> ## Content Index > Fetch the complete content index at: https://serpapi.com/blog/llms.txt > Use this file to discover other available public pages before exploring further. # Scrape Google Carousel Results with Python - URL: https://serpapi.com/blog/scrape-google-carousel-results-with-python/ - Published: 2021-10-22T06:36:49.000Z - Updated: 2023-05-01T20:05:32.000Z - Description: This blog post will show how to scrape the title, thumbnail, link, and extensions from Google Organic Carousel Results results using Python. - Author: Dmitriy Zub - Tags: Web Scraping, Google, Python ## What will be scraped ![image](https://user-images.githubusercontent.com/78694043/159895497-a4fbe5b1-427b-4398-8250-0aa6955cdb47.png) ## Prerequisites **Install libraries**: ```python pip install requests parsel google-search-results ``` **Basic knowledge scraping with CSS selectors** CSS selectors declare which part of the markup a style applies to, thus allowing you to extract data from matching tags and attributes. If you haven't scraped with CSS selectors, there's a dedicated blog post of mine about [how to use CSS selectors when web-scraping](https://serpapi.com/blog/web-scraping-with-css-selectors-using-python/) that covers what it is, pros and cons, and why they matter from a web-scraping perspective. 📌 Note: Only such layout will be covered in this blog post. There are at least 3 different Carousel results. ## Using [Google Top Carousel API](https://serpapi.com/top-carousel) SerpApi is a paid API with a free plan which allows end-user to forget about figuring out how to bypass blocks from search entities and focus on which data to extract. This section is to compare the DIY example solution below to our API. ```python from serpapi import GoogleSearch import json def serpapi_get_top_carousel(): params = { "api_key": "...", # https://serpapi.com/manage-api-key "engine": "google", # search engine "q": "dune actors", # search query "hl": "en", # language "gl": "us" # country } search = GoogleSearch(params) results = search.get_dict() for result in results['knowledge_graph']['cast']: print(json.dumps(result, indent=2)) serpapi_get_top_carousel() ``` Part of the output: ```json { "name": "Timothée Chalamet", "extensions": [ "Paul Atreides" ], "link": "https://www.google.com/search?hl=en&gl=us&q=Timoth%C3%A9e+Chalamet&stick=H4sIAAAAAAAAAONgFuLVT9c3NEzLqko2ii8xUOLSz9U3KDDKM0wr0BLKTrbST8vMyQUTVsmJxSWPGJcycgu8_HFPWGo246Q1J68xTmHkwqJOyJCLzTWvJLOkUkhQip8L1RIjEahAtll2hpFZXqHAwmWzGJWcjUx2XZp2jk1P8FkoA0Ndb4iDkiLnFCHrhswn7-wFXd__299ywsBBgkWBQYPB8JElq8P6KYwHtBgOMDI17VtxiI2Fg1GAwYpJg6mKiYOFZxGrUEhmbn5JxuGVqQrOGYk5ibmpJRPYGAHILgFT8gAAAA&sa=X&ved=2ahUKEwiMxLi-ksXzAhUAl2oFHf88AN0Q-BZ6BAgBEDQ", "image": "https://serpapi.com/searches/6165a3dcfa86759a4fa42ba4/images/94afec67f82aa614bb572a123ec09cf051cf10bde8e0bc8025daf21915c49798.jpeg" } ... other results ``` ## Full DIY Code ```python import requests, lxml, re, json from parsel import Selector # https://docs.python-requests.org/en/master/user/quickstart/#custom-headers headers = { "User-agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/99.0.4844.83 Safari/537.36" } params = { "q": "dune actors", # search query "gl": "us" # country to search from } def parsel_get_top_carousel(): html = requests.get('https://www.google.com/search', headers=headers, params=params) selector = Selector(text=html.text) carousel_name = selector.css(".yKMVIe::text").get() all_script_tags = selector.css("script::text").getall() data = {f"{carousel_name}": []} decoded_thumbnails = [] for _id in selector.css("img.d7ENZc::attr(id)").getall(): # https://regex101.com/r/YGtoJn/1 thumbnails = re.findall(r"var\s?s=\'([^']+)\'\;var\s?ii\=\['{_id}'\];".format(_id=_id), str(all_script_tags)) thumbnail = [ bytes(bytes(img, "ascii").decode("unicode-escape"), "ascii").decode("unicode-escape") for img in thumbnails ] decoded_thumbnails.append("".join(thumbnail)) for result, image in zip(selector.css('.QjXCXd.X8kvh'), decoded_thumbnails): title = result.css(".JjtOHd::text").get() link = f"https://www.google.com{result.css('.QjXCXd div a::attr(href)').get()}" extensions = result.css(".ellip.AqEFvb::text").getall() if title and link and extensions is not None: data[carousel_name].append({ "title": title, "link": link, "extensions": extensions, "thumbnail": image }) print(json.dumps(data, indent=2, ensure_ascii=False)) parsel_get_top_carousel() ``` ### Code Explanation #### Thumbnail extraction ![image](https://user-images.githubusercontent.com/78694043/159909214-de29333f-d717-4daf-be8b-ae4512fd19f3.png) Parsing thumbnails from `img.d7ENZc` CSS selector to grab `src` attribute will bring a 1x1 placeholder, instead of actual thumbnail. Thumbnails are located in the `