Hugo sometimes doesn't process relref / rel correctly and I end up with literally {% ... output

Yes. The entire site.

2 Likes

Guess you can skip the images…if your coding is defensive a site should build without.

It’s not defensive on purpose because I want builds to fail if I have a missing image. That means I either forgot an image or typo’d the file name. In either of those cases it’s an unexpected event.

Do you want the whole site or everything related to rendering 1 complete Markdown page? There’s no value in having 500 unique blog posts, 300 drafts and over 1,000 images right? Also I can skip all of the CSS / JS right since that’s independent of Hugo’s Markdown renderer?

I could provide a repo with my baseof and single blog post layout files along with the partials and shortcodes related to rendering a single blog post. Then quickly generate 500 posts and 300 drafts so the project is at least of similar size, the only exception is I’ll stick with 1 image referenced in all of them?

The objective is to provide a reproducible example. Whatever you provide, you must be able to reproduce the problem on your system before sharing it with us.

1 Like

I will try. In case this post gets locked due to inactivity I will invite both you and Irkode to the private repo on GitHub if I’m able to reproduce it.

I’ll create a tiny script that loops every half a second and then appends the datetime to one of the posts and then does a Hugo build. I’ll then grep for strings that indicate a link is broken. I should be able to run something like that overnight and we’ll see any failures.