Sanitation plugin

[edditler]

I'm not entirely sure if a plugin is the best way to go about this. This PR could also be a shell script, but I'm curious to hear what you think. My idea is that the plugin iterates over all (new) notes and makes replacements like

• Replace links to /dataobj/XYZ (bookmarks) by backlinks
• Replace curly with straight quotation marks and similar stuff
• Replace **Bold text**: by **Bold text:** (moving the colon)
• Replace frontmatter["path"] by the actual path
• Make sure that no dataobj["id"] are used twice!
• ...

The way I see it, such a plugin would be unavoidably super opinionated, so maybe an archivy linter is not really in scope for this project.

The last point I consider to be a bug actually: When you git pull your data, the database becomes inconsistent because of name": "max_id" in the database

[Uzay-G]

Hmm this could be interesting, but maybe not as an official archivy project (because opinionated). For the last point, maybe we could update the max_id by looking at all dataobjs when archivy starts, and making sure it's correct? I think we had that at some point.

[edditler]

re the max_id: that seems like a good approach. one could, in principle, git pull data into a running instance, but restarting at that point shouldn't be a problem

re the plugin: I think I'll start writing something up and see how it goes. The path issue annoys me enough that I need to write a script in any case.