The core mechanism of an image scraper involves three steps: fetching a webpage, parsing its HTML to extract image URLs, and downloading each image file. Below are two practical examples—one for a general gallery site and another for a specific article. These scripts rely on Python's urllib.request and re modules and were originally written in 2020; the URLs may no longer be active.

1. Scrapign images from a gallery page

import urllib.request
import re

class ImageScraper:
    def __init__(self, base_url, user_agent):
        self.base_url = base_url
        self.user_agent = user_agent

    def fetch_page(self):
        req = urllib.request.Request(self.base_url)
        req.add_header("User-Agent", self.user_agent)
        response = urllib.request.urlopen(req)
        return response.read()

    def extract_image_urls(self):
        raw_urls = []
        pattern = re.compile(b"https://pit1\\.maozhew\\.com/forum/.{32}\\.(?:jpg|jpeg|png)")
        matches = pattern.findall(self.fetch_page())
        for match in matches:
            raw_urls.append(match.decode("utf-8"))
        return raw_urls

    def download_images(self):
        counter = 0
        for img_url in self.extract_image_urls():
            counter += 1
            self.base_url = img_url  # reuse fetch with new URL
            with open(f"{counter}.jpg", "wb") as f:
                f.write(self.fetch_page())

scraper = ImageScraper(
    "https://www.topit.pro/",
    "Mozilla/5.0 (Windows NT 10.0; Win64; x64) "
    "AppleWebKit/537.36 (KHTML, like Gecko) Chrome/86.0.4240.198 Safari/537.36"
)
scraper.download_images()

The code defines a class that bundles fetching, parsing, and downloading. The regular expression looks for image URLs matching the gallery's pattern.

2. Scraping images from a news article

import urllib.request
import re

class ImageScraper:
    def __init__(self, base_url, user_agent):
        self.base_url = base_url
        self.user_agent = user_agent

    def fetch_page(self):
        req = urllib.request.Request(self.base_url)
        req.add_header("User-Agent", self.user_agent)
        response = urllib.request.urlopen(req)
        return response.read()

    def extract_image_urls(self):
        raw_urls = []
        pattern = re.compile(b"http://5b0988e595225\\.cdn\\.sohucs\\.com/images/.{41}\\.(?:jpg|jpeg|png)")
        matches = pattern.findall(self.fetch_page())
        for match in matches:
            raw_urls.append(match.decode("utf-8"))
        return raw_urls

    def download_images(self):
        counter = 0
        for img_url in self.extract_image_urls():
            counter += 1
            self.base_url = img_url
            with open(f"{counter}.jpg", "wb") as f:
                f.write(self.fetch_page())

scraper = ImageScraper(
    "https://www.sohu.com/a/255373346_721493",
    "Mozilla/5.0 (Windows NT 10.0; Win64; x64) "
    "AppleWebKit/537.36 (KHTML, like Gecko) Chrome/86.0.4240.198 Safari/537.36"
)
scraper.download_images()

This version adjusts the regex pattern to match the article's image CDN. Both scripts follow the same structure: request the page, extract URLs with regex, and write each image to a local file.