XML Folio / Notebook

Give the reading copy its own address.

A useful migration preserves the machine endpoint and makes the human path easy to test.

XML Folio editorial notes · Reviewed October 15, 2026

1. Separate the two responses

Keep the feed or sitemap at its existing XML URL with its existing appropriate content type. Put the generated HTML at a separate address such as /feed-reader.html. Link people to that page when they want to read, and retain an explicit XML subscription or source link. Do not redirect a machine consumer to the HTML just because it looks better in a browser.

XML Folio’s feed exports include an alternate-feed discovery link when you supply the source URL. Sitemap exports have a source link but do not present the HTML itself as a sitemap. The file is self-contained, so you do not need to deploy assets from this workbench or keep this domain available.

2. Review content before indexing

The exported page contains a noindex,follow meta tag by default. A duplicate reading copy may not deserve its own search result. If the page provides a distinct, useful reading experience and you want it indexed, review its title, content, canonical strategy and relationship to the original pages before changing that default.

Inspect the field manifest first. It contains titles, identifiers and source URLs, including query strings. Do not publish a private feed or report just because its XML parsed successfully. The manifest is intended for your release review; there is no requirement to host it publicly.

3. Make freshness explicit

A static export records its generation time and keeps that snapshot until you regenerate it. The optional refresh button adds a different behavior: after a click, the hosted page requests the configured source URL and updates its displayed records. That URL must share the page’s scheme, hostname and port.

The request omits credentials, refuses redirects, requires an XML response, stops after eight seconds and caps the body at 1 MiB. The same parser and field limits apply. If parsing fails or the profile changes, the last displayed copy remains. A successful refresh changes only the current tab; it does not rewrite the saved HTML file on your server. Its heading remains the chosen publication title.

When opened from disk or hosted on another origin, refresh is disabled and the snapshot still works. The generator does not fetch arbitrary URLs through its Cloudflare Worker. If your feed requires authentication, redirects or a server-side transformation, implement that in your own controlled publishing pipeline.

4. Test the actual human page

Open the deployed HTML with JavaScript disabled and verify titles, order, record count and several encoded URLs. Then test the optional refresh path separately. Keep screenshots of the same content before and after the change; a successful status code does not prove that the page is readable.

5. Follow the optional email path

If your own service emails feed-subscription instructions or a report link, send a non-sensitive staging message. Confirm the human link reaches the HTML reading copy and the subscription link still points to XML. This step is unnecessary for a sitemap with no email workflow.

WebTraceCheck, CleanTempMail and XML Folio share an operator. Both links are optional and include site/placement attribution. No XML or manifest content is attached to those links. Use your own browser and test inbox if you prefer.