Practice · Knowledge
How do I check which rel values my links carry?
Three zoom levels, three tools: the element inspector for a single link, the page source for one page, a crawler for the whole site – each answers a different question.
The classic pitfall first: judge the rendered front end, never the editor view – editors show the stored HTML, while attributes are typically added at render time.
Verify after every relevant change: plugin updates, template switches, new extensions that output links.
The three levels in practice
Level one, the single link: right-click → inspect element opens the browser's live view of the DOM; the a-tag shows its rel and target values as the browser actually sees them – the fastest answer to “did my rule catch this link?”. This is also where the editor pitfall resolves: Joomla content plugins (ours included) process article HTML at render time, so the article editor legitimately shows the link without attributes while the front end carries them – different layers, both telling the truth about their layer. Level two, the page: view source (or Ctrl+U) plus a search for rel= audits every link of one page at a glance – useful after configuring, to confirm the pattern (externals marked, internals untouched, the bypass exceptions intact) across a real article; the demo cases of this site double as the expected-output reference. Level three, the site: a desktop crawler walks every page and exports all links with their rel values and targets into a filterable table – the level where you find the forgotten corners: the old landing page from 2019, the template's footer partner link, the widget module nobody audited. Filters worth running there: external links without any rel marking (candidates for the net), external links with follow (should exactly match your deliberate exceptions), and internal links with nofollow (should be zero – the neighbouring question explains why). Cadence: the three levels after every plugin or template change, the crawl additionally once or twice a year – attributes drift silently, and silent drift is what routines are for.
Key facts
- Level 1 – inspector: the live DOM answer for a single link's rel/target.
- Pitfall: the editor shows stored HTML – attributes are added at render time; judge the front end.
- Level 2 – page source + search for rel=: one page audited at a glance.
- Level 3 – site crawler: export all links, filter for unmarked externals, follow externals, internal nofollow.
- Cadence: after every plugin/template change, plus a yearly full crawl.