-3-haiku-20240307")
|
||||
|
||||
elif self.provider == "qwen":
|
||||
self.client = AsyncOpenAI(
|
||||
api_key=os.getenv("QWEN_API_KEY"),
|
||||
base_url=os.getenv("QWEN_BASE_URL", "https://dashscope.aliyuncs.com/compatible-mode/v1")
|
||||
)
|
||||
self.model = os.getenv("QWEN_MODEL", "qwen-plus")
|
||||
|
||||
else:
|
||||
raise ValueError(f"Unsupported AI provider: {self.provider}")
|
||||
|
||||
async def parse_document(self, content: str) -> List[Dict[str, Any]]:
|
||||
"""
|
||||
Parse document content and extract questions.
|
||||
|
||||
Returns a list of dictionaries with question data:
|
||||
[
|
||||
{
|
||||
"content": "Question text",
|
||||
"type": "single/multiple/judge/short",
|
||||
"options": ["A. Option1", "B. Option2", ...], # For choice questions
|
||||
"answer": "Correct answer",
|
||||
"analysis": "Explanation"
|
||||
},
|
||||
...
|
||||
]
|
||||
"""
|
||||
prompt = """You are a professional question parser. Parse the given document and extract all questions.
|
||||
|
||||
For each question, identify:
|
||||
1. Question content (the question text)
|
||||
2. Question type: single (单选), multiple (多选), judge (判断), short (简答)
|
||||
3. Options (for choice questions only, format: ["A. Option1", "B. Option2", ...])
|
||||
4. Correct answer
|
||||
5. Analysis/Explanation (if available)
|
||||
|
||||
Return ONLY a JSON array of questions, with no additional text:
|
||||
[
|
||||
{
|
||||
"content": "question text",
|
||||
"type": "single",
|
||||
"options": ["A. Option1", "B. Option2", "C. Option3", "D. Option4"],
|
||||
"answer": "A",
|
||||
"analysis": "explanation"
|
||||
},
|
||||
...
|
||||
]
|
||||
|
||||
Document content:
|
||||
---
|
||||
{content}
|
||||
---
|
||||
|
||||
IMPORTANT: Return ONLY the JSON array, no markdown code blocks or explanations."""
|
||||
|
||||
try:
|
||||
if self.provider == "anthropic":
|
||||
response = await self.client.messages.create(
|
||||
model=self.model,
|
||||
max_tokens=4096,
|
||||
messages=[
|
||||
{"role": "user", "content": prompt.format(content=content)}
|
||||
]
|
||||
)
|
||||
result = response.content[0].text
|
||||
else: # OpenAI or Qwen
|
||||
response = await self.client.chat.completions.create(
|
||||
model=self.model,
|
||||
messages=[
|
||||
{"role": "system", "content": "You are a professional question parser. Return only JSON."},
|
||||
{"role": "user", "content": prompt.format(content=content)}
|
||||
],
|
||||
temperature=0.3,
|
||||
)
|
||||
result = response.choices[0].message.content
|
||||
|
||||
# Clean result and parse JSON
|
||||
result = result.strip()
|
||||
if result.startswith("```json"):
|
||||
result = result[7:]
|
||||
if result.startswith("```"):
|
||||
result = result[3:]
|
||||
if result.endswith("```"):
|
||||
result = result[:-3]
|
||||
result = result.strip()
|
||||
|
||||
questions = json.loads(result)
|
||||
|
||||
# Add content hash to each question
|
||||
for q in questions:
|
||||
q["content_hash"] = calculate_content_hash(q["content"])
|
||||
|
||||
return questions
|
||||
|
||||
except Exception as e:
|
||||
print(f"Error parsing document: {e}")
|
||||
raise Exception(f"Failed to parse document: {str(e)}")
|
||||
|
||||
async def grade_short_answer(
|
||||
self,
|
||||
question: str,
|
||||
correct_answer: str,
|
||||
user_answer: str
|
||||
) -> Dict[str, Any]:
|
||||
"""
|
||||
Grade a short answer question using AI.
|
||||
|
||||
Returns:
|
||||
{
|
||||
"score": 0.0-1.0,
|
||||
"feedback": "Detailed feedback"
|
||||
}
|
||||
"""
|
||||
prompt = f"""Grade the following short answer question.
|
||||
|
||||
Question: {question}
|
||||
|
||||
Standard Answer: {correct_answer}
|
||||
|
||||
Student Answer: {user_answer}
|
||||
|
||||
Provide a score from 0.0 to 1.0 (where 1.0 is perfect) and detailed feedback.
|
||||
|
||||
Return ONLY a JSON object:
|
||||
{{
|
||||
"score": 0.85,
|
||||
"feedback": "Your detailed feedback here"
|
||||
}}
|
||||
|
||||
Be fair but strict. Consider:
|
||||
1. Correctness of key points
|
||||
2. Completeness of answer
|
||||
3. Clarity of expression
|
||||
|
||||
Return ONLY the JSON object, no markdown or explanations."""
|
||||
|
||||
try:
|
||||
if self.provider == "anthropic":
|
||||
response = await self.client.messages.create(
|
||||
model=self.model,
|
||||
max_tokens=1024,
|
||||
messages=[
|
||||
{"role": "user", "content": prompt}
|
||||
]
|
||||
)
|
||||
result = response.content[0].text
|
||||
else: # OpenAI or Qwen
|
||||
response = await self.client.chat.completions.create(
|
||||
model=self.model,
|
||||
messages=[
|
||||
{"role": "system", "content": "You are a fair and strict grader. Return only JSON."},
|
||||
{"role": "user", "content": prompt}
|
||||
],
|
||||
temperature=0.5,
|
||||
)
|
||||
result = response.choices[0].message.content
|
||||
|
||||
# Clean and parse JSON
|
||||
result = result.strip()
|
||||
if result.startswith("```json"):
|
||||
result = result[7:]
|
||||
if result.startswith("```"):
|
||||
result = result[3:]
|
||||
if result.endswith("```"):
|
||||
result = result[:-3]
|
||||
result = result.strip()
|
||||
|
||||
grading = json.loads(result)
|
||||
return {
|
||||
"score": float(grading.get("score", 0.0)),
|
||||
"feedback": grading.get("feedback", "")
|
||||
}
|
||||
|
||||
except Exception as e:
|
||||
print(f"Error grading answer: {e}")
|
||||
# Return default grading on error
|
||||
return {
|
||||
"score": 0.0,
|
||||
"feedback": "Unable to grade answer due to an error."
|
||||
}
|
||||
|
||||
|
||||
# Singleton instance
|
||||
llm_service = LLMService()
|
||||
Reference in New Issue
Block a user