🌸 Perfume Scraper

A Python-based web scraping project designed to extract structured perfume product data from Liliome.com.
The scraper collects brand and product information and supports multiple database backends, allowing the same data pipeline to be stored in relational and NoSQL databases.

---

📌 Features

✔ Robust HTTP session

• Uses requests.Session with retry logic
• Handles connection failures gracefully (safe_get())

✔ Web scraping

• Extracts:
  • Brand name
  • English title
  • Persian title
  • Old price
  • New price
  • Product rating (Point)
  • Photo URL
• Automatically discovers all available brands and their product pages

✔ Pagination handling

• Detects number of pages for each brand using total_pages()

---

🗄️ Supported Databases

The project has been refactored to support multiple storage backends, making it easy to switch between databases:

• SQLite – lightweight local storage
• SQL Server – enterprise relational database
• PostgreSQL – open-source relational database
• MongoDB – NoSQL document-based storage

This design enables comparison between SQL and NoSQL data models using the same scraping logic. Two tables are created automatically:

Brands

Column	Type	Description
Brand_ID	INTEGER	Primary key
Brand_Link	TEXT	URL of brand page
Brand_Name	TEXT	Extracted brand name

Master

Column	Type	Description
ID	INTEGER	Primary key
Brand	TEXT	Brand slug
EnglishName	TEXT	Product English title
Name	TEXT	Product Persian title
Point	FLOAT	Product rating
OldPrice	INTEGER	Old price
NewPrice	INTEGER	New price
Photo	TEXT	Image URL

---

🛠 Technologies Used

• Python 3
• Requests
• BeautifulSoup4
• SQLite3
• SQL Server
• PostgreSQL
• MongoDB
• Retry & Timeout handling
• Regex for price cleanup

---

📁 Project Structure

Perfume_Scraper/
│
├── assets/
│ └── mongodb_brands.png
│ └── mongodb_master.png
│ └── postgres_brands.png
│ └── postgres_master.png
│ └── sqlite_brands.png
│ └── sqlite_master.png
│ └── sqlserver_brands.png
│ └── sqlserver_master.png
│
├── db/
│ └── Perfume.db # Automatically created database for SQLite
│
├── Scraper_MongoDB.py
├── Scraper_MongoDB_Safe.py
├── Scraper_PostgreSQL.py
├── Scraper_PostgreSQL_Safe.py
├── Scraper_SQL.py
├── Scraper_SQL_Safe.py
├── Scraper_SQLite.py
├── Scraper_SQLite_Safe.py
├── README.md
├── requirements.txt

---

🚀 How It Works

1️⃣ Load Liliome brand list

The script visits:

https://liliome.com/برندها-عطر-ادکلن-فروشگاه-عطر-لیلیوم

It finds all brand links and stores them in the Brands table.

---

2️⃣ For each brand:

• Detects how many pages of products exist
• Extracts products from each page
• Saves structured data into the Master table

---

▶️ How to Run

1. Clone the repository:

git clone https://github.com/SamiraSiavash/Perfume_Scraper_Multi_DB.git
cd Perfume_Scraper_Multi_DB

2. Install dependencies:

pip install -r requirements.txt

3. Run one of the scrapers:

Scraper_MongoDB.py
Scraper_MongoDB_Safe.py
Scraper_PostgreSQL.py
Scraper_PostgreSQL_Safe.py
Scraper_SQL.py
Scraper_SQL_Safe.py
Scraper_SQLite.py
Scraper_SQLite_Safe.py

---

🖼 Screenshots

SQLite

![Brands Table](assets/sqlite_brands.png)

![Master Table](assets/sqlite_master.png)

---

SQL Server

![Brands Table](assets/sqlserver_brands.png)

![Master Table](assets/sqlserver_master.png)

---

PostgreSQL

![Brands Table](assets/postgres_brands.png)

![Master Table](assets/postgres_master.png)

---

MongoDB

![Brands Collection](assets/mongodb_brands.png)

![Master Collection](assets/mongodb_master.png)

---

📝 Notes

• Adjust CSS selectors depending on website structure.
• Website layouts may change; update selectors accordingly.
• Always follow the target website’s Terms of Service.

---

📄 License

MIT License (optional)

---

✨ Author

Samira Siavash

🔗 GitHub: https://github.com/SamiraSiavash

🔗 LinkedIn: https://linkedin.com/in/samira-siavash