hrdl

At least some form of adjustment is probably desirable, as we'd end up at (point-min) otherwise. I will look into this as soon as I have time to see if...

I've been running with this change, which seems to be working fine for `file+head`: https://github.com/lennonhill/org-roam/commit/8ba4c8e00afdace9a6f7cd393c8855585b5de3de. I haven't tested `file` yet.

This is still relevant even with https://github.com/org-roam/org-roam/commit/4fabea2757824c0571d3dfd0614208b7e69afd2d, as the ID is inserted at `(point-min)` for all target types but `file+datetree`. This makes perfect sense for target `file`, but for all...

Sure. Thanks for taking over maintainership of this project!

I cannot reproduce this behaviour.

I've also run into this when migrating from ol-bibtex.el and zotero to papis very recently. I've implemented my own Formater for this: https://github.com/lennonhill/papis/commit/b9b9c6eaa3de159e1b210174ef49e90a89271eb8

Thanks for elaborating on this. I will check for which attributes this makes sense a submit a patch soon.

Those links worked fine for me, but it took around 14 seconds to get a reply from my wallabag instance using firefox. I occasionally observe the same with other articles...

I get an `DEBUG Wallabag: Server error: wantread` message, which we can easily check against: https://github.com/koreader/koreader/blob/118e063ae29eb86b35ffb331ac368137f67102c1/plugins/wallabag.koplugin/main.lua#L634-L641 Handling this is more difficult, as there's not a mechanism yet to indicate that...

There's also a deprecated macos 10.15 image: https://github.com/actions/runner-images#available-images