Actually, I can confirm on www.joomla.org
and developer.joomla.org
that there are CSS classes in the search index.
Status | New | ⇒ | Confirmed |
Did we need to re-run the index after the 3.7 update to get that change?
We don't have it on a cron to do regular updates so it relies on the plugins to update the index as we make content updates on the sites. I'm running an update through the CLI for the dev site now (that one takes some time because of the JoomlaCode archive), I'll check the result when that's finished.
thanks
The full reindex has finished on the developer site, span1
is no longer a search term there.
Status | Confirmed | ⇒ | Closed |
Closed_Date | 0000-00-00 00:00:00 | ⇒ | 2017-05-19 15:57:30 |
Closed_By | ⇒ | brianteeman |
Smart Search does remove CSS and JavaScript before parsing the text. It also ignores anything inside
<noscript>
tags and anything in the<head>
and it strips all HTML tags. You can see the code that does it here:https://github.com/joomla/joomla-cms/blob/staging/administrator/components/com_finder/helpers/indexer/parser/html.php
There must be something that is preventing that from happening in your case. Can you examine the HTML of the affected content items carefully and make sure that the markup is valid? Invalid markup is likely to confuse the parser and lead to leakage of class names into the index.