Sitemap Extractor & Validator

How this tool computes its result

Accepts a sitemap URL (fetched client-side, which can fail on CORS-restricted hosts) or raw pasted XML, and parses it with DOMParser, checking for a <parsererror> node to catch malformed markup. It reads whichever of <url> or <sitemap> elements exists (never both, and never recursively) to determine urlset vs sitemapindex type. For each node, a missing <loc> child is counted toward missingLocCount and the entry is skipped entirely from the URL list; present loc values are checked for an "http" prefix and run through `new URL()`, with failures added to invalidUrls. Duplicates are any URL string whose first occurrence index differs from its current index in the list.

Worked example

Pasting a urlset with 250 <url> entries where 5 loc values repeat exactly and 2 entries have no <loc> child at all: Total URLs reads 250 (loc-bearing entries only), Missing <loc> reads 2, Duplicates reads 5, and a separately computed Unique URLs stat reads 245.

When not to use this tool

This is a client-side XML structure check only -- it never verifies that listed URLs actually return 200, does not decompress .xml.gz files, and a CORS-blocked fetch fails with a generic "unable to fetch, paste XML below" message rather than surfacing the real network error.

Common mistakes

• - Pasting a sitemap index and expecting the tool to follow each child <sitemap> link and total their URLs -- it only reads whichever top-level node type is present in what you pasted, it never recurses.
• - Assuming Total URLs and Unique URLs measure the same thing -- Total counts every loc-bearing entry including duplicates, Unique is the deduplicated count, so the two diverging is expected behavior, not a bug.
• - Fetching a sitemap URL that redirects to a different domain which itself blocks CORS -- the fetch can fail on the redirect target even though the original URL loads fine in a normal browser tab.

Related tools