E-Commerce Product Image Scraper
Budget: ₹600 – ₹1,500 INR
I need a reliable script that can harvest every visual asset tied to a product on typical e-commerce sites. The job covers three image sizes—product thumbnails, full-size shots, and in-page detail images—and for each file I also have to capture practical parameters such as the original URL, width, height, and any available alt or title text.
Python is my preferred stack, and I’m already familiar with Scrapy, BeautifulSoup, Selenium, or Playwright, so feel free to lean on whichever of those best handles lazy-loaded images, infinite scroll, and pagination. Respect robots.txt where possible, but the solution still has to navigate basic anti-scraping measures like dynamic class names or background image CSS.
Deliverables
• Clean, well-commented source code
• A requirements.txt / environment.yml for painless setup
• A short README walking through execution and expected CLI arguments (target URL list, output folder, etc.)
• Sample output: downloaded images plus a CSV or JSON file containing URLs, dimensions, and any other captured attributes
Acceptance criteria: running the script against two sample e-commerce domains should populate the dataset with all three image types, and reported dimensions must match those obtained by a manual check.
Python is my preferred stack, and I’m already familiar with Scrapy, BeautifulSoup, Selenium, or Playwright, so feel free to lean on whichever of those best handles lazy-loaded images, infinite scroll, and pagination. Respect robots.txt where possible, but the solution still has to navigate basic anti-scraping measures like dynamic class names or background image CSS.
Deliverables
• Clean, well-commented source code
• A requirements.txt / environment.yml for painless setup
• A short README walking through execution and expected CLI arguments (target URL list, output folder, etc.)
• Sample output: downloaded images plus a CSV or JSON file containing URLs, dimensions, and any other captured attributes
Acceptance criteria: running the script against two sample e-commerce domains should populate the dataset with all three image types, and reported dimensions must match those obtained by a manual check.