Failures on not-yet-published blog posts' meta tags

[grahamc]

Our CI pipeline builds our static site and runs lychee on it. Most of our pages have a canonical rel , and an og:url, among others, that point to the final production domain:

<link rel="canonical" href="https://determinate.systems/posts/xxxx">
<meta property="og:url" content="https://determinate.systems/posts/xxxx">

and to solve this, we've taken the unfortunate step of blanket-excluding our own domain from our checks. This seems... sub-optimal, and I think I'd like it better if I could explicitly exclude tags like these?

[mre]

Yup. Although I wonder how such an option might look like. --exclude-tags rel,property would not be enough, as you're targeting rel="canonical" and property="og:url" specifically.
Open for suggestions on the syntax here.

The alternative would be to count these links as verbatim URLs and exclude them by default, unless --include-verbatim is set. Alternatively, we could have our own flag like --exclude-meta-urls or so, but I'm not sure if it warrants a separate CLI flag. It's quite specific to HTML as well.