6.3 Link Checker with relative links

[kc16079626]

We have a case where we have a domain that has content from AEM and other applications. So, there are times where an author will want to put in a relative link to content on the domain that isn't in AEM. The link checker appears to be thinking these are invalid and removing the link from the page.

I did see there was a way to disable the link checker, and that there is an option to add x-cq-linkchecker="skip" or x-cq-linkchecker="valid" to the <a tag.

I'm wondering, is there is a way to keep the link checker enabled, but have it not disable/remove the link? That way we could check if there are broken links, but the links would still be rendering on the site.

[Kunwar]

How are external links relative ? Are these static files hosted on Apache or webserver and you'd just display them if an end user clicks on it ?

if you can identify a pattern say all these files are under /mysite/docs/ on Apache you can adjust the 'Link Check Override Patterns' in the LinkChecker service to let linkchecker know not to strip them out.

[kc16079626]

There are several other appservers that are running apps on the same domain.

There are over 50 applications, and they don't follow a pattern.

Instead of identifying a pattern for the apps, I was thinking of identifying a pattern for AEM.

I was planning on adding ^/[^content] to the LinkChecker service override pattern, I figured that way it will only do verification for relative links that exist in AEM, and it won't check for external relative links.

[jeff_foxx · Accepted solution]

In the Day CQ Link Service OSGi configuration, there is an option to input Link Check Override Patterns.

You can put in the URL patterns that match your non-AEM content and it would be preserved.  This is for the external link checker.

You can leave the internal link checker active and it will flag the links as broken, but activating them will preserve the links.

[jeff_foxx]

I don't think that will work, because the override patterns match the URL in question.  It doesn't mean that it doesn't scan that content.