Amazon scrapers silently return wrong currencies, prices, and blocked pages
A developer spent a week systematically testing Amazon scraping across 18 marketplaces and documented several silent failure modes that corrupt price data. Amazon determines the currency based on the requester's IP address, meaning the same product can return euros, dong, or dollars depending on proxy location, while most parsers discard the currency symbol entirely. Using an i18n-prefs cookie to force a specific currency does not fetch native prices but instead applies a single exchange rate conversion, producing misleading results. Anti-bot interstitials were found to return HTTP 200 or 202 status codes with small HTML pages, causing standard retry logic to treat blocked requests as successful. Additional pitfalls include locale-specific decimal separators, trailing currency symbols in markets like Poland and Sweden, and a fullwidth yen sign on Amazon Japan that silently causes all prices to go unrecognized.
This is an AI-generated summary. ShortSingh links to the original source for the complete article.
Discussion (0)
Log in to join the discussion and vote.
Log in