kaxil opened a new pull request, #1500:
URL: https://github.com/apache/airflow-site/pull/1500
## Summary
The blog Atom feed at `/blog/index.xml` has been producing invalid XML since
it was introduced in #1391, reported in #1490. Three separate issues:
**1. Leading whitespace before XML declaration**
Hugo strips `{{/* */}}` template comments but preserves surrounding
newlines. This produced two blank lines before `<?xml?>`, which the XML spec
requires to be at byte 0. Fixed with Hugo's whitespace-trimming markers (`{{-
-}}`).
**2. HTML escaping of XML constructs**
Hugo's `html/template` engine escaped `<?xml` → `<?xml` and `<![CDATA[` →
`<![CDATA[`. Most browsers/readers partially recovered from this, but the
feed was technically invalid. Fixed by piping through `safeHTML`.
**3. Relative URLs instead of absolute**
Since `baseURL = "/"`, `.Permalink` produced relative paths like
`/blog/common-ai-provider/`. The Atom spec requires absolute URIs in `<link>`
and `<id>` elements (this was the original request in #223). Added a
`canonicalBaseURL` site param rather than changing the global `baseURL`, which
could have unintended side effects across the site.
## Verification
```bash
cd landing-pages/site
hugo --quiet
python3 -c "
import xml.etree.ElementTree as ET
tree = ET.parse('public/blog/index.xml')
ns = {'atom': 'http://www.w3.org/2005/Atom'}
entries = tree.findall('.//atom:entry', ns)
print(f'Valid XML with {len(entries)} entries')
link = entries[0].find('atom:link', ns).get('href')
print(f'Entry link: {link}')
"
```
Output:
```
Valid XML with 37 entries
Entry link: https://airflow.apache.org/blog/common-ai-provider/
```
Closes #1490
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
