2021-01-24 UTC
#
Zegnat I have a couple of site crawls (wget) from which I need to extract specific contents on a page-by-page basis, [schmarty]. And probably clean some in the middle elements, brs, etc. I think it will be easier to do scripting that than wrangling a cli tool into doing that, but I will have a look at pup!
