SCRAPE_BY_CSS_PATH

By default, SCRAPE_BY_CSS_PATH returns the visible text content of the matched element, which is what a user sees on the page. When you specify an attribute parameter, it extracts the value of that HTML attribute instead. For example, an anchor tag <a href="https://example.com" class="link">Click Here</a> would return "Click Here" without an attribute, but would return "https://example.com" if you set the attribute to "href". Common attributes include "href" for links, "src" for images, "alt" for image descriptions, "content" for meta tags, and any "data-*" custom attributes.

Enable renderJs when the content you want to scrape is loaded dynamically via JavaScript. Signs that you need it include: the function returns empty results or "No matches found" even though you can see the content in your browser; the website is a single-page application built with React, Vue, Angular, or similar frameworks; the content loads after scrolling or after a delay; or the page shows a loading spinner before content appears. Standard mode is faster and cheaper, so try without renderJs first and only enable it if needed.

Each call to SCRAPE_BY_CSS_PATH targets one CSS selector, but that selector can match multiple elements of the same type (all returned as rows). To scrape different types of elements (e.g., both titles and prices), make separate function calls with different selectors. You can combine CSS selectors with commas to match multiple element types in one call, for example ".title, .price", but the results will be interleaved. For structured data extraction where you need to keep titles paired with their prices, use separate columns with one selector per column.

The function returns up to 500 matched elements per call. If the page contains more than 500 matching elements, only the first 500 are returned. For most use cases this is more than sufficient. If you need to scrape pages with extremely large numbers of elements, consider using more specific CSS selectors to narrow down the results, or use pagination parameters if the target website supports them.

SCRAPE_BY_CSS_PATH accesses pages as an anonymous visitor without cookies or session data. It cannot scrape content behind login walls, paywalls, or authentication barriers. The scraper does not support sending cookies, headers, or credentials. For authenticated content, you would need to find a public API for that service or use an alternative data source. Some websites offer public JSON APIs or RSS feeds that may provide the data you need without authentication.

The scraper makes HTTP requests similar to a web browser and does not automatically check or enforce robots.txt rules. However, you should review and respect the target website's robots.txt file and terms of service before scraping. Excessive scraping can lead to your IP being blocked. Use the function responsibly, avoid scraping at high frequency, and respect rate limits. Unlimited Sheets applies its own rate limiting to prevent abuse.

SCRAPE_BY_CSS_PATH is a standalone function that replaces the need for IMPORTXML or IMPORTHTML in many cases. Unlike IMPORTXML, it uses CSS selectors (which are more intuitive for most users) rather than XPath expressions. Unlike IMPORTHTML, it can target any element on the page, not just tables and lists. You can use SCRAPE_BY_CSS_PATH alongside native Google Sheets functions in the same spreadsheet without conflicts.