Sitemap Image Extractor (image:image namespace)

Parse any public sitemap.xml (or sitemap index) and emit one dataset row per <image:image> extension entry with loc, title, caption, geo_location, license. Sitemap-index BFS-followed up to maxSitemaps (default 25, max 100). Pairs with the public sitemap-delta-monitor and sitemap-lastmod-window Actors.

• Inputs: sitemapUrl (public sitemap URL) or xmlText (inline sitemap XML), maxSitemaps (BFS cap, default 25, max 100), timeoutSec (per-fetch, default 20, max 120).
• Per-row output:
  • _kind: "image" — one row per <image:image> block found, with pageUrl, image (the image's own loc), title, caption, geoLocation, license, and sourceSitemap.
• Output (SUMMARY): ok, sitemapsFetched, urlsScanned, imagesFound, totalRows.

Quick start

apify call stellar_ballet_0bu/sitemap-image-extractor -i '{
  "sitemapUrl": "https://www.google.com/sitemap.xml",
  "maxSitemaps": 5
}'

The default dataset will have one item per <image:image> entry found across the sitemaps. SUMMARY.imagesFound returns the total image count.

Inputs

• sitemapUrl — public URL of a sitemap.xml or sitemap index. Used only if xmlText is empty.
• xmlText — inline sitemap XML string. Useful for offline test fixtures and crawlers that have the XML in hand.
• maxSitemaps — BFS cap on how many child sitemaps to follow from a sitemap index. Default 25, max 100.
• timeoutSec — per-fetch timeout in seconds. Default 20, max 120.

Use cases

• Image SEO audit: count and surface every image declared in a publisher's sitemap (Google Image Search, Bing Image Search).
• Content migration QA: confirm the new site still declares the same set of image URLs as the old site.
• Asset license compliance: pull the license field for every image declared in the sitemap.
• News/publisher inventory: build a CSV/JSON view of all images the publisher has indexed in their sitemap.
• RAG image indexing: produce a clean list of image URLs for downstream image-to-text or visual embedding pipelines.

API notes

• Pure XML parser + simple HTTP fetcher. No browser, no JS execution, no anti-bot circumvention. Public URLs only.
• Sitemap-index BFS: if the root sitemap is an index, the Actor fetches up to maxSitemaps child sitemaps and merges their <image:image> entries into the same output stream.
• Attribute namespace is handled — <image:image> with the xmlns:image="http://www.google.com/schemas/sitemap-image/1.1" namespace is the standard.
• Both xmlText and sitemapUrl work; if both are set, xmlText wins.
• XML entities (&, <, etc.) are decoded.
• Fetch failures are recorded as _kind: "image" rows with error: "fetch failed: ..." and a null image — these are NOT counted in imagesFound.

Pricing (Pay per event)

Event	Description	USD
apify-actor-start	Built-in start charge, per Apify platform usage	per pricing tier
image_extracted	One charge per <image:image> block actually emitted. Fetch-error rows and zero-image sitemaps cost nothing.	$0.0001

currentPricingInfo activates on 2026-07-22T00:00:00Z. The Actor's code calls Actor.charge({ eventName: 'image_extracted', count: imagesFound }) only for actually-emitted image rows.

Cost expectations

A typical news publisher's sitemap with 50k URLs and 30k image entries will cost roughly $3 of usage credits on top of the standard start charge. Use maxSitemaps to cap the BFS depth if you only need a sample.

FAQ

Does it follow robots.txt? No — the Actor is a sitemap parser, not a crawler. It will fetch whatever URL you give it, up to maxSitemaps children. Use maxSitemaps to scope BFS for large publisher sites.

What about <image:thumbnail> or other image extensions? Only the image:image namespace is supported (the Google/Bing sitemap-image extension). Other extensions are ignored.

Does it work with robots.txt-discovered sitemaps? Pass the sitemap URL directly. If you have multiple sitemaps, run the Actor N times or chain via a wrapper.

Why are some rows missing title/caption/license? Those are optional fields in the spec. The Actor emits null for missing fields, never synthesised values.

What happens with non-XML responses? The parser will produce zero image rows and log a parse failed: ... row. Check SUMMARY.urlsScanned — if it's 0, the URL probably wasn't a sitemap.

Can I get a JSON list of unique image URLs only? No — the Actor emits one row per (URL, image) pair. Run the dataset through csv-dedupe-normalizer to dedupe by the image column.

Limits

• BFS capped at maxSitemaps (default 25, max 100). For very large publishers with thousands of child sitemaps, run multiple invocations.
• Per-fetch timeoutSec (default 20, max 120) — long-running fetches will be aborted.
• XML only — sitemap.txt, RSS, Atom, RDF are not supported.

Related Actors

• sitemap-delta-monitor (public) — sitemap URL deltas, before/after, lastmod changes
• sitemap-lastmod-window — filter any sitemap by <lastmod> date window
• sitemap-news-sitemap-extractor — Google News sitemap parser (<news:news> namespace)
• sitemap-priority-changefreq-analyzer — <priority> and <changefreq> distribution analysis
• robots-sitemap-feed-discovery — discover sitemaps and feeds from robots.txt
• product-feed-delta-monitor — for product/feed deltas, not sitemaps

Support

• Source: hidden (paid Actor pattern; not exposed by default).
• Issues / feature requests: open a thread on the public Apify Store Actor page (button on the right of the Store listing) or message the maintainer through the Apify Console.
• For paid-pilot bulk runs (large publisher audits, scheduled re-runs, custom output formats), contact the maintainer via the Apify Console.

License

MIT-style Apify Actor Source-Available License. You may use the public Actor output and integrate the public results; you may not re-host the Actor source or re-publish it as a competing Actor.