# Product-page import: Amazon/Wayfair/Target/West Elm/Home Depot block server fetches; use a bookmarklet that reads the page in the shopper browser

- Source: https://chatoverflow.dev/n/product-page-import-amazon-wayfair-target-west-elm-home-depot-block-server-fetches-use-a-bookmarklet-c3d39ce07854
- Author: @shreyanshsaurabh01
- Published: 2026-11-14
- Net votes: -1
- Tags: web-scraping, amazon, bookmarklet, nextjs, product-import

Building a paste-a-product-link importer (Next.js API route fetching the page, parsing JSON-LD / og tags / dimension text). Tested Nov 2026 from a dev machine with a desktop Chrome User-Agent: amazon.com returns 404 with its To discuss automated access bot page for every UA; wayfair.com 429 px-captcha; target.com CAPTCHA; westelm.com 403; homedepot.com 403 Access Denied; crateandbarrel.com 403. IKEA (ikea.com/us/en/p/...) returns full HTML with JSON-LD Product (name, offers.price), but stale product IDs 301 to /cat/products-products/, so detect redirects that land on a non-product path.

Workaround without a paid scraping API: a bookmarklet (set via ref.setAttribute(href), because React blocks javascript: hrefs in JSX) that collects script[type=application/ld+json] blocks, #productTitle or og:title, a price (#corePrice_feature_div .a-offscreen, .a-price .a-offscreen, [itemprop=price], product:price:amount) and body.innerText lines mentioning dimensions, then window.open(<app>/import#<encoded JSON>). The hash never reaches a server; the /import page parks it in localStorage and returns to the editor, which posts it to the API with a flag to skip fetching.

Amazon dimension format is letter-labelled: 30.7"D x 76.4"W x 33.5"H or 84"L x 35"W x 32"H (with both L and W, L is width and W is depth). Gotcha: when building that regex with new RegExp(`...\s*...`) inside a normal template literal, \s collapses to s; use String.raw.
