PINGDOM_CHECK

The Modern Scrapy Developer's Guide (Part 1): Building Your First Spider


Scrapy can feel daunting. It's a massive, powerful framework, and the documentation can be overwhelming for a newcomer. Where do you even begin?


In this definitive guide, we will walk you through, step-by-step, how to build a real, multi-page crawling spider. You will go from an empty folder to a clean JSON file of structured data in about 15 minutes. We'll use modern, async/await Python and cover project setup, finding selectors, following links (crawling), and saving your data.


What We'll Build


We will build a Scrapy spider that crawls the "Fantasy" category on books.toscrape.com, follows the "Next" button to crawl every page in that category, follows the link for every book, and scrapes the name, price, and URL from all 48 books, saving the result to a clean books.json file.



Here's a preview of our final spider code:


# The final spider we'll build
import scrapy

class BooksSpider(scrapy.Spider):
    name = "books"
    allowed_domains = ["toscrape.com"]

    url: str = "https://books.toscrape.com/catalogue/category/books/fantasy_19/index.html"

    async def start(self):
        yield scrapy.Request(self.url, callback=self.parse_listpage)

    async def parse_listpage(self, response):
        product_urls = response.css("article.product_pod h3 a::attr(href)").getall()
        for url in product_urls:
            yield response.follow(url, callback=self.parse_book)

        next_page_url = response.css("li.next a::attr(href)").get()
        if next_page_url:
            yield response.follow(next_page_url, callback=self.parse_listpage)

    async def parse_book(self, response):
        yield {
            "name": response.css("h1::text").get(),
            "price": response.css("p.price_color::text").get(),
            "url": response.url
        }

On This Page (Table of Contents)

  • Prerequisites & Setup
  • Step 1: Initialize Your Project
  • Step 2: Configure Your Settings
  • Step 3: Finding Our Selectors (with scrapy shell)
  • Step 4: Building the Spider (Crawling & Parsing)
  • Step 5: Running The Spider & Saving Data
  • Conclusion & Next Steps

Prerequisites & Setup


Before we start, you'll need Python 3.x installed. We'll also be using a virtual environment to keep our dependencies clean. You can use standard pip or a modern package manager like uv.

First, let's create a project folder and activate a virtual environment.


# Create a new folder
mkdir scrapy_project
cd scrapy_project

# Option 1: Using standard pip + venv
python -m venv .venv
source .venv/bin/activate  # On Windows, use: .venv\Scripts\activate

# Option 2: Using uv (a fast, modern alternative)
uv init

Now, let's install Scrapy.


# Option 1: Using pip
pip install scrapy

# Option 2: Using uv
uv add scrapy
source .venv/bin/activate

Step 1: Initialize Your Project


With Scrapy installed, we can use its built-in command-line tools to generate our project boilerplate.

First, create the project itself.


# The 'scrapy startproject' command creates the project structure
# The '.' tells it to use the current folder
scrapy startproject tutorial .

You'll see a tutorial folder and a scrapy.cfg file appear. This folder contains all your project's logic.

Next, we'll generate our first spider.


# The 'genspider' command creates a new spider file
# Usage: scrapy genspider <spider_name> <allowed_domain>
scrapy genspider books toscrape.com

If you look in tutorial/spiders/, you'll now see books.py. This is where we'll write our code.


Step 2: Configure Your Settings


Before we write our spider, let's quickly adjust two settings in tutorial/settings.py.


ROBOTSTXT_OBEY

By default, Scrapy respects robots.txt files. This is a good practice, but our test site (toscrape.com) doesn't have one, which can cause a 404 error in our logs. We'll turn it off for this tutorial.


# tutorial/settings.py

# Find this line and change it to False
ROBOTSTXT_OBEY = False

Concurrency

Scrapy is polite by default and runs slowly. Since toscrape.com is a test site built for scraping, we can speed it up.


# tutorial/settings.py

# Uncomment or add these lines
CONCURRENT_REQUESTS = 16
DOWNLOAD_DELAY = 0

Warning: These settings are for this test site only. When scraping in the wild, you must be mindful of your target site and use respectful DOWNLOAD_DELAY and CONCURRENT_REQUESTS values.


Step 3: Finding Our Selectors (with scrapy shell)


To scrape a site, we need to tell Scrapy what data to get. We do this with CSS selectors. The scrapy shell is the best tool for this.

Let's launch the shell on our target category page:



scrapy shell https://books.toscrape.com/catalogue/category/books/fantasy_19/index.html

This will download the page and give you an interactive shell with a response object.

Let's find the data we need:



Find all Book Links:

By inspecting the page, we see each book is in an article.product_pod. The link is inside an h3.


# In scrapy shell:
>>> response.css("article.product_pod h3 a::attr(href)").getall()
[
  '../../../../the-host_979/index.html',
  '../../../../the-hunted_978/index.html',
  ...
]

Find the "Next" Page Link:

At the bottom, we find the "Next" button in an li.next.


# In scrapy shell:
>>> response.css("li.next a::attr(href)").get()
'page-2.html'

Find the Book Data (on a product page):

Finally, let's open a shell on a product page to find the selectors for our data.


# Exit the shell and open a new one:
scrapy shell https://books.toscrape.com/catalogue/the-host_979/index.html

# In scrapy shell:
>>> response.css("h1::text").get()
'The Host'

>>> response.css("p.price_color::text").get()
'£25.82'

Step 4: Building the Spider (Crawling & Parsing)


Now, let's open tutorial/spiders/books.py and write our spider. We'll use the user's provided code, as it's a clean, final version.

Delete the boilerplate in books.py and replace it with this:


# tutorial/spiders/books.py

import scrapy

class BooksSpider(scrapy.Spider):
    name = "books"
    allowed_domains = ["toscrape.com"]

    # This is our starting URL (the first page of the Fantasy category)
    url: str = "https://books.toscrape.com/catalogue/category/books/fantasy_19/index.html"

    # This is the modern, async version of 'start_requests'
    async def start(self):
        # We yield our first request, sending the response to 'parse_listpage'
        yield scrapy.Request(self.url, callback=self.parse_listpage)

    # This function handles the *category page*
    async def parse_listpage(self, response):

        # 1. Get all product URLs using the selector we found
        product_urls = response.css("article.product_pod h3 a::attr(href)").getall()

        # 2. For each product URL, follow it and send the response to 'parse_book'
        for url in product_urls:
            yield response.follow(url, callback=self.parse_book)

        # 3. Find the 'Next' page URL
        next_page_url = response.css("li.next a::attr(href)").get()

        # 4. If a 'Next' page exists, follow it and send the response
        if next_page_url:
            yield response.follow(next_page_url, callback=self.parse_listpage)

    # This function handles the *product page*
    async def parse_book(self, response):

        # We yield a dictionary of the data we want
        yield {
            "name": response.css("h1::text").get(),
            "price": response.css("p.price_color::text").get(),
            "url": response.url
        }

Step 5: Running The Spider & Saving Data


We're ready to run. Go to your terminal (at the project root) and run:



scrapy crawl books

You'll see Scrapy start up, and in the logs, you'll see all 48 items being scraped!

But we want to save this data. Scrapy has a built-in "Feed Exporter" that makes this easy. We just use the -o (output) flag.



scrapy crawl books -o books.json

This will run the spider again, but this time, you'll see a new books.json file in your project root, containing all 48 items, perfectly structured.


Conclusion & Next Steps


Today you built a powerful, modern, async Scrapy crawler. You learned how to set up a project, find selectors, follow links, and handle pagination.

This is just the starting block.


What's Next? Join the Community.


  • 💬 TALK: Stuck on this Scrapy code? Ask the maintainers and 5k+ devs in our Discord.
  • ▶️ WATCH: This post was based on our video! Watch the full walkthrough on our YouTube channel.
  • 📩 READ: Want more? In Part 2, we'll cover Scrapy Items and Pipelines. Get the Extract newsletter so you don't miss it.