Spaces:
Sleeping
Sleeping
Commit ·
2706084
0
Parent(s):
Initial backend commit
Browse files- .env.example +3 -0
- .gitignore +9 -0
- Dockerfile +22 -0
- README.md +37 -0
- app/db/database.py +19 -0
- app/main.py +357 -0
- app/models/models.py +36 -0
- app/schemas/schemas.py +67 -0
- app/services/scraper.py +117 -0
- requirements.txt +12 -0
- scripts/seed.py +73 -0
.env.example
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
GROQ_API_KEY=your_groq_api_key_here
|
| 2 |
+
CORS_ORIGINS=https://YOUR_PROJECT.vercel.app,http://localhost:5173
|
| 3 |
+
DATABASE_URL=sqlite:///./smart_reader.db
|
.gitignore
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
.env
|
| 2 |
+
__pycache__/
|
| 3 |
+
*.pyc
|
| 4 |
+
*.pyo
|
| 5 |
+
*.pyd
|
| 6 |
+
*.db
|
| 7 |
+
*.sqlite
|
| 8 |
+
.pytest_cache/
|
| 9 |
+
error.log
|
Dockerfile
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
FROM python:3.11-slim
|
| 2 |
+
|
| 3 |
+
# Set working directory
|
| 4 |
+
WORKDIR /app
|
| 5 |
+
|
| 6 |
+
# Install uv
|
| 7 |
+
COPY --from=ghcr.io/astral-sh/uv:latest /uv /uvx /bin/
|
| 8 |
+
|
| 9 |
+
# Copy requirements first to leverage Docker cache
|
| 10 |
+
COPY requirements.txt .
|
| 11 |
+
|
| 12 |
+
# Install dependencies using uv
|
| 13 |
+
RUN uv pip install --system --no-cache -r requirements.txt
|
| 14 |
+
|
| 15 |
+
# Copy the rest of the application code
|
| 16 |
+
COPY . .
|
| 17 |
+
|
| 18 |
+
# Expose port 7860 for HuggingFace Spaces
|
| 19 |
+
EXPOSE 7860
|
| 20 |
+
|
| 21 |
+
# Run Uvicorn on 0.0.0.0:7860
|
| 22 |
+
CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "7860"]
|
README.md
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Smart Reader Backend
|
| 2 |
+
|
| 3 |
+
This is the FastAPI backend for the Smart Reader application, designed to be deployed on HuggingFace Spaces.
|
| 4 |
+
|
| 5 |
+
## Requirements
|
| 6 |
+
- Python 3.11+
|
| 7 |
+
- Groq API Key
|
| 8 |
+
|
| 9 |
+
## Environment Variables
|
| 10 |
+
Create a `.env` file in the root directory and add the following variables:
|
| 11 |
+
```
|
| 12 |
+
GROQ_API_KEY=your_groq_api_key_here
|
| 13 |
+
CORS_ORIGINS=https://YOUR_PROJECT.vercel.app,http://localhost:5173
|
| 14 |
+
DATABASE_URL=sqlite:///./smart_reader.db
|
| 15 |
+
```
|
| 16 |
+
|
| 17 |
+
## Local Development
|
| 18 |
+
1. Install dependencies:
|
| 19 |
+
```bash
|
| 20 |
+
pip install -r requirements.txt
|
| 21 |
+
```
|
| 22 |
+
2. Run the application:
|
| 23 |
+
```bash
|
| 24 |
+
uvicorn main:app --reload --port 8080
|
| 25 |
+
```
|
| 26 |
+
|
| 27 |
+
## Docker
|
| 28 |
+
Build and run the Docker container locally:
|
| 29 |
+
```bash
|
| 30 |
+
docker build -t smart-reader-backend .
|
| 31 |
+
docker run -p 7860:7860 --env-file .env smart-reader-backend
|
| 32 |
+
```
|
| 33 |
+
|
| 34 |
+
## HuggingFace Deployment
|
| 35 |
+
This repository is configured for HuggingFace Spaces using Docker.
|
| 36 |
+
- The `Dockerfile` exposes port `7860`.
|
| 37 |
+
- Set your `GROQ_API_KEY` and `CORS_ORIGINS` in the HuggingFace Spaces secrets configuration.
|
app/db/database.py
ADDED
|
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from sqlalchemy import create_engine
|
| 2 |
+
from sqlalchemy.ext.declarative import declarative_base
|
| 3 |
+
from sqlalchemy.orm import sessionmaker
|
| 4 |
+
|
| 5 |
+
SQLALCHEMY_DATABASE_URL = "sqlite:///./smart_reader.db"
|
| 6 |
+
|
| 7 |
+
engine = create_engine(
|
| 8 |
+
SQLALCHEMY_DATABASE_URL, connect_args={"check_same_thread": False}
|
| 9 |
+
)
|
| 10 |
+
SessionLocal = sessionmaker(autocommit=False, autoflush=False, bind=engine)
|
| 11 |
+
|
| 12 |
+
Base = declarative_base()
|
| 13 |
+
|
| 14 |
+
def get_db():
|
| 15 |
+
db = SessionLocal()
|
| 16 |
+
try:
|
| 17 |
+
yield db
|
| 18 |
+
finally:
|
| 19 |
+
db.close()
|
app/main.py
ADDED
|
@@ -0,0 +1,357 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import os
|
| 2 |
+
import sys
|
| 3 |
+
from fastapi import FastAPI, Depends, HTTPException, UploadFile, File
|
| 4 |
+
from fastapi.middleware.cors import CORSMiddleware
|
| 5 |
+
from sqlalchemy.orm import Session
|
| 6 |
+
from typing import List
|
| 7 |
+
from dotenv import load_dotenv
|
| 8 |
+
|
| 9 |
+
# Force UTF-8 encoding for standard output to prevent Uvicorn from crashing when logging Arabic URLs on Windows
|
| 10 |
+
if sys.platform == 'win32' and sys.stdout:
|
| 11 |
+
sys.stdout.reconfigure(encoding='utf-8')
|
| 12 |
+
|
| 13 |
+
from app.db.database import engine, get_db
|
| 14 |
+
from app.models import models
|
| 15 |
+
from app.schemas import schemas
|
| 16 |
+
from app.services.scraper import scrape_article_url
|
| 17 |
+
|
| 18 |
+
# Load environment variables from .env file
|
| 19 |
+
load_dotenv()
|
| 20 |
+
|
| 21 |
+
# Initialize database tables
|
| 22 |
+
models.Base.metadata.create_all(bind=engine)
|
| 23 |
+
|
| 24 |
+
app = FastAPI(title="Smart Reader API", version="1.0.0")
|
| 25 |
+
|
| 26 |
+
# --- CORS Middleware Configuration ---
|
| 27 |
+
cors_origins_str = os.environ.get("CORS_ORIGINS", "")
|
| 28 |
+
if cors_origins_str:
|
| 29 |
+
allow_origins = [origin.strip() for origin in cors_origins_str.split(",")]
|
| 30 |
+
else:
|
| 31 |
+
allow_origins = ["*"] # Fallback for local development if not set
|
| 32 |
+
|
| 33 |
+
app.add_middleware(
|
| 34 |
+
CORSMiddleware,
|
| 35 |
+
allow_origins=allow_origins,
|
| 36 |
+
allow_credentials=True,
|
| 37 |
+
allow_methods=["*"],
|
| 38 |
+
allow_headers=["*"],
|
| 39 |
+
)
|
| 40 |
+
|
| 41 |
+
# --- Sentiment Analysis Helper ---
|
| 42 |
+
def analyze_sentiment(text: str) -> str:
|
| 43 |
+
"""
|
| 44 |
+
Analyzes the sentiment of a comment text.
|
| 45 |
+
Works for both Arabic and English using a keyword-based lexicon approach.
|
| 46 |
+
"""
|
| 47 |
+
t = text.lower()
|
| 48 |
+
positive_keywords = [
|
| 49 |
+
"جميل", "رائع", "ممتاز", "مفيد", "شكرا", "تحفة", "حب", "جيد", "حلو",
|
| 50 |
+
"عظيم", "أعجبني", "قوي", "سهل", "واضح",
|
| 51 |
+
"nice", "good", "great", "awesome", "love", "useful", "thanks", "like", "easy"
|
| 52 |
+
]
|
| 53 |
+
negative_keywords = [
|
| 54 |
+
"سيء", "صعب", "ملل", "خطأ", "فشل", "ضعيف", "لا يعجبني", "حزين", "أسف",
|
| 55 |
+
"وحش", "ركيك", "ممل", "معقد", "ناقص",
|
| 56 |
+
"bad", "worst", "boring", "error", "fail", "sad", "dislike", "hate", "hard", "difficult"
|
| 57 |
+
]
|
| 58 |
+
|
| 59 |
+
pos_count = sum(1 for word in positive_keywords if word in t)
|
| 60 |
+
neg_count = sum(1 for word in negative_keywords if word in t)
|
| 61 |
+
|
| 62 |
+
if pos_count > neg_count:
|
| 63 |
+
return "POSITIVE"
|
| 64 |
+
elif neg_count > pos_count:
|
| 65 |
+
return "NEGATIVE"
|
| 66 |
+
return "NEUTRAL"
|
| 67 |
+
|
| 68 |
+
# --- AI Summarization Helper ---
|
| 69 |
+
def generate_ai_summary(content: str) -> str:
|
| 70 |
+
"""
|
| 71 |
+
Generates a summary of the article using Groq API.
|
| 72 |
+
Falls back to a clean local extraction summary if no Groq API Key is set.
|
| 73 |
+
"""
|
| 74 |
+
api_key = os.environ.get("GROQ_API_KEY")
|
| 75 |
+
if api_key and api_key.strip():
|
| 76 |
+
try:
|
| 77 |
+
from groq import Groq
|
| 78 |
+
client = Groq(api_key=api_key)
|
| 79 |
+
response = client.chat.completions.create(
|
| 80 |
+
model="llama-3.3-70b-versatile",
|
| 81 |
+
messages=[
|
| 82 |
+
{"role": "system", "content": "You are a professional summarizer. You MUST output ONLY the raw summary text. No intro, no outro, no conversational filler. Summarize the text in the exact same language as the input text."},
|
| 83 |
+
{"role": "user", "content": f"Summarize this text in 2-3 sentences. Start immediately with the first word of the summary:\n\n{content[:4000]}"}
|
| 84 |
+
],
|
| 85 |
+
max_tokens=300,
|
| 86 |
+
)
|
| 87 |
+
summary = response.choices[0].message.content.strip()
|
| 88 |
+
|
| 89 |
+
# Programmatically strip common conversational fillers
|
| 90 |
+
fillers = [
|
| 91 |
+
"Here is a summary of the text",
|
| 92 |
+
"Here is a summary",
|
| 93 |
+
"Here is the summary",
|
| 94 |
+
"Sure, here is a summary",
|
| 95 |
+
"The text can be summarized as follows:",
|
| 96 |
+
"This text is about",
|
| 97 |
+
"إليك ملخص",
|
| 98 |
+
"فيما يلي ملخص"
|
| 99 |
+
]
|
| 100 |
+
for filler in fillers:
|
| 101 |
+
if summary.lower().startswith(filler.lower()):
|
| 102 |
+
# Find the end of the first line or colon and slice it
|
| 103 |
+
idx = summary.find("\n\n")
|
| 104 |
+
if idx != -1:
|
| 105 |
+
summary = summary[idx+2:].strip()
|
| 106 |
+
else:
|
| 107 |
+
idx = summary.find(":")
|
| 108 |
+
if idx != -1 and idx < 100:
|
| 109 |
+
summary = summary[idx+1:].strip()
|
| 110 |
+
return summary.strip('"\'- \n')
|
| 111 |
+
except Exception as e:
|
| 112 |
+
print(f"Error calling Groq API: {e}")
|
| 113 |
+
|
| 114 |
+
# Fallback / Local summarization (First 2 full sentences)
|
| 115 |
+
sentences = content.replace("\n", " ").split(".")
|
| 116 |
+
fallback_summary = ". ".join([s.strip() for s in sentences[:2] if s.strip()])
|
| 117 |
+
if fallback_summary:
|
| 118 |
+
return fallback_summary + "."
|
| 119 |
+
return content[:150] + "..."
|
| 120 |
+
|
| 121 |
+
# --- Endpoints ---
|
| 122 |
+
|
| 123 |
+
@app.get("/api/v1/articles", response_model=List[schemas.ArticleResponse])
|
| 124 |
+
def get_articles(db: Session = Depends(get_db)):
|
| 125 |
+
articles = db.query(models.Article).all()
|
| 126 |
+
return articles
|
| 127 |
+
|
| 128 |
+
@app.get("/api/v1/ai/articles/{article_id}/summary")
|
| 129 |
+
def get_article_summary(article_id: int, db: Session = Depends(get_db)):
|
| 130 |
+
article = db.query(models.Article).filter(models.Article.id == article_id).first()
|
| 131 |
+
if not article:
|
| 132 |
+
raise HTTPException(status_code=404, detail="المقال غير موجود")
|
| 133 |
+
|
| 134 |
+
# Generate live summary
|
| 135 |
+
summary = generate_ai_summary(article.content)
|
| 136 |
+
return summary
|
| 137 |
+
|
| 138 |
+
@app.post("/api/v1/comments", response_model=schemas.CommentResponse)
|
| 139 |
+
def create_comment(comment_in: schemas.CommentCreate, db: Session = Depends(get_db)):
|
| 140 |
+
user_id = comment_in.user.id if comment_in.user else 1
|
| 141 |
+
user = db.query(models.User).filter(models.User.id == user_id).first()
|
| 142 |
+
if not user:
|
| 143 |
+
user = models.User(id=user_id, name="مستخدم جديد")
|
| 144 |
+
db.add(user)
|
| 145 |
+
db.commit()
|
| 146 |
+
db.refresh(user)
|
| 147 |
+
|
| 148 |
+
article_id = None
|
| 149 |
+
if comment_in.article and hasattr(comment_in.article, 'id'):
|
| 150 |
+
article_id = comment_in.article.id
|
| 151 |
+
|
| 152 |
+
if not article_id:
|
| 153 |
+
raise HTTPException(status_code=400, detail="يجب تحديد المقال المرتبط بالتعليق")
|
| 154 |
+
|
| 155 |
+
article = db.query(models.Article).filter(models.Article.id == article_id).first()
|
| 156 |
+
if not article:
|
| 157 |
+
raise HTTPException(status_code=404, detail="المقال غير موجود")
|
| 158 |
+
|
| 159 |
+
sentiment = analyze_sentiment(comment_in.text)
|
| 160 |
+
|
| 161 |
+
db_comment = models.Comment(
|
| 162 |
+
text=comment_in.text,
|
| 163 |
+
sentiment=sentiment,
|
| 164 |
+
user_id=user.id,
|
| 165 |
+
article_id=article.id
|
| 166 |
+
)
|
| 167 |
+
db.add(db_comment)
|
| 168 |
+
db.commit()
|
| 169 |
+
db.refresh(db_comment)
|
| 170 |
+
|
| 171 |
+
return db_comment
|
| 172 |
+
|
| 173 |
+
# --- 🚀 Upgraded Scraping Endpoint ---
|
| 174 |
+
@app.post("/api/v1/articles/scrape", response_model=schemas.ArticleResponse)
|
| 175 |
+
def scrape_and_add_article(req: schemas.ScrapeRequest, db: Session = Depends(get_db)):
|
| 176 |
+
scraped_data = scrape_article_url(req.url)
|
| 177 |
+
|
| 178 |
+
# Auto-generate next ID
|
| 179 |
+
max_id_article = db.query(models.Article).order_by(models.Article.id.desc()).first()
|
| 180 |
+
next_id = (max_id_article.id + 1) if max_id_article else 1
|
| 181 |
+
|
| 182 |
+
db_article = models.Article(
|
| 183 |
+
id=next_id,
|
| 184 |
+
title=scraped_data["title"],
|
| 185 |
+
content=scraped_data["content"],
|
| 186 |
+
category=scraped_data["category"],
|
| 187 |
+
author=scraped_data["author"],
|
| 188 |
+
image=scraped_data["image"],
|
| 189 |
+
summary="" # Generate on demand
|
| 190 |
+
)
|
| 191 |
+
|
| 192 |
+
db.add(db_article)
|
| 193 |
+
db.commit()
|
| 194 |
+
db.refresh(db_article)
|
| 195 |
+
|
| 196 |
+
return db_article
|
| 197 |
+
|
| 198 |
+
# --- 📁 File Upload Endpoint ---
|
| 199 |
+
import io
|
| 200 |
+
import docx
|
| 201 |
+
from PyPDF2 import PdfReader
|
| 202 |
+
|
| 203 |
+
@app.post("/api/v1/articles/upload", response_model=schemas.ArticleResponse)
|
| 204 |
+
async def upload_document(file: UploadFile = File(...), db: Session = Depends(get_db)):
|
| 205 |
+
if not file.filename:
|
| 206 |
+
raise HTTPException(status_code=400, detail="الملف غير صالح")
|
| 207 |
+
|
| 208 |
+
ext = file.filename.split('.')[-1].lower()
|
| 209 |
+
content_text = ""
|
| 210 |
+
title = file.filename.rsplit('.', 1)[0]
|
| 211 |
+
|
| 212 |
+
try:
|
| 213 |
+
file_bytes = await file.read()
|
| 214 |
+
|
| 215 |
+
if ext == "txt":
|
| 216 |
+
content_text = file_bytes.decode('utf-8', errors='ignore')
|
| 217 |
+
elif ext == "pdf":
|
| 218 |
+
reader = PdfReader(io.BytesIO(file_bytes))
|
| 219 |
+
for page in reader.pages:
|
| 220 |
+
text = page.extract_text()
|
| 221 |
+
if text:
|
| 222 |
+
content_text += text + "\n"
|
| 223 |
+
elif ext == "docx":
|
| 224 |
+
doc = docx.Document(io.BytesIO(file_bytes))
|
| 225 |
+
for para in doc.paragraphs:
|
| 226 |
+
content_text += para.text + "\n"
|
| 227 |
+
else:
|
| 228 |
+
raise HTTPException(status_code=400, detail="نوع الملف غير مدعوم. يرجى رفع ملفات (pdf, docx, txt)")
|
| 229 |
+
|
| 230 |
+
if not content_text.strip():
|
| 231 |
+
raise HTTPException(status_code=400, detail="لم نتمكن من قراءة أي نص من الملف")
|
| 232 |
+
|
| 233 |
+
# Add to database as article
|
| 234 |
+
max_id_article = db.query(models.Article).order_by(models.Article.id.desc()).first()
|
| 235 |
+
next_id = (max_id_article.id + 1) if max_id_article else 1
|
| 236 |
+
|
| 237 |
+
db_article = models.Article(
|
| 238 |
+
id=next_id,
|
| 239 |
+
title=title,
|
| 240 |
+
content=content_text.strip(),
|
| 241 |
+
category="مستند مرفوع",
|
| 242 |
+
author="أنت",
|
| 243 |
+
image=None,
|
| 244 |
+
summary=""
|
| 245 |
+
)
|
| 246 |
+
db.add(db_article)
|
| 247 |
+
db.commit()
|
| 248 |
+
db.refresh(db_article)
|
| 249 |
+
return db_article
|
| 250 |
+
|
| 251 |
+
except HTTPException:
|
| 252 |
+
raise
|
| 253 |
+
except Exception as e:
|
| 254 |
+
raise HTTPException(status_code=500, detail=f"حدث خطأ أثناء معالجة الملف: {str(e)}")
|
| 255 |
+
|
| 256 |
+
# --- 🚀 Upgraded Chatbot (RAG) Endpoint ---
|
| 257 |
+
@app.post("/api/v1/ai/articles/{article_id}/chat")
|
| 258 |
+
def chat_with_article(article_id: int, chat_req: schemas.ChatRequest, db: Session = Depends(get_db)):
|
| 259 |
+
article = db.query(models.Article).filter(models.Article.id == article_id).first()
|
| 260 |
+
if not article:
|
| 261 |
+
raise HTTPException(status_code=404, detail="المقال غير موجود")
|
| 262 |
+
|
| 263 |
+
api_key = os.environ.get("GROQ_API_KEY")
|
| 264 |
+
if not api_key or not api_key.strip():
|
| 265 |
+
# Fallback offline replies
|
| 266 |
+
user_msg = chat_req.message.lower()
|
| 267 |
+
if any(greeting in user_msg for greeting in ["مرحباً", "hello", "hi", "سلام", "هلا"]):
|
| 268 |
+
return {"reply": "أهلاً بك! أنا مساعدك الذكي لقراءة وتلخيص هذا المقال. كيف يمكنني مساعدتك اليوم؟"}
|
| 269 |
+
|
| 270 |
+
if any(kw in user_msg for kw in ["لخص", "ملخص", "summary", "summarize"]):
|
| 271 |
+
summary_text = article.summary if article.summary else generate_ai_summary(article.content)
|
| 272 |
+
return {"reply": f"إليك ملخص سريع للمقال:\n{summary_text}"}
|
| 273 |
+
|
| 274 |
+
return {"reply": f"أهلاً بك! لم يتم ضبط مفتاح Groq API Key في ملف `.env` بعد. يمكنك الحصول عليه مجاناً من https://console.groq.com"}
|
| 275 |
+
|
| 276 |
+
try:
|
| 277 |
+
from groq import Groq
|
| 278 |
+
client = Groq(api_key=api_key)
|
| 279 |
+
|
| 280 |
+
system_prompt = f"""أنت مساعد قراءة ذكي وتفاعلي لموقع "Smart Reader".
|
| 281 |
+
وظيفتك الإجابة عن أي أسئلة يطرحها القارئ حول المقال التالي بدقة بالغة وبنفس لغة سؤال القارئ (العربية أو الإنجليزية).
|
| 282 |
+
تجنب تأليف معلومات ليست في المقال، وإذا سألك عن موضوع عام غير موجود بالمقال وضح له ذلك ثم أجب باختصار وبشكل مفيد.
|
| 283 |
+
|
| 284 |
+
محتوى المقال كمرجع لك:
|
| 285 |
+
العنوان: {article.title}
|
| 286 |
+
الكاتب: {article.author}
|
| 287 |
+
التصنيف: {article.category}
|
| 288 |
+
المحتوى:
|
| 289 |
+
{article.content[:4000]}"""
|
| 290 |
+
|
| 291 |
+
# Build messages from history
|
| 292 |
+
messages = [{"role": "system", "content": system_prompt}]
|
| 293 |
+
if chat_req.history:
|
| 294 |
+
for h in chat_req.history:
|
| 295 |
+
role = "user" if h.role == "user" else "assistant"
|
| 296 |
+
messages.append({"role": role, "content": h.content})
|
| 297 |
+
messages.append({"role": "user", "content": chat_req.message})
|
| 298 |
+
|
| 299 |
+
response = client.chat.completions.create(
|
| 300 |
+
model="llama-3.3-70b-versatile",
|
| 301 |
+
messages=messages,
|
| 302 |
+
max_tokens=1024,
|
| 303 |
+
)
|
| 304 |
+
return {"reply": response.choices[0].message.content.strip()}
|
| 305 |
+
except Exception as e:
|
| 306 |
+
print(f"Groq Chat error: {e}")
|
| 307 |
+
return {"reply": f"حدث خطأ أثناء الاتصال بالذكاء الاصطناعي: {str(e)}"}
|
| 308 |
+
|
| 309 |
+
# --- 🔍 Web Search Endpoint ---
|
| 310 |
+
@app.get("/api/v1/search", response_model=List[schemas.SearchResult])
|
| 311 |
+
def web_search(q: str, max_results: int = 8):
|
| 312 |
+
"""
|
| 313 |
+
Searches Wikipedia (Arabic or English depending on query) and returns article results.
|
| 314 |
+
"""
|
| 315 |
+
if not q or not q.strip():
|
| 316 |
+
raise HTTPException(status_code=400, detail="يجب إدخال كلمة بحث")
|
| 317 |
+
try:
|
| 318 |
+
import urllib.request
|
| 319 |
+
import urllib.parse
|
| 320 |
+
import json
|
| 321 |
+
import re
|
| 322 |
+
|
| 323 |
+
# Detect if query has Arabic characters
|
| 324 |
+
is_arabic = bool(re.search(r'[\u0600-\u06FF]', q))
|
| 325 |
+
lang = "ar" if is_arabic else "en"
|
| 326 |
+
|
| 327 |
+
# Wikipedia Search API
|
| 328 |
+
url = f"https://{lang}.wikipedia.org/w/api.php?action=query&list=search&srsearch={urllib.parse.quote(q)}&utf8=&format=json&srlimit={max_results}"
|
| 329 |
+
req = urllib.request.Request(url, headers={'User-Agent': 'SmartReader/1.0'})
|
| 330 |
+
|
| 331 |
+
with urllib.request.urlopen(req, timeout=10) as response:
|
| 332 |
+
data = json.loads(response.read().decode('utf-8'))
|
| 333 |
+
|
| 334 |
+
results = []
|
| 335 |
+
for item in data.get('query', {}).get('search', []):
|
| 336 |
+
title = item.get('title', '')
|
| 337 |
+
# Clean HTML tags from snippet
|
| 338 |
+
snippet_html = item.get('snippet', '')
|
| 339 |
+
snippet = re.sub(r'<[^>]+>', '', snippet_html)
|
| 340 |
+
|
| 341 |
+
# Construct Wikipedia URL
|
| 342 |
+
article_url = f"https://{lang}.wikipedia.org/wiki/{urllib.parse.quote(title.replace(' ', '_'))}"
|
| 343 |
+
|
| 344 |
+
results.append(schemas.SearchResult(
|
| 345 |
+
title=title,
|
| 346 |
+
url=article_url,
|
| 347 |
+
snippet=snippet,
|
| 348 |
+
image=None,
|
| 349 |
+
source="ويكيبيديا" if is_arabic else "Wikipedia"
|
| 350 |
+
))
|
| 351 |
+
|
| 352 |
+
return results
|
| 353 |
+
except Exception as e:
|
| 354 |
+
import traceback
|
| 355 |
+
with open("error.log", "w", encoding="utf-8") as f:
|
| 356 |
+
f.write(traceback.format_exc())
|
| 357 |
+
raise HTTPException(status_code=500, detail=f"حدث خطأ أثناء البحث: {str(e)}")
|
app/models/models.py
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from sqlalchemy import Column, Integer, String, Text, ForeignKey
|
| 2 |
+
from sqlalchemy.orm import relationship
|
| 3 |
+
from app.db.database import Base
|
| 4 |
+
|
| 5 |
+
class User(Base):
|
| 6 |
+
__tablename__ = "users"
|
| 7 |
+
|
| 8 |
+
id = Column(Integer, primary_key=True, index=True)
|
| 9 |
+
name = Column(String(100), nullable=False)
|
| 10 |
+
|
| 11 |
+
comments = relationship("Comment", back_populates="user")
|
| 12 |
+
|
| 13 |
+
class Article(Base):
|
| 14 |
+
__tablename__ = "articles"
|
| 15 |
+
|
| 16 |
+
id = Column(Integer, primary_key=True, index=True)
|
| 17 |
+
title = Column(String(255), nullable=False)
|
| 18 |
+
content = Column(Text, nullable=False)
|
| 19 |
+
category = Column(String(100))
|
| 20 |
+
author = Column(String(100))
|
| 21 |
+
image = Column(String(500))
|
| 22 |
+
summary = Column(Text, nullable=True)
|
| 23 |
+
|
| 24 |
+
comments = relationship("Comment", back_populates="article", cascade="all, delete-orphan")
|
| 25 |
+
|
| 26 |
+
class Comment(Base):
|
| 27 |
+
__tablename__ = "comments"
|
| 28 |
+
|
| 29 |
+
id = Column(Integer, primary_key=True, index=True)
|
| 30 |
+
text = Column(Text, nullable=False)
|
| 31 |
+
sentiment = Column(String(20), default="NEUTRAL") # POSITIVE, NEGATIVE, NEUTRAL
|
| 32 |
+
user_id = Column(Integer, ForeignKey("users.id"))
|
| 33 |
+
article_id = Column(Integer, ForeignKey("articles.id"))
|
| 34 |
+
|
| 35 |
+
user = relationship("User", back_populates="comments")
|
| 36 |
+
article = relationship("Article", back_populates="comments")
|
app/schemas/schemas.py
ADDED
|
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from pydantic import BaseModel
|
| 2 |
+
from typing import List, Optional
|
| 3 |
+
|
| 4 |
+
# --- User Schemas ---
|
| 5 |
+
class UserBase(BaseModel):
|
| 6 |
+
id: int
|
| 7 |
+
name: str
|
| 8 |
+
|
| 9 |
+
class Config:
|
| 10 |
+
from_attributes = True
|
| 11 |
+
|
| 12 |
+
# --- Comment Schemas ---
|
| 13 |
+
class CommentCreate(BaseModel):
|
| 14 |
+
text: str
|
| 15 |
+
user: Optional[UserBase] = None
|
| 16 |
+
article: Optional[BaseModel] = None # We will handle nested article ID parsing in main
|
| 17 |
+
|
| 18 |
+
class Config:
|
| 19 |
+
from_attributes = True
|
| 20 |
+
|
| 21 |
+
class CommentResponse(BaseModel):
|
| 22 |
+
id: int
|
| 23 |
+
text: str
|
| 24 |
+
sentiment: str
|
| 25 |
+
user: Optional[UserBase] = None
|
| 26 |
+
|
| 27 |
+
class Config:
|
| 28 |
+
from_attributes = True
|
| 29 |
+
|
| 30 |
+
# --- Article Schemas ---
|
| 31 |
+
class ArticleBase(BaseModel):
|
| 32 |
+
id: int
|
| 33 |
+
title: str
|
| 34 |
+
content: str
|
| 35 |
+
category: Optional[str] = None
|
| 36 |
+
author: Optional[str] = None
|
| 37 |
+
image: Optional[str] = None
|
| 38 |
+
summary: Optional[str] = None
|
| 39 |
+
|
| 40 |
+
class Config:
|
| 41 |
+
from_attributes = True
|
| 42 |
+
|
| 43 |
+
class ArticleResponse(ArticleBase):
|
| 44 |
+
comments: List[CommentResponse] = []
|
| 45 |
+
|
| 46 |
+
class Config:
|
| 47 |
+
from_attributes = True
|
| 48 |
+
|
| 49 |
+
# --- Chat & Scraping Upgrades ---
|
| 50 |
+
class ScrapeRequest(BaseModel):
|
| 51 |
+
url: str
|
| 52 |
+
|
| 53 |
+
class ChatMessage(BaseModel):
|
| 54 |
+
role: str # "user" or "model"
|
| 55 |
+
content: str
|
| 56 |
+
|
| 57 |
+
class ChatRequest(BaseModel):
|
| 58 |
+
message: str
|
| 59 |
+
history: Optional[List[ChatMessage]] = None
|
| 60 |
+
|
| 61 |
+
# --- Web Search Schema ---
|
| 62 |
+
class SearchResult(BaseModel):
|
| 63 |
+
title: str
|
| 64 |
+
url: str
|
| 65 |
+
snippet: Optional[str] = None
|
| 66 |
+
image: Optional[str] = None
|
| 67 |
+
source: Optional[str] = None
|
app/services/scraper.py
ADDED
|
@@ -0,0 +1,117 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import requests
|
| 2 |
+
from bs4 import BeautifulSoup
|
| 3 |
+
import re
|
| 4 |
+
from urllib.parse import urljoin, urlparse
|
| 5 |
+
|
| 6 |
+
def scrape_article_url(url: str) -> dict:
|
| 7 |
+
"""
|
| 8 |
+
Scrapes a webpage URL and extracts:
|
| 9 |
+
- title
|
| 10 |
+
- content (main body text)
|
| 11 |
+
- author
|
| 12 |
+
- image (cover image)
|
| 13 |
+
- category
|
| 14 |
+
"""
|
| 15 |
+
headers = {
|
| 16 |
+
"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36"
|
| 17 |
+
}
|
| 18 |
+
|
| 19 |
+
try:
|
| 20 |
+
response = requests.get(url, headers=headers, timeout=10)
|
| 21 |
+
response.raise_for_status()
|
| 22 |
+
|
| 23 |
+
# Detect encoding
|
| 24 |
+
if response.encoding == 'ISO-8859-1':
|
| 25 |
+
response.encoding = response.apparent_encoding
|
| 26 |
+
|
| 27 |
+
soup = BeautifulSoup(response.text, "html.parser")
|
| 28 |
+
|
| 29 |
+
# 1. Extract Title
|
| 30 |
+
title = ""
|
| 31 |
+
# Try og:title first
|
| 32 |
+
og_title = soup.find("meta", property="og:title")
|
| 33 |
+
if og_title and og_title.get("content"):
|
| 34 |
+
title = og_title["content"]
|
| 35 |
+
else:
|
| 36 |
+
# Try <h1>
|
| 37 |
+
h1 = soup.find("h1")
|
| 38 |
+
if h1:
|
| 39 |
+
title = h1.get_text().strip()
|
| 40 |
+
else:
|
| 41 |
+
title_tag = soup.find("title")
|
| 42 |
+
if title_tag:
|
| 43 |
+
title = title_tag.get_text().strip()
|
| 44 |
+
|
| 45 |
+
if not title:
|
| 46 |
+
title = "مقال مستخلص من الويب"
|
| 47 |
+
|
| 48 |
+
# 2. Extract Author
|
| 49 |
+
author = "كاتب ويب"
|
| 50 |
+
author_meta = soup.find("meta", attrs={"name": "author"}) or soup.find("meta", property="article:author")
|
| 51 |
+
if author_meta and author_meta.get("content"):
|
| 52 |
+
author = author_meta["content"].strip()
|
| 53 |
+
else:
|
| 54 |
+
# Search for typical author classes
|
| 55 |
+
author_tag = soup.find(class_=re.compile(r"author|byline|writer", re.I))
|
| 56 |
+
if author_tag:
|
| 57 |
+
author = author_tag.get_text().strip()
|
| 58 |
+
|
| 59 |
+
# 3. Extract Cover Image
|
| 60 |
+
image_url = "https://images.unsplash.com/photo-1451187580459-43490279c0fa?w=600&auto=format&fit=crop&q=60" # default
|
| 61 |
+
og_image = soup.find("meta", property="og:image")
|
| 62 |
+
if og_image and og_image.get("content"):
|
| 63 |
+
image_url = og_image["content"]
|
| 64 |
+
else:
|
| 65 |
+
# Try finding the first large image in body
|
| 66 |
+
for img in soup.find_all("img"):
|
| 67 |
+
src = img.get("src")
|
| 68 |
+
if src and not src.endswith(".gif") and not src.endswith(".svg"):
|
| 69 |
+
# Resolve relative url
|
| 70 |
+
image_url = urljoin(url, src)
|
| 71 |
+
break
|
| 72 |
+
|
| 73 |
+
# 4. Extract Main Content
|
| 74 |
+
# Remove noisy elements
|
| 75 |
+
for element in soup(["script", "style", "nav", "footer", "header", "aside", "form"]):
|
| 76 |
+
element.extract()
|
| 77 |
+
|
| 78 |
+
# Find paragraphs
|
| 79 |
+
paragraphs = soup.find_all("p")
|
| 80 |
+
text_blocks = []
|
| 81 |
+
for p in paragraphs:
|
| 82 |
+
text = p.get_text().strip()
|
| 83 |
+
# Ignore short/noise paragraphs (less than 30 characters)
|
| 84 |
+
if len(text) > 30:
|
| 85 |
+
text_blocks.append(text)
|
| 86 |
+
|
| 87 |
+
content = "\n\n".join(text_blocks)
|
| 88 |
+
|
| 89 |
+
if not content:
|
| 90 |
+
# Fallback: get raw body text if no paragraphs are found
|
| 91 |
+
content = soup.body.get_text(separator="\n\n").strip() if soup.body else "تعذر استخلاص محتوى النص من هذا الموقع."
|
| 92 |
+
# Limit length if it's too raw and full of noise
|
| 93 |
+
content = content[:3000]
|
| 94 |
+
|
| 95 |
+
# 5. Extract/Guess Category or Domain Name
|
| 96 |
+
domain = urlparse(url).netloc.replace("www.", "")
|
| 97 |
+
category = domain.split(".")[0].capitalize()
|
| 98 |
+
|
| 99 |
+
return {
|
| 100 |
+
"title": title,
|
| 101 |
+
"content": content,
|
| 102 |
+
"author": author,
|
| 103 |
+
"image": image_url,
|
| 104 |
+
"category": category
|
| 105 |
+
}
|
| 106 |
+
|
| 107 |
+
except Exception as e:
|
| 108 |
+
import traceback
|
| 109 |
+
traceback.print_exc()
|
| 110 |
+
print(f"Scraping error: {e}")
|
| 111 |
+
return {
|
| 112 |
+
"title": "فشل جلب المقال",
|
| 113 |
+
"content": f"حدث خطأ أثناء محاولة الاتصال بالموقع أو جلب محتواه:\n{str(e)}",
|
| 114 |
+
"author": "خطأ النظام",
|
| 115 |
+
"image": "https://images.unsplash.com/photo-1594322436404-5a0526db4d13?w=600&auto=format&fit=crop&q=60",
|
| 116 |
+
"category": "خطأ"
|
| 117 |
+
}
|
requirements.txt
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
fastapi==0.115.8
|
| 2 |
+
uvicorn==0.34.0
|
| 3 |
+
sqlalchemy==2.0.38
|
| 4 |
+
pydantic==2.10.6
|
| 5 |
+
google-generativeai==0.8.4
|
| 6 |
+
beautifulsoup4==4.12.3
|
| 7 |
+
python-dotenv==1.0.1
|
| 8 |
+
requests==2.32.3
|
| 9 |
+
groq==0.13.1
|
| 10 |
+
python-multipart==0.0.20
|
| 11 |
+
PyPDF2==3.0.1
|
| 12 |
+
python-docx==1.1.2
|
scripts/seed.py
ADDED
|
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import sys
|
| 2 |
+
import os
|
| 3 |
+
|
| 4 |
+
# Add parent directory to path so database modules can be imported
|
| 5 |
+
sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
|
| 6 |
+
|
| 7 |
+
from app.db.database import engine, SessionLocal, Base
|
| 8 |
+
from app.models import models
|
| 9 |
+
|
| 10 |
+
def seed_db():
|
| 11 |
+
# Recreate tables
|
| 12 |
+
Base.metadata.drop_all(bind=engine)
|
| 13 |
+
Base.metadata.create_all(bind=engine)
|
| 14 |
+
|
| 15 |
+
db = SessionLocal()
|
| 16 |
+
try:
|
| 17 |
+
# Create a test user with ID 1 (required by the frontend comment payload)
|
| 18 |
+
test_user = models.User(id=1, name="أحمد علي")
|
| 19 |
+
db.add(test_user)
|
| 20 |
+
db.commit()
|
| 21 |
+
|
| 22 |
+
# Add sample articles
|
| 23 |
+
articles = [
|
| 24 |
+
models.Article(
|
| 25 |
+
id=1,
|
| 26 |
+
title="مستقبل الذكاء الاصطناعي في الطب",
|
| 27 |
+
content="""شهد قطاع الطب ثورة هائلة بفضل تقنيات الذكاء الاصطناعي في الآونة الأخيرة.
|
| 28 |
+
تُستخدم خوارزميات التعلم العميق الآن في تشخيص الأمراض المعقدة مثل السرطان بدقة تتفوق أحياناً على أمهر الأطباء.
|
| 29 |
+
علاوة على ذلك، يساهم الذكاء الاصطناعي في تسريع عملية تطوير الأدوية الجديدة من خلال التنبؤ بكيفية تفاعل الجزيئات الكيميائية، مما يوفر سنوات من البحث السريري والملايين من الدولارات.
|
| 30 |
+
في المستقبل القريب، سنرى روبوتات جراحية تعمل بدعم كامل من الذكاء الاصطناعي لتقليل الأخطاء البشرية أثناء العمليات الحساسة.""",
|
| 31 |
+
category="تقنية وصحة",
|
| 32 |
+
author="د. سامي الجمال",
|
| 33 |
+
image="https://images.unsplash.com/photo-1526374965328-7f61d4dc18c5?w=600&auto=format&fit=crop&q=60",
|
| 34 |
+
summary="يستعرض المقال كيف يُحدث الذكاء الاصطناعي ثورة في قطاع الرعاية الصحية من خلال التشخيص الدقيق للأمراض وتطوير الأدوية وتوجيه الروبوتات الجراحية."
|
| 35 |
+
),
|
| 36 |
+
models.Article(
|
| 37 |
+
id=2,
|
| 38 |
+
title="أسرار النوم الصحي وأثره على الإنتاجية",
|
| 39 |
+
content="""يعتقد الكثيرون أن النوم هو مجرد وقت مستقطع من اليوم للراحة، لكن الأبحاث الحديثة تؤكد أنه عملية حيوية بالغة الأهمية لتنظيف الدماغ من السموم المتراكمة طوال النهار.
|
| 40 |
+
الحصول على 7 إلى 8 ساعات من النوم العميق ليلاً يساعد على تحسين الذاكرة قصيرة المدى وزيادة القدرة على التركيز واتخاذ القرارات الصائبة في اليوم التالي.
|
| 41 |
+
على النقيض من ذلك، يؤدي الحرمان المزمن من النوم إلى تدهور الصحة النفسية وزيادة خطر الإصابة بأمراض القلب والسكري.
|
| 42 |
+
للحصول على نوم مثالي، يُنصح بالابتعاد عن الشاشات الزرقاء قبل ساعة من النوم وتثبيت موعد النوم والاستيقاظ يومياً.""",
|
| 43 |
+
category="نمط حياة وصحة",
|
| 44 |
+
author="منى الصاوي",
|
| 45 |
+
image="https://images.unsplash.com/photo-1511295742364-92767fa62d9f?w=600&auto=format&fit=crop&q=60",
|
| 46 |
+
summary="يتناول المقال الأهمية البيولوجية للنوم الصحي في تحسين التركيز والذاكرة وحماية الصحة العامة، مع تقديم نصائح عملية لنوم أفضل."
|
| 47 |
+
),
|
| 48 |
+
models.Article(
|
| 49 |
+
id=3,
|
| 50 |
+
title="استكشاف الكواكب البعيدة: هل سنصل إليها يوماً؟",
|
| 51 |
+
content="""لطالما كان السفر عبر النجوم حلماً يراود البشرية منذ عقود.
|
| 52 |
+
مع اكتشاف آلاف الكواكب خارج مجموعتنا الشمسية (Exoplanets)، بدأ العلماء في البحث عن كواكب تقع في النطاق الصالح للحياة (Goldilocks zone) حيث يمكن للماء السائل أن يتواجد.
|
| 53 |
+
العقبة الكبرى التي تواجهنا هي المسافات الشاسعة؛ فأقرب كوكب يحتمل أن يكون صالحاً للحياة يبعد عنا حوالي 4 سنوات ضوئية، وهي مسافة تتطلب آلاف السنين للوصول إليها باستخدام تكنولوجيا الصواريخ الحالية.
|
| 54 |
+
تجرى حالياً أبحاث حول محركات الدفع الضوئي والمركبات النانوية فائقة السرعة التي قد تمهد الطريق لإرسال أولى المسابر البشرية إلى تلك العوالم البعيدة خلال هذا القرن.""",
|
| 55 |
+
category="علوم وفضاء",
|
| 56 |
+
author="م. رامي كمال",
|
| 57 |
+
image="https://images.unsplash.com/photo-1451187580459-43490279c0fa?w=600&auto=format&fit=crop&q=60",
|
| 58 |
+
summary="يناقش المقال التحديات والآمال المتعلقة بالسفر لاستكشاف الكواكب البعيدة الصالحة للحياة خارج نظامنا الشمسي والتكنولوجيا المستقبلية المقترحة."
|
| 59 |
+
)
|
| 60 |
+
]
|
| 61 |
+
|
| 62 |
+
db.add_all(articles)
|
| 63 |
+
db.commit()
|
| 64 |
+
print("Database seeded successfully with test data!")
|
| 65 |
+
|
| 66 |
+
except Exception as e:
|
| 67 |
+
print(f"Error seeding database: {e}")
|
| 68 |
+
db.rollback()
|
| 69 |
+
finally:
|
| 70 |
+
db.close()
|
| 71 |
+
|
| 72 |
+
if __name__ == "__main__":
|
| 73 |
+
seed_db()
|