Generating a valid feed

Building feed XML safely from application data, validating it, and advertising it so readers can find it.

Never build XML by concatenation

Hand-written string templates are how feeds break. One headline containing an ampersand, an emoji, or a stray tag produces invalid XML. Use a serialiser that escapes text nodes for you.

from xml.etree import ElementTree as ET
from email.utils import format_datetime
from datetime import datetime, timezone

rss = ET.Element("rss", {"version": "2.0"})
ch = ET.SubElement(rss, "channel")
ET.SubElement(ch, "title").text = "Example Dev Blog"
ET.SubElement(ch, "link").text = "https://example.com/"
ET.SubElement(ch, "description").text = "Notes on building web systems."

for post in posts:                       # each post: dict with title/url/id/published/body
    it = ET.SubElement(ch, "item")
    ET.SubElement(it, "title").text = post["title"]
    ET.SubElement(it, "link").text = post["url"]
    ET.SubElement(it, "guid", {"isPermaLink": "false"}).text = post["id"]
    ET.SubElement(it, "pubDate").text = format_datetime(
        post["published"].astimezone(timezone.utc))
    ET.SubElement(it, "description").text = post["body"]   # escaping handled here

ET.ElementTree(rss).write("feed.xml", encoding="utf-8", xml_declaration=True)

const esc = (s) => String(s)
  .replace(/&/g, "&amp;").replace(/</g, "&lt;")
  .replace(/>/g, "&gt;").replace(/"/g, "&quot;");

const item = (p) => [
  "<item>",
  "  <title>" + esc(p.title) + "</title>",
  "  <link>" + esc(p.url) + "</link>",
  '  <guid isPermaLink="false">' + esc(p.id) + "</guid>",
  "  <pubDate>" + p.published.toUTCString() + "</pubDate>",
  "  <description>" + esc(p.summary) + "</description>",
  "</item>"
].join("\n");

Validate and serve correctly

# well-formedness, then structure against the RSS 2.0 schema
xmllint --noout feed.xml
curl -sSL https://www.rssboard.org/files/schema.rng -o rss.rng
xmllint --noout --relaxng rss.rng feed.xml

# check what the reader will actually see
curl -sSI https://example.com/feed.xml | grep -i content-type
curl -s https://example.com/feed.xml | xmllint --format - | head -30

HTTP/1.1 200 OK
Content-Type: application/rss+xml; charset=utf-8
Cache-Control: max-age=300
ETag: "feed-2026-09-18-01"
Last-Modified: Fri, 18 Sep 2026 09:00:00 GMT

Advertise the feed in the document head so browsers and readers discover it, and keep the URL stable forever — subscribers are lost when a feed moves without a redirect.

<link rel="alternate" type="application/rss+xml"
      title="Example Dev Blog" href="https://example.com/feed.xml"/>

Publishing reliably

• Generate the feed in the same build as the pages, so it can never be newer or older than the content.
• Sort items newest first, and cap the list to a fixed number so the file stays small.
• Emit lastBuildDate from the newest item, not from the current time — otherwise the feed looks changed on every regeneration.
• Keep a stable guid derived from the item's canonical URL or a content id, never from a timestamp.
• Serve the feed with a short Cache-Control and an ETag; readers poll often and a 304 costs almost nothing.
• Set a realistic ttl or document your update cadence rather than relying on readers guessing.

FAQ

Related

The RSS 2.0 feed format Atom and how readers consume feeds

Last refreshed 2026-09-18.