Web Scraping with Scrapy for Data Extraction

Overview

Web scraping is the process of extracting data from websites and web applications. It is a common technique used by data scientists, researchers, and developers to collect data for analysis, machine learning, and natural language processing.

Some Important Terms

• Items: The main goal in scraping is to extract structured data from unstructured sources, typically web pages. Spiders are used to gather the extracted data and present it as items, which are Python objects that consist of key-value pairs.
• Item Pipeline: The responsibility of the Item Pipeline is to handle the processing of items once the spiders have extracted them. This involves performing tasks such as cleansing, validation, and persistence, such as storing the item in a database.
• Spiders: Spiders are custom classes written by users to parse responses and extract items from them or additional requests to follow.
• Spider Middlewares: Spider middlewares are specialized hooks positioned between the Engine and the Spiders, enabling them to process both the input (responses) and output (items and requests) of the spider.

Architecture Overview

The following diagram shows an overview of the Scrapy architecture with its components and an outline of the data flow inside the system.

Source: Scrapy architecture

The execution engine controls the data flow in Scrapy and goes like this:

• The Engine receives the initial request to initiate the crawling process from the Spider.
• The Engine adds the request to the Scheduler and instructs it to initiate the crawling process.
• The Scheduler provides the next request to the Engine.
• The Engine sends the request to the Downloader, which passes through the Downloader Middleware.
• Once the download is complete, the Downloader generates a response and sends it back to the Engine through the Downloader Middleware.
• The Engine receives the response and forwards it to the Spider for further processing, which involves passing through the Spider Middleware.
• After processing the response, the Spider returns scraped items and new requests (for subsequent crawling) to the Engine, passing through the Spider Middleware.
• After processing the response, the Spider returns scraped items and new requests (for subsequent crawling) to the Engine, passing through the Spider Middleware.
• The process repeats (from step 3) until there are no more requests from the Scheduler.

Demo

In the demo, we will scrap a website containing a list of books. Here we will extract each book’s name, image link, and price using Scrapy.

Link of the website to scrap – All products | Books to Scrape – Sandbox

Example HTML for a single book

We will use CMD/Terminal or Anaconda Prompt for the following.

1. Install Scrapy using – pip install scrapy
2. Then, go to a folder where the project is needed to be created and run the following command

scrapy startproject bookscraper

“bookscraper” is the project name

3. Following is the structure of our project

4. Inside “spiders” folder, we create a file “scarpybooks.py”

5. Inside the file, the following codes

6. At the root, we will create another folder called “items” containing our item classes.

7. Inside “item” folder, we will create “book_item.py” file with the following codes

8. After that we are going to run the following line.

scrapy crawl books -O books.json

This will create a JSON file and store all the extracted data

Conclusion

Scrapy is a powerful and flexible web crawling and scraping framework that enables users to extract structured data from websites and web applications efficiently. Its spider-based crawling architecture, support for asynchronous and concurrent requests, and convenient item pipelines make it a popular tool for data scientists, researchers, and developers. By following best practices for web scraping and using Scrapy, we can collect and process data for analysis, machine learning, and natural language processing.

About CloudThat